prompt
large_stringlengths 72
9.34k
| completion
large_stringlengths 0
7.61k
|
|---|---|
<|file_name|>main_ctrl.py<|end_file_name|><｜fim▁begin｜>from controllers.job_ctrl import JobController
from models.job_model import JobModel
from views.job_view import JobView
class MainController(object):
def __init__(self, main_model):
self.main_view = None
self.main_model = main_model
self.main_model.begin_job_fetch.connect(self.on_begin_job_fetch)
self.main_model.update_job_fetch_progress.connect(self.on_job_fetch_update)
self.main_model.fetched_job.connect(self.on_fetched_job)
def <|fim_middle|>(self, main_view):
self.main_view = main_view
self.init_hotkeys()
def init_hotkeys(self):
self.main_model.hotkey_model.add_hotkey(["Lcontrol", "Lmenu", "J"], self.main_view.focus_job_num_edit)
self.main_model.hotkey_model.add_hotkey(["Lcontrol", "Lmenu", "O"], self.main_view.open_current_job_folder)
self.main_model.hotkey_model.add_hotkey(["Lcontrol", "Lmenu", "B"], self.main_view.open_current_job_basecamp)
self.main_model.hotkey_model.start_detection()
def fetch_job(self):
job_num = self.main_view.job_num
if self.main_model.job_exists(job_num):
self.main_view.show_job_already_exists_dialog()
return
self.main_model.fetch_job(job_num)
def cancel_job_fetch(self):
self.main_model.cancel_job_fetch()
def on_begin_job_fetch(self, max):
self.main_view.show_job_fetch_progress_dialog(max)
def on_job_fetch_update(self, progress):
self.main_view.update_job_fetch_progress_dialog(progress)
def on_fetched_job(self, job_num, base_folder):
job = JobModel(job_num,
base_folder,
self.main_model.settings_model.basecamp_email,
self.main_model.settings_model.basecamp_password,
self.main_model.settings_model.google_maps_js_api_key,
self.main_model.settings_model.google_maps_static_api_key,
self.main_model.settings_model.google_earth_exe_path,
self.main_model.settings_model.scene_exe_path)
self.main_model.jobs[job.job_num] = job
found = bool(job.base_folder)
self.main_view.close_job_fetch_progress_dialog()
if not found:
open_anyway = self.main_view.show_job_not_found_dialog()
if not open_anyway:
return
job_view = JobView(JobController(job))
job_view.request_minimize.connect(self.main_view.close)
self.main_view.add_tab(job_view, job.job_name)
def remove_job(self, index):
job_num = int(self.main_view.ui.jobs_tab_widget.tabText(index)[1:])
self.main_model.jobs.pop(job_num, None)
self.main_view.remove_tab(index)
<｜fim▁end｜> | init_ui |
<|file_name|>main_ctrl.py<|end_file_name|><｜fim▁begin｜>from controllers.job_ctrl import JobController
from models.job_model import JobModel
from views.job_view import JobView
class MainController(object):
def __init__(self, main_model):
self.main_view = None
self.main_model = main_model
self.main_model.begin_job_fetch.connect(self.on_begin_job_fetch)
self.main_model.update_job_fetch_progress.connect(self.on_job_fetch_update)
self.main_model.fetched_job.connect(self.on_fetched_job)
def init_ui(self, main_view):
self.main_view = main_view
self.init_hotkeys()
def <|fim_middle|>(self):
self.main_model.hotkey_model.add_hotkey(["Lcontrol", "Lmenu", "J"], self.main_view.focus_job_num_edit)
self.main_model.hotkey_model.add_hotkey(["Lcontrol", "Lmenu", "O"], self.main_view.open_current_job_folder)
self.main_model.hotkey_model.add_hotkey(["Lcontrol", "Lmenu", "B"], self.main_view.open_current_job_basecamp)
self.main_model.hotkey_model.start_detection()
def fetch_job(self):
job_num = self.main_view.job_num
if self.main_model.job_exists(job_num):
self.main_view.show_job_already_exists_dialog()
return
self.main_model.fetch_job(job_num)
def cancel_job_fetch(self):
self.main_model.cancel_job_fetch()
def on_begin_job_fetch(self, max):
self.main_view.show_job_fetch_progress_dialog(max)
def on_job_fetch_update(self, progress):
self.main_view.update_job_fetch_progress_dialog(progress)
def on_fetched_job(self, job_num, base_folder):
job = JobModel(job_num,
base_folder,
self.main_model.settings_model.basecamp_email,
self.main_model.settings_model.basecamp_password,
self.main_model.settings_model.google_maps_js_api_key,
self.main_model.settings_model.google_maps_static_api_key,
self.main_model.settings_model.google_earth_exe_path,
self.main_model.settings_model.scene_exe_path)
self.main_model.jobs[job.job_num] = job
found = bool(job.base_folder)
self.main_view.close_job_fetch_progress_dialog()
if not found:
open_anyway = self.main_view.show_job_not_found_dialog()
if not open_anyway:
return
job_view = JobView(JobController(job))
job_view.request_minimize.connect(self.main_view.close)
self.main_view.add_tab(job_view, job.job_name)
def remove_job(self, index):
job_num = int(self.main_view.ui.jobs_tab_widget.tabText(index)[1:])
self.main_model.jobs.pop(job_num, None)
self.main_view.remove_tab(index)
<｜fim▁end｜> | init_hotkeys |
<|file_name|>main_ctrl.py<|end_file_name|><｜fim▁begin｜>from controllers.job_ctrl import JobController
from models.job_model import JobModel
from views.job_view import JobView
class MainController(object):
def __init__(self, main_model):
self.main_view = None
self.main_model = main_model
self.main_model.begin_job_fetch.connect(self.on_begin_job_fetch)
self.main_model.update_job_fetch_progress.connect(self.on_job_fetch_update)
self.main_model.fetched_job.connect(self.on_fetched_job)
def init_ui(self, main_view):
self.main_view = main_view
self.init_hotkeys()
def init_hotkeys(self):
self.main_model.hotkey_model.add_hotkey(["Lcontrol", "Lmenu", "J"], self.main_view.focus_job_num_edit)
self.main_model.hotkey_model.add_hotkey(["Lcontrol", "Lmenu", "O"], self.main_view.open_current_job_folder)
self.main_model.hotkey_model.add_hotkey(["Lcontrol", "Lmenu", "B"], self.main_view.open_current_job_basecamp)
self.main_model.hotkey_model.start_detection()
def <|fim_middle|>(self):
job_num = self.main_view.job_num
if self.main_model.job_exists(job_num):
self.main_view.show_job_already_exists_dialog()
return
self.main_model.fetch_job(job_num)
def cancel_job_fetch(self):
self.main_model.cancel_job_fetch()
def on_begin_job_fetch(self, max):
self.main_view.show_job_fetch_progress_dialog(max)
def on_job_fetch_update(self, progress):
self.main_view.update_job_fetch_progress_dialog(progress)
def on_fetched_job(self, job_num, base_folder):
job = JobModel(job_num,
base_folder,
self.main_model.settings_model.basecamp_email,
self.main_model.settings_model.basecamp_password,
self.main_model.settings_model.google_maps_js_api_key,
self.main_model.settings_model.google_maps_static_api_key,
self.main_model.settings_model.google_earth_exe_path,
self.main_model.settings_model.scene_exe_path)
self.main_model.jobs[job.job_num] = job
found = bool(job.base_folder)
self.main_view.close_job_fetch_progress_dialog()
if not found:
open_anyway = self.main_view.show_job_not_found_dialog()
if not open_anyway:
return
job_view = JobView(JobController(job))
job_view.request_minimize.connect(self.main_view.close)
self.main_view.add_tab(job_view, job.job_name)
def remove_job(self, index):
job_num = int(self.main_view.ui.jobs_tab_widget.tabText(index)[1:])
self.main_model.jobs.pop(job_num, None)
self.main_view.remove_tab(index)
<｜fim▁end｜> | fetch_job |
<|file_name|>main_ctrl.py<|end_file_name|><｜fim▁begin｜>from controllers.job_ctrl import JobController
from models.job_model import JobModel
from views.job_view import JobView
class MainController(object):
def __init__(self, main_model):
self.main_view = None
self.main_model = main_model
self.main_model.begin_job_fetch.connect(self.on_begin_job_fetch)
self.main_model.update_job_fetch_progress.connect(self.on_job_fetch_update)
self.main_model.fetched_job.connect(self.on_fetched_job)
def init_ui(self, main_view):
self.main_view = main_view
self.init_hotkeys()
def init_hotkeys(self):
self.main_model.hotkey_model.add_hotkey(["Lcontrol", "Lmenu", "J"], self.main_view.focus_job_num_edit)
self.main_model.hotkey_model.add_hotkey(["Lcontrol", "Lmenu", "O"], self.main_view.open_current_job_folder)
self.main_model.hotkey_model.add_hotkey(["Lcontrol", "Lmenu", "B"], self.main_view.open_current_job_basecamp)
self.main_model.hotkey_model.start_detection()
def fetch_job(self):
job_num = self.main_view.job_num
if self.main_model.job_exists(job_num):
self.main_view.show_job_already_exists_dialog()
return
self.main_model.fetch_job(job_num)
def <|fim_middle|>(self):
self.main_model.cancel_job_fetch()
def on_begin_job_fetch(self, max):
self.main_view.show_job_fetch_progress_dialog(max)
def on_job_fetch_update(self, progress):
self.main_view.update_job_fetch_progress_dialog(progress)
def on_fetched_job(self, job_num, base_folder):
job = JobModel(job_num,
base_folder,
self.main_model.settings_model.basecamp_email,
self.main_model.settings_model.basecamp_password,
self.main_model.settings_model.google_maps_js_api_key,
self.main_model.settings_model.google_maps_static_api_key,
self.main_model.settings_model.google_earth_exe_path,
self.main_model.settings_model.scene_exe_path)
self.main_model.jobs[job.job_num] = job
found = bool(job.base_folder)
self.main_view.close_job_fetch_progress_dialog()
if not found:
open_anyway = self.main_view.show_job_not_found_dialog()
if not open_anyway:
return
job_view = JobView(JobController(job))
job_view.request_minimize.connect(self.main_view.close)
self.main_view.add_tab(job_view, job.job_name)
def remove_job(self, index):
job_num = int(self.main_view.ui.jobs_tab_widget.tabText(index)[1:])
self.main_model.jobs.pop(job_num, None)
self.main_view.remove_tab(index)
<｜fim▁end｜> | cancel_job_fetch |
<|file_name|>main_ctrl.py<|end_file_name|><｜fim▁begin｜>from controllers.job_ctrl import JobController
from models.job_model import JobModel
from views.job_view import JobView
class MainController(object):
def __init__(self, main_model):
self.main_view = None
self.main_model = main_model
self.main_model.begin_job_fetch.connect(self.on_begin_job_fetch)
self.main_model.update_job_fetch_progress.connect(self.on_job_fetch_update)
self.main_model.fetched_job.connect(self.on_fetched_job)
def init_ui(self, main_view):
self.main_view = main_view
self.init_hotkeys()
def init_hotkeys(self):
self.main_model.hotkey_model.add_hotkey(["Lcontrol", "Lmenu", "J"], self.main_view.focus_job_num_edit)
self.main_model.hotkey_model.add_hotkey(["Lcontrol", "Lmenu", "O"], self.main_view.open_current_job_folder)
self.main_model.hotkey_model.add_hotkey(["Lcontrol", "Lmenu", "B"], self.main_view.open_current_job_basecamp)
self.main_model.hotkey_model.start_detection()
def fetch_job(self):
job_num = self.main_view.job_num
if self.main_model.job_exists(job_num):
self.main_view.show_job_already_exists_dialog()
return
self.main_model.fetch_job(job_num)
def cancel_job_fetch(self):
self.main_model.cancel_job_fetch()
def <|fim_middle|>(self, max):
self.main_view.show_job_fetch_progress_dialog(max)
def on_job_fetch_update(self, progress):
self.main_view.update_job_fetch_progress_dialog(progress)
def on_fetched_job(self, job_num, base_folder):
job = JobModel(job_num,
base_folder,
self.main_model.settings_model.basecamp_email,
self.main_model.settings_model.basecamp_password,
self.main_model.settings_model.google_maps_js_api_key,
self.main_model.settings_model.google_maps_static_api_key,
self.main_model.settings_model.google_earth_exe_path,
self.main_model.settings_model.scene_exe_path)
self.main_model.jobs[job.job_num] = job
found = bool(job.base_folder)
self.main_view.close_job_fetch_progress_dialog()
if not found:
open_anyway = self.main_view.show_job_not_found_dialog()
if not open_anyway:
return
job_view = JobView(JobController(job))
job_view.request_minimize.connect(self.main_view.close)
self.main_view.add_tab(job_view, job.job_name)
def remove_job(self, index):
job_num = int(self.main_view.ui.jobs_tab_widget.tabText(index)[1:])
self.main_model.jobs.pop(job_num, None)
self.main_view.remove_tab(index)
<｜fim▁end｜> | on_begin_job_fetch |
<|file_name|>main_ctrl.py<|end_file_name|><｜fim▁begin｜>from controllers.job_ctrl import JobController
from models.job_model import JobModel
from views.job_view import JobView
class MainController(object):
def __init__(self, main_model):
self.main_view = None
self.main_model = main_model
self.main_model.begin_job_fetch.connect(self.on_begin_job_fetch)
self.main_model.update_job_fetch_progress.connect(self.on_job_fetch_update)
self.main_model.fetched_job.connect(self.on_fetched_job)
def init_ui(self, main_view):
self.main_view = main_view
self.init_hotkeys()
def init_hotkeys(self):
self.main_model.hotkey_model.add_hotkey(["Lcontrol", "Lmenu", "J"], self.main_view.focus_job_num_edit)
self.main_model.hotkey_model.add_hotkey(["Lcontrol", "Lmenu", "O"], self.main_view.open_current_job_folder)
self.main_model.hotkey_model.add_hotkey(["Lcontrol", "Lmenu", "B"], self.main_view.open_current_job_basecamp)
self.main_model.hotkey_model.start_detection()
def fetch_job(self):
job_num = self.main_view.job_num
if self.main_model.job_exists(job_num):
self.main_view.show_job_already_exists_dialog()
return
self.main_model.fetch_job(job_num)
def cancel_job_fetch(self):
self.main_model.cancel_job_fetch()
def on_begin_job_fetch(self, max):
self.main_view.show_job_fetch_progress_dialog(max)
def <|fim_middle|>(self, progress):
self.main_view.update_job_fetch_progress_dialog(progress)
def on_fetched_job(self, job_num, base_folder):
job = JobModel(job_num,
base_folder,
self.main_model.settings_model.basecamp_email,
self.main_model.settings_model.basecamp_password,
self.main_model.settings_model.google_maps_js_api_key,
self.main_model.settings_model.google_maps_static_api_key,
self.main_model.settings_model.google_earth_exe_path,
self.main_model.settings_model.scene_exe_path)
self.main_model.jobs[job.job_num] = job
found = bool(job.base_folder)
self.main_view.close_job_fetch_progress_dialog()
if not found:
open_anyway = self.main_view.show_job_not_found_dialog()
if not open_anyway:
return
job_view = JobView(JobController(job))
job_view.request_minimize.connect(self.main_view.close)
self.main_view.add_tab(job_view, job.job_name)
def remove_job(self, index):
job_num = int(self.main_view.ui.jobs_tab_widget.tabText(index)[1:])
self.main_model.jobs.pop(job_num, None)
self.main_view.remove_tab(index)
<｜fim▁end｜> | on_job_fetch_update |
<|file_name|>main_ctrl.py<|end_file_name|><｜fim▁begin｜>from controllers.job_ctrl import JobController
from models.job_model import JobModel
from views.job_view import JobView
class MainController(object):
def __init__(self, main_model):
self.main_view = None
self.main_model = main_model
self.main_model.begin_job_fetch.connect(self.on_begin_job_fetch)
self.main_model.update_job_fetch_progress.connect(self.on_job_fetch_update)
self.main_model.fetched_job.connect(self.on_fetched_job)
def init_ui(self, main_view):
self.main_view = main_view
self.init_hotkeys()
def init_hotkeys(self):
self.main_model.hotkey_model.add_hotkey(["Lcontrol", "Lmenu", "J"], self.main_view.focus_job_num_edit)
self.main_model.hotkey_model.add_hotkey(["Lcontrol", "Lmenu", "O"], self.main_view.open_current_job_folder)
self.main_model.hotkey_model.add_hotkey(["Lcontrol", "Lmenu", "B"], self.main_view.open_current_job_basecamp)
self.main_model.hotkey_model.start_detection()
def fetch_job(self):
job_num = self.main_view.job_num
if self.main_model.job_exists(job_num):
self.main_view.show_job_already_exists_dialog()
return
self.main_model.fetch_job(job_num)
def cancel_job_fetch(self):
self.main_model.cancel_job_fetch()
def on_begin_job_fetch(self, max):
self.main_view.show_job_fetch_progress_dialog(max)
def on_job_fetch_update(self, progress):
self.main_view.update_job_fetch_progress_dialog(progress)
def <|fim_middle|>(self, job_num, base_folder):
job = JobModel(job_num,
base_folder,
self.main_model.settings_model.basecamp_email,
self.main_model.settings_model.basecamp_password,
self.main_model.settings_model.google_maps_js_api_key,
self.main_model.settings_model.google_maps_static_api_key,
self.main_model.settings_model.google_earth_exe_path,
self.main_model.settings_model.scene_exe_path)
self.main_model.jobs[job.job_num] = job
found = bool(job.base_folder)
self.main_view.close_job_fetch_progress_dialog()
if not found:
open_anyway = self.main_view.show_job_not_found_dialog()
if not open_anyway:
return
job_view = JobView(JobController(job))
job_view.request_minimize.connect(self.main_view.close)
self.main_view.add_tab(job_view, job.job_name)
def remove_job(self, index):
job_num = int(self.main_view.ui.jobs_tab_widget.tabText(index)[1:])
self.main_model.jobs.pop(job_num, None)
self.main_view.remove_tab(index)
<｜fim▁end｜> | on_fetched_job |
<|file_name|>main_ctrl.py<|end_file_name|><｜fim▁begin｜>from controllers.job_ctrl import JobController
from models.job_model import JobModel
from views.job_view import JobView
class MainController(object):
def __init__(self, main_model):
self.main_view = None
self.main_model = main_model
self.main_model.begin_job_fetch.connect(self.on_begin_job_fetch)
self.main_model.update_job_fetch_progress.connect(self.on_job_fetch_update)
self.main_model.fetched_job.connect(self.on_fetched_job)
def init_ui(self, main_view):
self.main_view = main_view
self.init_hotkeys()
def init_hotkeys(self):
self.main_model.hotkey_model.add_hotkey(["Lcontrol", "Lmenu", "J"], self.main_view.focus_job_num_edit)
self.main_model.hotkey_model.add_hotkey(["Lcontrol", "Lmenu", "O"], self.main_view.open_current_job_folder)
self.main_model.hotkey_model.add_hotkey(["Lcontrol", "Lmenu", "B"], self.main_view.open_current_job_basecamp)
self.main_model.hotkey_model.start_detection()
def fetch_job(self):
job_num = self.main_view.job_num
if self.main_model.job_exists(job_num):
self.main_view.show_job_already_exists_dialog()
return
self.main_model.fetch_job(job_num)
def cancel_job_fetch(self):
self.main_model.cancel_job_fetch()
def on_begin_job_fetch(self, max):
self.main_view.show_job_fetch_progress_dialog(max)
def on_job_fetch_update(self, progress):
self.main_view.update_job_fetch_progress_dialog(progress)
def on_fetched_job(self, job_num, base_folder):
job = JobModel(job_num,
base_folder,
self.main_model.settings_model.basecamp_email,
self.main_model.settings_model.basecamp_password,
self.main_model.settings_model.google_maps_js_api_key,
self.main_model.settings_model.google_maps_static_api_key,
self.main_model.settings_model.google_earth_exe_path,
self.main_model.settings_model.scene_exe_path)
self.main_model.jobs[job.job_num] = job
found = bool(job.base_folder)
self.main_view.close_job_fetch_progress_dialog()
if not found:
open_anyway = self.main_view.show_job_not_found_dialog()
if not open_anyway:
return
job_view = JobView(JobController(job))
job_view.request_minimize.connect(self.main_view.close)
self.main_view.add_tab(job_view, job.job_name)
def <|fim_middle|>(self, index):
job_num = int(self.main_view.ui.jobs_tab_widget.tabText(index)[1:])
self.main_model.jobs.pop(job_num, None)
self.main_view.remove_tab(index)
<｜fim▁end｜> | remove_job |
<|file_name|>CategoryTree.py<|end_file_name|><｜fim▁begin｜>import json
from collections import (
Counter,
defaultdict as deft
)
from copy import deepcopy as cp
# from cPickle import (
# dump as to_pickle,
# load as from_pickle
# )
from StringIO import StringIO
from TfIdfMatrix import TfIdfMatrix
from Tools import from_csv
class CategoryTree:
def __init__(self, categories_by_concept, terms,
categories, tfidf, max_depth=5, min_df=20
):
self.min_df = min_df
self.path_categories_by_concept = categories_by_concept
self.path_categories = categories
self.path_terms = terms
self.max_depth = max_depth
self.observed_category = deft(bool)
self.id_by_concept = dict([])
self.concept_by_id = dict([])
self.term_is_category = deft(bool)
self.parents_by_category = dict([])
self.parents_by_concept = deft(list)
self.id_by_term = dict([])
self.term_by_id = dict([])
self.has_parents = deft(bool)
self.tfidf = tfidf
self.pulling = set([])
self.vector_by_category = deft(Counter)
self.contributors_by_category = deft(set)
self.projected = Counter()
def build(self):
for i, c in enumerate(self.concept_by_id.values()):
self(c)
if not i % 100:
t = float(len(self.concept_by_id.keys()))
print i, int(t), round(i / t, 2)
# if i >= 5000:
# break
def dump(self):
# Simulate a file with StringIO
out = open('vector.dump.txt', 'wb')
for i, (_id, projections) in enumerate(self.projected.items()):
if not i % 100:
print i, len(self.projected.keys())
if not projections:
continue
features = [
(self.tfidf.word_by_id[wid], round(weight, 4))
for wid, weight in self.vector_by_category[_id].most_common()
if round(weight, 4)
]
record = (
_id,
self.concept_by_id[_id],
features
)
out.write('%s\n' % str(record))
out.close()
def __call__(self, category):
self.pulling = set([])
return self.__pull(None, 0, category, dict([]))
def __get_parents(self, _id):
parents = []
name = self.concept_by_id[_id]
if (
not self.observed_category[name] or
not self.observed_category[_id] or
not self.has_parents[_id]<｜fim▁hole｜> if not self.observed_category[i]:
continue
_name = self.concept_by_id[i]
parents.append(_name)
return set(parents) - self.pulling
def __pull(self, vector, depth, category, tree):
_id = self.id_by_concept[category]
if not self.pulling:
# print
# print
# print category, _id
# print [self.term_by_id[x] for x in self.contributors_by_category[_id]]
# print self.vector_by_category[_id].most_common(20)
vector = self.vector_by_category[_id]
if not self.observed_category[category]:
return dict([])
parents = self.__get_parents(_id)
if not parents or depth >= self.max_depth:
tree[category] = dict([])
else:
subtree = dict([])
self.pulling.update(parents)
for parent in parents:
subtree = self.__pull(vector, depth + 1, parent, subtree)
tree[category] = subtree
self.__project(vector, tree)
return tree
def __project(self, vector, tree):
if not tree.keys():
return
else:
for key, subtree in tree.items():
_id = self.id_by_concept[key]
self.projected[_id] += 1
self.__add2vec(vector, _id)
self.__project(vector, subtree)
def __add2vec(self, vector, _id):
# for w, weight in vector.items():
# __id = self.tfidf.id_by_word[w]
for __id, weight in vector.items():
self.vector_by_category[_id][__id] += weight
def load(self):
self.__load_terms()
self.__load_categories()
self.__load_assignments()
def __load_categories(self):
for concept, _id in from_csv(self.path_categories):
_id = int(_id)
self.id_by_concept[concept] = _id
self.concept_by_id[_id] = concept
self.observed_category[_id] = True
self.observed_category[concept] = True
# print concept, _id, len(self.id_by_concept.keys())
# exit()
def __load_terms(self):
for term, _id in from_csv(self.path_terms):
_id = int(_id)
self.term_by_id[_id] = term
self.id_by_term[term] = _id
if not term.startswith('Category:'):
continue
self.term_is_category[term] = True
self.term_is_category[_id] = True
def __load_assignments(self):
for row in from_csv(self.path_categories_by_concept):
ints = [int(field) for field in row]
term_id = ints[0]
term = self.term_by_id[term_id]
if self.term_is_category[term_id] and \
self.observed_category[term]:
term = self.term_by_id[term_id]
cat_id = self.id_by_concept[term]
assignments = [i for i in ints[1:] if self.observed_category[i]]
self.parents_by_category[cat_id] = assignments
self.has_parents[cat_id] = True
else:
vector = self.tfidf.content(term_id)
assignments = [i for i in ints[1:] if self.observed_category[i]]
self.parents_by_concept[term_id] = assignments
for a_id in assignments:
for w, weight in vector:
if self.tfidf.df[w] < self.min_df:
continue
#print term, term_id, self.concept_by_id[a_id], w, self.vector_by_category[a_id][w], '\t+%f' % weight
self.vector_by_category[a_id][w] += weight
self.contributors_by_category[a_id].update([term_id])
if __name__ == '__main__':
import random
from random import shuffle as randomize
tfidf = TfIdfMatrix()
tfidf.load_features('bkp.big.out/vector.term.csv')
tfidf.load_distribution('bkp.big.out/vector.index.csv')
# tfidf.load_features('vector.term.csv')
# tfidf.load_distribution('vector.index.csv')
ctree = CategoryTree(
'bkp.big.out/category.index.csv',
'bkp.big.out/term.csv',
'bkp.big.out/category.csv',
# 'category.index.csv',
# 'term.csv',
# 'category.csv',
tfidf,
max_depth=1
)
ctree.load()
ctree.build()
ctree.dump()<｜fim▁end｜> | ):
return []
else:
for i in self.parents_by_category[_id]: |
<|file_name|>CategoryTree.py<|end_file_name|><｜fim▁begin｜>import json
from collections import (
Counter,
defaultdict as deft
)
from copy import deepcopy as cp
# from cPickle import (
# dump as to_pickle,
# load as from_pickle
# )
from StringIO import StringIO
from TfIdfMatrix import TfIdfMatrix
from Tools import from_csv
class CategoryTree:
<|fim_middle|>
if __name__ == '__main__':
import random
from random import shuffle as randomize
tfidf = TfIdfMatrix()
tfidf.load_features('bkp.big.out/vector.term.csv')
tfidf.load_distribution('bkp.big.out/vector.index.csv')
# tfidf.load_features('vector.term.csv')
# tfidf.load_distribution('vector.index.csv')
ctree = CategoryTree(
'bkp.big.out/category.index.csv',
'bkp.big.out/term.csv',
'bkp.big.out/category.csv',
# 'category.index.csv',
# 'term.csv',
# 'category.csv',
tfidf,
max_depth=1
)
ctree.load()
ctree.build()
ctree.dump()
<｜fim▁end｜> | def __init__(self, categories_by_concept, terms,
categories, tfidf, max_depth=5, min_df=20
):
self.min_df = min_df
self.path_categories_by_concept = categories_by_concept
self.path_categories = categories
self.path_terms = terms
self.max_depth = max_depth
self.observed_category = deft(bool)
self.id_by_concept = dict([])
self.concept_by_id = dict([])
self.term_is_category = deft(bool)
self.parents_by_category = dict([])
self.parents_by_concept = deft(list)
self.id_by_term = dict([])
self.term_by_id = dict([])
self.has_parents = deft(bool)
self.tfidf = tfidf
self.pulling = set([])
self.vector_by_category = deft(Counter)
self.contributors_by_category = deft(set)
self.projected = Counter()
def build(self):
for i, c in enumerate(self.concept_by_id.values()):
self(c)
if not i % 100:
t = float(len(self.concept_by_id.keys()))
print i, int(t), round(i / t, 2)
# if i >= 5000:
# break
def dump(self):
# Simulate a file with StringIO
out = open('vector.dump.txt', 'wb')
for i, (_id, projections) in enumerate(self.projected.items()):
if not i % 100:
print i, len(self.projected.keys())
if not projections:
continue
features = [
(self.tfidf.word_by_id[wid], round(weight, 4))
for wid, weight in self.vector_by_category[_id].most_common()
if round(weight, 4)
]
record = (
_id,
self.concept_by_id[_id],
features
)
out.write('%s\n' % str(record))
out.close()
def __call__(self, category):
self.pulling = set([])
return self.__pull(None, 0, category, dict([]))
def __get_parents(self, _id):
parents = []
name = self.concept_by_id[_id]
if (
not self.observed_category[name] or
not self.observed_category[_id] or
not self.has_parents[_id]
):
return []
else:
for i in self.parents_by_category[_id]:
if not self.observed_category[i]:
continue
_name = self.concept_by_id[i]
parents.append(_name)
return set(parents) - self.pulling
def __pull(self, vector, depth, category, tree):
_id = self.id_by_concept[category]
if not self.pulling:
# print
# print
# print category, _id
# print [self.term_by_id[x] for x in self.contributors_by_category[_id]]
# print self.vector_by_category[_id].most_common(20)
vector = self.vector_by_category[_id]
if not self.observed_category[category]:
return dict([])
parents = self.__get_parents(_id)
if not parents or depth >= self.max_depth:
tree[category] = dict([])
else:
subtree = dict([])
self.pulling.update(parents)
for parent in parents:
subtree = self.__pull(vector, depth + 1, parent, subtree)
tree[category] = subtree
self.__project(vector, tree)
return tree
def __project(self, vector, tree):
if not tree.keys():
return
else:
for key, subtree in tree.items():
_id = self.id_by_concept[key]
self.projected[_id] += 1
self.__add2vec(vector, _id)
self.__project(vector, subtree)
def __add2vec(self, vector, _id):
# for w, weight in vector.items():
# __id = self.tfidf.id_by_word[w]
for __id, weight in vector.items():
self.vector_by_category[_id][__id] += weight
def load(self):
self.__load_terms()
self.__load_categories()
self.__load_assignments()
def __load_categories(self):
for concept, _id in from_csv(self.path_categories):
_id = int(_id)
self.id_by_concept[concept] = _id
self.concept_by_id[_id] = concept
self.observed_category[_id] = True
self.observed_category[concept] = True
# print concept, _id, len(self.id_by_concept.keys())
# exit()
def __load_terms(self):
for term, _id in from_csv(self.path_terms):
_id = int(_id)
self.term_by_id[_id] = term
self.id_by_term[term] = _id
if not term.startswith('Category:'):
continue
self.term_is_category[term] = True
self.term_is_category[_id] = True
def __load_assignments(self):
for row in from_csv(self.path_categories_by_concept):
ints = [int(field) for field in row]
term_id = ints[0]
term = self.term_by_id[term_id]
if self.term_is_category[term_id] and \
self.observed_category[term]:
term = self.term_by_id[term_id]
cat_id = self.id_by_concept[term]
assignments = [i for i in ints[1:] if self.observed_category[i]]
self.parents_by_category[cat_id] = assignments
self.has_parents[cat_id] = True
else:
vector = self.tfidf.content(term_id)
assignments = [i for i in ints[1:] if self.observed_category[i]]
self.parents_by_concept[term_id] = assignments
for a_id in assignments:
for w, weight in vector:
if self.tfidf.df[w] < self.min_df:
continue
#print term, term_id, self.concept_by_id[a_id], w, self.vector_by_category[a_id][w], '\t+%f' % weight
self.vector_by_category[a_id][w] += weight
self.contributors_by_category[a_id].update([term_id]) |
<|file_name|>CategoryTree.py<|end_file_name|><｜fim▁begin｜>import json
from collections import (
Counter,
defaultdict as deft
)
from copy import deepcopy as cp
# from cPickle import (
# dump as to_pickle,
# load as from_pickle
# )
from StringIO import StringIO
from TfIdfMatrix import TfIdfMatrix
from Tools import from_csv
class CategoryTree:
def __init__(self, categories_by_concept, terms,
categories, tfidf, max_depth=5, min_df=20
):
<|fim_middle|>
def build(self):
for i, c in enumerate(self.concept_by_id.values()):
self(c)
if not i % 100:
t = float(len(self.concept_by_id.keys()))
print i, int(t), round(i / t, 2)
# if i >= 5000:
# break
def dump(self):
# Simulate a file with StringIO
out = open('vector.dump.txt', 'wb')
for i, (_id, projections) in enumerate(self.projected.items()):
if not i % 100:
print i, len(self.projected.keys())
if not projections:
continue
features = [
(self.tfidf.word_by_id[wid], round(weight, 4))
for wid, weight in self.vector_by_category[_id].most_common()
if round(weight, 4)
]
record = (
_id,
self.concept_by_id[_id],
features
)
out.write('%s\n' % str(record))
out.close()
def __call__(self, category):
self.pulling = set([])
return self.__pull(None, 0, category, dict([]))
def __get_parents(self, _id):
parents = []
name = self.concept_by_id[_id]
if (
not self.observed_category[name] or
not self.observed_category[_id] or
not self.has_parents[_id]
):
return []
else:
for i in self.parents_by_category[_id]:
if not self.observed_category[i]:
continue
_name = self.concept_by_id[i]
parents.append(_name)
return set(parents) - self.pulling
def __pull(self, vector, depth, category, tree):
_id = self.id_by_concept[category]
if not self.pulling:
# print
# print
# print category, _id
# print [self.term_by_id[x] for x in self.contributors_by_category[_id]]
# print self.vector_by_category[_id].most_common(20)
vector = self.vector_by_category[_id]
if not self.observed_category[category]:
return dict([])
parents = self.__get_parents(_id)
if not parents or depth >= self.max_depth:
tree[category] = dict([])
else:
subtree = dict([])
self.pulling.update(parents)
for parent in parents:
subtree = self.__pull(vector, depth + 1, parent, subtree)
tree[category] = subtree
self.__project(vector, tree)
return tree
def __project(self, vector, tree):
if not tree.keys():
return
else:
for key, subtree in tree.items():
_id = self.id_by_concept[key]
self.projected[_id] += 1
self.__add2vec(vector, _id)
self.__project(vector, subtree)
def __add2vec(self, vector, _id):
# for w, weight in vector.items():
# __id = self.tfidf.id_by_word[w]
for __id, weight in vector.items():
self.vector_by_category[_id][__id] += weight
def load(self):
self.__load_terms()
self.__load_categories()
self.__load_assignments()
def __load_categories(self):
for concept, _id in from_csv(self.path_categories):
_id = int(_id)
self.id_by_concept[concept] = _id
self.concept_by_id[_id] = concept
self.observed_category[_id] = True
self.observed_category[concept] = True
# print concept, _id, len(self.id_by_concept.keys())
# exit()
def __load_terms(self):
for term, _id in from_csv(self.path_terms):
_id = int(_id)
self.term_by_id[_id] = term
self.id_by_term[term] = _id
if not term.startswith('Category:'):
continue
self.term_is_category[term] = True
self.term_is_category[_id] = True
def __load_assignments(self):
for row in from_csv(self.path_categories_by_concept):
ints = [int(field) for field in row]
term_id = ints[0]
term = self.term_by_id[term_id]
if self.term_is_category[term_id] and \
self.observed_category[term]:
term = self.term_by_id[term_id]
cat_id = self.id_by_concept[term]
assignments = [i for i in ints[1:] if self.observed_category[i]]
self.parents_by_category[cat_id] = assignments
self.has_parents[cat_id] = True
else:
vector = self.tfidf.content(term_id)
assignments = [i for i in ints[1:] if self.observed_category[i]]
self.parents_by_concept[term_id] = assignments
for a_id in assignments:
for w, weight in vector:
if self.tfidf.df[w] < self.min_df:
continue
#print term, term_id, self.concept_by_id[a_id], w, self.vector_by_category[a_id][w], '\t+%f' % weight
self.vector_by_category[a_id][w] += weight
self.contributors_by_category[a_id].update([term_id])
if __name__ == '__main__':
import random
from random import shuffle as randomize
tfidf = TfIdfMatrix()
tfidf.load_features('bkp.big.out/vector.term.csv')
tfidf.load_distribution('bkp.big.out/vector.index.csv')
# tfidf.load_features('vector.term.csv')
# tfidf.load_distribution('vector.index.csv')
ctree = CategoryTree(
'bkp.big.out/category.index.csv',
'bkp.big.out/term.csv',
'bkp.big.out/category.csv',
# 'category.index.csv',
# 'term.csv',
# 'category.csv',
tfidf,
max_depth=1
)
ctree.load()
ctree.build()
ctree.dump()
<｜fim▁end｜> | self.min_df = min_df
self.path_categories_by_concept = categories_by_concept
self.path_categories = categories
self.path_terms = terms
self.max_depth = max_depth
self.observed_category = deft(bool)
self.id_by_concept = dict([])
self.concept_by_id = dict([])
self.term_is_category = deft(bool)
self.parents_by_category = dict([])
self.parents_by_concept = deft(list)
self.id_by_term = dict([])
self.term_by_id = dict([])
self.has_parents = deft(bool)
self.tfidf = tfidf
self.pulling = set([])
self.vector_by_category = deft(Counter)
self.contributors_by_category = deft(set)
self.projected = Counter() |
<|file_name|>CategoryTree.py<|end_file_name|><｜fim▁begin｜>import json
from collections import (
Counter,
defaultdict as deft
)
from copy import deepcopy as cp
# from cPickle import (
# dump as to_pickle,
# load as from_pickle
# )
from StringIO import StringIO
from TfIdfMatrix import TfIdfMatrix
from Tools import from_csv
class CategoryTree:
def __init__(self, categories_by_concept, terms,
categories, tfidf, max_depth=5, min_df=20
):
self.min_df = min_df
self.path_categories_by_concept = categories_by_concept
self.path_categories = categories
self.path_terms = terms
self.max_depth = max_depth
self.observed_category = deft(bool)
self.id_by_concept = dict([])
self.concept_by_id = dict([])
self.term_is_category = deft(bool)
self.parents_by_category = dict([])
self.parents_by_concept = deft(list)
self.id_by_term = dict([])
self.term_by_id = dict([])
self.has_parents = deft(bool)
self.tfidf = tfidf
self.pulling = set([])
self.vector_by_category = deft(Counter)
self.contributors_by_category = deft(set)
self.projected = Counter()
def build(self):
<|fim_middle|>
# if i >= 5000:
# break
def dump(self):
# Simulate a file with StringIO
out = open('vector.dump.txt', 'wb')
for i, (_id, projections) in enumerate(self.projected.items()):
if not i % 100:
print i, len(self.projected.keys())
if not projections:
continue
features = [
(self.tfidf.word_by_id[wid], round(weight, 4))
for wid, weight in self.vector_by_category[_id].most_common()
if round(weight, 4)
]
record = (
_id,
self.concept_by_id[_id],
features
)
out.write('%s\n' % str(record))
out.close()
def __call__(self, category):
self.pulling = set([])
return self.__pull(None, 0, category, dict([]))
def __get_parents(self, _id):
parents = []
name = self.concept_by_id[_id]
if (
not self.observed_category[name] or
not self.observed_category[_id] or
not self.has_parents[_id]
):
return []
else:
for i in self.parents_by_category[_id]:
if not self.observed_category[i]:
continue
_name = self.concept_by_id[i]
parents.append(_name)
return set(parents) - self.pulling
def __pull(self, vector, depth, category, tree):
_id = self.id_by_concept[category]
if not self.pulling:
# print
# print
# print category, _id
# print [self.term_by_id[x] for x in self.contributors_by_category[_id]]
# print self.vector_by_category[_id].most_common(20)
vector = self.vector_by_category[_id]
if not self.observed_category[category]:
return dict([])
parents = self.__get_parents(_id)
if not parents or depth >= self.max_depth:
tree[category] = dict([])
else:
subtree = dict([])
self.pulling.update(parents)
for parent in parents:
subtree = self.__pull(vector, depth + 1, parent, subtree)
tree[category] = subtree
self.__project(vector, tree)
return tree
def __project(self, vector, tree):
if not tree.keys():
return
else:
for key, subtree in tree.items():
_id = self.id_by_concept[key]
self.projected[_id] += 1
self.__add2vec(vector, _id)
self.__project(vector, subtree)
def __add2vec(self, vector, _id):
# for w, weight in vector.items():
# __id = self.tfidf.id_by_word[w]
for __id, weight in vector.items():
self.vector_by_category[_id][__id] += weight
def load(self):
self.__load_terms()
self.__load_categories()
self.__load_assignments()
def __load_categories(self):
for concept, _id in from_csv(self.path_categories):
_id = int(_id)
self.id_by_concept[concept] = _id
self.concept_by_id[_id] = concept
self.observed_category[_id] = True
self.observed_category[concept] = True
# print concept, _id, len(self.id_by_concept.keys())
# exit()
def __load_terms(self):
for term, _id in from_csv(self.path_terms):
_id = int(_id)
self.term_by_id[_id] = term
self.id_by_term[term] = _id
if not term.startswith('Category:'):
continue
self.term_is_category[term] = True
self.term_is_category[_id] = True
def __load_assignments(self):
for row in from_csv(self.path_categories_by_concept):
ints = [int(field) for field in row]
term_id = ints[0]
term = self.term_by_id[term_id]
if self.term_is_category[term_id] and \
self.observed_category[term]:
term = self.term_by_id[term_id]
cat_id = self.id_by_concept[term]
assignments = [i for i in ints[1:] if self.observed_category[i]]
self.parents_by_category[cat_id] = assignments
self.has_parents[cat_id] = True
else:
vector = self.tfidf.content(term_id)
assignments = [i for i in ints[1:] if self.observed_category[i]]
self.parents_by_concept[term_id] = assignments
for a_id in assignments:
for w, weight in vector:
if self.tfidf.df[w] < self.min_df:
continue
#print term, term_id, self.concept_by_id[a_id], w, self.vector_by_category[a_id][w], '\t+%f' % weight
self.vector_by_category[a_id][w] += weight
self.contributors_by_category[a_id].update([term_id])
if __name__ == '__main__':
import random
from random import shuffle as randomize
tfidf = TfIdfMatrix()
tfidf.load_features('bkp.big.out/vector.term.csv')
tfidf.load_distribution('bkp.big.out/vector.index.csv')
# tfidf.load_features('vector.term.csv')
# tfidf.load_distribution('vector.index.csv')
ctree = CategoryTree(
'bkp.big.out/category.index.csv',
'bkp.big.out/term.csv',
'bkp.big.out/category.csv',
# 'category.index.csv',
# 'term.csv',
# 'category.csv',
tfidf,
max_depth=1
)
ctree.load()
ctree.build()
ctree.dump()
<｜fim▁end｜> | for i, c in enumerate(self.concept_by_id.values()):
self(c)
if not i % 100:
t = float(len(self.concept_by_id.keys()))
print i, int(t), round(i / t, 2) |
<|file_name|>CategoryTree.py<|end_file_name|><｜fim▁begin｜>import json
from collections import (
Counter,
defaultdict as deft
)
from copy import deepcopy as cp
# from cPickle import (
# dump as to_pickle,
# load as from_pickle
# )
from StringIO import StringIO
from TfIdfMatrix import TfIdfMatrix
from Tools import from_csv
class CategoryTree:
def __init__(self, categories_by_concept, terms,
categories, tfidf, max_depth=5, min_df=20
):
self.min_df = min_df
self.path_categories_by_concept = categories_by_concept
self.path_categories = categories
self.path_terms = terms
self.max_depth = max_depth
self.observed_category = deft(bool)
self.id_by_concept = dict([])
self.concept_by_id = dict([])
self.term_is_category = deft(bool)
self.parents_by_category = dict([])
self.parents_by_concept = deft(list)
self.id_by_term = dict([])
self.term_by_id = dict([])
self.has_parents = deft(bool)
self.tfidf = tfidf
self.pulling = set([])
self.vector_by_category = deft(Counter)
self.contributors_by_category = deft(set)
self.projected = Counter()
def build(self):
for i, c in enumerate(self.concept_by_id.values()):
self(c)
if not i % 100:
t = float(len(self.concept_by_id.keys()))
print i, int(t), round(i / t, 2)
# if i >= 5000:
# break
def dump(self):
# Simulate a file with StringIO
<|fim_middle|>
def __call__(self, category):
self.pulling = set([])
return self.__pull(None, 0, category, dict([]))
def __get_parents(self, _id):
parents = []
name = self.concept_by_id[_id]
if (
not self.observed_category[name] or
not self.observed_category[_id] or
not self.has_parents[_id]
):
return []
else:
for i in self.parents_by_category[_id]:
if not self.observed_category[i]:
continue
_name = self.concept_by_id[i]
parents.append(_name)
return set(parents) - self.pulling
def __pull(self, vector, depth, category, tree):
_id = self.id_by_concept[category]
if not self.pulling:
# print
# print
# print category, _id
# print [self.term_by_id[x] for x in self.contributors_by_category[_id]]
# print self.vector_by_category[_id].most_common(20)
vector = self.vector_by_category[_id]
if not self.observed_category[category]:
return dict([])
parents = self.__get_parents(_id)
if not parents or depth >= self.max_depth:
tree[category] = dict([])
else:
subtree = dict([])
self.pulling.update(parents)
for parent in parents:
subtree = self.__pull(vector, depth + 1, parent, subtree)
tree[category] = subtree
self.__project(vector, tree)
return tree
def __project(self, vector, tree):
if not tree.keys():
return
else:
for key, subtree in tree.items():
_id = self.id_by_concept[key]
self.projected[_id] += 1
self.__add2vec(vector, _id)
self.__project(vector, subtree)
def __add2vec(self, vector, _id):
# for w, weight in vector.items():
# __id = self.tfidf.id_by_word[w]
for __id, weight in vector.items():
self.vector_by_category[_id][__id] += weight
def load(self):
self.__load_terms()
self.__load_categories()
self.__load_assignments()
def __load_categories(self):
for concept, _id in from_csv(self.path_categories):
_id = int(_id)
self.id_by_concept[concept] = _id
self.concept_by_id[_id] = concept
self.observed_category[_id] = True
self.observed_category[concept] = True
# print concept, _id, len(self.id_by_concept.keys())
# exit()
def __load_terms(self):
for term, _id in from_csv(self.path_terms):
_id = int(_id)
self.term_by_id[_id] = term
self.id_by_term[term] = _id
if not term.startswith('Category:'):
continue
self.term_is_category[term] = True
self.term_is_category[_id] = True
def __load_assignments(self):
for row in from_csv(self.path_categories_by_concept):
ints = [int(field) for field in row]
term_id = ints[0]
term = self.term_by_id[term_id]
if self.term_is_category[term_id] and \
self.observed_category[term]:
term = self.term_by_id[term_id]
cat_id = self.id_by_concept[term]
assignments = [i for i in ints[1:] if self.observed_category[i]]
self.parents_by_category[cat_id] = assignments
self.has_parents[cat_id] = True
else:
vector = self.tfidf.content(term_id)
assignments = [i for i in ints[1:] if self.observed_category[i]]
self.parents_by_concept[term_id] = assignments
for a_id in assignments:
for w, weight in vector:
if self.tfidf.df[w] < self.min_df:
continue
#print term, term_id, self.concept_by_id[a_id], w, self.vector_by_category[a_id][w], '\t+%f' % weight
self.vector_by_category[a_id][w] += weight
self.contributors_by_category[a_id].update([term_id])
if __name__ == '__main__':
import random
from random import shuffle as randomize
tfidf = TfIdfMatrix()
tfidf.load_features('bkp.big.out/vector.term.csv')
tfidf.load_distribution('bkp.big.out/vector.index.csv')
# tfidf.load_features('vector.term.csv')
# tfidf.load_distribution('vector.index.csv')
ctree = CategoryTree(
'bkp.big.out/category.index.csv',
'bkp.big.out/term.csv',
'bkp.big.out/category.csv',
# 'category.index.csv',
# 'term.csv',
# 'category.csv',
tfidf,
max_depth=1
)
ctree.load()
ctree.build()
ctree.dump()
<｜fim▁end｜> | out = open('vector.dump.txt', 'wb')
for i, (_id, projections) in enumerate(self.projected.items()):
if not i % 100:
print i, len(self.projected.keys())
if not projections:
continue
features = [
(self.tfidf.word_by_id[wid], round(weight, 4))
for wid, weight in self.vector_by_category[_id].most_common()
if round(weight, 4)
]
record = (
_id,
self.concept_by_id[_id],
features
)
out.write('%s\n' % str(record))
out.close() |
<|file_name|>CategoryTree.py<|end_file_name|><｜fim▁begin｜>import json
from collections import (
Counter,
defaultdict as deft
)
from copy import deepcopy as cp
# from cPickle import (
# dump as to_pickle,
# load as from_pickle
# )
from StringIO import StringIO
from TfIdfMatrix import TfIdfMatrix
from Tools import from_csv
class CategoryTree:
def __init__(self, categories_by_concept, terms,
categories, tfidf, max_depth=5, min_df=20
):
self.min_df = min_df
self.path_categories_by_concept = categories_by_concept
self.path_categories = categories
self.path_terms = terms
self.max_depth = max_depth
self.observed_category = deft(bool)
self.id_by_concept = dict([])
self.concept_by_id = dict([])
self.term_is_category = deft(bool)
self.parents_by_category = dict([])
self.parents_by_concept = deft(list)
self.id_by_term = dict([])
self.term_by_id = dict([])
self.has_parents = deft(bool)
self.tfidf = tfidf
self.pulling = set([])
self.vector_by_category = deft(Counter)
self.contributors_by_category = deft(set)
self.projected = Counter()
def build(self):
for i, c in enumerate(self.concept_by_id.values()):
self(c)
if not i % 100:
t = float(len(self.concept_by_id.keys()))
print i, int(t), round(i / t, 2)
# if i >= 5000:
# break
def dump(self):
# Simulate a file with StringIO
out = open('vector.dump.txt', 'wb')
for i, (_id, projections) in enumerate(self.projected.items()):
if not i % 100:
print i, len(self.projected.keys())
if not projections:
continue
features = [
(self.tfidf.word_by_id[wid], round(weight, 4))
for wid, weight in self.vector_by_category[_id].most_common()
if round(weight, 4)
]
record = (
_id,
self.concept_by_id[_id],
features
)
out.write('%s\n' % str(record))
out.close()
def __call__(self, category):
<|fim_middle|>
def __get_parents(self, _id):
parents = []
name = self.concept_by_id[_id]
if (
not self.observed_category[name] or
not self.observed_category[_id] or
not self.has_parents[_id]
):
return []
else:
for i in self.parents_by_category[_id]:
if not self.observed_category[i]:
continue
_name = self.concept_by_id[i]
parents.append(_name)
return set(parents) - self.pulling
def __pull(self, vector, depth, category, tree):
_id = self.id_by_concept[category]
if not self.pulling:
# print
# print
# print category, _id
# print [self.term_by_id[x] for x in self.contributors_by_category[_id]]
# print self.vector_by_category[_id].most_common(20)
vector = self.vector_by_category[_id]
if not self.observed_category[category]:
return dict([])
parents = self.__get_parents(_id)
if not parents or depth >= self.max_depth:
tree[category] = dict([])
else:
subtree = dict([])
self.pulling.update(parents)
for parent in parents:
subtree = self.__pull(vector, depth + 1, parent, subtree)
tree[category] = subtree
self.__project(vector, tree)
return tree
def __project(self, vector, tree):
if not tree.keys():
return
else:
for key, subtree in tree.items():
_id = self.id_by_concept[key]
self.projected[_id] += 1
self.__add2vec(vector, _id)
self.__project(vector, subtree)
def __add2vec(self, vector, _id):
# for w, weight in vector.items():
# __id = self.tfidf.id_by_word[w]
for __id, weight in vector.items():
self.vector_by_category[_id][__id] += weight
def load(self):
self.__load_terms()
self.__load_categories()
self.__load_assignments()
def __load_categories(self):
for concept, _id in from_csv(self.path_categories):
_id = int(_id)
self.id_by_concept[concept] = _id
self.concept_by_id[_id] = concept
self.observed_category[_id] = True
self.observed_category[concept] = True
# print concept, _id, len(self.id_by_concept.keys())
# exit()
def __load_terms(self):
for term, _id in from_csv(self.path_terms):
_id = int(_id)
self.term_by_id[_id] = term
self.id_by_term[term] = _id
if not term.startswith('Category:'):
continue
self.term_is_category[term] = True
self.term_is_category[_id] = True
def __load_assignments(self):
for row in from_csv(self.path_categories_by_concept):
ints = [int(field) for field in row]
term_id = ints[0]
term = self.term_by_id[term_id]
if self.term_is_category[term_id] and \
self.observed_category[term]:
term = self.term_by_id[term_id]
cat_id = self.id_by_concept[term]
assignments = [i for i in ints[1:] if self.observed_category[i]]
self.parents_by_category[cat_id] = assignments
self.has_parents[cat_id] = True
else:
vector = self.tfidf.content(term_id)
assignments = [i for i in ints[1:] if self.observed_category[i]]
self.parents_by_concept[term_id] = assignments
for a_id in assignments:
for w, weight in vector:
if self.tfidf.df[w] < self.min_df:
continue
#print term, term_id, self.concept_by_id[a_id], w, self.vector_by_category[a_id][w], '\t+%f' % weight
self.vector_by_category[a_id][w] += weight
self.contributors_by_category[a_id].update([term_id])
if __name__ == '__main__':
import random
from random import shuffle as randomize
tfidf = TfIdfMatrix()
tfidf.load_features('bkp.big.out/vector.term.csv')
tfidf.load_distribution('bkp.big.out/vector.index.csv')
# tfidf.load_features('vector.term.csv')
# tfidf.load_distribution('vector.index.csv')
ctree = CategoryTree(
'bkp.big.out/category.index.csv',
'bkp.big.out/term.csv',
'bkp.big.out/category.csv',
# 'category.index.csv',
# 'term.csv',
# 'category.csv',
tfidf,
max_depth=1
)
ctree.load()
ctree.build()
ctree.dump()
<｜fim▁end｜> | self.pulling = set([])
return self.__pull(None, 0, category, dict([])) |
<|file_name|>CategoryTree.py<|end_file_name|><｜fim▁begin｜>import json
from collections import (
Counter,
defaultdict as deft
)
from copy import deepcopy as cp
# from cPickle import (
# dump as to_pickle,
# load as from_pickle
# )
from StringIO import StringIO
from TfIdfMatrix import TfIdfMatrix
from Tools import from_csv
class CategoryTree:
def __init__(self, categories_by_concept, terms,
categories, tfidf, max_depth=5, min_df=20
):
self.min_df = min_df
self.path_categories_by_concept = categories_by_concept
self.path_categories = categories
self.path_terms = terms
self.max_depth = max_depth
self.observed_category = deft(bool)
self.id_by_concept = dict([])
self.concept_by_id = dict([])
self.term_is_category = deft(bool)
self.parents_by_category = dict([])
self.parents_by_concept = deft(list)
self.id_by_term = dict([])
self.term_by_id = dict([])
self.has_parents = deft(bool)
self.tfidf = tfidf
self.pulling = set([])
self.vector_by_category = deft(Counter)
self.contributors_by_category = deft(set)
self.projected = Counter()
def build(self):
for i, c in enumerate(self.concept_by_id.values()):
self(c)
if not i % 100:
t = float(len(self.concept_by_id.keys()))
print i, int(t), round(i / t, 2)
# if i >= 5000:
# break
def dump(self):
# Simulate a file with StringIO
out = open('vector.dump.txt', 'wb')
for i, (_id, projections) in enumerate(self.projected.items()):
if not i % 100:
print i, len(self.projected.keys())
if not projections:
continue
features = [
(self.tfidf.word_by_id[wid], round(weight, 4))
for wid, weight in self.vector_by_category[_id].most_common()
if round(weight, 4)
]
record = (
_id,
self.concept_by_id[_id],
features
)
out.write('%s\n' % str(record))
out.close()
def __call__(self, category):
self.pulling = set([])
return self.__pull(None, 0, category, dict([]))
def __get_parents(self, _id):
<|fim_middle|>
def __pull(self, vector, depth, category, tree):
_id = self.id_by_concept[category]
if not self.pulling:
# print
# print
# print category, _id
# print [self.term_by_id[x] for x in self.contributors_by_category[_id]]
# print self.vector_by_category[_id].most_common(20)
vector = self.vector_by_category[_id]
if not self.observed_category[category]:
return dict([])
parents = self.__get_parents(_id)
if not parents or depth >= self.max_depth:
tree[category] = dict([])
else:
subtree = dict([])
self.pulling.update(parents)
for parent in parents:
subtree = self.__pull(vector, depth + 1, parent, subtree)
tree[category] = subtree
self.__project(vector, tree)
return tree
def __project(self, vector, tree):
if not tree.keys():
return
else:
for key, subtree in tree.items():
_id = self.id_by_concept[key]
self.projected[_id] += 1
self.__add2vec(vector, _id)
self.__project(vector, subtree)
def __add2vec(self, vector, _id):
# for w, weight in vector.items():
# __id = self.tfidf.id_by_word[w]
for __id, weight in vector.items():
self.vector_by_category[_id][__id] += weight
def load(self):
self.__load_terms()
self.__load_categories()
self.__load_assignments()
def __load_categories(self):
for concept, _id in from_csv(self.path_categories):
_id = int(_id)
self.id_by_concept[concept] = _id
self.concept_by_id[_id] = concept
self.observed_category[_id] = True
self.observed_category[concept] = True
# print concept, _id, len(self.id_by_concept.keys())
# exit()
def __load_terms(self):
for term, _id in from_csv(self.path_terms):
_id = int(_id)
self.term_by_id[_id] = term
self.id_by_term[term] = _id
if not term.startswith('Category:'):
continue
self.term_is_category[term] = True
self.term_is_category[_id] = True
def __load_assignments(self):
for row in from_csv(self.path_categories_by_concept):
ints = [int(field) for field in row]
term_id = ints[0]
term = self.term_by_id[term_id]
if self.term_is_category[term_id] and \
self.observed_category[term]:
term = self.term_by_id[term_id]
cat_id = self.id_by_concept[term]
assignments = [i for i in ints[1:] if self.observed_category[i]]
self.parents_by_category[cat_id] = assignments
self.has_parents[cat_id] = True
else:
vector = self.tfidf.content(term_id)
assignments = [i for i in ints[1:] if self.observed_category[i]]
self.parents_by_concept[term_id] = assignments
for a_id in assignments:
for w, weight in vector:
if self.tfidf.df[w] < self.min_df:
continue
#print term, term_id, self.concept_by_id[a_id], w, self.vector_by_category[a_id][w], '\t+%f' % weight
self.vector_by_category[a_id][w] += weight
self.contributors_by_category[a_id].update([term_id])
if __name__ == '__main__':
import random
from random import shuffle as randomize
tfidf = TfIdfMatrix()
tfidf.load_features('bkp.big.out/vector.term.csv')
tfidf.load_distribution('bkp.big.out/vector.index.csv')
# tfidf.load_features('vector.term.csv')
# tfidf.load_distribution('vector.index.csv')
ctree = CategoryTree(
'bkp.big.out/category.index.csv',
'bkp.big.out/term.csv',
'bkp.big.out/category.csv',
# 'category.index.csv',
# 'term.csv',
# 'category.csv',
tfidf,
max_depth=1
)
ctree.load()
ctree.build()
ctree.dump()
<｜fim▁end｜> | parents = []
name = self.concept_by_id[_id]
if (
not self.observed_category[name] or
not self.observed_category[_id] or
not self.has_parents[_id]
):
return []
else:
for i in self.parents_by_category[_id]:
if not self.observed_category[i]:
continue
_name = self.concept_by_id[i]
parents.append(_name)
return set(parents) - self.pulling |
<|file_name|>CategoryTree.py<|end_file_name|><｜fim▁begin｜>import json
from collections import (
Counter,
defaultdict as deft
)
from copy import deepcopy as cp
# from cPickle import (
# dump as to_pickle,
# load as from_pickle
# )
from StringIO import StringIO
from TfIdfMatrix import TfIdfMatrix
from Tools import from_csv
class CategoryTree:
def __init__(self, categories_by_concept, terms,
categories, tfidf, max_depth=5, min_df=20
):
self.min_df = min_df
self.path_categories_by_concept = categories_by_concept
self.path_categories = categories
self.path_terms = terms
self.max_depth = max_depth
self.observed_category = deft(bool)
self.id_by_concept = dict([])
self.concept_by_id = dict([])
self.term_is_category = deft(bool)
self.parents_by_category = dict([])
self.parents_by_concept = deft(list)
self.id_by_term = dict([])
self.term_by_id = dict([])
self.has_parents = deft(bool)
self.tfidf = tfidf
self.pulling = set([])
self.vector_by_category = deft(Counter)
self.contributors_by_category = deft(set)
self.projected = Counter()
def build(self):
for i, c in enumerate(self.concept_by_id.values()):
self(c)
if not i % 100:
t = float(len(self.concept_by_id.keys()))
print i, int(t), round(i / t, 2)
# if i >= 5000:
# break
def dump(self):
# Simulate a file with StringIO
out = open('vector.dump.txt', 'wb')
for i, (_id, projections) in enumerate(self.projected.items()):
if not i % 100:
print i, len(self.projected.keys())
if not projections:
continue
features = [
(self.tfidf.word_by_id[wid], round(weight, 4))
for wid, weight in self.vector_by_category[_id].most_common()
if round(weight, 4)
]
record = (
_id,
self.concept_by_id[_id],
features
)
out.write('%s\n' % str(record))
out.close()
def __call__(self, category):
self.pulling = set([])
return self.__pull(None, 0, category, dict([]))
def __get_parents(self, _id):
parents = []
name = self.concept_by_id[_id]
if (
not self.observed_category[name] or
not self.observed_category[_id] or
not self.has_parents[_id]
):
return []
else:
for i in self.parents_by_category[_id]:
if not self.observed_category[i]:
continue
_name = self.concept_by_id[i]
parents.append(_name)
return set(parents) - self.pulling
def __pull(self, vector, depth, category, tree):
<|fim_middle|>
def __project(self, vector, tree):
if not tree.keys():
return
else:
for key, subtree in tree.items():
_id = self.id_by_concept[key]
self.projected[_id] += 1
self.__add2vec(vector, _id)
self.__project(vector, subtree)
def __add2vec(self, vector, _id):
# for w, weight in vector.items():
# __id = self.tfidf.id_by_word[w]
for __id, weight in vector.items():
self.vector_by_category[_id][__id] += weight
def load(self):
self.__load_terms()
self.__load_categories()
self.__load_assignments()
def __load_categories(self):
for concept, _id in from_csv(self.path_categories):
_id = int(_id)
self.id_by_concept[concept] = _id
self.concept_by_id[_id] = concept
self.observed_category[_id] = True
self.observed_category[concept] = True
# print concept, _id, len(self.id_by_concept.keys())
# exit()
def __load_terms(self):
for term, _id in from_csv(self.path_terms):
_id = int(_id)
self.term_by_id[_id] = term
self.id_by_term[term] = _id
if not term.startswith('Category:'):
continue
self.term_is_category[term] = True
self.term_is_category[_id] = True
def __load_assignments(self):
for row in from_csv(self.path_categories_by_concept):
ints = [int(field) for field in row]
term_id = ints[0]
term = self.term_by_id[term_id]
if self.term_is_category[term_id] and \
self.observed_category[term]:
term = self.term_by_id[term_id]
cat_id = self.id_by_concept[term]
assignments = [i for i in ints[1:] if self.observed_category[i]]
self.parents_by_category[cat_id] = assignments
self.has_parents[cat_id] = True
else:
vector = self.tfidf.content(term_id)
assignments = [i for i in ints[1:] if self.observed_category[i]]
self.parents_by_concept[term_id] = assignments
for a_id in assignments:
for w, weight in vector:
if self.tfidf.df[w] < self.min_df:
continue
#print term, term_id, self.concept_by_id[a_id], w, self.vector_by_category[a_id][w], '\t+%f' % weight
self.vector_by_category[a_id][w] += weight
self.contributors_by_category[a_id].update([term_id])
if __name__ == '__main__':
import random
from random import shuffle as randomize
tfidf = TfIdfMatrix()
tfidf.load_features('bkp.big.out/vector.term.csv')
tfidf.load_distribution('bkp.big.out/vector.index.csv')
# tfidf.load_features('vector.term.csv')
# tfidf.load_distribution('vector.index.csv')
ctree = CategoryTree(
'bkp.big.out/category.index.csv',
'bkp.big.out/term.csv',
'bkp.big.out/category.csv',
# 'category.index.csv',
# 'term.csv',
# 'category.csv',
tfidf,
max_depth=1
)
ctree.load()
ctree.build()
ctree.dump()
<｜fim▁end｜> | _id = self.id_by_concept[category]
if not self.pulling:
# print
# print
# print category, _id
# print [self.term_by_id[x] for x in self.contributors_by_category[_id]]
# print self.vector_by_category[_id].most_common(20)
vector = self.vector_by_category[_id]
if not self.observed_category[category]:
return dict([])
parents = self.__get_parents(_id)
if not parents or depth >= self.max_depth:
tree[category] = dict([])
else:
subtree = dict([])
self.pulling.update(parents)
for parent in parents:
subtree = self.__pull(vector, depth + 1, parent, subtree)
tree[category] = subtree
self.__project(vector, tree)
return tree |
<|file_name|>CategoryTree.py<|end_file_name|><｜fim▁begin｜>import json
from collections import (
Counter,
defaultdict as deft
)
from copy import deepcopy as cp
# from cPickle import (
# dump as to_pickle,
# load as from_pickle
# )
from StringIO import StringIO
from TfIdfMatrix import TfIdfMatrix
from Tools import from_csv
class CategoryTree:
def __init__(self, categories_by_concept, terms,
categories, tfidf, max_depth=5, min_df=20
):
self.min_df = min_df
self.path_categories_by_concept = categories_by_concept
self.path_categories = categories
self.path_terms = terms
self.max_depth = max_depth
self.observed_category = deft(bool)
self.id_by_concept = dict([])
self.concept_by_id = dict([])
self.term_is_category = deft(bool)
self.parents_by_category = dict([])
self.parents_by_concept = deft(list)
self.id_by_term = dict([])
self.term_by_id = dict([])
self.has_parents = deft(bool)
self.tfidf = tfidf
self.pulling = set([])
self.vector_by_category = deft(Counter)
self.contributors_by_category = deft(set)
self.projected = Counter()
def build(self):
for i, c in enumerate(self.concept_by_id.values()):
self(c)
if not i % 100:
t = float(len(self.concept_by_id.keys()))
print i, int(t), round(i / t, 2)
# if i >= 5000:
# break
def dump(self):
# Simulate a file with StringIO
out = open('vector.dump.txt', 'wb')
for i, (_id, projections) in enumerate(self.projected.items()):
if not i % 100:
print i, len(self.projected.keys())
if not projections:
continue
features = [
(self.tfidf.word_by_id[wid], round(weight, 4))
for wid, weight in self.vector_by_category[_id].most_common()
if round(weight, 4)
]
record = (
_id,
self.concept_by_id[_id],
features
)
out.write('%s\n' % str(record))
out.close()
def __call__(self, category):
self.pulling = set([])
return self.__pull(None, 0, category, dict([]))
def __get_parents(self, _id):
parents = []
name = self.concept_by_id[_id]
if (
not self.observed_category[name] or
not self.observed_category[_id] or
not self.has_parents[_id]
):
return []
else:
for i in self.parents_by_category[_id]:
if not self.observed_category[i]:
continue
_name = self.concept_by_id[i]
parents.append(_name)
return set(parents) - self.pulling
def __pull(self, vector, depth, category, tree):
_id = self.id_by_concept[category]
if not self.pulling:
# print
# print
# print category, _id
# print [self.term_by_id[x] for x in self.contributors_by_category[_id]]
# print self.vector_by_category[_id].most_common(20)
vector = self.vector_by_category[_id]
if not self.observed_category[category]:
return dict([])
parents = self.__get_parents(_id)
if not parents or depth >= self.max_depth:
tree[category] = dict([])
else:
subtree = dict([])
self.pulling.update(parents)
for parent in parents:
subtree = self.__pull(vector, depth + 1, parent, subtree)
tree[category] = subtree
self.__project(vector, tree)
return tree
def __project(self, vector, tree):
<|fim_middle|>
def __add2vec(self, vector, _id):
# for w, weight in vector.items():
# __id = self.tfidf.id_by_word[w]
for __id, weight in vector.items():
self.vector_by_category[_id][__id] += weight
def load(self):
self.__load_terms()
self.__load_categories()
self.__load_assignments()
def __load_categories(self):
for concept, _id in from_csv(self.path_categories):
_id = int(_id)
self.id_by_concept[concept] = _id
self.concept_by_id[_id] = concept
self.observed_category[_id] = True
self.observed_category[concept] = True
# print concept, _id, len(self.id_by_concept.keys())
# exit()
def __load_terms(self):
for term, _id in from_csv(self.path_terms):
_id = int(_id)
self.term_by_id[_id] = term
self.id_by_term[term] = _id
if not term.startswith('Category:'):
continue
self.term_is_category[term] = True
self.term_is_category[_id] = True
def __load_assignments(self):
for row in from_csv(self.path_categories_by_concept):
ints = [int(field) for field in row]
term_id = ints[0]
term = self.term_by_id[term_id]
if self.term_is_category[term_id] and \
self.observed_category[term]:
term = self.term_by_id[term_id]
cat_id = self.id_by_concept[term]
assignments = [i for i in ints[1:] if self.observed_category[i]]
self.parents_by_category[cat_id] = assignments
self.has_parents[cat_id] = True
else:
vector = self.tfidf.content(term_id)
assignments = [i for i in ints[1:] if self.observed_category[i]]
self.parents_by_concept[term_id] = assignments
for a_id in assignments:
for w, weight in vector:
if self.tfidf.df[w] < self.min_df:
continue
#print term, term_id, self.concept_by_id[a_id], w, self.vector_by_category[a_id][w], '\t+%f' % weight
self.vector_by_category[a_id][w] += weight
self.contributors_by_category[a_id].update([term_id])
if __name__ == '__main__':
import random
from random import shuffle as randomize
tfidf = TfIdfMatrix()
tfidf.load_features('bkp.big.out/vector.term.csv')
tfidf.load_distribution('bkp.big.out/vector.index.csv')
# tfidf.load_features('vector.term.csv')
# tfidf.load_distribution('vector.index.csv')
ctree = CategoryTree(
'bkp.big.out/category.index.csv',
'bkp.big.out/term.csv',
'bkp.big.out/category.csv',
# 'category.index.csv',
# 'term.csv',
# 'category.csv',
tfidf,
max_depth=1
)
ctree.load()
ctree.build()
ctree.dump()
<｜fim▁end｜> | if not tree.keys():
return
else:
for key, subtree in tree.items():
_id = self.id_by_concept[key]
self.projected[_id] += 1
self.__add2vec(vector, _id)
self.__project(vector, subtree) |
<|file_name|>CategoryTree.py<|end_file_name|><｜fim▁begin｜>import json
from collections import (
Counter,
defaultdict as deft
)
from copy import deepcopy as cp
# from cPickle import (
# dump as to_pickle,
# load as from_pickle
# )
from StringIO import StringIO
from TfIdfMatrix import TfIdfMatrix
from Tools import from_csv
class CategoryTree:
def __init__(self, categories_by_concept, terms,
categories, tfidf, max_depth=5, min_df=20
):
self.min_df = min_df
self.path_categories_by_concept = categories_by_concept
self.path_categories = categories
self.path_terms = terms
self.max_depth = max_depth
self.observed_category = deft(bool)
self.id_by_concept = dict([])
self.concept_by_id = dict([])
self.term_is_category = deft(bool)
self.parents_by_category = dict([])
self.parents_by_concept = deft(list)
self.id_by_term = dict([])
self.term_by_id = dict([])
self.has_parents = deft(bool)
self.tfidf = tfidf
self.pulling = set([])
self.vector_by_category = deft(Counter)
self.contributors_by_category = deft(set)
self.projected = Counter()
def build(self):
for i, c in enumerate(self.concept_by_id.values()):
self(c)
if not i % 100:
t = float(len(self.concept_by_id.keys()))
print i, int(t), round(i / t, 2)
# if i >= 5000:
# break
def dump(self):
# Simulate a file with StringIO
out = open('vector.dump.txt', 'wb')
for i, (_id, projections) in enumerate(self.projected.items()):
if not i % 100:
print i, len(self.projected.keys())
if not projections:
continue
features = [
(self.tfidf.word_by_id[wid], round(weight, 4))
for wid, weight in self.vector_by_category[_id].most_common()
if round(weight, 4)
]
record = (
_id,
self.concept_by_id[_id],
features
)
out.write('%s\n' % str(record))
out.close()
def __call__(self, category):
self.pulling = set([])
return self.__pull(None, 0, category, dict([]))
def __get_parents(self, _id):
parents = []
name = self.concept_by_id[_id]
if (
not self.observed_category[name] or
not self.observed_category[_id] or
not self.has_parents[_id]
):
return []
else:
for i in self.parents_by_category[_id]:
if not self.observed_category[i]:
continue
_name = self.concept_by_id[i]
parents.append(_name)
return set(parents) - self.pulling
def __pull(self, vector, depth, category, tree):
_id = self.id_by_concept[category]
if not self.pulling:
# print
# print
# print category, _id
# print [self.term_by_id[x] for x in self.contributors_by_category[_id]]
# print self.vector_by_category[_id].most_common(20)
vector = self.vector_by_category[_id]
if not self.observed_category[category]:
return dict([])
parents = self.__get_parents(_id)
if not parents or depth >= self.max_depth:
tree[category] = dict([])
else:
subtree = dict([])
self.pulling.update(parents)
for parent in parents:
subtree = self.__pull(vector, depth + 1, parent, subtree)
tree[category] = subtree
self.__project(vector, tree)
return tree
def __project(self, vector, tree):
if not tree.keys():
return
else:
for key, subtree in tree.items():
_id = self.id_by_concept[key]
self.projected[_id] += 1
self.__add2vec(vector, _id)
self.__project(vector, subtree)
def __add2vec(self, vector, _id):
# for w, weight in vector.items():
# __id = self.tfidf.id_by_word[w]
<|fim_middle|>
def load(self):
self.__load_terms()
self.__load_categories()
self.__load_assignments()
def __load_categories(self):
for concept, _id in from_csv(self.path_categories):
_id = int(_id)
self.id_by_concept[concept] = _id
self.concept_by_id[_id] = concept
self.observed_category[_id] = True
self.observed_category[concept] = True
# print concept, _id, len(self.id_by_concept.keys())
# exit()
def __load_terms(self):
for term, _id in from_csv(self.path_terms):
_id = int(_id)
self.term_by_id[_id] = term
self.id_by_term[term] = _id
if not term.startswith('Category:'):
continue
self.term_is_category[term] = True
self.term_is_category[_id] = True
def __load_assignments(self):
for row in from_csv(self.path_categories_by_concept):
ints = [int(field) for field in row]
term_id = ints[0]
term = self.term_by_id[term_id]
if self.term_is_category[term_id] and \
self.observed_category[term]:
term = self.term_by_id[term_id]
cat_id = self.id_by_concept[term]
assignments = [i for i in ints[1:] if self.observed_category[i]]
self.parents_by_category[cat_id] = assignments
self.has_parents[cat_id] = True
else:
vector = self.tfidf.content(term_id)
assignments = [i for i in ints[1:] if self.observed_category[i]]
self.parents_by_concept[term_id] = assignments
for a_id in assignments:
for w, weight in vector:
if self.tfidf.df[w] < self.min_df:
continue
#print term, term_id, self.concept_by_id[a_id], w, self.vector_by_category[a_id][w], '\t+%f' % weight
self.vector_by_category[a_id][w] += weight
self.contributors_by_category[a_id].update([term_id])
if __name__ == '__main__':
import random
from random import shuffle as randomize
tfidf = TfIdfMatrix()
tfidf.load_features('bkp.big.out/vector.term.csv')
tfidf.load_distribution('bkp.big.out/vector.index.csv')
# tfidf.load_features('vector.term.csv')
# tfidf.load_distribution('vector.index.csv')
ctree = CategoryTree(
'bkp.big.out/category.index.csv',
'bkp.big.out/term.csv',
'bkp.big.out/category.csv',
# 'category.index.csv',
# 'term.csv',
# 'category.csv',
tfidf,
max_depth=1
)
ctree.load()
ctree.build()
ctree.dump()
<｜fim▁end｜> | for __id, weight in vector.items():
self.vector_by_category[_id][__id] += weight |
<|file_name|>CategoryTree.py<|end_file_name|><｜fim▁begin｜>import json
from collections import (
Counter,
defaultdict as deft
)
from copy import deepcopy as cp
# from cPickle import (
# dump as to_pickle,
# load as from_pickle
# )
from StringIO import StringIO
from TfIdfMatrix import TfIdfMatrix
from Tools import from_csv
class CategoryTree:
def __init__(self, categories_by_concept, terms,
categories, tfidf, max_depth=5, min_df=20
):
self.min_df = min_df
self.path_categories_by_concept = categories_by_concept
self.path_categories = categories
self.path_terms = terms
self.max_depth = max_depth
self.observed_category = deft(bool)
self.id_by_concept = dict([])
self.concept_by_id = dict([])
self.term_is_category = deft(bool)
self.parents_by_category = dict([])
self.parents_by_concept = deft(list)
self.id_by_term = dict([])
self.term_by_id = dict([])
self.has_parents = deft(bool)
self.tfidf = tfidf
self.pulling = set([])
self.vector_by_category = deft(Counter)
self.contributors_by_category = deft(set)
self.projected = Counter()
def build(self):
for i, c in enumerate(self.concept_by_id.values()):
self(c)
if not i % 100:
t = float(len(self.concept_by_id.keys()))
print i, int(t), round(i / t, 2)
# if i >= 5000:
# break
def dump(self):
# Simulate a file with StringIO
out = open('vector.dump.txt', 'wb')
for i, (_id, projections) in enumerate(self.projected.items()):
if not i % 100:
print i, len(self.projected.keys())
if not projections:
continue
features = [
(self.tfidf.word_by_id[wid], round(weight, 4))
for wid, weight in self.vector_by_category[_id].most_common()
if round(weight, 4)
]
record = (
_id,
self.concept_by_id[_id],
features
)
out.write('%s\n' % str(record))
out.close()
def __call__(self, category):
self.pulling = set([])
return self.__pull(None, 0, category, dict([]))
def __get_parents(self, _id):
parents = []
name = self.concept_by_id[_id]
if (
not self.observed_category[name] or
not self.observed_category[_id] or
not self.has_parents[_id]
):
return []
else:
for i in self.parents_by_category[_id]:
if not self.observed_category[i]:
continue
_name = self.concept_by_id[i]
parents.append(_name)
return set(parents) - self.pulling
def __pull(self, vector, depth, category, tree):
_id = self.id_by_concept[category]
if not self.pulling:
# print
# print
# print category, _id
# print [self.term_by_id[x] for x in self.contributors_by_category[_id]]
# print self.vector_by_category[_id].most_common(20)
vector = self.vector_by_category[_id]
if not self.observed_category[category]:
return dict([])
parents = self.__get_parents(_id)
if not parents or depth >= self.max_depth:
tree[category] = dict([])
else:
subtree = dict([])
self.pulling.update(parents)
for parent in parents:
subtree = self.__pull(vector, depth + 1, parent, subtree)
tree[category] = subtree
self.__project(vector, tree)
return tree
def __project(self, vector, tree):
if not tree.keys():
return
else:
for key, subtree in tree.items():
_id = self.id_by_concept[key]
self.projected[_id] += 1
self.__add2vec(vector, _id)
self.__project(vector, subtree)
def __add2vec(self, vector, _id):
# for w, weight in vector.items():
# __id = self.tfidf.id_by_word[w]
for __id, weight in vector.items():
self.vector_by_category[_id][__id] += weight
def load(self):
<|fim_middle|>
def __load_categories(self):
for concept, _id in from_csv(self.path_categories):
_id = int(_id)
self.id_by_concept[concept] = _id
self.concept_by_id[_id] = concept
self.observed_category[_id] = True
self.observed_category[concept] = True
# print concept, _id, len(self.id_by_concept.keys())
# exit()
def __load_terms(self):
for term, _id in from_csv(self.path_terms):
_id = int(_id)
self.term_by_id[_id] = term
self.id_by_term[term] = _id
if not term.startswith('Category:'):
continue
self.term_is_category[term] = True
self.term_is_category[_id] = True
def __load_assignments(self):
for row in from_csv(self.path_categories_by_concept):
ints = [int(field) for field in row]
term_id = ints[0]
term = self.term_by_id[term_id]
if self.term_is_category[term_id] and \
self.observed_category[term]:
term = self.term_by_id[term_id]
cat_id = self.id_by_concept[term]
assignments = [i for i in ints[1:] if self.observed_category[i]]
self.parents_by_category[cat_id] = assignments
self.has_parents[cat_id] = True
else:
vector = self.tfidf.content(term_id)
assignments = [i for i in ints[1:] if self.observed_category[i]]
self.parents_by_concept[term_id] = assignments
for a_id in assignments:
for w, weight in vector:
if self.tfidf.df[w] < self.min_df:
continue
#print term, term_id, self.concept_by_id[a_id], w, self.vector_by_category[a_id][w], '\t+%f' % weight
self.vector_by_category[a_id][w] += weight
self.contributors_by_category[a_id].update([term_id])
if __name__ == '__main__':
import random
from random import shuffle as randomize
tfidf = TfIdfMatrix()
tfidf.load_features('bkp.big.out/vector.term.csv')
tfidf.load_distribution('bkp.big.out/vector.index.csv')
# tfidf.load_features('vector.term.csv')
# tfidf.load_distribution('vector.index.csv')
ctree = CategoryTree(
'bkp.big.out/category.index.csv',
'bkp.big.out/term.csv',
'bkp.big.out/category.csv',
# 'category.index.csv',
# 'term.csv',
# 'category.csv',
tfidf,
max_depth=1
)
ctree.load()
ctree.build()
ctree.dump()
<｜fim▁end｜> | self.__load_terms()
self.__load_categories()
self.__load_assignments() |
<|file_name|>CategoryTree.py<|end_file_name|><｜fim▁begin｜>import json
from collections import (
Counter,
defaultdict as deft
)
from copy import deepcopy as cp
# from cPickle import (
# dump as to_pickle,
# load as from_pickle
# )
from StringIO import StringIO
from TfIdfMatrix import TfIdfMatrix
from Tools import from_csv
class CategoryTree:
def __init__(self, categories_by_concept, terms,
categories, tfidf, max_depth=5, min_df=20
):
self.min_df = min_df
self.path_categories_by_concept = categories_by_concept
self.path_categories = categories
self.path_terms = terms
self.max_depth = max_depth
self.observed_category = deft(bool)
self.id_by_concept = dict([])
self.concept_by_id = dict([])
self.term_is_category = deft(bool)
self.parents_by_category = dict([])
self.parents_by_concept = deft(list)
self.id_by_term = dict([])
self.term_by_id = dict([])
self.has_parents = deft(bool)
self.tfidf = tfidf
self.pulling = set([])
self.vector_by_category = deft(Counter)
self.contributors_by_category = deft(set)
self.projected = Counter()
def build(self):
for i, c in enumerate(self.concept_by_id.values()):
self(c)
if not i % 100:
t = float(len(self.concept_by_id.keys()))
print i, int(t), round(i / t, 2)
# if i >= 5000:
# break
def dump(self):
# Simulate a file with StringIO
out = open('vector.dump.txt', 'wb')
for i, (_id, projections) in enumerate(self.projected.items()):
if not i % 100:
print i, len(self.projected.keys())
if not projections:
continue
features = [
(self.tfidf.word_by_id[wid], round(weight, 4))
for wid, weight in self.vector_by_category[_id].most_common()
if round(weight, 4)
]
record = (
_id,
self.concept_by_id[_id],
features
)
out.write('%s\n' % str(record))
out.close()
def __call__(self, category):
self.pulling = set([])
return self.__pull(None, 0, category, dict([]))
def __get_parents(self, _id):
parents = []
name = self.concept_by_id[_id]
if (
not self.observed_category[name] or
not self.observed_category[_id] or
not self.has_parents[_id]
):
return []
else:
for i in self.parents_by_category[_id]:
if not self.observed_category[i]:
continue
_name = self.concept_by_id[i]
parents.append(_name)
return set(parents) - self.pulling
def __pull(self, vector, depth, category, tree):
_id = self.id_by_concept[category]
if not self.pulling:
# print
# print
# print category, _id
# print [self.term_by_id[x] for x in self.contributors_by_category[_id]]
# print self.vector_by_category[_id].most_common(20)
vector = self.vector_by_category[_id]
if not self.observed_category[category]:
return dict([])
parents = self.__get_parents(_id)
if not parents or depth >= self.max_depth:
tree[category] = dict([])
else:
subtree = dict([])
self.pulling.update(parents)
for parent in parents:
subtree = self.__pull(vector, depth + 1, parent, subtree)
tree[category] = subtree
self.__project(vector, tree)
return tree
def __project(self, vector, tree):
if not tree.keys():
return
else:
for key, subtree in tree.items():
_id = self.id_by_concept[key]
self.projected[_id] += 1
self.__add2vec(vector, _id)
self.__project(vector, subtree)
def __add2vec(self, vector, _id):
# for w, weight in vector.items():
# __id = self.tfidf.id_by_word[w]
for __id, weight in vector.items():
self.vector_by_category[_id][__id] += weight
def load(self):
self.__load_terms()
self.__load_categories()
self.__load_assignments()
def __load_categories(self):
<|fim_middle|>
# print concept, _id, len(self.id_by_concept.keys())
# exit()
def __load_terms(self):
for term, _id in from_csv(self.path_terms):
_id = int(_id)
self.term_by_id[_id] = term
self.id_by_term[term] = _id
if not term.startswith('Category:'):
continue
self.term_is_category[term] = True
self.term_is_category[_id] = True
def __load_assignments(self):
for row in from_csv(self.path_categories_by_concept):
ints = [int(field) for field in row]
term_id = ints[0]
term = self.term_by_id[term_id]
if self.term_is_category[term_id] and \
self.observed_category[term]:
term = self.term_by_id[term_id]
cat_id = self.id_by_concept[term]
assignments = [i for i in ints[1:] if self.observed_category[i]]
self.parents_by_category[cat_id] = assignments
self.has_parents[cat_id] = True
else:
vector = self.tfidf.content(term_id)
assignments = [i for i in ints[1:] if self.observed_category[i]]
self.parents_by_concept[term_id] = assignments
for a_id in assignments:
for w, weight in vector:
if self.tfidf.df[w] < self.min_df:
continue
#print term, term_id, self.concept_by_id[a_id], w, self.vector_by_category[a_id][w], '\t+%f' % weight
self.vector_by_category[a_id][w] += weight
self.contributors_by_category[a_id].update([term_id])
if __name__ == '__main__':
import random
from random import shuffle as randomize
tfidf = TfIdfMatrix()
tfidf.load_features('bkp.big.out/vector.term.csv')
tfidf.load_distribution('bkp.big.out/vector.index.csv')
# tfidf.load_features('vector.term.csv')
# tfidf.load_distribution('vector.index.csv')
ctree = CategoryTree(
'bkp.big.out/category.index.csv',
'bkp.big.out/term.csv',
'bkp.big.out/category.csv',
# 'category.index.csv',
# 'term.csv',
# 'category.csv',
tfidf,
max_depth=1
)
ctree.load()
ctree.build()
ctree.dump()
<｜fim▁end｜> | for concept, _id in from_csv(self.path_categories):
_id = int(_id)
self.id_by_concept[concept] = _id
self.concept_by_id[_id] = concept
self.observed_category[_id] = True
self.observed_category[concept] = True |
<|file_name|>CategoryTree.py<|end_file_name|><｜fim▁begin｜>import json
from collections import (
Counter,
defaultdict as deft
)
from copy import deepcopy as cp
# from cPickle import (
# dump as to_pickle,
# load as from_pickle
# )
from StringIO import StringIO
from TfIdfMatrix import TfIdfMatrix
from Tools import from_csv
class CategoryTree:
def __init__(self, categories_by_concept, terms,
categories, tfidf, max_depth=5, min_df=20
):
self.min_df = min_df
self.path_categories_by_concept = categories_by_concept
self.path_categories = categories
self.path_terms = terms
self.max_depth = max_depth
self.observed_category = deft(bool)
self.id_by_concept = dict([])
self.concept_by_id = dict([])
self.term_is_category = deft(bool)
self.parents_by_category = dict([])
self.parents_by_concept = deft(list)
self.id_by_term = dict([])
self.term_by_id = dict([])
self.has_parents = deft(bool)
self.tfidf = tfidf
self.pulling = set([])
self.vector_by_category = deft(Counter)
self.contributors_by_category = deft(set)
self.projected = Counter()
def build(self):
for i, c in enumerate(self.concept_by_id.values()):
self(c)
if not i % 100:
t = float(len(self.concept_by_id.keys()))
print i, int(t), round(i / t, 2)
# if i >= 5000:
# break
def dump(self):
# Simulate a file with StringIO
out = open('vector.dump.txt', 'wb')
for i, (_id, projections) in enumerate(self.projected.items()):
if not i % 100:
print i, len(self.projected.keys())
if not projections:
continue
features = [
(self.tfidf.word_by_id[wid], round(weight, 4))
for wid, weight in self.vector_by_category[_id].most_common()
if round(weight, 4)
]
record = (
_id,
self.concept_by_id[_id],
features
)
out.write('%s\n' % str(record))
out.close()
def __call__(self, category):
self.pulling = set([])
return self.__pull(None, 0, category, dict([]))
def __get_parents(self, _id):
parents = []
name = self.concept_by_id[_id]
if (
not self.observed_category[name] or
not self.observed_category[_id] or
not self.has_parents[_id]
):
return []
else:
for i in self.parents_by_category[_id]:
if not self.observed_category[i]:
continue
_name = self.concept_by_id[i]
parents.append(_name)
return set(parents) - self.pulling
def __pull(self, vector, depth, category, tree):
_id = self.id_by_concept[category]
if not self.pulling:
# print
# print
# print category, _id
# print [self.term_by_id[x] for x in self.contributors_by_category[_id]]
# print self.vector_by_category[_id].most_common(20)
vector = self.vector_by_category[_id]
if not self.observed_category[category]:
return dict([])
parents = self.__get_parents(_id)
if not parents or depth >= self.max_depth:
tree[category] = dict([])
else:
subtree = dict([])
self.pulling.update(parents)
for parent in parents:
subtree = self.__pull(vector, depth + 1, parent, subtree)
tree[category] = subtree
self.__project(vector, tree)
return tree
def __project(self, vector, tree):
if not tree.keys():
return
else:
for key, subtree in tree.items():
_id = self.id_by_concept[key]
self.projected[_id] += 1
self.__add2vec(vector, _id)
self.__project(vector, subtree)
def __add2vec(self, vector, _id):
# for w, weight in vector.items():
# __id = self.tfidf.id_by_word[w]
for __id, weight in vector.items():
self.vector_by_category[_id][__id] += weight
def load(self):
self.__load_terms()
self.__load_categories()
self.__load_assignments()
def __load_categories(self):
for concept, _id in from_csv(self.path_categories):
_id = int(_id)
self.id_by_concept[concept] = _id
self.concept_by_id[_id] = concept
self.observed_category[_id] = True
self.observed_category[concept] = True
# print concept, _id, len(self.id_by_concept.keys())
# exit()
def __load_terms(self):
<|fim_middle|>
def __load_assignments(self):
for row in from_csv(self.path_categories_by_concept):
ints = [int(field) for field in row]
term_id = ints[0]
term = self.term_by_id[term_id]
if self.term_is_category[term_id] and \
self.observed_category[term]:
term = self.term_by_id[term_id]
cat_id = self.id_by_concept[term]
assignments = [i for i in ints[1:] if self.observed_category[i]]
self.parents_by_category[cat_id] = assignments
self.has_parents[cat_id] = True
else:
vector = self.tfidf.content(term_id)
assignments = [i for i in ints[1:] if self.observed_category[i]]
self.parents_by_concept[term_id] = assignments
for a_id in assignments:
for w, weight in vector:
if self.tfidf.df[w] < self.min_df:
continue
#print term, term_id, self.concept_by_id[a_id], w, self.vector_by_category[a_id][w], '\t+%f' % weight
self.vector_by_category[a_id][w] += weight
self.contributors_by_category[a_id].update([term_id])
if __name__ == '__main__':
import random
from random import shuffle as randomize
tfidf = TfIdfMatrix()
tfidf.load_features('bkp.big.out/vector.term.csv')
tfidf.load_distribution('bkp.big.out/vector.index.csv')
# tfidf.load_features('vector.term.csv')
# tfidf.load_distribution('vector.index.csv')
ctree = CategoryTree(
'bkp.big.out/category.index.csv',
'bkp.big.out/term.csv',
'bkp.big.out/category.csv',
# 'category.index.csv',
# 'term.csv',
# 'category.csv',
tfidf,
max_depth=1
)
ctree.load()
ctree.build()
ctree.dump()
<｜fim▁end｜> | for term, _id in from_csv(self.path_terms):
_id = int(_id)
self.term_by_id[_id] = term
self.id_by_term[term] = _id
if not term.startswith('Category:'):
continue
self.term_is_category[term] = True
self.term_is_category[_id] = True |
<|file_name|>CategoryTree.py<|end_file_name|><｜fim▁begin｜>import json
from collections import (
Counter,
defaultdict as deft
)
from copy import deepcopy as cp
# from cPickle import (
# dump as to_pickle,
# load as from_pickle
# )
from StringIO import StringIO
from TfIdfMatrix import TfIdfMatrix
from Tools import from_csv
class CategoryTree:
def __init__(self, categories_by_concept, terms,
categories, tfidf, max_depth=5, min_df=20
):
self.min_df = min_df
self.path_categories_by_concept = categories_by_concept
self.path_categories = categories
self.path_terms = terms
self.max_depth = max_depth
self.observed_category = deft(bool)
self.id_by_concept = dict([])
self.concept_by_id = dict([])
self.term_is_category = deft(bool)
self.parents_by_category = dict([])
self.parents_by_concept = deft(list)
self.id_by_term = dict([])
self.term_by_id = dict([])
self.has_parents = deft(bool)
self.tfidf = tfidf
self.pulling = set([])
self.vector_by_category = deft(Counter)
self.contributors_by_category = deft(set)
self.projected = Counter()
def build(self):
for i, c in enumerate(self.concept_by_id.values()):
self(c)
if not i % 100:
t = float(len(self.concept_by_id.keys()))
print i, int(t), round(i / t, 2)
# if i >= 5000:
# break
def dump(self):
# Simulate a file with StringIO
out = open('vector.dump.txt', 'wb')
for i, (_id, projections) in enumerate(self.projected.items()):
if not i % 100:
print i, len(self.projected.keys())
if not projections:
continue
features = [
(self.tfidf.word_by_id[wid], round(weight, 4))
for wid, weight in self.vector_by_category[_id].most_common()
if round(weight, 4)
]
record = (
_id,
self.concept_by_id[_id],
features
)
out.write('%s\n' % str(record))
out.close()
def __call__(self, category):
self.pulling = set([])
return self.__pull(None, 0, category, dict([]))
def __get_parents(self, _id):
parents = []
name = self.concept_by_id[_id]
if (
not self.observed_category[name] or
not self.observed_category[_id] or
not self.has_parents[_id]
):
return []
else:
for i in self.parents_by_category[_id]:
if not self.observed_category[i]:
continue
_name = self.concept_by_id[i]
parents.append(_name)
return set(parents) - self.pulling
def __pull(self, vector, depth, category, tree):
_id = self.id_by_concept[category]
if not self.pulling:
# print
# print
# print category, _id
# print [self.term_by_id[x] for x in self.contributors_by_category[_id]]
# print self.vector_by_category[_id].most_common(20)
vector = self.vector_by_category[_id]
if not self.observed_category[category]:
return dict([])
parents = self.__get_parents(_id)
if not parents or depth >= self.max_depth:
tree[category] = dict([])
else:
subtree = dict([])
self.pulling.update(parents)
for parent in parents:
subtree = self.__pull(vector, depth + 1, parent, subtree)
tree[category] = subtree
self.__project(vector, tree)
return tree
def __project(self, vector, tree):
if not tree.keys():
return
else:
for key, subtree in tree.items():
_id = self.id_by_concept[key]
self.projected[_id] += 1
self.__add2vec(vector, _id)
self.__project(vector, subtree)
def __add2vec(self, vector, _id):
# for w, weight in vector.items():
# __id = self.tfidf.id_by_word[w]
for __id, weight in vector.items():
self.vector_by_category[_id][__id] += weight
def load(self):
self.__load_terms()
self.__load_categories()
self.__load_assignments()
def __load_categories(self):
for concept, _id in from_csv(self.path_categories):
_id = int(_id)
self.id_by_concept[concept] = _id
self.concept_by_id[_id] = concept
self.observed_category[_id] = True
self.observed_category[concept] = True
# print concept, _id, len(self.id_by_concept.keys())
# exit()
def __load_terms(self):
for term, _id in from_csv(self.path_terms):
_id = int(_id)
self.term_by_id[_id] = term
self.id_by_term[term] = _id
if not term.startswith('Category:'):
continue
self.term_is_category[term] = True
self.term_is_category[_id] = True
def __load_assignments(self):
<|fim_middle|>
if __name__ == '__main__':
import random
from random import shuffle as randomize
tfidf = TfIdfMatrix()
tfidf.load_features('bkp.big.out/vector.term.csv')
tfidf.load_distribution('bkp.big.out/vector.index.csv')
# tfidf.load_features('vector.term.csv')
# tfidf.load_distribution('vector.index.csv')
ctree = CategoryTree(
'bkp.big.out/category.index.csv',
'bkp.big.out/term.csv',
'bkp.big.out/category.csv',
# 'category.index.csv',
# 'term.csv',
# 'category.csv',
tfidf,
max_depth=1
)
ctree.load()
ctree.build()
ctree.dump()
<｜fim▁end｜> | for row in from_csv(self.path_categories_by_concept):
ints = [int(field) for field in row]
term_id = ints[0]
term = self.term_by_id[term_id]
if self.term_is_category[term_id] and \
self.observed_category[term]:
term = self.term_by_id[term_id]
cat_id = self.id_by_concept[term]
assignments = [i for i in ints[1:] if self.observed_category[i]]
self.parents_by_category[cat_id] = assignments
self.has_parents[cat_id] = True
else:
vector = self.tfidf.content(term_id)
assignments = [i for i in ints[1:] if self.observed_category[i]]
self.parents_by_concept[term_id] = assignments
for a_id in assignments:
for w, weight in vector:
if self.tfidf.df[w] < self.min_df:
continue
#print term, term_id, self.concept_by_id[a_id], w, self.vector_by_category[a_id][w], '\t+%f' % weight
self.vector_by_category[a_id][w] += weight
self.contributors_by_category[a_id].update([term_id]) |
<|file_name|>CategoryTree.py<|end_file_name|><｜fim▁begin｜>import json
from collections import (
Counter,
defaultdict as deft
)
from copy import deepcopy as cp
# from cPickle import (
# dump as to_pickle,
# load as from_pickle
# )
from StringIO import StringIO
from TfIdfMatrix import TfIdfMatrix
from Tools import from_csv
class CategoryTree:
def __init__(self, categories_by_concept, terms,
categories, tfidf, max_depth=5, min_df=20
):
self.min_df = min_df
self.path_categories_by_concept = categories_by_concept
self.path_categories = categories
self.path_terms = terms
self.max_depth = max_depth
self.observed_category = deft(bool)
self.id_by_concept = dict([])
self.concept_by_id = dict([])
self.term_is_category = deft(bool)
self.parents_by_category = dict([])
self.parents_by_concept = deft(list)
self.id_by_term = dict([])
self.term_by_id = dict([])
self.has_parents = deft(bool)
self.tfidf = tfidf
self.pulling = set([])
self.vector_by_category = deft(Counter)
self.contributors_by_category = deft(set)
self.projected = Counter()
def build(self):
for i, c in enumerate(self.concept_by_id.values()):
self(c)
if not i % 100:
<|fim_middle|>
# if i >= 5000:
# break
def dump(self):
# Simulate a file with StringIO
out = open('vector.dump.txt', 'wb')
for i, (_id, projections) in enumerate(self.projected.items()):
if not i % 100:
print i, len(self.projected.keys())
if not projections:
continue
features = [
(self.tfidf.word_by_id[wid], round(weight, 4))
for wid, weight in self.vector_by_category[_id].most_common()
if round(weight, 4)
]
record = (
_id,
self.concept_by_id[_id],
features
)
out.write('%s\n' % str(record))
out.close()
def __call__(self, category):
self.pulling = set([])
return self.__pull(None, 0, category, dict([]))
def __get_parents(self, _id):
parents = []
name = self.concept_by_id[_id]
if (
not self.observed_category[name] or
not self.observed_category[_id] or
not self.has_parents[_id]
):
return []
else:
for i in self.parents_by_category[_id]:
if not self.observed_category[i]:
continue
_name = self.concept_by_id[i]
parents.append(_name)
return set(parents) - self.pulling
def __pull(self, vector, depth, category, tree):
_id = self.id_by_concept[category]
if not self.pulling:
# print
# print
# print category, _id
# print [self.term_by_id[x] for x in self.contributors_by_category[_id]]
# print self.vector_by_category[_id].most_common(20)
vector = self.vector_by_category[_id]
if not self.observed_category[category]:
return dict([])
parents = self.__get_parents(_id)
if not parents or depth >= self.max_depth:
tree[category] = dict([])
else:
subtree = dict([])
self.pulling.update(parents)
for parent in parents:
subtree = self.__pull(vector, depth + 1, parent, subtree)
tree[category] = subtree
self.__project(vector, tree)
return tree
def __project(self, vector, tree):
if not tree.keys():
return
else:
for key, subtree in tree.items():
_id = self.id_by_concept[key]
self.projected[_id] += 1
self.__add2vec(vector, _id)
self.__project(vector, subtree)
def __add2vec(self, vector, _id):
# for w, weight in vector.items():
# __id = self.tfidf.id_by_word[w]
for __id, weight in vector.items():
self.vector_by_category[_id][__id] += weight
def load(self):
self.__load_terms()
self.__load_categories()
self.__load_assignments()
def __load_categories(self):
for concept, _id in from_csv(self.path_categories):
_id = int(_id)
self.id_by_concept[concept] = _id
self.concept_by_id[_id] = concept
self.observed_category[_id] = True
self.observed_category[concept] = True
# print concept, _id, len(self.id_by_concept.keys())
# exit()
def __load_terms(self):
for term, _id in from_csv(self.path_terms):
_id = int(_id)
self.term_by_id[_id] = term
self.id_by_term[term] = _id
if not term.startswith('Category:'):
continue
self.term_is_category[term] = True
self.term_is_category[_id] = True
def __load_assignments(self):
for row in from_csv(self.path_categories_by_concept):
ints = [int(field) for field in row]
term_id = ints[0]
term = self.term_by_id[term_id]
if self.term_is_category[term_id] and \
self.observed_category[term]:
term = self.term_by_id[term_id]
cat_id = self.id_by_concept[term]
assignments = [i for i in ints[1:] if self.observed_category[i]]
self.parents_by_category[cat_id] = assignments
self.has_parents[cat_id] = True
else:
vector = self.tfidf.content(term_id)
assignments = [i for i in ints[1:] if self.observed_category[i]]
self.parents_by_concept[term_id] = assignments
for a_id in assignments:
for w, weight in vector:
if self.tfidf.df[w] < self.min_df:
continue
#print term, term_id, self.concept_by_id[a_id], w, self.vector_by_category[a_id][w], '\t+%f' % weight
self.vector_by_category[a_id][w] += weight
self.contributors_by_category[a_id].update([term_id])
if __name__ == '__main__':
import random
from random import shuffle as randomize
tfidf = TfIdfMatrix()
tfidf.load_features('bkp.big.out/vector.term.csv')
tfidf.load_distribution('bkp.big.out/vector.index.csv')
# tfidf.load_features('vector.term.csv')
# tfidf.load_distribution('vector.index.csv')
ctree = CategoryTree(
'bkp.big.out/category.index.csv',
'bkp.big.out/term.csv',
'bkp.big.out/category.csv',
# 'category.index.csv',
# 'term.csv',
# 'category.csv',
tfidf,
max_depth=1
)
ctree.load()
ctree.build()
ctree.dump()
<｜fim▁end｜> | t = float(len(self.concept_by_id.keys()))
print i, int(t), round(i / t, 2) |
<|file_name|>CategoryTree.py<|end_file_name|><｜fim▁begin｜>import json
from collections import (
Counter,
defaultdict as deft
)
from copy import deepcopy as cp
# from cPickle import (
# dump as to_pickle,
# load as from_pickle
# )
from StringIO import StringIO
from TfIdfMatrix import TfIdfMatrix
from Tools import from_csv
class CategoryTree:
def __init__(self, categories_by_concept, terms,
categories, tfidf, max_depth=5, min_df=20
):
self.min_df = min_df
self.path_categories_by_concept = categories_by_concept
self.path_categories = categories
self.path_terms = terms
self.max_depth = max_depth
self.observed_category = deft(bool)
self.id_by_concept = dict([])
self.concept_by_id = dict([])
self.term_is_category = deft(bool)
self.parents_by_category = dict([])
self.parents_by_concept = deft(list)
self.id_by_term = dict([])
self.term_by_id = dict([])
self.has_parents = deft(bool)
self.tfidf = tfidf
self.pulling = set([])
self.vector_by_category = deft(Counter)
self.contributors_by_category = deft(set)
self.projected = Counter()
def build(self):
for i, c in enumerate(self.concept_by_id.values()):
self(c)
if not i % 100:
t = float(len(self.concept_by_id.keys()))
print i, int(t), round(i / t, 2)
# if i >= 5000:
# break
def dump(self):
# Simulate a file with StringIO
out = open('vector.dump.txt', 'wb')
for i, (_id, projections) in enumerate(self.projected.items()):
if not i % 100:
<|fim_middle|>
if not projections:
continue
features = [
(self.tfidf.word_by_id[wid], round(weight, 4))
for wid, weight in self.vector_by_category[_id].most_common()
if round(weight, 4)
]
record = (
_id,
self.concept_by_id[_id],
features
)
out.write('%s\n' % str(record))
out.close()
def __call__(self, category):
self.pulling = set([])
return self.__pull(None, 0, category, dict([]))
def __get_parents(self, _id):
parents = []
name = self.concept_by_id[_id]
if (
not self.observed_category[name] or
not self.observed_category[_id] or
not self.has_parents[_id]
):
return []
else:
for i in self.parents_by_category[_id]:
if not self.observed_category[i]:
continue
_name = self.concept_by_id[i]
parents.append(_name)
return set(parents) - self.pulling
def __pull(self, vector, depth, category, tree):
_id = self.id_by_concept[category]
if not self.pulling:
# print
# print
# print category, _id
# print [self.term_by_id[x] for x in self.contributors_by_category[_id]]
# print self.vector_by_category[_id].most_common(20)
vector = self.vector_by_category[_id]
if not self.observed_category[category]:
return dict([])
parents = self.__get_parents(_id)
if not parents or depth >= self.max_depth:
tree[category] = dict([])
else:
subtree = dict([])
self.pulling.update(parents)
for parent in parents:
subtree = self.__pull(vector, depth + 1, parent, subtree)
tree[category] = subtree
self.__project(vector, tree)
return tree
def __project(self, vector, tree):
if not tree.keys():
return
else:
for key, subtree in tree.items():
_id = self.id_by_concept[key]
self.projected[_id] += 1
self.__add2vec(vector, _id)
self.__project(vector, subtree)
def __add2vec(self, vector, _id):
# for w, weight in vector.items():
# __id = self.tfidf.id_by_word[w]
for __id, weight in vector.items():
self.vector_by_category[_id][__id] += weight
def load(self):
self.__load_terms()
self.__load_categories()
self.__load_assignments()
def __load_categories(self):
for concept, _id in from_csv(self.path_categories):
_id = int(_id)
self.id_by_concept[concept] = _id
self.concept_by_id[_id] = concept
self.observed_category[_id] = True
self.observed_category[concept] = True
# print concept, _id, len(self.id_by_concept.keys())
# exit()
def __load_terms(self):
for term, _id in from_csv(self.path_terms):
_id = int(_id)
self.term_by_id[_id] = term
self.id_by_term[term] = _id
if not term.startswith('Category:'):
continue
self.term_is_category[term] = True
self.term_is_category[_id] = True
def __load_assignments(self):
for row in from_csv(self.path_categories_by_concept):
ints = [int(field) for field in row]
term_id = ints[0]
term = self.term_by_id[term_id]
if self.term_is_category[term_id] and \
self.observed_category[term]:
term = self.term_by_id[term_id]
cat_id = self.id_by_concept[term]
assignments = [i for i in ints[1:] if self.observed_category[i]]
self.parents_by_category[cat_id] = assignments
self.has_parents[cat_id] = True
else:
vector = self.tfidf.content(term_id)
assignments = [i for i in ints[1:] if self.observed_category[i]]
self.parents_by_concept[term_id] = assignments
for a_id in assignments:
for w, weight in vector:
if self.tfidf.df[w] < self.min_df:
continue
#print term, term_id, self.concept_by_id[a_id], w, self.vector_by_category[a_id][w], '\t+%f' % weight
self.vector_by_category[a_id][w] += weight
self.contributors_by_category[a_id].update([term_id])
if __name__ == '__main__':
import random
from random import shuffle as randomize
tfidf = TfIdfMatrix()
tfidf.load_features('bkp.big.out/vector.term.csv')
tfidf.load_distribution('bkp.big.out/vector.index.csv')
# tfidf.load_features('vector.term.csv')
# tfidf.load_distribution('vector.index.csv')
ctree = CategoryTree(
'bkp.big.out/category.index.csv',
'bkp.big.out/term.csv',
'bkp.big.out/category.csv',
# 'category.index.csv',
# 'term.csv',
# 'category.csv',
tfidf,
max_depth=1
)
ctree.load()
ctree.build()
ctree.dump()
<｜fim▁end｜> | print i, len(self.projected.keys()) |
<|file_name|>CategoryTree.py<|end_file_name|><｜fim▁begin｜>import json
from collections import (
Counter,
defaultdict as deft
)
from copy import deepcopy as cp
# from cPickle import (
# dump as to_pickle,
# load as from_pickle
# )
from StringIO import StringIO
from TfIdfMatrix import TfIdfMatrix
from Tools import from_csv
class CategoryTree:
def __init__(self, categories_by_concept, terms,
categories, tfidf, max_depth=5, min_df=20
):
self.min_df = min_df
self.path_categories_by_concept = categories_by_concept
self.path_categories = categories
self.path_terms = terms
self.max_depth = max_depth
self.observed_category = deft(bool)
self.id_by_concept = dict([])
self.concept_by_id = dict([])
self.term_is_category = deft(bool)
self.parents_by_category = dict([])
self.parents_by_concept = deft(list)
self.id_by_term = dict([])
self.term_by_id = dict([])
self.has_parents = deft(bool)
self.tfidf = tfidf
self.pulling = set([])
self.vector_by_category = deft(Counter)
self.contributors_by_category = deft(set)
self.projected = Counter()
def build(self):
for i, c in enumerate(self.concept_by_id.values()):
self(c)
if not i % 100:
t = float(len(self.concept_by_id.keys()))
print i, int(t), round(i / t, 2)
# if i >= 5000:
# break
def dump(self):
# Simulate a file with StringIO
out = open('vector.dump.txt', 'wb')
for i, (_id, projections) in enumerate(self.projected.items()):
if not i % 100:
print i, len(self.projected.keys())
if not projections:
<|fim_middle|>
features = [
(self.tfidf.word_by_id[wid], round(weight, 4))
for wid, weight in self.vector_by_category[_id].most_common()
if round(weight, 4)
]
record = (
_id,
self.concept_by_id[_id],
features
)
out.write('%s\n' % str(record))
out.close()
def __call__(self, category):
self.pulling = set([])
return self.__pull(None, 0, category, dict([]))
def __get_parents(self, _id):
parents = []
name = self.concept_by_id[_id]
if (
not self.observed_category[name] or
not self.observed_category[_id] or
not self.has_parents[_id]
):
return []
else:
for i in self.parents_by_category[_id]:
if not self.observed_category[i]:
continue
_name = self.concept_by_id[i]
parents.append(_name)
return set(parents) - self.pulling
def __pull(self, vector, depth, category, tree):
_id = self.id_by_concept[category]
if not self.pulling:
# print
# print
# print category, _id
# print [self.term_by_id[x] for x in self.contributors_by_category[_id]]
# print self.vector_by_category[_id].most_common(20)
vector = self.vector_by_category[_id]
if not self.observed_category[category]:
return dict([])
parents = self.__get_parents(_id)
if not parents or depth >= self.max_depth:
tree[category] = dict([])
else:
subtree = dict([])
self.pulling.update(parents)
for parent in parents:
subtree = self.__pull(vector, depth + 1, parent, subtree)
tree[category] = subtree
self.__project(vector, tree)
return tree
def __project(self, vector, tree):
if not tree.keys():
return
else:
for key, subtree in tree.items():
_id = self.id_by_concept[key]
self.projected[_id] += 1
self.__add2vec(vector, _id)
self.__project(vector, subtree)
def __add2vec(self, vector, _id):
# for w, weight in vector.items():
# __id = self.tfidf.id_by_word[w]
for __id, weight in vector.items():
self.vector_by_category[_id][__id] += weight
def load(self):
self.__load_terms()
self.__load_categories()
self.__load_assignments()
def __load_categories(self):
for concept, _id in from_csv(self.path_categories):
_id = int(_id)
self.id_by_concept[concept] = _id
self.concept_by_id[_id] = concept
self.observed_category[_id] = True
self.observed_category[concept] = True
# print concept, _id, len(self.id_by_concept.keys())
# exit()
def __load_terms(self):
for term, _id in from_csv(self.path_terms):
_id = int(_id)
self.term_by_id[_id] = term
self.id_by_term[term] = _id
if not term.startswith('Category:'):
continue
self.term_is_category[term] = True
self.term_is_category[_id] = True
def __load_assignments(self):
for row in from_csv(self.path_categories_by_concept):
ints = [int(field) for field in row]
term_id = ints[0]
term = self.term_by_id[term_id]
if self.term_is_category[term_id] and \
self.observed_category[term]:
term = self.term_by_id[term_id]
cat_id = self.id_by_concept[term]
assignments = [i for i in ints[1:] if self.observed_category[i]]
self.parents_by_category[cat_id] = assignments
self.has_parents[cat_id] = True
else:
vector = self.tfidf.content(term_id)
assignments = [i for i in ints[1:] if self.observed_category[i]]
self.parents_by_concept[term_id] = assignments
for a_id in assignments:
for w, weight in vector:
if self.tfidf.df[w] < self.min_df:
continue
#print term, term_id, self.concept_by_id[a_id], w, self.vector_by_category[a_id][w], '\t+%f' % weight
self.vector_by_category[a_id][w] += weight
self.contributors_by_category[a_id].update([term_id])
if __name__ == '__main__':
import random
from random import shuffle as randomize
tfidf = TfIdfMatrix()
tfidf.load_features('bkp.big.out/vector.term.csv')
tfidf.load_distribution('bkp.big.out/vector.index.csv')
# tfidf.load_features('vector.term.csv')
# tfidf.load_distribution('vector.index.csv')
ctree = CategoryTree(
'bkp.big.out/category.index.csv',
'bkp.big.out/term.csv',
'bkp.big.out/category.csv',
# 'category.index.csv',
# 'term.csv',
# 'category.csv',
tfidf,
max_depth=1
)
ctree.load()
ctree.build()
ctree.dump()
<｜fim▁end｜> | continue |
<|file_name|>CategoryTree.py<|end_file_name|><｜fim▁begin｜>import json
from collections import (
Counter,
defaultdict as deft
)
from copy import deepcopy as cp
# from cPickle import (
# dump as to_pickle,
# load as from_pickle
# )
from StringIO import StringIO
from TfIdfMatrix import TfIdfMatrix
from Tools import from_csv
class CategoryTree:
def __init__(self, categories_by_concept, terms,
categories, tfidf, max_depth=5, min_df=20
):
self.min_df = min_df
self.path_categories_by_concept = categories_by_concept
self.path_categories = categories
self.path_terms = terms
self.max_depth = max_depth
self.observed_category = deft(bool)
self.id_by_concept = dict([])
self.concept_by_id = dict([])
self.term_is_category = deft(bool)
self.parents_by_category = dict([])
self.parents_by_concept = deft(list)
self.id_by_term = dict([])
self.term_by_id = dict([])
self.has_parents = deft(bool)
self.tfidf = tfidf
self.pulling = set([])
self.vector_by_category = deft(Counter)
self.contributors_by_category = deft(set)
self.projected = Counter()
def build(self):
for i, c in enumerate(self.concept_by_id.values()):
self(c)
if not i % 100:
t = float(len(self.concept_by_id.keys()))
print i, int(t), round(i / t, 2)
# if i >= 5000:
# break
def dump(self):
# Simulate a file with StringIO
out = open('vector.dump.txt', 'wb')
for i, (_id, projections) in enumerate(self.projected.items()):
if not i % 100:
print i, len(self.projected.keys())
if not projections:
continue
features = [
(self.tfidf.word_by_id[wid], round(weight, 4))
for wid, weight in self.vector_by_category[_id].most_common()
if round(weight, 4)
]
record = (
_id,
self.concept_by_id[_id],
features
)
out.write('%s\n' % str(record))
out.close()
def __call__(self, category):
self.pulling = set([])
return self.__pull(None, 0, category, dict([]))
def __get_parents(self, _id):
parents = []
name = self.concept_by_id[_id]
if (
not self.observed_category[name] or
not self.observed_category[_id] or
not self.has_parents[_id]
):
<|fim_middle|>
else:
for i in self.parents_by_category[_id]:
if not self.observed_category[i]:
continue
_name = self.concept_by_id[i]
parents.append(_name)
return set(parents) - self.pulling
def __pull(self, vector, depth, category, tree):
_id = self.id_by_concept[category]
if not self.pulling:
# print
# print
# print category, _id
# print [self.term_by_id[x] for x in self.contributors_by_category[_id]]
# print self.vector_by_category[_id].most_common(20)
vector = self.vector_by_category[_id]
if not self.observed_category[category]:
return dict([])
parents = self.__get_parents(_id)
if not parents or depth >= self.max_depth:
tree[category] = dict([])
else:
subtree = dict([])
self.pulling.update(parents)
for parent in parents:
subtree = self.__pull(vector, depth + 1, parent, subtree)
tree[category] = subtree
self.__project(vector, tree)
return tree
def __project(self, vector, tree):
if not tree.keys():
return
else:
for key, subtree in tree.items():
_id = self.id_by_concept[key]
self.projected[_id] += 1
self.__add2vec(vector, _id)
self.__project(vector, subtree)
def __add2vec(self, vector, _id):
# for w, weight in vector.items():
# __id = self.tfidf.id_by_word[w]
for __id, weight in vector.items():
self.vector_by_category[_id][__id] += weight
def load(self):
self.__load_terms()
self.__load_categories()
self.__load_assignments()
def __load_categories(self):
for concept, _id in from_csv(self.path_categories):
_id = int(_id)
self.id_by_concept[concept] = _id
self.concept_by_id[_id] = concept
self.observed_category[_id] = True
self.observed_category[concept] = True
# print concept, _id, len(self.id_by_concept.keys())
# exit()
def __load_terms(self):
for term, _id in from_csv(self.path_terms):
_id = int(_id)
self.term_by_id[_id] = term
self.id_by_term[term] = _id
if not term.startswith('Category:'):
continue
self.term_is_category[term] = True
self.term_is_category[_id] = True
def __load_assignments(self):
for row in from_csv(self.path_categories_by_concept):
ints = [int(field) for field in row]
term_id = ints[0]
term = self.term_by_id[term_id]
if self.term_is_category[term_id] and \
self.observed_category[term]:
term = self.term_by_id[term_id]
cat_id = self.id_by_concept[term]
assignments = [i for i in ints[1:] if self.observed_category[i]]
self.parents_by_category[cat_id] = assignments
self.has_parents[cat_id] = True
else:
vector = self.tfidf.content(term_id)
assignments = [i for i in ints[1:] if self.observed_category[i]]
self.parents_by_concept[term_id] = assignments
for a_id in assignments:
for w, weight in vector:
if self.tfidf.df[w] < self.min_df:
continue
#print term, term_id, self.concept_by_id[a_id], w, self.vector_by_category[a_id][w], '\t+%f' % weight
self.vector_by_category[a_id][w] += weight
self.contributors_by_category[a_id].update([term_id])
if __name__ == '__main__':
import random
from random import shuffle as randomize
tfidf = TfIdfMatrix()
tfidf.load_features('bkp.big.out/vector.term.csv')
tfidf.load_distribution('bkp.big.out/vector.index.csv')
# tfidf.load_features('vector.term.csv')
# tfidf.load_distribution('vector.index.csv')
ctree = CategoryTree(
'bkp.big.out/category.index.csv',
'bkp.big.out/term.csv',
'bkp.big.out/category.csv',
# 'category.index.csv',
# 'term.csv',
# 'category.csv',
tfidf,
max_depth=1
)
ctree.load()
ctree.build()
ctree.dump()
<｜fim▁end｜> | return [] |
<|file_name|>CategoryTree.py<|end_file_name|><｜fim▁begin｜>import json
from collections import (
Counter,
defaultdict as deft
)
from copy import deepcopy as cp
# from cPickle import (
# dump as to_pickle,
# load as from_pickle
# )
from StringIO import StringIO
from TfIdfMatrix import TfIdfMatrix
from Tools import from_csv
class CategoryTree:
def __init__(self, categories_by_concept, terms,
categories, tfidf, max_depth=5, min_df=20
):
self.min_df = min_df
self.path_categories_by_concept = categories_by_concept
self.path_categories = categories
self.path_terms = terms
self.max_depth = max_depth
self.observed_category = deft(bool)
self.id_by_concept = dict([])
self.concept_by_id = dict([])
self.term_is_category = deft(bool)
self.parents_by_category = dict([])
self.parents_by_concept = deft(list)
self.id_by_term = dict([])
self.term_by_id = dict([])
self.has_parents = deft(bool)
self.tfidf = tfidf
self.pulling = set([])
self.vector_by_category = deft(Counter)
self.contributors_by_category = deft(set)
self.projected = Counter()
def build(self):
for i, c in enumerate(self.concept_by_id.values()):
self(c)
if not i % 100:
t = float(len(self.concept_by_id.keys()))
print i, int(t), round(i / t, 2)
# if i >= 5000:
# break
def dump(self):
# Simulate a file with StringIO
out = open('vector.dump.txt', 'wb')
for i, (_id, projections) in enumerate(self.projected.items()):
if not i % 100:
print i, len(self.projected.keys())
if not projections:
continue
features = [
(self.tfidf.word_by_id[wid], round(weight, 4))
for wid, weight in self.vector_by_category[_id].most_common()
if round(weight, 4)
]
record = (
_id,
self.concept_by_id[_id],
features
)
out.write('%s\n' % str(record))
out.close()
def __call__(self, category):
self.pulling = set([])
return self.__pull(None, 0, category, dict([]))
def __get_parents(self, _id):
parents = []
name = self.concept_by_id[_id]
if (
not self.observed_category[name] or
not self.observed_category[_id] or
not self.has_parents[_id]
):
return []
else:
<|fim_middle|>
def __pull(self, vector, depth, category, tree):
_id = self.id_by_concept[category]
if not self.pulling:
# print
# print
# print category, _id
# print [self.term_by_id[x] for x in self.contributors_by_category[_id]]
# print self.vector_by_category[_id].most_common(20)
vector = self.vector_by_category[_id]
if not self.observed_category[category]:
return dict([])
parents = self.__get_parents(_id)
if not parents or depth >= self.max_depth:
tree[category] = dict([])
else:
subtree = dict([])
self.pulling.update(parents)
for parent in parents:
subtree = self.__pull(vector, depth + 1, parent, subtree)
tree[category] = subtree
self.__project(vector, tree)
return tree
def __project(self, vector, tree):
if not tree.keys():
return
else:
for key, subtree in tree.items():
_id = self.id_by_concept[key]
self.projected[_id] += 1
self.__add2vec(vector, _id)
self.__project(vector, subtree)
def __add2vec(self, vector, _id):
# for w, weight in vector.items():
# __id = self.tfidf.id_by_word[w]
for __id, weight in vector.items():
self.vector_by_category[_id][__id] += weight
def load(self):
self.__load_terms()
self.__load_categories()
self.__load_assignments()
def __load_categories(self):
for concept, _id in from_csv(self.path_categories):
_id = int(_id)
self.id_by_concept[concept] = _id
self.concept_by_id[_id] = concept
self.observed_category[_id] = True
self.observed_category[concept] = True
# print concept, _id, len(self.id_by_concept.keys())
# exit()
def __load_terms(self):
for term, _id in from_csv(self.path_terms):
_id = int(_id)
self.term_by_id[_id] = term
self.id_by_term[term] = _id
if not term.startswith('Category:'):
continue
self.term_is_category[term] = True
self.term_is_category[_id] = True
def __load_assignments(self):
for row in from_csv(self.path_categories_by_concept):
ints = [int(field) for field in row]
term_id = ints[0]
term = self.term_by_id[term_id]
if self.term_is_category[term_id] and \
self.observed_category[term]:
term = self.term_by_id[term_id]
cat_id = self.id_by_concept[term]
assignments = [i for i in ints[1:] if self.observed_category[i]]
self.parents_by_category[cat_id] = assignments
self.has_parents[cat_id] = True
else:
vector = self.tfidf.content(term_id)
assignments = [i for i in ints[1:] if self.observed_category[i]]
self.parents_by_concept[term_id] = assignments
for a_id in assignments:
for w, weight in vector:
if self.tfidf.df[w] < self.min_df:
continue
#print term, term_id, self.concept_by_id[a_id], w, self.vector_by_category[a_id][w], '\t+%f' % weight
self.vector_by_category[a_id][w] += weight
self.contributors_by_category[a_id].update([term_id])
if __name__ == '__main__':
import random
from random import shuffle as randomize
tfidf = TfIdfMatrix()
tfidf.load_features('bkp.big.out/vector.term.csv')
tfidf.load_distribution('bkp.big.out/vector.index.csv')
# tfidf.load_features('vector.term.csv')
# tfidf.load_distribution('vector.index.csv')
ctree = CategoryTree(
'bkp.big.out/category.index.csv',
'bkp.big.out/term.csv',
'bkp.big.out/category.csv',
# 'category.index.csv',
# 'term.csv',
# 'category.csv',
tfidf,
max_depth=1
)
ctree.load()
ctree.build()
ctree.dump()
<｜fim▁end｜> | for i in self.parents_by_category[_id]:
if not self.observed_category[i]:
continue
_name = self.concept_by_id[i]
parents.append(_name)
return set(parents) - self.pulling |
<|file_name|>CategoryTree.py<|end_file_name|><｜fim▁begin｜>import json
from collections import (
Counter,
defaultdict as deft
)
from copy import deepcopy as cp
# from cPickle import (
# dump as to_pickle,
# load as from_pickle
# )
from StringIO import StringIO
from TfIdfMatrix import TfIdfMatrix
from Tools import from_csv
class CategoryTree:
def __init__(self, categories_by_concept, terms,
categories, tfidf, max_depth=5, min_df=20
):
self.min_df = min_df
self.path_categories_by_concept = categories_by_concept
self.path_categories = categories
self.path_terms = terms
self.max_depth = max_depth
self.observed_category = deft(bool)
self.id_by_concept = dict([])
self.concept_by_id = dict([])
self.term_is_category = deft(bool)
self.parents_by_category = dict([])
self.parents_by_concept = deft(list)
self.id_by_term = dict([])
self.term_by_id = dict([])
self.has_parents = deft(bool)
self.tfidf = tfidf
self.pulling = set([])
self.vector_by_category = deft(Counter)
self.contributors_by_category = deft(set)
self.projected = Counter()
def build(self):
for i, c in enumerate(self.concept_by_id.values()):
self(c)
if not i % 100:
t = float(len(self.concept_by_id.keys()))
print i, int(t), round(i / t, 2)
# if i >= 5000:
# break
def dump(self):
# Simulate a file with StringIO
out = open('vector.dump.txt', 'wb')
for i, (_id, projections) in enumerate(self.projected.items()):
if not i % 100:
print i, len(self.projected.keys())
if not projections:
continue
features = [
(self.tfidf.word_by_id[wid], round(weight, 4))
for wid, weight in self.vector_by_category[_id].most_common()
if round(weight, 4)
]
record = (
_id,
self.concept_by_id[_id],
features
)
out.write('%s\n' % str(record))
out.close()
def __call__(self, category):
self.pulling = set([])
return self.__pull(None, 0, category, dict([]))
def __get_parents(self, _id):
parents = []
name = self.concept_by_id[_id]
if (
not self.observed_category[name] or
not self.observed_category[_id] or
not self.has_parents[_id]
):
return []
else:
for i in self.parents_by_category[_id]:
if not self.observed_category[i]:
<|fim_middle|>
_name = self.concept_by_id[i]
parents.append(_name)
return set(parents) - self.pulling
def __pull(self, vector, depth, category, tree):
_id = self.id_by_concept[category]
if not self.pulling:
# print
# print
# print category, _id
# print [self.term_by_id[x] for x in self.contributors_by_category[_id]]
# print self.vector_by_category[_id].most_common(20)
vector = self.vector_by_category[_id]
if not self.observed_category[category]:
return dict([])
parents = self.__get_parents(_id)
if not parents or depth >= self.max_depth:
tree[category] = dict([])
else:
subtree = dict([])
self.pulling.update(parents)
for parent in parents:
subtree = self.__pull(vector, depth + 1, parent, subtree)
tree[category] = subtree
self.__project(vector, tree)
return tree
def __project(self, vector, tree):
if not tree.keys():
return
else:
for key, subtree in tree.items():
_id = self.id_by_concept[key]
self.projected[_id] += 1
self.__add2vec(vector, _id)
self.__project(vector, subtree)
def __add2vec(self, vector, _id):
# for w, weight in vector.items():
# __id = self.tfidf.id_by_word[w]
for __id, weight in vector.items():
self.vector_by_category[_id][__id] += weight
def load(self):
self.__load_terms()
self.__load_categories()
self.__load_assignments()
def __load_categories(self):
for concept, _id in from_csv(self.path_categories):
_id = int(_id)
self.id_by_concept[concept] = _id
self.concept_by_id[_id] = concept
self.observed_category[_id] = True
self.observed_category[concept] = True
# print concept, _id, len(self.id_by_concept.keys())
# exit()
def __load_terms(self):
for term, _id in from_csv(self.path_terms):
_id = int(_id)
self.term_by_id[_id] = term
self.id_by_term[term] = _id
if not term.startswith('Category:'):
continue
self.term_is_category[term] = True
self.term_is_category[_id] = True
def __load_assignments(self):
for row in from_csv(self.path_categories_by_concept):
ints = [int(field) for field in row]
term_id = ints[0]
term = self.term_by_id[term_id]
if self.term_is_category[term_id] and \
self.observed_category[term]:
term = self.term_by_id[term_id]
cat_id = self.id_by_concept[term]
assignments = [i for i in ints[1:] if self.observed_category[i]]
self.parents_by_category[cat_id] = assignments
self.has_parents[cat_id] = True
else:
vector = self.tfidf.content(term_id)
assignments = [i for i in ints[1:] if self.observed_category[i]]
self.parents_by_concept[term_id] = assignments
for a_id in assignments:
for w, weight in vector:
if self.tfidf.df[w] < self.min_df:
continue
#print term, term_id, self.concept_by_id[a_id], w, self.vector_by_category[a_id][w], '\t+%f' % weight
self.vector_by_category[a_id][w] += weight
self.contributors_by_category[a_id].update([term_id])
if __name__ == '__main__':
import random
from random import shuffle as randomize
tfidf = TfIdfMatrix()
tfidf.load_features('bkp.big.out/vector.term.csv')
tfidf.load_distribution('bkp.big.out/vector.index.csv')
# tfidf.load_features('vector.term.csv')
# tfidf.load_distribution('vector.index.csv')
ctree = CategoryTree(
'bkp.big.out/category.index.csv',
'bkp.big.out/term.csv',
'bkp.big.out/category.csv',
# 'category.index.csv',
# 'term.csv',
# 'category.csv',
tfidf,
max_depth=1
)
ctree.load()
ctree.build()
ctree.dump()
<｜fim▁end｜> | continue |
<|file_name|>CategoryTree.py<|end_file_name|><｜fim▁begin｜>import json
from collections import (
Counter,
defaultdict as deft
)
from copy import deepcopy as cp
# from cPickle import (
# dump as to_pickle,
# load as from_pickle
# )
from StringIO import StringIO
from TfIdfMatrix import TfIdfMatrix
from Tools import from_csv
class CategoryTree:
def __init__(self, categories_by_concept, terms,
categories, tfidf, max_depth=5, min_df=20
):
self.min_df = min_df
self.path_categories_by_concept = categories_by_concept
self.path_categories = categories
self.path_terms = terms
self.max_depth = max_depth
self.observed_category = deft(bool)
self.id_by_concept = dict([])
self.concept_by_id = dict([])
self.term_is_category = deft(bool)
self.parents_by_category = dict([])
self.parents_by_concept = deft(list)
self.id_by_term = dict([])
self.term_by_id = dict([])
self.has_parents = deft(bool)
self.tfidf = tfidf
self.pulling = set([])
self.vector_by_category = deft(Counter)
self.contributors_by_category = deft(set)
self.projected = Counter()
def build(self):
for i, c in enumerate(self.concept_by_id.values()):
self(c)
if not i % 100:
t = float(len(self.concept_by_id.keys()))
print i, int(t), round(i / t, 2)
# if i >= 5000:
# break
def dump(self):
# Simulate a file with StringIO
out = open('vector.dump.txt', 'wb')
for i, (_id, projections) in enumerate(self.projected.items()):
if not i % 100:
print i, len(self.projected.keys())
if not projections:
continue
features = [
(self.tfidf.word_by_id[wid], round(weight, 4))
for wid, weight in self.vector_by_category[_id].most_common()
if round(weight, 4)
]
record = (
_id,
self.concept_by_id[_id],
features
)
out.write('%s\n' % str(record))
out.close()
def __call__(self, category):
self.pulling = set([])
return self.__pull(None, 0, category, dict([]))
def __get_parents(self, _id):
parents = []
name = self.concept_by_id[_id]
if (
not self.observed_category[name] or
not self.observed_category[_id] or
not self.has_parents[_id]
):
return []
else:
for i in self.parents_by_category[_id]:
if not self.observed_category[i]:
continue
_name = self.concept_by_id[i]
parents.append(_name)
return set(parents) - self.pulling
def __pull(self, vector, depth, category, tree):
_id = self.id_by_concept[category]
if not self.pulling:
# print
# print
# print category, _id
# print [self.term_by_id[x] for x in self.contributors_by_category[_id]]
# print self.vector_by_category[_id].most_common(20)
<|fim_middle|>
if not self.observed_category[category]:
return dict([])
parents = self.__get_parents(_id)
if not parents or depth >= self.max_depth:
tree[category] = dict([])
else:
subtree = dict([])
self.pulling.update(parents)
for parent in parents:
subtree = self.__pull(vector, depth + 1, parent, subtree)
tree[category] = subtree
self.__project(vector, tree)
return tree
def __project(self, vector, tree):
if not tree.keys():
return
else:
for key, subtree in tree.items():
_id = self.id_by_concept[key]
self.projected[_id] += 1
self.__add2vec(vector, _id)
self.__project(vector, subtree)
def __add2vec(self, vector, _id):
# for w, weight in vector.items():
# __id = self.tfidf.id_by_word[w]
for __id, weight in vector.items():
self.vector_by_category[_id][__id] += weight
def load(self):
self.__load_terms()
self.__load_categories()
self.__load_assignments()
def __load_categories(self):
for concept, _id in from_csv(self.path_categories):
_id = int(_id)
self.id_by_concept[concept] = _id
self.concept_by_id[_id] = concept
self.observed_category[_id] = True
self.observed_category[concept] = True
# print concept, _id, len(self.id_by_concept.keys())
# exit()
def __load_terms(self):
for term, _id in from_csv(self.path_terms):
_id = int(_id)
self.term_by_id[_id] = term
self.id_by_term[term] = _id
if not term.startswith('Category:'):
continue
self.term_is_category[term] = True
self.term_is_category[_id] = True
def __load_assignments(self):
for row in from_csv(self.path_categories_by_concept):
ints = [int(field) for field in row]
term_id = ints[0]
term = self.term_by_id[term_id]
if self.term_is_category[term_id] and \
self.observed_category[term]:
term = self.term_by_id[term_id]
cat_id = self.id_by_concept[term]
assignments = [i for i in ints[1:] if self.observed_category[i]]
self.parents_by_category[cat_id] = assignments
self.has_parents[cat_id] = True
else:
vector = self.tfidf.content(term_id)
assignments = [i for i in ints[1:] if self.observed_category[i]]
self.parents_by_concept[term_id] = assignments
for a_id in assignments:
for w, weight in vector:
if self.tfidf.df[w] < self.min_df:
continue
#print term, term_id, self.concept_by_id[a_id], w, self.vector_by_category[a_id][w], '\t+%f' % weight
self.vector_by_category[a_id][w] += weight
self.contributors_by_category[a_id].update([term_id])
if __name__ == '__main__':
import random
from random import shuffle as randomize
tfidf = TfIdfMatrix()
tfidf.load_features('bkp.big.out/vector.term.csv')
tfidf.load_distribution('bkp.big.out/vector.index.csv')
# tfidf.load_features('vector.term.csv')
# tfidf.load_distribution('vector.index.csv')
ctree = CategoryTree(
'bkp.big.out/category.index.csv',
'bkp.big.out/term.csv',
'bkp.big.out/category.csv',
# 'category.index.csv',
# 'term.csv',
# 'category.csv',
tfidf,
max_depth=1
)
ctree.load()
ctree.build()
ctree.dump()
<｜fim▁end｜> | vector = self.vector_by_category[_id] |
<|file_name|>CategoryTree.py<|end_file_name|><｜fim▁begin｜>import json
from collections import (
Counter,
defaultdict as deft
)
from copy import deepcopy as cp
# from cPickle import (
# dump as to_pickle,
# load as from_pickle
# )
from StringIO import StringIO
from TfIdfMatrix import TfIdfMatrix
from Tools import from_csv
class CategoryTree:
def __init__(self, categories_by_concept, terms,
categories, tfidf, max_depth=5, min_df=20
):
self.min_df = min_df
self.path_categories_by_concept = categories_by_concept
self.path_categories = categories
self.path_terms = terms
self.max_depth = max_depth
self.observed_category = deft(bool)
self.id_by_concept = dict([])
self.concept_by_id = dict([])
self.term_is_category = deft(bool)
self.parents_by_category = dict([])
self.parents_by_concept = deft(list)
self.id_by_term = dict([])
self.term_by_id = dict([])
self.has_parents = deft(bool)
self.tfidf = tfidf
self.pulling = set([])
self.vector_by_category = deft(Counter)
self.contributors_by_category = deft(set)
self.projected = Counter()
def build(self):
for i, c in enumerate(self.concept_by_id.values()):
self(c)
if not i % 100:
t = float(len(self.concept_by_id.keys()))
print i, int(t), round(i / t, 2)
# if i >= 5000:
# break
def dump(self):
# Simulate a file with StringIO
out = open('vector.dump.txt', 'wb')
for i, (_id, projections) in enumerate(self.projected.items()):
if not i % 100:
print i, len(self.projected.keys())
if not projections:
continue
features = [
(self.tfidf.word_by_id[wid], round(weight, 4))
for wid, weight in self.vector_by_category[_id].most_common()
if round(weight, 4)
]
record = (
_id,
self.concept_by_id[_id],
features
)
out.write('%s\n' % str(record))
out.close()
def __call__(self, category):
self.pulling = set([])
return self.__pull(None, 0, category, dict([]))
def __get_parents(self, _id):
parents = []
name = self.concept_by_id[_id]
if (
not self.observed_category[name] or
not self.observed_category[_id] or
not self.has_parents[_id]
):
return []
else:
for i in self.parents_by_category[_id]:
if not self.observed_category[i]:
continue
_name = self.concept_by_id[i]
parents.append(_name)
return set(parents) - self.pulling
def __pull(self, vector, depth, category, tree):
_id = self.id_by_concept[category]
if not self.pulling:
# print
# print
# print category, _id
# print [self.term_by_id[x] for x in self.contributors_by_category[_id]]
# print self.vector_by_category[_id].most_common(20)
vector = self.vector_by_category[_id]
if not self.observed_category[category]:
<|fim_middle|>
parents = self.__get_parents(_id)
if not parents or depth >= self.max_depth:
tree[category] = dict([])
else:
subtree = dict([])
self.pulling.update(parents)
for parent in parents:
subtree = self.__pull(vector, depth + 1, parent, subtree)
tree[category] = subtree
self.__project(vector, tree)
return tree
def __project(self, vector, tree):
if not tree.keys():
return
else:
for key, subtree in tree.items():
_id = self.id_by_concept[key]
self.projected[_id] += 1
self.__add2vec(vector, _id)
self.__project(vector, subtree)
def __add2vec(self, vector, _id):
# for w, weight in vector.items():
# __id = self.tfidf.id_by_word[w]
for __id, weight in vector.items():
self.vector_by_category[_id][__id] += weight
def load(self):
self.__load_terms()
self.__load_categories()
self.__load_assignments()
def __load_categories(self):
for concept, _id in from_csv(self.path_categories):
_id = int(_id)
self.id_by_concept[concept] = _id
self.concept_by_id[_id] = concept
self.observed_category[_id] = True
self.observed_category[concept] = True
# print concept, _id, len(self.id_by_concept.keys())
# exit()
def __load_terms(self):
for term, _id in from_csv(self.path_terms):
_id = int(_id)
self.term_by_id[_id] = term
self.id_by_term[term] = _id
if not term.startswith('Category:'):
continue
self.term_is_category[term] = True
self.term_is_category[_id] = True
def __load_assignments(self):
for row in from_csv(self.path_categories_by_concept):
ints = [int(field) for field in row]
term_id = ints[0]
term = self.term_by_id[term_id]
if self.term_is_category[term_id] and \
self.observed_category[term]:
term = self.term_by_id[term_id]
cat_id = self.id_by_concept[term]
assignments = [i for i in ints[1:] if self.observed_category[i]]
self.parents_by_category[cat_id] = assignments
self.has_parents[cat_id] = True
else:
vector = self.tfidf.content(term_id)
assignments = [i for i in ints[1:] if self.observed_category[i]]
self.parents_by_concept[term_id] = assignments
for a_id in assignments:
for w, weight in vector:
if self.tfidf.df[w] < self.min_df:
continue
#print term, term_id, self.concept_by_id[a_id], w, self.vector_by_category[a_id][w], '\t+%f' % weight
self.vector_by_category[a_id][w] += weight
self.contributors_by_category[a_id].update([term_id])
if __name__ == '__main__':
import random
from random import shuffle as randomize
tfidf = TfIdfMatrix()
tfidf.load_features('bkp.big.out/vector.term.csv')
tfidf.load_distribution('bkp.big.out/vector.index.csv')
# tfidf.load_features('vector.term.csv')
# tfidf.load_distribution('vector.index.csv')
ctree = CategoryTree(
'bkp.big.out/category.index.csv',
'bkp.big.out/term.csv',
'bkp.big.out/category.csv',
# 'category.index.csv',
# 'term.csv',
# 'category.csv',
tfidf,
max_depth=1
)
ctree.load()
ctree.build()
ctree.dump()
<｜fim▁end｜> | return dict([]) |
<|file_name|>CategoryTree.py<|end_file_name|><｜fim▁begin｜>import json
from collections import (
Counter,
defaultdict as deft
)
from copy import deepcopy as cp
# from cPickle import (
# dump as to_pickle,
# load as from_pickle
# )
from StringIO import StringIO
from TfIdfMatrix import TfIdfMatrix
from Tools import from_csv
class CategoryTree:
def __init__(self, categories_by_concept, terms,
categories, tfidf, max_depth=5, min_df=20
):
self.min_df = min_df
self.path_categories_by_concept = categories_by_concept
self.path_categories = categories
self.path_terms = terms
self.max_depth = max_depth
self.observed_category = deft(bool)
self.id_by_concept = dict([])
self.concept_by_id = dict([])
self.term_is_category = deft(bool)
self.parents_by_category = dict([])
self.parents_by_concept = deft(list)
self.id_by_term = dict([])
self.term_by_id = dict([])
self.has_parents = deft(bool)
self.tfidf = tfidf
self.pulling = set([])
self.vector_by_category = deft(Counter)
self.contributors_by_category = deft(set)
self.projected = Counter()
def build(self):
for i, c in enumerate(self.concept_by_id.values()):
self(c)
if not i % 100:
t = float(len(self.concept_by_id.keys()))
print i, int(t), round(i / t, 2)
# if i >= 5000:
# break
def dump(self):
# Simulate a file with StringIO
out = open('vector.dump.txt', 'wb')
for i, (_id, projections) in enumerate(self.projected.items()):
if not i % 100:
print i, len(self.projected.keys())
if not projections:
continue
features = [
(self.tfidf.word_by_id[wid], round(weight, 4))
for wid, weight in self.vector_by_category[_id].most_common()
if round(weight, 4)
]
record = (
_id,
self.concept_by_id[_id],
features
)
out.write('%s\n' % str(record))
out.close()
def __call__(self, category):
self.pulling = set([])
return self.__pull(None, 0, category, dict([]))
def __get_parents(self, _id):
parents = []
name = self.concept_by_id[_id]
if (
not self.observed_category[name] or
not self.observed_category[_id] or
not self.has_parents[_id]
):
return []
else:
for i in self.parents_by_category[_id]:
if not self.observed_category[i]:
continue
_name = self.concept_by_id[i]
parents.append(_name)
return set(parents) - self.pulling
def __pull(self, vector, depth, category, tree):
_id = self.id_by_concept[category]
if not self.pulling:
# print
# print
# print category, _id
# print [self.term_by_id[x] for x in self.contributors_by_category[_id]]
# print self.vector_by_category[_id].most_common(20)
vector = self.vector_by_category[_id]
if not self.observed_category[category]:
return dict([])
parents = self.__get_parents(_id)
if not parents or depth >= self.max_depth:
<|fim_middle|>
else:
subtree = dict([])
self.pulling.update(parents)
for parent in parents:
subtree = self.__pull(vector, depth + 1, parent, subtree)
tree[category] = subtree
self.__project(vector, tree)
return tree
def __project(self, vector, tree):
if not tree.keys():
return
else:
for key, subtree in tree.items():
_id = self.id_by_concept[key]
self.projected[_id] += 1
self.__add2vec(vector, _id)
self.__project(vector, subtree)
def __add2vec(self, vector, _id):
# for w, weight in vector.items():
# __id = self.tfidf.id_by_word[w]
for __id, weight in vector.items():
self.vector_by_category[_id][__id] += weight
def load(self):
self.__load_terms()
self.__load_categories()
self.__load_assignments()
def __load_categories(self):
for concept, _id in from_csv(self.path_categories):
_id = int(_id)
self.id_by_concept[concept] = _id
self.concept_by_id[_id] = concept
self.observed_category[_id] = True
self.observed_category[concept] = True
# print concept, _id, len(self.id_by_concept.keys())
# exit()
def __load_terms(self):
for term, _id in from_csv(self.path_terms):
_id = int(_id)
self.term_by_id[_id] = term
self.id_by_term[term] = _id
if not term.startswith('Category:'):
continue
self.term_is_category[term] = True
self.term_is_category[_id] = True
def __load_assignments(self):
for row in from_csv(self.path_categories_by_concept):
ints = [int(field) for field in row]
term_id = ints[0]
term = self.term_by_id[term_id]
if self.term_is_category[term_id] and \
self.observed_category[term]:
term = self.term_by_id[term_id]
cat_id = self.id_by_concept[term]
assignments = [i for i in ints[1:] if self.observed_category[i]]
self.parents_by_category[cat_id] = assignments
self.has_parents[cat_id] = True
else:
vector = self.tfidf.content(term_id)
assignments = [i for i in ints[1:] if self.observed_category[i]]
self.parents_by_concept[term_id] = assignments
for a_id in assignments:
for w, weight in vector:
if self.tfidf.df[w] < self.min_df:
continue
#print term, term_id, self.concept_by_id[a_id], w, self.vector_by_category[a_id][w], '\t+%f' % weight
self.vector_by_category[a_id][w] += weight
self.contributors_by_category[a_id].update([term_id])
if __name__ == '__main__':
import random
from random import shuffle as randomize
tfidf = TfIdfMatrix()
tfidf.load_features('bkp.big.out/vector.term.csv')
tfidf.load_distribution('bkp.big.out/vector.index.csv')
# tfidf.load_features('vector.term.csv')
# tfidf.load_distribution('vector.index.csv')
ctree = CategoryTree(
'bkp.big.out/category.index.csv',
'bkp.big.out/term.csv',
'bkp.big.out/category.csv',
# 'category.index.csv',
# 'term.csv',
# 'category.csv',
tfidf,
max_depth=1
)
ctree.load()
ctree.build()
ctree.dump()
<｜fim▁end｜> | tree[category] = dict([]) |
<|file_name|>CategoryTree.py<|end_file_name|><｜fim▁begin｜>import json
from collections import (
Counter,
defaultdict as deft
)
from copy import deepcopy as cp
# from cPickle import (
# dump as to_pickle,
# load as from_pickle
# )
from StringIO import StringIO
from TfIdfMatrix import TfIdfMatrix
from Tools import from_csv
class CategoryTree:
def __init__(self, categories_by_concept, terms,
categories, tfidf, max_depth=5, min_df=20
):
self.min_df = min_df
self.path_categories_by_concept = categories_by_concept
self.path_categories = categories
self.path_terms = terms
self.max_depth = max_depth
self.observed_category = deft(bool)
self.id_by_concept = dict([])
self.concept_by_id = dict([])
self.term_is_category = deft(bool)
self.parents_by_category = dict([])
self.parents_by_concept = deft(list)
self.id_by_term = dict([])
self.term_by_id = dict([])
self.has_parents = deft(bool)
self.tfidf = tfidf
self.pulling = set([])
self.vector_by_category = deft(Counter)
self.contributors_by_category = deft(set)
self.projected = Counter()
def build(self):
for i, c in enumerate(self.concept_by_id.values()):
self(c)
if not i % 100:
t = float(len(self.concept_by_id.keys()))
print i, int(t), round(i / t, 2)
# if i >= 5000:
# break
def dump(self):
# Simulate a file with StringIO
out = open('vector.dump.txt', 'wb')
for i, (_id, projections) in enumerate(self.projected.items()):
if not i % 100:
print i, len(self.projected.keys())
if not projections:
continue
features = [
(self.tfidf.word_by_id[wid], round(weight, 4))
for wid, weight in self.vector_by_category[_id].most_common()
if round(weight, 4)
]
record = (
_id,
self.concept_by_id[_id],
features
)
out.write('%s\n' % str(record))
out.close()
def __call__(self, category):
self.pulling = set([])
return self.__pull(None, 0, category, dict([]))
def __get_parents(self, _id):
parents = []
name = self.concept_by_id[_id]
if (
not self.observed_category[name] or
not self.observed_category[_id] or
not self.has_parents[_id]
):
return []
else:
for i in self.parents_by_category[_id]:
if not self.observed_category[i]:
continue
_name = self.concept_by_id[i]
parents.append(_name)
return set(parents) - self.pulling
def __pull(self, vector, depth, category, tree):
_id = self.id_by_concept[category]
if not self.pulling:
# print
# print
# print category, _id
# print [self.term_by_id[x] for x in self.contributors_by_category[_id]]
# print self.vector_by_category[_id].most_common(20)
vector = self.vector_by_category[_id]
if not self.observed_category[category]:
return dict([])
parents = self.__get_parents(_id)
if not parents or depth >= self.max_depth:
tree[category] = dict([])
else:
<|fim_middle|>
self.__project(vector, tree)
return tree
def __project(self, vector, tree):
if not tree.keys():
return
else:
for key, subtree in tree.items():
_id = self.id_by_concept[key]
self.projected[_id] += 1
self.__add2vec(vector, _id)
self.__project(vector, subtree)
def __add2vec(self, vector, _id):
# for w, weight in vector.items():
# __id = self.tfidf.id_by_word[w]
for __id, weight in vector.items():
self.vector_by_category[_id][__id] += weight
def load(self):
self.__load_terms()
self.__load_categories()
self.__load_assignments()
def __load_categories(self):
for concept, _id in from_csv(self.path_categories):
_id = int(_id)
self.id_by_concept[concept] = _id
self.concept_by_id[_id] = concept
self.observed_category[_id] = True
self.observed_category[concept] = True
# print concept, _id, len(self.id_by_concept.keys())
# exit()
def __load_terms(self):
for term, _id in from_csv(self.path_terms):
_id = int(_id)
self.term_by_id[_id] = term
self.id_by_term[term] = _id
if not term.startswith('Category:'):
continue
self.term_is_category[term] = True
self.term_is_category[_id] = True
def __load_assignments(self):
for row in from_csv(self.path_categories_by_concept):
ints = [int(field) for field in row]
term_id = ints[0]
term = self.term_by_id[term_id]
if self.term_is_category[term_id] and \
self.observed_category[term]:
term = self.term_by_id[term_id]
cat_id = self.id_by_concept[term]
assignments = [i for i in ints[1:] if self.observed_category[i]]
self.parents_by_category[cat_id] = assignments
self.has_parents[cat_id] = True
else:
vector = self.tfidf.content(term_id)
assignments = [i for i in ints[1:] if self.observed_category[i]]
self.parents_by_concept[term_id] = assignments
for a_id in assignments:
for w, weight in vector:
if self.tfidf.df[w] < self.min_df:
continue
#print term, term_id, self.concept_by_id[a_id], w, self.vector_by_category[a_id][w], '\t+%f' % weight
self.vector_by_category[a_id][w] += weight
self.contributors_by_category[a_id].update([term_id])
if __name__ == '__main__':
import random
from random import shuffle as randomize
tfidf = TfIdfMatrix()
tfidf.load_features('bkp.big.out/vector.term.csv')
tfidf.load_distribution('bkp.big.out/vector.index.csv')
# tfidf.load_features('vector.term.csv')
# tfidf.load_distribution('vector.index.csv')
ctree = CategoryTree(
'bkp.big.out/category.index.csv',
'bkp.big.out/term.csv',
'bkp.big.out/category.csv',
# 'category.index.csv',
# 'term.csv',
# 'category.csv',
tfidf,
max_depth=1
)
ctree.load()
ctree.build()
ctree.dump()
<｜fim▁end｜> | subtree = dict([])
self.pulling.update(parents)
for parent in parents:
subtree = self.__pull(vector, depth + 1, parent, subtree)
tree[category] = subtree |
<|file_name|>CategoryTree.py<|end_file_name|><｜fim▁begin｜>import json
from collections import (
Counter,
defaultdict as deft
)
from copy import deepcopy as cp
# from cPickle import (
# dump as to_pickle,
# load as from_pickle
# )
from StringIO import StringIO
from TfIdfMatrix import TfIdfMatrix
from Tools import from_csv
class CategoryTree:
def __init__(self, categories_by_concept, terms,
categories, tfidf, max_depth=5, min_df=20
):
self.min_df = min_df
self.path_categories_by_concept = categories_by_concept
self.path_categories = categories
self.path_terms = terms
self.max_depth = max_depth
self.observed_category = deft(bool)
self.id_by_concept = dict([])
self.concept_by_id = dict([])
self.term_is_category = deft(bool)
self.parents_by_category = dict([])
self.parents_by_concept = deft(list)
self.id_by_term = dict([])
self.term_by_id = dict([])
self.has_parents = deft(bool)
self.tfidf = tfidf
self.pulling = set([])
self.vector_by_category = deft(Counter)
self.contributors_by_category = deft(set)
self.projected = Counter()
def build(self):
for i, c in enumerate(self.concept_by_id.values()):
self(c)
if not i % 100:
t = float(len(self.concept_by_id.keys()))
print i, int(t), round(i / t, 2)
# if i >= 5000:
# break
def dump(self):
# Simulate a file with StringIO
out = open('vector.dump.txt', 'wb')
for i, (_id, projections) in enumerate(self.projected.items()):
if not i % 100:
print i, len(self.projected.keys())
if not projections:
continue
features = [
(self.tfidf.word_by_id[wid], round(weight, 4))
for wid, weight in self.vector_by_category[_id].most_common()
if round(weight, 4)
]
record = (
_id,
self.concept_by_id[_id],
features
)
out.write('%s\n' % str(record))
out.close()
def __call__(self, category):
self.pulling = set([])
return self.__pull(None, 0, category, dict([]))
def __get_parents(self, _id):
parents = []
name = self.concept_by_id[_id]
if (
not self.observed_category[name] or
not self.observed_category[_id] or
not self.has_parents[_id]
):
return []
else:
for i in self.parents_by_category[_id]:
if not self.observed_category[i]:
continue
_name = self.concept_by_id[i]
parents.append(_name)
return set(parents) - self.pulling
def __pull(self, vector, depth, category, tree):
_id = self.id_by_concept[category]
if not self.pulling:
# print
# print
# print category, _id
# print [self.term_by_id[x] for x in self.contributors_by_category[_id]]
# print self.vector_by_category[_id].most_common(20)
vector = self.vector_by_category[_id]
if not self.observed_category[category]:
return dict([])
parents = self.__get_parents(_id)
if not parents or depth >= self.max_depth:
tree[category] = dict([])
else:
subtree = dict([])
self.pulling.update(parents)
for parent in parents:
subtree = self.__pull(vector, depth + 1, parent, subtree)
tree[category] = subtree
self.__project(vector, tree)
return tree
def __project(self, vector, tree):
if not tree.keys():
<|fim_middle|>
else:
for key, subtree in tree.items():
_id = self.id_by_concept[key]
self.projected[_id] += 1
self.__add2vec(vector, _id)
self.__project(vector, subtree)
def __add2vec(self, vector, _id):
# for w, weight in vector.items():
# __id = self.tfidf.id_by_word[w]
for __id, weight in vector.items():
self.vector_by_category[_id][__id] += weight
def load(self):
self.__load_terms()
self.__load_categories()
self.__load_assignments()
def __load_categories(self):
for concept, _id in from_csv(self.path_categories):
_id = int(_id)
self.id_by_concept[concept] = _id
self.concept_by_id[_id] = concept
self.observed_category[_id] = True
self.observed_category[concept] = True
# print concept, _id, len(self.id_by_concept.keys())
# exit()
def __load_terms(self):
for term, _id in from_csv(self.path_terms):
_id = int(_id)
self.term_by_id[_id] = term
self.id_by_term[term] = _id
if not term.startswith('Category:'):
continue
self.term_is_category[term] = True
self.term_is_category[_id] = True
def __load_assignments(self):
for row in from_csv(self.path_categories_by_concept):
ints = [int(field) for field in row]
term_id = ints[0]
term = self.term_by_id[term_id]
if self.term_is_category[term_id] and \
self.observed_category[term]:
term = self.term_by_id[term_id]
cat_id = self.id_by_concept[term]
assignments = [i for i in ints[1:] if self.observed_category[i]]
self.parents_by_category[cat_id] = assignments
self.has_parents[cat_id] = True
else:
vector = self.tfidf.content(term_id)
assignments = [i for i in ints[1:] if self.observed_category[i]]
self.parents_by_concept[term_id] = assignments
for a_id in assignments:
for w, weight in vector:
if self.tfidf.df[w] < self.min_df:
continue
#print term, term_id, self.concept_by_id[a_id], w, self.vector_by_category[a_id][w], '\t+%f' % weight
self.vector_by_category[a_id][w] += weight
self.contributors_by_category[a_id].update([term_id])
if __name__ == '__main__':
import random
from random import shuffle as randomize
tfidf = TfIdfMatrix()
tfidf.load_features('bkp.big.out/vector.term.csv')
tfidf.load_distribution('bkp.big.out/vector.index.csv')
# tfidf.load_features('vector.term.csv')
# tfidf.load_distribution('vector.index.csv')
ctree = CategoryTree(
'bkp.big.out/category.index.csv',
'bkp.big.out/term.csv',
'bkp.big.out/category.csv',
# 'category.index.csv',
# 'term.csv',
# 'category.csv',
tfidf,
max_depth=1
)
ctree.load()
ctree.build()
ctree.dump()
<｜fim▁end｜> | return |
<|file_name|>CategoryTree.py<|end_file_name|><｜fim▁begin｜>import json
from collections import (
Counter,
defaultdict as deft
)
from copy import deepcopy as cp
# from cPickle import (
# dump as to_pickle,
# load as from_pickle
# )
from StringIO import StringIO
from TfIdfMatrix import TfIdfMatrix
from Tools import from_csv
class CategoryTree:
def __init__(self, categories_by_concept, terms,
categories, tfidf, max_depth=5, min_df=20
):
self.min_df = min_df
self.path_categories_by_concept = categories_by_concept
self.path_categories = categories
self.path_terms = terms
self.max_depth = max_depth
self.observed_category = deft(bool)
self.id_by_concept = dict([])
self.concept_by_id = dict([])
self.term_is_category = deft(bool)
self.parents_by_category = dict([])
self.parents_by_concept = deft(list)
self.id_by_term = dict([])
self.term_by_id = dict([])
self.has_parents = deft(bool)
self.tfidf = tfidf
self.pulling = set([])
self.vector_by_category = deft(Counter)
self.contributors_by_category = deft(set)
self.projected = Counter()
def build(self):
for i, c in enumerate(self.concept_by_id.values()):
self(c)
if not i % 100:
t = float(len(self.concept_by_id.keys()))
print i, int(t), round(i / t, 2)
# if i >= 5000:
# break
def dump(self):
# Simulate a file with StringIO
out = open('vector.dump.txt', 'wb')
for i, (_id, projections) in enumerate(self.projected.items()):
if not i % 100:
print i, len(self.projected.keys())
if not projections:
continue
features = [
(self.tfidf.word_by_id[wid], round(weight, 4))
for wid, weight in self.vector_by_category[_id].most_common()
if round(weight, 4)
]
record = (
_id,
self.concept_by_id[_id],
features
)
out.write('%s\n' % str(record))
out.close()
def __call__(self, category):
self.pulling = set([])
return self.__pull(None, 0, category, dict([]))
def __get_parents(self, _id):
parents = []
name = self.concept_by_id[_id]
if (
not self.observed_category[name] or
not self.observed_category[_id] or
not self.has_parents[_id]
):
return []
else:
for i in self.parents_by_category[_id]:
if not self.observed_category[i]:
continue
_name = self.concept_by_id[i]
parents.append(_name)
return set(parents) - self.pulling
def __pull(self, vector, depth, category, tree):
_id = self.id_by_concept[category]
if not self.pulling:
# print
# print
# print category, _id
# print [self.term_by_id[x] for x in self.contributors_by_category[_id]]
# print self.vector_by_category[_id].most_common(20)
vector = self.vector_by_category[_id]
if not self.observed_category[category]:
return dict([])
parents = self.__get_parents(_id)
if not parents or depth >= self.max_depth:
tree[category] = dict([])
else:
subtree = dict([])
self.pulling.update(parents)
for parent in parents:
subtree = self.__pull(vector, depth + 1, parent, subtree)
tree[category] = subtree
self.__project(vector, tree)
return tree
def __project(self, vector, tree):
if not tree.keys():
return
else:
<|fim_middle|>
def __add2vec(self, vector, _id):
# for w, weight in vector.items():
# __id = self.tfidf.id_by_word[w]
for __id, weight in vector.items():
self.vector_by_category[_id][__id] += weight
def load(self):
self.__load_terms()
self.__load_categories()
self.__load_assignments()
def __load_categories(self):
for concept, _id in from_csv(self.path_categories):
_id = int(_id)
self.id_by_concept[concept] = _id
self.concept_by_id[_id] = concept
self.observed_category[_id] = True
self.observed_category[concept] = True
# print concept, _id, len(self.id_by_concept.keys())
# exit()
def __load_terms(self):
for term, _id in from_csv(self.path_terms):
_id = int(_id)
self.term_by_id[_id] = term
self.id_by_term[term] = _id
if not term.startswith('Category:'):
continue
self.term_is_category[term] = True
self.term_is_category[_id] = True
def __load_assignments(self):
for row in from_csv(self.path_categories_by_concept):
ints = [int(field) for field in row]
term_id = ints[0]
term = self.term_by_id[term_id]
if self.term_is_category[term_id] and \
self.observed_category[term]:
term = self.term_by_id[term_id]
cat_id = self.id_by_concept[term]
assignments = [i for i in ints[1:] if self.observed_category[i]]
self.parents_by_category[cat_id] = assignments
self.has_parents[cat_id] = True
else:
vector = self.tfidf.content(term_id)
assignments = [i for i in ints[1:] if self.observed_category[i]]
self.parents_by_concept[term_id] = assignments
for a_id in assignments:
for w, weight in vector:
if self.tfidf.df[w] < self.min_df:
continue
#print term, term_id, self.concept_by_id[a_id], w, self.vector_by_category[a_id][w], '\t+%f' % weight
self.vector_by_category[a_id][w] += weight
self.contributors_by_category[a_id].update([term_id])
if __name__ == '__main__':
import random
from random import shuffle as randomize
tfidf = TfIdfMatrix()
tfidf.load_features('bkp.big.out/vector.term.csv')
tfidf.load_distribution('bkp.big.out/vector.index.csv')
# tfidf.load_features('vector.term.csv')
# tfidf.load_distribution('vector.index.csv')
ctree = CategoryTree(
'bkp.big.out/category.index.csv',
'bkp.big.out/term.csv',
'bkp.big.out/category.csv',
# 'category.index.csv',
# 'term.csv',
# 'category.csv',
tfidf,
max_depth=1
)
ctree.load()
ctree.build()
ctree.dump()
<｜fim▁end｜> | for key, subtree in tree.items():
_id = self.id_by_concept[key]
self.projected[_id] += 1
self.__add2vec(vector, _id)
self.__project(vector, subtree) |
<|file_name|>CategoryTree.py<|end_file_name|><｜fim▁begin｜>import json
from collections import (
Counter,
defaultdict as deft
)
from copy import deepcopy as cp
# from cPickle import (
# dump as to_pickle,
# load as from_pickle
# )
from StringIO import StringIO
from TfIdfMatrix import TfIdfMatrix
from Tools import from_csv
class CategoryTree:
def __init__(self, categories_by_concept, terms,
categories, tfidf, max_depth=5, min_df=20
):
self.min_df = min_df
self.path_categories_by_concept = categories_by_concept
self.path_categories = categories
self.path_terms = terms
self.max_depth = max_depth
self.observed_category = deft(bool)
self.id_by_concept = dict([])
self.concept_by_id = dict([])
self.term_is_category = deft(bool)
self.parents_by_category = dict([])
self.parents_by_concept = deft(list)
self.id_by_term = dict([])
self.term_by_id = dict([])
self.has_parents = deft(bool)
self.tfidf = tfidf
self.pulling = set([])
self.vector_by_category = deft(Counter)
self.contributors_by_category = deft(set)
self.projected = Counter()
def build(self):
for i, c in enumerate(self.concept_by_id.values()):
self(c)
if not i % 100:
t = float(len(self.concept_by_id.keys()))
print i, int(t), round(i / t, 2)
# if i >= 5000:
# break
def dump(self):
# Simulate a file with StringIO
out = open('vector.dump.txt', 'wb')
for i, (_id, projections) in enumerate(self.projected.items()):
if not i % 100:
print i, len(self.projected.keys())
if not projections:
continue
features = [
(self.tfidf.word_by_id[wid], round(weight, 4))
for wid, weight in self.vector_by_category[_id].most_common()
if round(weight, 4)
]
record = (
_id,
self.concept_by_id[_id],
features
)
out.write('%s\n' % str(record))
out.close()
def __call__(self, category):
self.pulling = set([])
return self.__pull(None, 0, category, dict([]))
def __get_parents(self, _id):
parents = []
name = self.concept_by_id[_id]
if (
not self.observed_category[name] or
not self.observed_category[_id] or
not self.has_parents[_id]
):
return []
else:
for i in self.parents_by_category[_id]:
if not self.observed_category[i]:
continue
_name = self.concept_by_id[i]
parents.append(_name)
return set(parents) - self.pulling
def __pull(self, vector, depth, category, tree):
_id = self.id_by_concept[category]
if not self.pulling:
# print
# print
# print category, _id
# print [self.term_by_id[x] for x in self.contributors_by_category[_id]]
# print self.vector_by_category[_id].most_common(20)
vector = self.vector_by_category[_id]
if not self.observed_category[category]:
return dict([])
parents = self.__get_parents(_id)
if not parents or depth >= self.max_depth:
tree[category] = dict([])
else:
subtree = dict([])
self.pulling.update(parents)
for parent in parents:
subtree = self.__pull(vector, depth + 1, parent, subtree)
tree[category] = subtree
self.__project(vector, tree)
return tree
def __project(self, vector, tree):
if not tree.keys():
return
else:
for key, subtree in tree.items():
_id = self.id_by_concept[key]
self.projected[_id] += 1
self.__add2vec(vector, _id)
self.__project(vector, subtree)
def __add2vec(self, vector, _id):
# for w, weight in vector.items():
# __id = self.tfidf.id_by_word[w]
for __id, weight in vector.items():
self.vector_by_category[_id][__id] += weight
def load(self):
self.__load_terms()
self.__load_categories()
self.__load_assignments()
def __load_categories(self):
for concept, _id in from_csv(self.path_categories):
_id = int(_id)
self.id_by_concept[concept] = _id
self.concept_by_id[_id] = concept
self.observed_category[_id] = True
self.observed_category[concept] = True
# print concept, _id, len(self.id_by_concept.keys())
# exit()
def __load_terms(self):
for term, _id in from_csv(self.path_terms):
_id = int(_id)
self.term_by_id[_id] = term
self.id_by_term[term] = _id
if not term.startswith('Category:'):
<|fim_middle|>
self.term_is_category[term] = True
self.term_is_category[_id] = True
def __load_assignments(self):
for row in from_csv(self.path_categories_by_concept):
ints = [int(field) for field in row]
term_id = ints[0]
term = self.term_by_id[term_id]
if self.term_is_category[term_id] and \
self.observed_category[term]:
term = self.term_by_id[term_id]
cat_id = self.id_by_concept[term]
assignments = [i for i in ints[1:] if self.observed_category[i]]
self.parents_by_category[cat_id] = assignments
self.has_parents[cat_id] = True
else:
vector = self.tfidf.content(term_id)
assignments = [i for i in ints[1:] if self.observed_category[i]]
self.parents_by_concept[term_id] = assignments
for a_id in assignments:
for w, weight in vector:
if self.tfidf.df[w] < self.min_df:
continue
#print term, term_id, self.concept_by_id[a_id], w, self.vector_by_category[a_id][w], '\t+%f' % weight
self.vector_by_category[a_id][w] += weight
self.contributors_by_category[a_id].update([term_id])
if __name__ == '__main__':
import random
from random import shuffle as randomize
tfidf = TfIdfMatrix()
tfidf.load_features('bkp.big.out/vector.term.csv')
tfidf.load_distribution('bkp.big.out/vector.index.csv')
# tfidf.load_features('vector.term.csv')
# tfidf.load_distribution('vector.index.csv')
ctree = CategoryTree(
'bkp.big.out/category.index.csv',
'bkp.big.out/term.csv',
'bkp.big.out/category.csv',
# 'category.index.csv',
# 'term.csv',
# 'category.csv',
tfidf,
max_depth=1
)
ctree.load()
ctree.build()
ctree.dump()
<｜fim▁end｜> | continue |
<|file_name|>CategoryTree.py<|end_file_name|><｜fim▁begin｜>import json
from collections import (
Counter,
defaultdict as deft
)
from copy import deepcopy as cp
# from cPickle import (
# dump as to_pickle,
# load as from_pickle
# )
from StringIO import StringIO
from TfIdfMatrix import TfIdfMatrix
from Tools import from_csv
class CategoryTree:
def __init__(self, categories_by_concept, terms,
categories, tfidf, max_depth=5, min_df=20
):
self.min_df = min_df
self.path_categories_by_concept = categories_by_concept
self.path_categories = categories
self.path_terms = terms
self.max_depth = max_depth
self.observed_category = deft(bool)
self.id_by_concept = dict([])
self.concept_by_id = dict([])
self.term_is_category = deft(bool)
self.parents_by_category = dict([])
self.parents_by_concept = deft(list)
self.id_by_term = dict([])
self.term_by_id = dict([])
self.has_parents = deft(bool)
self.tfidf = tfidf
self.pulling = set([])
self.vector_by_category = deft(Counter)
self.contributors_by_category = deft(set)
self.projected = Counter()
def build(self):
for i, c in enumerate(self.concept_by_id.values()):
self(c)
if not i % 100:
t = float(len(self.concept_by_id.keys()))
print i, int(t), round(i / t, 2)
# if i >= 5000:
# break
def dump(self):
# Simulate a file with StringIO
out = open('vector.dump.txt', 'wb')
for i, (_id, projections) in enumerate(self.projected.items()):
if not i % 100:
print i, len(self.projected.keys())
if not projections:
continue
features = [
(self.tfidf.word_by_id[wid], round(weight, 4))
for wid, weight in self.vector_by_category[_id].most_common()
if round(weight, 4)
]
record = (
_id,
self.concept_by_id[_id],
features
)
out.write('%s\n' % str(record))
out.close()
def __call__(self, category):
self.pulling = set([])
return self.__pull(None, 0, category, dict([]))
def __get_parents(self, _id):
parents = []
name = self.concept_by_id[_id]
if (
not self.observed_category[name] or
not self.observed_category[_id] or
not self.has_parents[_id]
):
return []
else:
for i in self.parents_by_category[_id]:
if not self.observed_category[i]:
continue
_name = self.concept_by_id[i]
parents.append(_name)
return set(parents) - self.pulling
def __pull(self, vector, depth, category, tree):
_id = self.id_by_concept[category]
if not self.pulling:
# print
# print
# print category, _id
# print [self.term_by_id[x] for x in self.contributors_by_category[_id]]
# print self.vector_by_category[_id].most_common(20)
vector = self.vector_by_category[_id]
if not self.observed_category[category]:
return dict([])
parents = self.__get_parents(_id)
if not parents or depth >= self.max_depth:
tree[category] = dict([])
else:
subtree = dict([])
self.pulling.update(parents)
for parent in parents:
subtree = self.__pull(vector, depth + 1, parent, subtree)
tree[category] = subtree
self.__project(vector, tree)
return tree
def __project(self, vector, tree):
if not tree.keys():
return
else:
for key, subtree in tree.items():
_id = self.id_by_concept[key]
self.projected[_id] += 1
self.__add2vec(vector, _id)
self.__project(vector, subtree)
def __add2vec(self, vector, _id):
# for w, weight in vector.items():
# __id = self.tfidf.id_by_word[w]
for __id, weight in vector.items():
self.vector_by_category[_id][__id] += weight
def load(self):
self.__load_terms()
self.__load_categories()
self.__load_assignments()
def __load_categories(self):
for concept, _id in from_csv(self.path_categories):
_id = int(_id)
self.id_by_concept[concept] = _id
self.concept_by_id[_id] = concept
self.observed_category[_id] = True
self.observed_category[concept] = True
# print concept, _id, len(self.id_by_concept.keys())
# exit()
def __load_terms(self):
for term, _id in from_csv(self.path_terms):
_id = int(_id)
self.term_by_id[_id] = term
self.id_by_term[term] = _id
if not term.startswith('Category:'):
continue
self.term_is_category[term] = True
self.term_is_category[_id] = True
def __load_assignments(self):
for row in from_csv(self.path_categories_by_concept):
ints = [int(field) for field in row]
term_id = ints[0]
term = self.term_by_id[term_id]
if self.term_is_category[term_id] and \
self.observed_category[term]:
<|fim_middle|>
else:
vector = self.tfidf.content(term_id)
assignments = [i for i in ints[1:] if self.observed_category[i]]
self.parents_by_concept[term_id] = assignments
for a_id in assignments:
for w, weight in vector:
if self.tfidf.df[w] < self.min_df:
continue
#print term, term_id, self.concept_by_id[a_id], w, self.vector_by_category[a_id][w], '\t+%f' % weight
self.vector_by_category[a_id][w] += weight
self.contributors_by_category[a_id].update([term_id])
if __name__ == '__main__':
import random
from random import shuffle as randomize
tfidf = TfIdfMatrix()
tfidf.load_features('bkp.big.out/vector.term.csv')
tfidf.load_distribution('bkp.big.out/vector.index.csv')
# tfidf.load_features('vector.term.csv')
# tfidf.load_distribution('vector.index.csv')
ctree = CategoryTree(
'bkp.big.out/category.index.csv',
'bkp.big.out/term.csv',
'bkp.big.out/category.csv',
# 'category.index.csv',
# 'term.csv',
# 'category.csv',
tfidf,
max_depth=1
)
ctree.load()
ctree.build()
ctree.dump()
<｜fim▁end｜> | term = self.term_by_id[term_id]
cat_id = self.id_by_concept[term]
assignments = [i for i in ints[1:] if self.observed_category[i]]
self.parents_by_category[cat_id] = assignments
self.has_parents[cat_id] = True |
<|file_name|>CategoryTree.py<|end_file_name|><｜fim▁begin｜>import json
from collections import (
Counter,
defaultdict as deft
)
from copy import deepcopy as cp
# from cPickle import (
# dump as to_pickle,
# load as from_pickle
# )
from StringIO import StringIO
from TfIdfMatrix import TfIdfMatrix
from Tools import from_csv
class CategoryTree:
def __init__(self, categories_by_concept, terms,
categories, tfidf, max_depth=5, min_df=20
):
self.min_df = min_df
self.path_categories_by_concept = categories_by_concept
self.path_categories = categories
self.path_terms = terms
self.max_depth = max_depth
self.observed_category = deft(bool)
self.id_by_concept = dict([])
self.concept_by_id = dict([])
self.term_is_category = deft(bool)
self.parents_by_category = dict([])
self.parents_by_concept = deft(list)
self.id_by_term = dict([])
self.term_by_id = dict([])
self.has_parents = deft(bool)
self.tfidf = tfidf
self.pulling = set([])
self.vector_by_category = deft(Counter)
self.contributors_by_category = deft(set)
self.projected = Counter()
def build(self):
for i, c in enumerate(self.concept_by_id.values()):
self(c)
if not i % 100:
t = float(len(self.concept_by_id.keys()))
print i, int(t), round(i / t, 2)
# if i >= 5000:
# break
def dump(self):
# Simulate a file with StringIO
out = open('vector.dump.txt', 'wb')
for i, (_id, projections) in enumerate(self.projected.items()):
if not i % 100:
print i, len(self.projected.keys())
if not projections:
continue
features = [
(self.tfidf.word_by_id[wid], round(weight, 4))
for wid, weight in self.vector_by_category[_id].most_common()
if round(weight, 4)
]
record = (
_id,
self.concept_by_id[_id],
features
)
out.write('%s\n' % str(record))
out.close()
def __call__(self, category):
self.pulling = set([])
return self.__pull(None, 0, category, dict([]))
def __get_parents(self, _id):
parents = []
name = self.concept_by_id[_id]
if (
not self.observed_category[name] or
not self.observed_category[_id] or
not self.has_parents[_id]
):
return []
else:
for i in self.parents_by_category[_id]:
if not self.observed_category[i]:
continue
_name = self.concept_by_id[i]
parents.append(_name)
return set(parents) - self.pulling
def __pull(self, vector, depth, category, tree):
_id = self.id_by_concept[category]
if not self.pulling:
# print
# print
# print category, _id
# print [self.term_by_id[x] for x in self.contributors_by_category[_id]]
# print self.vector_by_category[_id].most_common(20)
vector = self.vector_by_category[_id]
if not self.observed_category[category]:
return dict([])
parents = self.__get_parents(_id)
if not parents or depth >= self.max_depth:
tree[category] = dict([])
else:
subtree = dict([])
self.pulling.update(parents)
for parent in parents:
subtree = self.__pull(vector, depth + 1, parent, subtree)
tree[category] = subtree
self.__project(vector, tree)
return tree
def __project(self, vector, tree):
if not tree.keys():
return
else:
for key, subtree in tree.items():
_id = self.id_by_concept[key]
self.projected[_id] += 1
self.__add2vec(vector, _id)
self.__project(vector, subtree)
def __add2vec(self, vector, _id):
# for w, weight in vector.items():
# __id = self.tfidf.id_by_word[w]
for __id, weight in vector.items():
self.vector_by_category[_id][__id] += weight
def load(self):
self.__load_terms()
self.__load_categories()
self.__load_assignments()
def __load_categories(self):
for concept, _id in from_csv(self.path_categories):
_id = int(_id)
self.id_by_concept[concept] = _id
self.concept_by_id[_id] = concept
self.observed_category[_id] = True
self.observed_category[concept] = True
# print concept, _id, len(self.id_by_concept.keys())
# exit()
def __load_terms(self):
for term, _id in from_csv(self.path_terms):
_id = int(_id)
self.term_by_id[_id] = term
self.id_by_term[term] = _id
if not term.startswith('Category:'):
continue
self.term_is_category[term] = True
self.term_is_category[_id] = True
def __load_assignments(self):
for row in from_csv(self.path_categories_by_concept):
ints = [int(field) for field in row]
term_id = ints[0]
term = self.term_by_id[term_id]
if self.term_is_category[term_id] and \
self.observed_category[term]:
term = self.term_by_id[term_id]
cat_id = self.id_by_concept[term]
assignments = [i for i in ints[1:] if self.observed_category[i]]
self.parents_by_category[cat_id] = assignments
self.has_parents[cat_id] = True
else:
<|fim_middle|>
if __name__ == '__main__':
import random
from random import shuffle as randomize
tfidf = TfIdfMatrix()
tfidf.load_features('bkp.big.out/vector.term.csv')
tfidf.load_distribution('bkp.big.out/vector.index.csv')
# tfidf.load_features('vector.term.csv')
# tfidf.load_distribution('vector.index.csv')
ctree = CategoryTree(
'bkp.big.out/category.index.csv',
'bkp.big.out/term.csv',
'bkp.big.out/category.csv',
# 'category.index.csv',
# 'term.csv',
# 'category.csv',
tfidf,
max_depth=1
)
ctree.load()
ctree.build()
ctree.dump()
<｜fim▁end｜> | vector = self.tfidf.content(term_id)
assignments = [i for i in ints[1:] if self.observed_category[i]]
self.parents_by_concept[term_id] = assignments
for a_id in assignments:
for w, weight in vector:
if self.tfidf.df[w] < self.min_df:
continue
#print term, term_id, self.concept_by_id[a_id], w, self.vector_by_category[a_id][w], '\t+%f' % weight
self.vector_by_category[a_id][w] += weight
self.contributors_by_category[a_id].update([term_id]) |
<|file_name|>CategoryTree.py<|end_file_name|><｜fim▁begin｜>import json
from collections import (
Counter,
defaultdict as deft
)
from copy import deepcopy as cp
# from cPickle import (
# dump as to_pickle,
# load as from_pickle
# )
from StringIO import StringIO
from TfIdfMatrix import TfIdfMatrix
from Tools import from_csv
class CategoryTree:
def __init__(self, categories_by_concept, terms,
categories, tfidf, max_depth=5, min_df=20
):
self.min_df = min_df
self.path_categories_by_concept = categories_by_concept
self.path_categories = categories
self.path_terms = terms
self.max_depth = max_depth
self.observed_category = deft(bool)
self.id_by_concept = dict([])
self.concept_by_id = dict([])
self.term_is_category = deft(bool)
self.parents_by_category = dict([])
self.parents_by_concept = deft(list)
self.id_by_term = dict([])
self.term_by_id = dict([])
self.has_parents = deft(bool)
self.tfidf = tfidf
self.pulling = set([])
self.vector_by_category = deft(Counter)
self.contributors_by_category = deft(set)
self.projected = Counter()
def build(self):
for i, c in enumerate(self.concept_by_id.values()):
self(c)
if not i % 100:
t = float(len(self.concept_by_id.keys()))
print i, int(t), round(i / t, 2)
# if i >= 5000:
# break
def dump(self):
# Simulate a file with StringIO
out = open('vector.dump.txt', 'wb')
for i, (_id, projections) in enumerate(self.projected.items()):
if not i % 100:
print i, len(self.projected.keys())
if not projections:
continue
features = [
(self.tfidf.word_by_id[wid], round(weight, 4))
for wid, weight in self.vector_by_category[_id].most_common()
if round(weight, 4)
]
record = (
_id,
self.concept_by_id[_id],
features
)
out.write('%s\n' % str(record))
out.close()
def __call__(self, category):
self.pulling = set([])
return self.__pull(None, 0, category, dict([]))
def __get_parents(self, _id):
parents = []
name = self.concept_by_id[_id]
if (
not self.observed_category[name] or
not self.observed_category[_id] or
not self.has_parents[_id]
):
return []
else:
for i in self.parents_by_category[_id]:
if not self.observed_category[i]:
continue
_name = self.concept_by_id[i]
parents.append(_name)
return set(parents) - self.pulling
def __pull(self, vector, depth, category, tree):
_id = self.id_by_concept[category]
if not self.pulling:
# print
# print
# print category, _id
# print [self.term_by_id[x] for x in self.contributors_by_category[_id]]
# print self.vector_by_category[_id].most_common(20)
vector = self.vector_by_category[_id]
if not self.observed_category[category]:
return dict([])
parents = self.__get_parents(_id)
if not parents or depth >= self.max_depth:
tree[category] = dict([])
else:
subtree = dict([])
self.pulling.update(parents)
for parent in parents:
subtree = self.__pull(vector, depth + 1, parent, subtree)
tree[category] = subtree
self.__project(vector, tree)
return tree
def __project(self, vector, tree):
if not tree.keys():
return
else:
for key, subtree in tree.items():
_id = self.id_by_concept[key]
self.projected[_id] += 1
self.__add2vec(vector, _id)
self.__project(vector, subtree)
def __add2vec(self, vector, _id):
# for w, weight in vector.items():
# __id = self.tfidf.id_by_word[w]
for __id, weight in vector.items():
self.vector_by_category[_id][__id] += weight
def load(self):
self.__load_terms()
self.__load_categories()
self.__load_assignments()
def __load_categories(self):
for concept, _id in from_csv(self.path_categories):
_id = int(_id)
self.id_by_concept[concept] = _id
self.concept_by_id[_id] = concept
self.observed_category[_id] = True
self.observed_category[concept] = True
# print concept, _id, len(self.id_by_concept.keys())
# exit()
def __load_terms(self):
for term, _id in from_csv(self.path_terms):
_id = int(_id)
self.term_by_id[_id] = term
self.id_by_term[term] = _id
if not term.startswith('Category:'):
continue
self.term_is_category[term] = True
self.term_is_category[_id] = True
def __load_assignments(self):
for row in from_csv(self.path_categories_by_concept):
ints = [int(field) for field in row]
term_id = ints[0]
term = self.term_by_id[term_id]
if self.term_is_category[term_id] and \
self.observed_category[term]:
term = self.term_by_id[term_id]
cat_id = self.id_by_concept[term]
assignments = [i for i in ints[1:] if self.observed_category[i]]
self.parents_by_category[cat_id] = assignments
self.has_parents[cat_id] = True
else:
vector = self.tfidf.content(term_id)
assignments = [i for i in ints[1:] if self.observed_category[i]]
self.parents_by_concept[term_id] = assignments
for a_id in assignments:
for w, weight in vector:
if self.tfidf.df[w] < self.min_df:
<|fim_middle|>
#print term, term_id, self.concept_by_id[a_id], w, self.vector_by_category[a_id][w], '\t+%f' % weight
self.vector_by_category[a_id][w] += weight
self.contributors_by_category[a_id].update([term_id])
if __name__ == '__main__':
import random
from random import shuffle as randomize
tfidf = TfIdfMatrix()
tfidf.load_features('bkp.big.out/vector.term.csv')
tfidf.load_distribution('bkp.big.out/vector.index.csv')
# tfidf.load_features('vector.term.csv')
# tfidf.load_distribution('vector.index.csv')
ctree = CategoryTree(
'bkp.big.out/category.index.csv',
'bkp.big.out/term.csv',
'bkp.big.out/category.csv',
# 'category.index.csv',
# 'term.csv',
# 'category.csv',
tfidf,
max_depth=1
)
ctree.load()
ctree.build()
ctree.dump()
<｜fim▁end｜> | continue |
<|file_name|>CategoryTree.py<|end_file_name|><｜fim▁begin｜>import json
from collections import (
Counter,
defaultdict as deft
)
from copy import deepcopy as cp
# from cPickle import (
# dump as to_pickle,
# load as from_pickle
# )
from StringIO import StringIO
from TfIdfMatrix import TfIdfMatrix
from Tools import from_csv
class CategoryTree:
def __init__(self, categories_by_concept, terms,
categories, tfidf, max_depth=5, min_df=20
):
self.min_df = min_df
self.path_categories_by_concept = categories_by_concept
self.path_categories = categories
self.path_terms = terms
self.max_depth = max_depth
self.observed_category = deft(bool)
self.id_by_concept = dict([])
self.concept_by_id = dict([])
self.term_is_category = deft(bool)
self.parents_by_category = dict([])
self.parents_by_concept = deft(list)
self.id_by_term = dict([])
self.term_by_id = dict([])
self.has_parents = deft(bool)
self.tfidf = tfidf
self.pulling = set([])
self.vector_by_category = deft(Counter)
self.contributors_by_category = deft(set)
self.projected = Counter()
def build(self):
for i, c in enumerate(self.concept_by_id.values()):
self(c)
if not i % 100:
t = float(len(self.concept_by_id.keys()))
print i, int(t), round(i / t, 2)
# if i >= 5000:
# break
def dump(self):
# Simulate a file with StringIO
out = open('vector.dump.txt', 'wb')
for i, (_id, projections) in enumerate(self.projected.items()):
if not i % 100:
print i, len(self.projected.keys())
if not projections:
continue
features = [
(self.tfidf.word_by_id[wid], round(weight, 4))
for wid, weight in self.vector_by_category[_id].most_common()
if round(weight, 4)
]
record = (
_id,
self.concept_by_id[_id],
features
)
out.write('%s\n' % str(record))
out.close()
def __call__(self, category):
self.pulling = set([])
return self.__pull(None, 0, category, dict([]))
def __get_parents(self, _id):
parents = []
name = self.concept_by_id[_id]
if (
not self.observed_category[name] or
not self.observed_category[_id] or
not self.has_parents[_id]
):
return []
else:
for i in self.parents_by_category[_id]:
if not self.observed_category[i]:
continue
_name = self.concept_by_id[i]
parents.append(_name)
return set(parents) - self.pulling
def __pull(self, vector, depth, category, tree):
_id = self.id_by_concept[category]
if not self.pulling:
# print
# print
# print category, _id
# print [self.term_by_id[x] for x in self.contributors_by_category[_id]]
# print self.vector_by_category[_id].most_common(20)
vector = self.vector_by_category[_id]
if not self.observed_category[category]:
return dict([])
parents = self.__get_parents(_id)
if not parents or depth >= self.max_depth:
tree[category] = dict([])
else:
subtree = dict([])
self.pulling.update(parents)
for parent in parents:
subtree = self.__pull(vector, depth + 1, parent, subtree)
tree[category] = subtree
self.__project(vector, tree)
return tree
def __project(self, vector, tree):
if not tree.keys():
return
else:
for key, subtree in tree.items():
_id = self.id_by_concept[key]
self.projected[_id] += 1
self.__add2vec(vector, _id)
self.__project(vector, subtree)
def __add2vec(self, vector, _id):
# for w, weight in vector.items():
# __id = self.tfidf.id_by_word[w]
for __id, weight in vector.items():
self.vector_by_category[_id][__id] += weight
def load(self):
self.__load_terms()
self.__load_categories()
self.__load_assignments()
def __load_categories(self):
for concept, _id in from_csv(self.path_categories):
_id = int(_id)
self.id_by_concept[concept] = _id
self.concept_by_id[_id] = concept
self.observed_category[_id] = True
self.observed_category[concept] = True
# print concept, _id, len(self.id_by_concept.keys())
# exit()
def __load_terms(self):
for term, _id in from_csv(self.path_terms):
_id = int(_id)
self.term_by_id[_id] = term
self.id_by_term[term] = _id
if not term.startswith('Category:'):
continue
self.term_is_category[term] = True
self.term_is_category[_id] = True
def __load_assignments(self):
for row in from_csv(self.path_categories_by_concept):
ints = [int(field) for field in row]
term_id = ints[0]
term = self.term_by_id[term_id]
if self.term_is_category[term_id] and \
self.observed_category[term]:
term = self.term_by_id[term_id]
cat_id = self.id_by_concept[term]
assignments = [i for i in ints[1:] if self.observed_category[i]]
self.parents_by_category[cat_id] = assignments
self.has_parents[cat_id] = True
else:
vector = self.tfidf.content(term_id)
assignments = [i for i in ints[1:] if self.observed_category[i]]
self.parents_by_concept[term_id] = assignments
for a_id in assignments:
for w, weight in vector:
if self.tfidf.df[w] < self.min_df:
continue
#print term, term_id, self.concept_by_id[a_id], w, self.vector_by_category[a_id][w], '\t+%f' % weight
self.vector_by_category[a_id][w] += weight
self.contributors_by_category[a_id].update([term_id])
if __name__ == '__main__':
<|fim_middle|>
<｜fim▁end｜> | import random
from random import shuffle as randomize
tfidf = TfIdfMatrix()
tfidf.load_features('bkp.big.out/vector.term.csv')
tfidf.load_distribution('bkp.big.out/vector.index.csv')
# tfidf.load_features('vector.term.csv')
# tfidf.load_distribution('vector.index.csv')
ctree = CategoryTree(
'bkp.big.out/category.index.csv',
'bkp.big.out/term.csv',
'bkp.big.out/category.csv',
# 'category.index.csv',
# 'term.csv',
# 'category.csv',
tfidf,
max_depth=1
)
ctree.load()
ctree.build()
ctree.dump() |
<|file_name|>CategoryTree.py<|end_file_name|><｜fim▁begin｜>import json
from collections import (
Counter,
defaultdict as deft
)
from copy import deepcopy as cp
# from cPickle import (
# dump as to_pickle,
# load as from_pickle
# )
from StringIO import StringIO
from TfIdfMatrix import TfIdfMatrix
from Tools import from_csv
class CategoryTree:
def <|fim_middle|>(self, categories_by_concept, terms,
categories, tfidf, max_depth=5, min_df=20
):
self.min_df = min_df
self.path_categories_by_concept = categories_by_concept
self.path_categories = categories
self.path_terms = terms
self.max_depth = max_depth
self.observed_category = deft(bool)
self.id_by_concept = dict([])
self.concept_by_id = dict([])
self.term_is_category = deft(bool)
self.parents_by_category = dict([])
self.parents_by_concept = deft(list)
self.id_by_term = dict([])
self.term_by_id = dict([])
self.has_parents = deft(bool)
self.tfidf = tfidf
self.pulling = set([])
self.vector_by_category = deft(Counter)
self.contributors_by_category = deft(set)
self.projected = Counter()
def build(self):
for i, c in enumerate(self.concept_by_id.values()):
self(c)
if not i % 100:
t = float(len(self.concept_by_id.keys()))
print i, int(t), round(i / t, 2)
# if i >= 5000:
# break
def dump(self):
# Simulate a file with StringIO
out = open('vector.dump.txt', 'wb')
for i, (_id, projections) in enumerate(self.projected.items()):
if not i % 100:
print i, len(self.projected.keys())
if not projections:
continue
features = [
(self.tfidf.word_by_id[wid], round(weight, 4))
for wid, weight in self.vector_by_category[_id].most_common()
if round(weight, 4)
]
record = (
_id,
self.concept_by_id[_id],
features
)
out.write('%s\n' % str(record))
out.close()
def __call__(self, category):
self.pulling = set([])
return self.__pull(None, 0, category, dict([]))
def __get_parents(self, _id):
parents = []
name = self.concept_by_id[_id]
if (
not self.observed_category[name] or
not self.observed_category[_id] or
not self.has_parents[_id]
):
return []
else:
for i in self.parents_by_category[_id]:
if not self.observed_category[i]:
continue
_name = self.concept_by_id[i]
parents.append(_name)
return set(parents) - self.pulling
def __pull(self, vector, depth, category, tree):
_id = self.id_by_concept[category]
if not self.pulling:
# print
# print
# print category, _id
# print [self.term_by_id[x] for x in self.contributors_by_category[_id]]
# print self.vector_by_category[_id].most_common(20)
vector = self.vector_by_category[_id]
if not self.observed_category[category]:
return dict([])
parents = self.__get_parents(_id)
if not parents or depth >= self.max_depth:
tree[category] = dict([])
else:
subtree = dict([])
self.pulling.update(parents)
for parent in parents:
subtree = self.__pull(vector, depth + 1, parent, subtree)
tree[category] = subtree
self.__project(vector, tree)
return tree
def __project(self, vector, tree):
if not tree.keys():
return
else:
for key, subtree in tree.items():
_id = self.id_by_concept[key]
self.projected[_id] += 1
self.__add2vec(vector, _id)
self.__project(vector, subtree)
def __add2vec(self, vector, _id):
# for w, weight in vector.items():
# __id = self.tfidf.id_by_word[w]
for __id, weight in vector.items():
self.vector_by_category[_id][__id] += weight
def load(self):
self.__load_terms()
self.__load_categories()
self.__load_assignments()
def __load_categories(self):
for concept, _id in from_csv(self.path_categories):
_id = int(_id)
self.id_by_concept[concept] = _id
self.concept_by_id[_id] = concept
self.observed_category[_id] = True
self.observed_category[concept] = True
# print concept, _id, len(self.id_by_concept.keys())
# exit()
def __load_terms(self):
for term, _id in from_csv(self.path_terms):
_id = int(_id)
self.term_by_id[_id] = term
self.id_by_term[term] = _id
if not term.startswith('Category:'):
continue
self.term_is_category[term] = True
self.term_is_category[_id] = True
def __load_assignments(self):
for row in from_csv(self.path_categories_by_concept):
ints = [int(field) for field in row]
term_id = ints[0]
term = self.term_by_id[term_id]
if self.term_is_category[term_id] and \
self.observed_category[term]:
term = self.term_by_id[term_id]
cat_id = self.id_by_concept[term]
assignments = [i for i in ints[1:] if self.observed_category[i]]
self.parents_by_category[cat_id] = assignments
self.has_parents[cat_id] = True
else:
vector = self.tfidf.content(term_id)
assignments = [i for i in ints[1:] if self.observed_category[i]]
self.parents_by_concept[term_id] = assignments
for a_id in assignments:
for w, weight in vector:
if self.tfidf.df[w] < self.min_df:
continue
#print term, term_id, self.concept_by_id[a_id], w, self.vector_by_category[a_id][w], '\t+%f' % weight
self.vector_by_category[a_id][w] += weight
self.contributors_by_category[a_id].update([term_id])
if __name__ == '__main__':
import random
from random import shuffle as randomize
tfidf = TfIdfMatrix()
tfidf.load_features('bkp.big.out/vector.term.csv')
tfidf.load_distribution('bkp.big.out/vector.index.csv')
# tfidf.load_features('vector.term.csv')
# tfidf.load_distribution('vector.index.csv')
ctree = CategoryTree(
'bkp.big.out/category.index.csv',
'bkp.big.out/term.csv',
'bkp.big.out/category.csv',
# 'category.index.csv',
# 'term.csv',
# 'category.csv',
tfidf,
max_depth=1
)
ctree.load()
ctree.build()
ctree.dump()
<｜fim▁end｜> | __init__ |
<|file_name|>CategoryTree.py<|end_file_name|><｜fim▁begin｜>import json
from collections import (
Counter,
defaultdict as deft
)
from copy import deepcopy as cp
# from cPickle import (
# dump as to_pickle,
# load as from_pickle
# )
from StringIO import StringIO
from TfIdfMatrix import TfIdfMatrix
from Tools import from_csv
class CategoryTree:
def __init__(self, categories_by_concept, terms,
categories, tfidf, max_depth=5, min_df=20
):
self.min_df = min_df
self.path_categories_by_concept = categories_by_concept
self.path_categories = categories
self.path_terms = terms
self.max_depth = max_depth
self.observed_category = deft(bool)
self.id_by_concept = dict([])
self.concept_by_id = dict([])
self.term_is_category = deft(bool)
self.parents_by_category = dict([])
self.parents_by_concept = deft(list)
self.id_by_term = dict([])
self.term_by_id = dict([])
self.has_parents = deft(bool)
self.tfidf = tfidf
self.pulling = set([])
self.vector_by_category = deft(Counter)
self.contributors_by_category = deft(set)
self.projected = Counter()
def <|fim_middle|>(self):
for i, c in enumerate(self.concept_by_id.values()):
self(c)
if not i % 100:
t = float(len(self.concept_by_id.keys()))
print i, int(t), round(i / t, 2)
# if i >= 5000:
# break
def dump(self):
# Simulate a file with StringIO
out = open('vector.dump.txt', 'wb')
for i, (_id, projections) in enumerate(self.projected.items()):
if not i % 100:
print i, len(self.projected.keys())
if not projections:
continue
features = [
(self.tfidf.word_by_id[wid], round(weight, 4))
for wid, weight in self.vector_by_category[_id].most_common()
if round(weight, 4)
]
record = (
_id,
self.concept_by_id[_id],
features
)
out.write('%s\n' % str(record))
out.close()
def __call__(self, category):
self.pulling = set([])
return self.__pull(None, 0, category, dict([]))
def __get_parents(self, _id):
parents = []
name = self.concept_by_id[_id]
if (
not self.observed_category[name] or
not self.observed_category[_id] or
not self.has_parents[_id]
):
return []
else:
for i in self.parents_by_category[_id]:
if not self.observed_category[i]:
continue
_name = self.concept_by_id[i]
parents.append(_name)
return set(parents) - self.pulling
def __pull(self, vector, depth, category, tree):
_id = self.id_by_concept[category]
if not self.pulling:
# print
# print
# print category, _id
# print [self.term_by_id[x] for x in self.contributors_by_category[_id]]
# print self.vector_by_category[_id].most_common(20)
vector = self.vector_by_category[_id]
if not self.observed_category[category]:
return dict([])
parents = self.__get_parents(_id)
if not parents or depth >= self.max_depth:
tree[category] = dict([])
else:
subtree = dict([])
self.pulling.update(parents)
for parent in parents:
subtree = self.__pull(vector, depth + 1, parent, subtree)
tree[category] = subtree
self.__project(vector, tree)
return tree
def __project(self, vector, tree):
if not tree.keys():
return
else:
for key, subtree in tree.items():
_id = self.id_by_concept[key]
self.projected[_id] += 1
self.__add2vec(vector, _id)
self.__project(vector, subtree)
def __add2vec(self, vector, _id):
# for w, weight in vector.items():
# __id = self.tfidf.id_by_word[w]
for __id, weight in vector.items():
self.vector_by_category[_id][__id] += weight
def load(self):
self.__load_terms()
self.__load_categories()
self.__load_assignments()
def __load_categories(self):
for concept, _id in from_csv(self.path_categories):
_id = int(_id)
self.id_by_concept[concept] = _id
self.concept_by_id[_id] = concept
self.observed_category[_id] = True
self.observed_category[concept] = True
# print concept, _id, len(self.id_by_concept.keys())
# exit()
def __load_terms(self):
for term, _id in from_csv(self.path_terms):
_id = int(_id)
self.term_by_id[_id] = term
self.id_by_term[term] = _id
if not term.startswith('Category:'):
continue
self.term_is_category[term] = True
self.term_is_category[_id] = True
def __load_assignments(self):
for row in from_csv(self.path_categories_by_concept):
ints = [int(field) for field in row]
term_id = ints[0]
term = self.term_by_id[term_id]
if self.term_is_category[term_id] and \
self.observed_category[term]:
term = self.term_by_id[term_id]
cat_id = self.id_by_concept[term]
assignments = [i for i in ints[1:] if self.observed_category[i]]
self.parents_by_category[cat_id] = assignments
self.has_parents[cat_id] = True
else:
vector = self.tfidf.content(term_id)
assignments = [i for i in ints[1:] if self.observed_category[i]]
self.parents_by_concept[term_id] = assignments
for a_id in assignments:
for w, weight in vector:
if self.tfidf.df[w] < self.min_df:
continue
#print term, term_id, self.concept_by_id[a_id], w, self.vector_by_category[a_id][w], '\t+%f' % weight
self.vector_by_category[a_id][w] += weight
self.contributors_by_category[a_id].update([term_id])
if __name__ == '__main__':
import random
from random import shuffle as randomize
tfidf = TfIdfMatrix()
tfidf.load_features('bkp.big.out/vector.term.csv')
tfidf.load_distribution('bkp.big.out/vector.index.csv')
# tfidf.load_features('vector.term.csv')
# tfidf.load_distribution('vector.index.csv')
ctree = CategoryTree(
'bkp.big.out/category.index.csv',
'bkp.big.out/term.csv',
'bkp.big.out/category.csv',
# 'category.index.csv',
# 'term.csv',
# 'category.csv',
tfidf,
max_depth=1
)
ctree.load()
ctree.build()
ctree.dump()
<｜fim▁end｜> | build |
<|file_name|>CategoryTree.py<|end_file_name|><｜fim▁begin｜>import json
from collections import (
Counter,
defaultdict as deft
)
from copy import deepcopy as cp
# from cPickle import (
# dump as to_pickle,
# load as from_pickle
# )
from StringIO import StringIO
from TfIdfMatrix import TfIdfMatrix
from Tools import from_csv
class CategoryTree:
def __init__(self, categories_by_concept, terms,
categories, tfidf, max_depth=5, min_df=20
):
self.min_df = min_df
self.path_categories_by_concept = categories_by_concept
self.path_categories = categories
self.path_terms = terms
self.max_depth = max_depth
self.observed_category = deft(bool)
self.id_by_concept = dict([])
self.concept_by_id = dict([])
self.term_is_category = deft(bool)
self.parents_by_category = dict([])
self.parents_by_concept = deft(list)
self.id_by_term = dict([])
self.term_by_id = dict([])
self.has_parents = deft(bool)
self.tfidf = tfidf
self.pulling = set([])
self.vector_by_category = deft(Counter)
self.contributors_by_category = deft(set)
self.projected = Counter()
def build(self):
for i, c in enumerate(self.concept_by_id.values()):
self(c)
if not i % 100:
t = float(len(self.concept_by_id.keys()))
print i, int(t), round(i / t, 2)
# if i >= 5000:
# break
def <|fim_middle|>(self):
# Simulate a file with StringIO
out = open('vector.dump.txt', 'wb')
for i, (_id, projections) in enumerate(self.projected.items()):
if not i % 100:
print i, len(self.projected.keys())
if not projections:
continue
features = [
(self.tfidf.word_by_id[wid], round(weight, 4))
for wid, weight in self.vector_by_category[_id].most_common()
if round(weight, 4)
]
record = (
_id,
self.concept_by_id[_id],
features
)
out.write('%s\n' % str(record))
out.close()
def __call__(self, category):
self.pulling = set([])
return self.__pull(None, 0, category, dict([]))
def __get_parents(self, _id):
parents = []
name = self.concept_by_id[_id]
if (
not self.observed_category[name] or
not self.observed_category[_id] or
not self.has_parents[_id]
):
return []
else:
for i in self.parents_by_category[_id]:
if not self.observed_category[i]:
continue
_name = self.concept_by_id[i]
parents.append(_name)
return set(parents) - self.pulling
def __pull(self, vector, depth, category, tree):
_id = self.id_by_concept[category]
if not self.pulling:
# print
# print
# print category, _id
# print [self.term_by_id[x] for x in self.contributors_by_category[_id]]
# print self.vector_by_category[_id].most_common(20)
vector = self.vector_by_category[_id]
if not self.observed_category[category]:
return dict([])
parents = self.__get_parents(_id)
if not parents or depth >= self.max_depth:
tree[category] = dict([])
else:
subtree = dict([])
self.pulling.update(parents)
for parent in parents:
subtree = self.__pull(vector, depth + 1, parent, subtree)
tree[category] = subtree
self.__project(vector, tree)
return tree
def __project(self, vector, tree):
if not tree.keys():
return
else:
for key, subtree in tree.items():
_id = self.id_by_concept[key]
self.projected[_id] += 1
self.__add2vec(vector, _id)
self.__project(vector, subtree)
def __add2vec(self, vector, _id):
# for w, weight in vector.items():
# __id = self.tfidf.id_by_word[w]
for __id, weight in vector.items():
self.vector_by_category[_id][__id] += weight
def load(self):
self.__load_terms()
self.__load_categories()
self.__load_assignments()
def __load_categories(self):
for concept, _id in from_csv(self.path_categories):
_id = int(_id)
self.id_by_concept[concept] = _id
self.concept_by_id[_id] = concept
self.observed_category[_id] = True
self.observed_category[concept] = True
# print concept, _id, len(self.id_by_concept.keys())
# exit()
def __load_terms(self):
for term, _id in from_csv(self.path_terms):
_id = int(_id)
self.term_by_id[_id] = term
self.id_by_term[term] = _id
if not term.startswith('Category:'):
continue
self.term_is_category[term] = True
self.term_is_category[_id] = True
def __load_assignments(self):
for row in from_csv(self.path_categories_by_concept):
ints = [int(field) for field in row]
term_id = ints[0]
term = self.term_by_id[term_id]
if self.term_is_category[term_id] and \
self.observed_category[term]:
term = self.term_by_id[term_id]
cat_id = self.id_by_concept[term]
assignments = [i for i in ints[1:] if self.observed_category[i]]
self.parents_by_category[cat_id] = assignments
self.has_parents[cat_id] = True
else:
vector = self.tfidf.content(term_id)
assignments = [i for i in ints[1:] if self.observed_category[i]]
self.parents_by_concept[term_id] = assignments
for a_id in assignments:
for w, weight in vector:
if self.tfidf.df[w] < self.min_df:
continue
#print term, term_id, self.concept_by_id[a_id], w, self.vector_by_category[a_id][w], '\t+%f' % weight
self.vector_by_category[a_id][w] += weight
self.contributors_by_category[a_id].update([term_id])
if __name__ == '__main__':
import random
from random import shuffle as randomize
tfidf = TfIdfMatrix()
tfidf.load_features('bkp.big.out/vector.term.csv')
tfidf.load_distribution('bkp.big.out/vector.index.csv')
# tfidf.load_features('vector.term.csv')
# tfidf.load_distribution('vector.index.csv')
ctree = CategoryTree(
'bkp.big.out/category.index.csv',
'bkp.big.out/term.csv',
'bkp.big.out/category.csv',
# 'category.index.csv',
# 'term.csv',
# 'category.csv',
tfidf,
max_depth=1
)
ctree.load()
ctree.build()
ctree.dump()
<｜fim▁end｜> | dump |
<|file_name|>CategoryTree.py<|end_file_name|><｜fim▁begin｜>import json
from collections import (
Counter,
defaultdict as deft
)
from copy import deepcopy as cp
# from cPickle import (
# dump as to_pickle,
# load as from_pickle
# )
from StringIO import StringIO
from TfIdfMatrix import TfIdfMatrix
from Tools import from_csv
class CategoryTree:
def __init__(self, categories_by_concept, terms,
categories, tfidf, max_depth=5, min_df=20
):
self.min_df = min_df
self.path_categories_by_concept = categories_by_concept
self.path_categories = categories
self.path_terms = terms
self.max_depth = max_depth
self.observed_category = deft(bool)
self.id_by_concept = dict([])
self.concept_by_id = dict([])
self.term_is_category = deft(bool)
self.parents_by_category = dict([])
self.parents_by_concept = deft(list)
self.id_by_term = dict([])
self.term_by_id = dict([])
self.has_parents = deft(bool)
self.tfidf = tfidf
self.pulling = set([])
self.vector_by_category = deft(Counter)
self.contributors_by_category = deft(set)
self.projected = Counter()
def build(self):
for i, c in enumerate(self.concept_by_id.values()):
self(c)
if not i % 100:
t = float(len(self.concept_by_id.keys()))
print i, int(t), round(i / t, 2)
# if i >= 5000:
# break
def dump(self):
# Simulate a file with StringIO
out = open('vector.dump.txt', 'wb')
for i, (_id, projections) in enumerate(self.projected.items()):
if not i % 100:
print i, len(self.projected.keys())
if not projections:
continue
features = [
(self.tfidf.word_by_id[wid], round(weight, 4))
for wid, weight in self.vector_by_category[_id].most_common()
if round(weight, 4)
]
record = (
_id,
self.concept_by_id[_id],
features
)
out.write('%s\n' % str(record))
out.close()
def <|fim_middle|>(self, category):
self.pulling = set([])
return self.__pull(None, 0, category, dict([]))
def __get_parents(self, _id):
parents = []
name = self.concept_by_id[_id]
if (
not self.observed_category[name] or
not self.observed_category[_id] or
not self.has_parents[_id]
):
return []
else:
for i in self.parents_by_category[_id]:
if not self.observed_category[i]:
continue
_name = self.concept_by_id[i]
parents.append(_name)
return set(parents) - self.pulling
def __pull(self, vector, depth, category, tree):
_id = self.id_by_concept[category]
if not self.pulling:
# print
# print
# print category, _id
# print [self.term_by_id[x] for x in self.contributors_by_category[_id]]
# print self.vector_by_category[_id].most_common(20)
vector = self.vector_by_category[_id]
if not self.observed_category[category]:
return dict([])
parents = self.__get_parents(_id)
if not parents or depth >= self.max_depth:
tree[category] = dict([])
else:
subtree = dict([])
self.pulling.update(parents)
for parent in parents:
subtree = self.__pull(vector, depth + 1, parent, subtree)
tree[category] = subtree
self.__project(vector, tree)
return tree
def __project(self, vector, tree):
if not tree.keys():
return
else:
for key, subtree in tree.items():
_id = self.id_by_concept[key]
self.projected[_id] += 1
self.__add2vec(vector, _id)
self.__project(vector, subtree)
def __add2vec(self, vector, _id):
# for w, weight in vector.items():
# __id = self.tfidf.id_by_word[w]
for __id, weight in vector.items():
self.vector_by_category[_id][__id] += weight
def load(self):
self.__load_terms()
self.__load_categories()
self.__load_assignments()
def __load_categories(self):
for concept, _id in from_csv(self.path_categories):
_id = int(_id)
self.id_by_concept[concept] = _id
self.concept_by_id[_id] = concept
self.observed_category[_id] = True
self.observed_category[concept] = True
# print concept, _id, len(self.id_by_concept.keys())
# exit()
def __load_terms(self):
for term, _id in from_csv(self.path_terms):
_id = int(_id)
self.term_by_id[_id] = term
self.id_by_term[term] = _id
if not term.startswith('Category:'):
continue
self.term_is_category[term] = True
self.term_is_category[_id] = True
def __load_assignments(self):
for row in from_csv(self.path_categories_by_concept):
ints = [int(field) for field in row]
term_id = ints[0]
term = self.term_by_id[term_id]
if self.term_is_category[term_id] and \
self.observed_category[term]:
term = self.term_by_id[term_id]
cat_id = self.id_by_concept[term]
assignments = [i for i in ints[1:] if self.observed_category[i]]
self.parents_by_category[cat_id] = assignments
self.has_parents[cat_id] = True
else:
vector = self.tfidf.content(term_id)
assignments = [i for i in ints[1:] if self.observed_category[i]]
self.parents_by_concept[term_id] = assignments
for a_id in assignments:
for w, weight in vector:
if self.tfidf.df[w] < self.min_df:
continue
#print term, term_id, self.concept_by_id[a_id], w, self.vector_by_category[a_id][w], '\t+%f' % weight
self.vector_by_category[a_id][w] += weight
self.contributors_by_category[a_id].update([term_id])
if __name__ == '__main__':
import random
from random import shuffle as randomize
tfidf = TfIdfMatrix()
tfidf.load_features('bkp.big.out/vector.term.csv')
tfidf.load_distribution('bkp.big.out/vector.index.csv')
# tfidf.load_features('vector.term.csv')
# tfidf.load_distribution('vector.index.csv')
ctree = CategoryTree(
'bkp.big.out/category.index.csv',
'bkp.big.out/term.csv',
'bkp.big.out/category.csv',
# 'category.index.csv',
# 'term.csv',
# 'category.csv',
tfidf,
max_depth=1
)
ctree.load()
ctree.build()
ctree.dump()
<｜fim▁end｜> | __call__ |
<|file_name|>CategoryTree.py<|end_file_name|><｜fim▁begin｜>import json
from collections import (
Counter,
defaultdict as deft
)
from copy import deepcopy as cp
# from cPickle import (
# dump as to_pickle,
# load as from_pickle
# )
from StringIO import StringIO
from TfIdfMatrix import TfIdfMatrix
from Tools import from_csv
class CategoryTree:
def __init__(self, categories_by_concept, terms,
categories, tfidf, max_depth=5, min_df=20
):
self.min_df = min_df
self.path_categories_by_concept = categories_by_concept
self.path_categories = categories
self.path_terms = terms
self.max_depth = max_depth
self.observed_category = deft(bool)
self.id_by_concept = dict([])
self.concept_by_id = dict([])
self.term_is_category = deft(bool)
self.parents_by_category = dict([])
self.parents_by_concept = deft(list)
self.id_by_term = dict([])
self.term_by_id = dict([])
self.has_parents = deft(bool)
self.tfidf = tfidf
self.pulling = set([])
self.vector_by_category = deft(Counter)
self.contributors_by_category = deft(set)
self.projected = Counter()
def build(self):
for i, c in enumerate(self.concept_by_id.values()):
self(c)
if not i % 100:
t = float(len(self.concept_by_id.keys()))
print i, int(t), round(i / t, 2)
# if i >= 5000:
# break
def dump(self):
# Simulate a file with StringIO
out = open('vector.dump.txt', 'wb')
for i, (_id, projections) in enumerate(self.projected.items()):
if not i % 100:
print i, len(self.projected.keys())
if not projections:
continue
features = [
(self.tfidf.word_by_id[wid], round(weight, 4))
for wid, weight in self.vector_by_category[_id].most_common()
if round(weight, 4)
]
record = (
_id,
self.concept_by_id[_id],
features
)
out.write('%s\n' % str(record))
out.close()
def __call__(self, category):
self.pulling = set([])
return self.__pull(None, 0, category, dict([]))
def <|fim_middle|>(self, _id):
parents = []
name = self.concept_by_id[_id]
if (
not self.observed_category[name] or
not self.observed_category[_id] or
not self.has_parents[_id]
):
return []
else:
for i in self.parents_by_category[_id]:
if not self.observed_category[i]:
continue
_name = self.concept_by_id[i]
parents.append(_name)
return set(parents) - self.pulling
def __pull(self, vector, depth, category, tree):
_id = self.id_by_concept[category]
if not self.pulling:
# print
# print
# print category, _id
# print [self.term_by_id[x] for x in self.contributors_by_category[_id]]
# print self.vector_by_category[_id].most_common(20)
vector = self.vector_by_category[_id]
if not self.observed_category[category]:
return dict([])
parents = self.__get_parents(_id)
if not parents or depth >= self.max_depth:
tree[category] = dict([])
else:
subtree = dict([])
self.pulling.update(parents)
for parent in parents:
subtree = self.__pull(vector, depth + 1, parent, subtree)
tree[category] = subtree
self.__project(vector, tree)
return tree
def __project(self, vector, tree):
if not tree.keys():
return
else:
for key, subtree in tree.items():
_id = self.id_by_concept[key]
self.projected[_id] += 1
self.__add2vec(vector, _id)
self.__project(vector, subtree)
def __add2vec(self, vector, _id):
# for w, weight in vector.items():
# __id = self.tfidf.id_by_word[w]
for __id, weight in vector.items():
self.vector_by_category[_id][__id] += weight
def load(self):
self.__load_terms()
self.__load_categories()
self.__load_assignments()
def __load_categories(self):
for concept, _id in from_csv(self.path_categories):
_id = int(_id)
self.id_by_concept[concept] = _id
self.concept_by_id[_id] = concept
self.observed_category[_id] = True
self.observed_category[concept] = True
# print concept, _id, len(self.id_by_concept.keys())
# exit()
def __load_terms(self):
for term, _id in from_csv(self.path_terms):
_id = int(_id)
self.term_by_id[_id] = term
self.id_by_term[term] = _id
if not term.startswith('Category:'):
continue
self.term_is_category[term] = True
self.term_is_category[_id] = True
def __load_assignments(self):
for row in from_csv(self.path_categories_by_concept):
ints = [int(field) for field in row]
term_id = ints[0]
term = self.term_by_id[term_id]
if self.term_is_category[term_id] and \
self.observed_category[term]:
term = self.term_by_id[term_id]
cat_id = self.id_by_concept[term]
assignments = [i for i in ints[1:] if self.observed_category[i]]
self.parents_by_category[cat_id] = assignments
self.has_parents[cat_id] = True
else:
vector = self.tfidf.content(term_id)
assignments = [i for i in ints[1:] if self.observed_category[i]]
self.parents_by_concept[term_id] = assignments
for a_id in assignments:
for w, weight in vector:
if self.tfidf.df[w] < self.min_df:
continue
#print term, term_id, self.concept_by_id[a_id], w, self.vector_by_category[a_id][w], '\t+%f' % weight
self.vector_by_category[a_id][w] += weight
self.contributors_by_category[a_id].update([term_id])
if __name__ == '__main__':
import random
from random import shuffle as randomize
tfidf = TfIdfMatrix()
tfidf.load_features('bkp.big.out/vector.term.csv')
tfidf.load_distribution('bkp.big.out/vector.index.csv')
# tfidf.load_features('vector.term.csv')
# tfidf.load_distribution('vector.index.csv')
ctree = CategoryTree(
'bkp.big.out/category.index.csv',
'bkp.big.out/term.csv',
'bkp.big.out/category.csv',
# 'category.index.csv',
# 'term.csv',
# 'category.csv',
tfidf,
max_depth=1
)
ctree.load()
ctree.build()
ctree.dump()
<｜fim▁end｜> | __get_parents |
<|file_name|>CategoryTree.py<|end_file_name|><｜fim▁begin｜>import json
from collections import (
Counter,
defaultdict as deft
)
from copy import deepcopy as cp
# from cPickle import (
# dump as to_pickle,
# load as from_pickle
# )
from StringIO import StringIO
from TfIdfMatrix import TfIdfMatrix
from Tools import from_csv
class CategoryTree:
def __init__(self, categories_by_concept, terms,
categories, tfidf, max_depth=5, min_df=20
):
self.min_df = min_df
self.path_categories_by_concept = categories_by_concept
self.path_categories = categories
self.path_terms = terms
self.max_depth = max_depth
self.observed_category = deft(bool)
self.id_by_concept = dict([])
self.concept_by_id = dict([])
self.term_is_category = deft(bool)
self.parents_by_category = dict([])
self.parents_by_concept = deft(list)
self.id_by_term = dict([])
self.term_by_id = dict([])
self.has_parents = deft(bool)
self.tfidf = tfidf
self.pulling = set([])
self.vector_by_category = deft(Counter)
self.contributors_by_category = deft(set)
self.projected = Counter()
def build(self):
for i, c in enumerate(self.concept_by_id.values()):
self(c)
if not i % 100:
t = float(len(self.concept_by_id.keys()))
print i, int(t), round(i / t, 2)
# if i >= 5000:
# break
def dump(self):
# Simulate a file with StringIO
out = open('vector.dump.txt', 'wb')
for i, (_id, projections) in enumerate(self.projected.items()):
if not i % 100:
print i, len(self.projected.keys())
if not projections:
continue
features = [
(self.tfidf.word_by_id[wid], round(weight, 4))
for wid, weight in self.vector_by_category[_id].most_common()
if round(weight, 4)
]
record = (
_id,
self.concept_by_id[_id],
features
)
out.write('%s\n' % str(record))
out.close()
def __call__(self, category):
self.pulling = set([])
return self.__pull(None, 0, category, dict([]))
def __get_parents(self, _id):
parents = []
name = self.concept_by_id[_id]
if (
not self.observed_category[name] or
not self.observed_category[_id] or
not self.has_parents[_id]
):
return []
else:
for i in self.parents_by_category[_id]:
if not self.observed_category[i]:
continue
_name = self.concept_by_id[i]
parents.append(_name)
return set(parents) - self.pulling
def <|fim_middle|>(self, vector, depth, category, tree):
_id = self.id_by_concept[category]
if not self.pulling:
# print
# print
# print category, _id
# print [self.term_by_id[x] for x in self.contributors_by_category[_id]]
# print self.vector_by_category[_id].most_common(20)
vector = self.vector_by_category[_id]
if not self.observed_category[category]:
return dict([])
parents = self.__get_parents(_id)
if not parents or depth >= self.max_depth:
tree[category] = dict([])
else:
subtree = dict([])
self.pulling.update(parents)
for parent in parents:
subtree = self.__pull(vector, depth + 1, parent, subtree)
tree[category] = subtree
self.__project(vector, tree)
return tree
def __project(self, vector, tree):
if not tree.keys():
return
else:
for key, subtree in tree.items():
_id = self.id_by_concept[key]
self.projected[_id] += 1
self.__add2vec(vector, _id)
self.__project(vector, subtree)
def __add2vec(self, vector, _id):
# for w, weight in vector.items():
# __id = self.tfidf.id_by_word[w]
for __id, weight in vector.items():
self.vector_by_category[_id][__id] += weight
def load(self):
self.__load_terms()
self.__load_categories()
self.__load_assignments()
def __load_categories(self):
for concept, _id in from_csv(self.path_categories):
_id = int(_id)
self.id_by_concept[concept] = _id
self.concept_by_id[_id] = concept
self.observed_category[_id] = True
self.observed_category[concept] = True
# print concept, _id, len(self.id_by_concept.keys())
# exit()
def __load_terms(self):
for term, _id in from_csv(self.path_terms):
_id = int(_id)
self.term_by_id[_id] = term
self.id_by_term[term] = _id
if not term.startswith('Category:'):
continue
self.term_is_category[term] = True
self.term_is_category[_id] = True
def __load_assignments(self):
for row in from_csv(self.path_categories_by_concept):
ints = [int(field) for field in row]
term_id = ints[0]
term = self.term_by_id[term_id]
if self.term_is_category[term_id] and \
self.observed_category[term]:
term = self.term_by_id[term_id]
cat_id = self.id_by_concept[term]
assignments = [i for i in ints[1:] if self.observed_category[i]]
self.parents_by_category[cat_id] = assignments
self.has_parents[cat_id] = True
else:
vector = self.tfidf.content(term_id)
assignments = [i for i in ints[1:] if self.observed_category[i]]
self.parents_by_concept[term_id] = assignments
for a_id in assignments:
for w, weight in vector:
if self.tfidf.df[w] < self.min_df:
continue
#print term, term_id, self.concept_by_id[a_id], w, self.vector_by_category[a_id][w], '\t+%f' % weight
self.vector_by_category[a_id][w] += weight
self.contributors_by_category[a_id].update([term_id])
if __name__ == '__main__':
import random
from random import shuffle as randomize
tfidf = TfIdfMatrix()
tfidf.load_features('bkp.big.out/vector.term.csv')
tfidf.load_distribution('bkp.big.out/vector.index.csv')
# tfidf.load_features('vector.term.csv')
# tfidf.load_distribution('vector.index.csv')
ctree = CategoryTree(
'bkp.big.out/category.index.csv',
'bkp.big.out/term.csv',
'bkp.big.out/category.csv',
# 'category.index.csv',
# 'term.csv',
# 'category.csv',
tfidf,
max_depth=1
)
ctree.load()
ctree.build()
ctree.dump()
<｜fim▁end｜> | __pull |
<|file_name|>CategoryTree.py<|end_file_name|><｜fim▁begin｜>import json
from collections import (
Counter,
defaultdict as deft
)
from copy import deepcopy as cp
# from cPickle import (
# dump as to_pickle,
# load as from_pickle
# )
from StringIO import StringIO
from TfIdfMatrix import TfIdfMatrix
from Tools import from_csv
class CategoryTree:
def __init__(self, categories_by_concept, terms,
categories, tfidf, max_depth=5, min_df=20
):
self.min_df = min_df
self.path_categories_by_concept = categories_by_concept
self.path_categories = categories
self.path_terms = terms
self.max_depth = max_depth
self.observed_category = deft(bool)
self.id_by_concept = dict([])
self.concept_by_id = dict([])
self.term_is_category = deft(bool)
self.parents_by_category = dict([])
self.parents_by_concept = deft(list)
self.id_by_term = dict([])
self.term_by_id = dict([])
self.has_parents = deft(bool)
self.tfidf = tfidf
self.pulling = set([])
self.vector_by_category = deft(Counter)
self.contributors_by_category = deft(set)
self.projected = Counter()
def build(self):
for i, c in enumerate(self.concept_by_id.values()):
self(c)
if not i % 100:
t = float(len(self.concept_by_id.keys()))
print i, int(t), round(i / t, 2)
# if i >= 5000:
# break
def dump(self):
# Simulate a file with StringIO
out = open('vector.dump.txt', 'wb')
for i, (_id, projections) in enumerate(self.projected.items()):
if not i % 100:
print i, len(self.projected.keys())
if not projections:
continue
features = [
(self.tfidf.word_by_id[wid], round(weight, 4))
for wid, weight in self.vector_by_category[_id].most_common()
if round(weight, 4)
]
record = (
_id,
self.concept_by_id[_id],
features
)
out.write('%s\n' % str(record))
out.close()
def __call__(self, category):
self.pulling = set([])
return self.__pull(None, 0, category, dict([]))
def __get_parents(self, _id):
parents = []
name = self.concept_by_id[_id]
if (
not self.observed_category[name] or
not self.observed_category[_id] or
not self.has_parents[_id]
):
return []
else:
for i in self.parents_by_category[_id]:
if not self.observed_category[i]:
continue
_name = self.concept_by_id[i]
parents.append(_name)
return set(parents) - self.pulling
def __pull(self, vector, depth, category, tree):
_id = self.id_by_concept[category]
if not self.pulling:
# print
# print
# print category, _id
# print [self.term_by_id[x] for x in self.contributors_by_category[_id]]
# print self.vector_by_category[_id].most_common(20)
vector = self.vector_by_category[_id]
if not self.observed_category[category]:
return dict([])
parents = self.__get_parents(_id)
if not parents or depth >= self.max_depth:
tree[category] = dict([])
else:
subtree = dict([])
self.pulling.update(parents)
for parent in parents:
subtree = self.__pull(vector, depth + 1, parent, subtree)
tree[category] = subtree
self.__project(vector, tree)
return tree
def <|fim_middle|>(self, vector, tree):
if not tree.keys():
return
else:
for key, subtree in tree.items():
_id = self.id_by_concept[key]
self.projected[_id] += 1
self.__add2vec(vector, _id)
self.__project(vector, subtree)
def __add2vec(self, vector, _id):
# for w, weight in vector.items():
# __id = self.tfidf.id_by_word[w]
for __id, weight in vector.items():
self.vector_by_category[_id][__id] += weight
def load(self):
self.__load_terms()
self.__load_categories()
self.__load_assignments()
def __load_categories(self):
for concept, _id in from_csv(self.path_categories):
_id = int(_id)
self.id_by_concept[concept] = _id
self.concept_by_id[_id] = concept
self.observed_category[_id] = True
self.observed_category[concept] = True
# print concept, _id, len(self.id_by_concept.keys())
# exit()
def __load_terms(self):
for term, _id in from_csv(self.path_terms):
_id = int(_id)
self.term_by_id[_id] = term
self.id_by_term[term] = _id
if not term.startswith('Category:'):
continue
self.term_is_category[term] = True
self.term_is_category[_id] = True
def __load_assignments(self):
for row in from_csv(self.path_categories_by_concept):
ints = [int(field) for field in row]
term_id = ints[0]
term = self.term_by_id[term_id]
if self.term_is_category[term_id] and \
self.observed_category[term]:
term = self.term_by_id[term_id]
cat_id = self.id_by_concept[term]
assignments = [i for i in ints[1:] if self.observed_category[i]]
self.parents_by_category[cat_id] = assignments
self.has_parents[cat_id] = True
else:
vector = self.tfidf.content(term_id)
assignments = [i for i in ints[1:] if self.observed_category[i]]
self.parents_by_concept[term_id] = assignments
for a_id in assignments:
for w, weight in vector:
if self.tfidf.df[w] < self.min_df:
continue
#print term, term_id, self.concept_by_id[a_id], w, self.vector_by_category[a_id][w], '\t+%f' % weight
self.vector_by_category[a_id][w] += weight
self.contributors_by_category[a_id].update([term_id])
if __name__ == '__main__':
import random
from random import shuffle as randomize
tfidf = TfIdfMatrix()
tfidf.load_features('bkp.big.out/vector.term.csv')
tfidf.load_distribution('bkp.big.out/vector.index.csv')
# tfidf.load_features('vector.term.csv')
# tfidf.load_distribution('vector.index.csv')
ctree = CategoryTree(
'bkp.big.out/category.index.csv',
'bkp.big.out/term.csv',
'bkp.big.out/category.csv',
# 'category.index.csv',
# 'term.csv',
# 'category.csv',
tfidf,
max_depth=1
)
ctree.load()
ctree.build()
ctree.dump()
<｜fim▁end｜> | __project |
<|file_name|>CategoryTree.py<|end_file_name|><｜fim▁begin｜>import json
from collections import (
Counter,
defaultdict as deft
)
from copy import deepcopy as cp
# from cPickle import (
# dump as to_pickle,
# load as from_pickle
# )
from StringIO import StringIO
from TfIdfMatrix import TfIdfMatrix
from Tools import from_csv
class CategoryTree:
def __init__(self, categories_by_concept, terms,
categories, tfidf, max_depth=5, min_df=20
):
self.min_df = min_df
self.path_categories_by_concept = categories_by_concept
self.path_categories = categories
self.path_terms = terms
self.max_depth = max_depth
self.observed_category = deft(bool)
self.id_by_concept = dict([])
self.concept_by_id = dict([])
self.term_is_category = deft(bool)
self.parents_by_category = dict([])
self.parents_by_concept = deft(list)
self.id_by_term = dict([])
self.term_by_id = dict([])
self.has_parents = deft(bool)
self.tfidf = tfidf
self.pulling = set([])
self.vector_by_category = deft(Counter)
self.contributors_by_category = deft(set)
self.projected = Counter()
def build(self):
for i, c in enumerate(self.concept_by_id.values()):
self(c)
if not i % 100:
t = float(len(self.concept_by_id.keys()))
print i, int(t), round(i / t, 2)
# if i >= 5000:
# break
def dump(self):
# Simulate a file with StringIO
out = open('vector.dump.txt', 'wb')
for i, (_id, projections) in enumerate(self.projected.items()):
if not i % 100:
print i, len(self.projected.keys())
if not projections:
continue
features = [
(self.tfidf.word_by_id[wid], round(weight, 4))
for wid, weight in self.vector_by_category[_id].most_common()
if round(weight, 4)
]
record = (
_id,
self.concept_by_id[_id],
features
)
out.write('%s\n' % str(record))
out.close()
def __call__(self, category):
self.pulling = set([])
return self.__pull(None, 0, category, dict([]))
def __get_parents(self, _id):
parents = []
name = self.concept_by_id[_id]
if (
not self.observed_category[name] or
not self.observed_category[_id] or
not self.has_parents[_id]
):
return []
else:
for i in self.parents_by_category[_id]:
if not self.observed_category[i]:
continue
_name = self.concept_by_id[i]
parents.append(_name)
return set(parents) - self.pulling
def __pull(self, vector, depth, category, tree):
_id = self.id_by_concept[category]
if not self.pulling:
# print
# print
# print category, _id
# print [self.term_by_id[x] for x in self.contributors_by_category[_id]]
# print self.vector_by_category[_id].most_common(20)
vector = self.vector_by_category[_id]
if not self.observed_category[category]:
return dict([])
parents = self.__get_parents(_id)
if not parents or depth >= self.max_depth:
tree[category] = dict([])
else:
subtree = dict([])
self.pulling.update(parents)
for parent in parents:
subtree = self.__pull(vector, depth + 1, parent, subtree)
tree[category] = subtree
self.__project(vector, tree)
return tree
def __project(self, vector, tree):
if not tree.keys():
return
else:
for key, subtree in tree.items():
_id = self.id_by_concept[key]
self.projected[_id] += 1
self.__add2vec(vector, _id)
self.__project(vector, subtree)
def <|fim_middle|>(self, vector, _id):
# for w, weight in vector.items():
# __id = self.tfidf.id_by_word[w]
for __id, weight in vector.items():
self.vector_by_category[_id][__id] += weight
def load(self):
self.__load_terms()
self.__load_categories()
self.__load_assignments()
def __load_categories(self):
for concept, _id in from_csv(self.path_categories):
_id = int(_id)
self.id_by_concept[concept] = _id
self.concept_by_id[_id] = concept
self.observed_category[_id] = True
self.observed_category[concept] = True
# print concept, _id, len(self.id_by_concept.keys())
# exit()
def __load_terms(self):
for term, _id in from_csv(self.path_terms):
_id = int(_id)
self.term_by_id[_id] = term
self.id_by_term[term] = _id
if not term.startswith('Category:'):
continue
self.term_is_category[term] = True
self.term_is_category[_id] = True
def __load_assignments(self):
for row in from_csv(self.path_categories_by_concept):
ints = [int(field) for field in row]
term_id = ints[0]
term = self.term_by_id[term_id]
if self.term_is_category[term_id] and \
self.observed_category[term]:
term = self.term_by_id[term_id]
cat_id = self.id_by_concept[term]
assignments = [i for i in ints[1:] if self.observed_category[i]]
self.parents_by_category[cat_id] = assignments
self.has_parents[cat_id] = True
else:
vector = self.tfidf.content(term_id)
assignments = [i for i in ints[1:] if self.observed_category[i]]
self.parents_by_concept[term_id] = assignments
for a_id in assignments:
for w, weight in vector:
if self.tfidf.df[w] < self.min_df:
continue
#print term, term_id, self.concept_by_id[a_id], w, self.vector_by_category[a_id][w], '\t+%f' % weight
self.vector_by_category[a_id][w] += weight
self.contributors_by_category[a_id].update([term_id])
if __name__ == '__main__':
import random
from random import shuffle as randomize
tfidf = TfIdfMatrix()
tfidf.load_features('bkp.big.out/vector.term.csv')
tfidf.load_distribution('bkp.big.out/vector.index.csv')
# tfidf.load_features('vector.term.csv')
# tfidf.load_distribution('vector.index.csv')
ctree = CategoryTree(
'bkp.big.out/category.index.csv',
'bkp.big.out/term.csv',
'bkp.big.out/category.csv',
# 'category.index.csv',
# 'term.csv',
# 'category.csv',
tfidf,
max_depth=1
)
ctree.load()
ctree.build()
ctree.dump()
<｜fim▁end｜> | __add2vec |
<|file_name|>CategoryTree.py<|end_file_name|><｜fim▁begin｜>import json
from collections import (
Counter,
defaultdict as deft
)
from copy import deepcopy as cp
# from cPickle import (
# dump as to_pickle,
# load as from_pickle
# )
from StringIO import StringIO
from TfIdfMatrix import TfIdfMatrix
from Tools import from_csv
class CategoryTree:
def __init__(self, categories_by_concept, terms,
categories, tfidf, max_depth=5, min_df=20
):
self.min_df = min_df
self.path_categories_by_concept = categories_by_concept
self.path_categories = categories
self.path_terms = terms
self.max_depth = max_depth
self.observed_category = deft(bool)
self.id_by_concept = dict([])
self.concept_by_id = dict([])
self.term_is_category = deft(bool)
self.parents_by_category = dict([])
self.parents_by_concept = deft(list)
self.id_by_term = dict([])
self.term_by_id = dict([])
self.has_parents = deft(bool)
self.tfidf = tfidf
self.pulling = set([])
self.vector_by_category = deft(Counter)
self.contributors_by_category = deft(set)
self.projected = Counter()
def build(self):
for i, c in enumerate(self.concept_by_id.values()):
self(c)
if not i % 100:
t = float(len(self.concept_by_id.keys()))
print i, int(t), round(i / t, 2)
# if i >= 5000:
# break
def dump(self):
# Simulate a file with StringIO
out = open('vector.dump.txt', 'wb')
for i, (_id, projections) in enumerate(self.projected.items()):
if not i % 100:
print i, len(self.projected.keys())
if not projections:
continue
features = [
(self.tfidf.word_by_id[wid], round(weight, 4))
for wid, weight in self.vector_by_category[_id].most_common()
if round(weight, 4)
]
record = (
_id,
self.concept_by_id[_id],
features
)
out.write('%s\n' % str(record))
out.close()
def __call__(self, category):
self.pulling = set([])
return self.__pull(None, 0, category, dict([]))
def __get_parents(self, _id):
parents = []
name = self.concept_by_id[_id]
if (
not self.observed_category[name] or
not self.observed_category[_id] or
not self.has_parents[_id]
):
return []
else:
for i in self.parents_by_category[_id]:
if not self.observed_category[i]:
continue
_name = self.concept_by_id[i]
parents.append(_name)
return set(parents) - self.pulling
def __pull(self, vector, depth, category, tree):
_id = self.id_by_concept[category]
if not self.pulling:
# print
# print
# print category, _id
# print [self.term_by_id[x] for x in self.contributors_by_category[_id]]
# print self.vector_by_category[_id].most_common(20)
vector = self.vector_by_category[_id]
if not self.observed_category[category]:
return dict([])
parents = self.__get_parents(_id)
if not parents or depth >= self.max_depth:
tree[category] = dict([])
else:
subtree = dict([])
self.pulling.update(parents)
for parent in parents:
subtree = self.__pull(vector, depth + 1, parent, subtree)
tree[category] = subtree
self.__project(vector, tree)
return tree
def __project(self, vector, tree):
if not tree.keys():
return
else:
for key, subtree in tree.items():
_id = self.id_by_concept[key]
self.projected[_id] += 1
self.__add2vec(vector, _id)
self.__project(vector, subtree)
def __add2vec(self, vector, _id):
# for w, weight in vector.items():
# __id = self.tfidf.id_by_word[w]
for __id, weight in vector.items():
self.vector_by_category[_id][__id] += weight
def <|fim_middle|>(self):
self.__load_terms()
self.__load_categories()
self.__load_assignments()
def __load_categories(self):
for concept, _id in from_csv(self.path_categories):
_id = int(_id)
self.id_by_concept[concept] = _id
self.concept_by_id[_id] = concept
self.observed_category[_id] = True
self.observed_category[concept] = True
# print concept, _id, len(self.id_by_concept.keys())
# exit()
def __load_terms(self):
for term, _id in from_csv(self.path_terms):
_id = int(_id)
self.term_by_id[_id] = term
self.id_by_term[term] = _id
if not term.startswith('Category:'):
continue
self.term_is_category[term] = True
self.term_is_category[_id] = True
def __load_assignments(self):
for row in from_csv(self.path_categories_by_concept):
ints = [int(field) for field in row]
term_id = ints[0]
term = self.term_by_id[term_id]
if self.term_is_category[term_id] and \
self.observed_category[term]:
term = self.term_by_id[term_id]
cat_id = self.id_by_concept[term]
assignments = [i for i in ints[1:] if self.observed_category[i]]
self.parents_by_category[cat_id] = assignments
self.has_parents[cat_id] = True
else:
vector = self.tfidf.content(term_id)
assignments = [i for i in ints[1:] if self.observed_category[i]]
self.parents_by_concept[term_id] = assignments
for a_id in assignments:
for w, weight in vector:
if self.tfidf.df[w] < self.min_df:
continue
#print term, term_id, self.concept_by_id[a_id], w, self.vector_by_category[a_id][w], '\t+%f' % weight
self.vector_by_category[a_id][w] += weight
self.contributors_by_category[a_id].update([term_id])
if __name__ == '__main__':
import random
from random import shuffle as randomize
tfidf = TfIdfMatrix()
tfidf.load_features('bkp.big.out/vector.term.csv')
tfidf.load_distribution('bkp.big.out/vector.index.csv')
# tfidf.load_features('vector.term.csv')
# tfidf.load_distribution('vector.index.csv')
ctree = CategoryTree(
'bkp.big.out/category.index.csv',
'bkp.big.out/term.csv',
'bkp.big.out/category.csv',
# 'category.index.csv',
# 'term.csv',
# 'category.csv',
tfidf,
max_depth=1
)
ctree.load()
ctree.build()
ctree.dump()
<｜fim▁end｜> | load |
<|file_name|>CategoryTree.py<|end_file_name|><｜fim▁begin｜>import json
from collections import (
Counter,
defaultdict as deft
)
from copy import deepcopy as cp
# from cPickle import (
# dump as to_pickle,
# load as from_pickle
# )
from StringIO import StringIO
from TfIdfMatrix import TfIdfMatrix
from Tools import from_csv
class CategoryTree:
def __init__(self, categories_by_concept, terms,
categories, tfidf, max_depth=5, min_df=20
):
self.min_df = min_df
self.path_categories_by_concept = categories_by_concept
self.path_categories = categories
self.path_terms = terms
self.max_depth = max_depth
self.observed_category = deft(bool)
self.id_by_concept = dict([])
self.concept_by_id = dict([])
self.term_is_category = deft(bool)
self.parents_by_category = dict([])
self.parents_by_concept = deft(list)
self.id_by_term = dict([])
self.term_by_id = dict([])
self.has_parents = deft(bool)
self.tfidf = tfidf
self.pulling = set([])
self.vector_by_category = deft(Counter)
self.contributors_by_category = deft(set)
self.projected = Counter()
def build(self):
for i, c in enumerate(self.concept_by_id.values()):
self(c)
if not i % 100:
t = float(len(self.concept_by_id.keys()))
print i, int(t), round(i / t, 2)
# if i >= 5000:
# break
def dump(self):
# Simulate a file with StringIO
out = open('vector.dump.txt', 'wb')
for i, (_id, projections) in enumerate(self.projected.items()):
if not i % 100:
print i, len(self.projected.keys())
if not projections:
continue
features = [
(self.tfidf.word_by_id[wid], round(weight, 4))
for wid, weight in self.vector_by_category[_id].most_common()
if round(weight, 4)
]
record = (
_id,
self.concept_by_id[_id],
features
)
out.write('%s\n' % str(record))
out.close()
def __call__(self, category):
self.pulling = set([])
return self.__pull(None, 0, category, dict([]))
def __get_parents(self, _id):
parents = []
name = self.concept_by_id[_id]
if (
not self.observed_category[name] or
not self.observed_category[_id] or
not self.has_parents[_id]
):
return []
else:
for i in self.parents_by_category[_id]:
if not self.observed_category[i]:
continue
_name = self.concept_by_id[i]
parents.append(_name)
return set(parents) - self.pulling
def __pull(self, vector, depth, category, tree):
_id = self.id_by_concept[category]
if not self.pulling:
# print
# print
# print category, _id
# print [self.term_by_id[x] for x in self.contributors_by_category[_id]]
# print self.vector_by_category[_id].most_common(20)
vector = self.vector_by_category[_id]
if not self.observed_category[category]:
return dict([])
parents = self.__get_parents(_id)
if not parents or depth >= self.max_depth:
tree[category] = dict([])
else:
subtree = dict([])
self.pulling.update(parents)
for parent in parents:
subtree = self.__pull(vector, depth + 1, parent, subtree)
tree[category] = subtree
self.__project(vector, tree)
return tree
def __project(self, vector, tree):
if not tree.keys():
return
else:
for key, subtree in tree.items():
_id = self.id_by_concept[key]
self.projected[_id] += 1
self.__add2vec(vector, _id)
self.__project(vector, subtree)
def __add2vec(self, vector, _id):
# for w, weight in vector.items():
# __id = self.tfidf.id_by_word[w]
for __id, weight in vector.items():
self.vector_by_category[_id][__id] += weight
def load(self):
self.__load_terms()
self.__load_categories()
self.__load_assignments()
def <|fim_middle|>(self):
for concept, _id in from_csv(self.path_categories):
_id = int(_id)
self.id_by_concept[concept] = _id
self.concept_by_id[_id] = concept
self.observed_category[_id] = True
self.observed_category[concept] = True
# print concept, _id, len(self.id_by_concept.keys())
# exit()
def __load_terms(self):
for term, _id in from_csv(self.path_terms):
_id = int(_id)
self.term_by_id[_id] = term
self.id_by_term[term] = _id
if not term.startswith('Category:'):
continue
self.term_is_category[term] = True
self.term_is_category[_id] = True
def __load_assignments(self):
for row in from_csv(self.path_categories_by_concept):
ints = [int(field) for field in row]
term_id = ints[0]
term = self.term_by_id[term_id]
if self.term_is_category[term_id] and \
self.observed_category[term]:
term = self.term_by_id[term_id]
cat_id = self.id_by_concept[term]
assignments = [i for i in ints[1:] if self.observed_category[i]]
self.parents_by_category[cat_id] = assignments
self.has_parents[cat_id] = True
else:
vector = self.tfidf.content(term_id)
assignments = [i for i in ints[1:] if self.observed_category[i]]
self.parents_by_concept[term_id] = assignments
for a_id in assignments:
for w, weight in vector:
if self.tfidf.df[w] < self.min_df:
continue
#print term, term_id, self.concept_by_id[a_id], w, self.vector_by_category[a_id][w], '\t+%f' % weight
self.vector_by_category[a_id][w] += weight
self.contributors_by_category[a_id].update([term_id])
if __name__ == '__main__':
import random
from random import shuffle as randomize
tfidf = TfIdfMatrix()
tfidf.load_features('bkp.big.out/vector.term.csv')
tfidf.load_distribution('bkp.big.out/vector.index.csv')
# tfidf.load_features('vector.term.csv')
# tfidf.load_distribution('vector.index.csv')
ctree = CategoryTree(
'bkp.big.out/category.index.csv',
'bkp.big.out/term.csv',
'bkp.big.out/category.csv',
# 'category.index.csv',
# 'term.csv',
# 'category.csv',
tfidf,
max_depth=1
)
ctree.load()
ctree.build()
ctree.dump()
<｜fim▁end｜> | __load_categories |
<|file_name|>CategoryTree.py<|end_file_name|><｜fim▁begin｜>import json
from collections import (
Counter,
defaultdict as deft
)
from copy import deepcopy as cp
# from cPickle import (
# dump as to_pickle,
# load as from_pickle
# )
from StringIO import StringIO
from TfIdfMatrix import TfIdfMatrix
from Tools import from_csv
class CategoryTree:
def __init__(self, categories_by_concept, terms,
categories, tfidf, max_depth=5, min_df=20
):
self.min_df = min_df
self.path_categories_by_concept = categories_by_concept
self.path_categories = categories
self.path_terms = terms
self.max_depth = max_depth
self.observed_category = deft(bool)
self.id_by_concept = dict([])
self.concept_by_id = dict([])
self.term_is_category = deft(bool)
self.parents_by_category = dict([])
self.parents_by_concept = deft(list)
self.id_by_term = dict([])
self.term_by_id = dict([])
self.has_parents = deft(bool)
self.tfidf = tfidf
self.pulling = set([])
self.vector_by_category = deft(Counter)
self.contributors_by_category = deft(set)
self.projected = Counter()
def build(self):
for i, c in enumerate(self.concept_by_id.values()):
self(c)
if not i % 100:
t = float(len(self.concept_by_id.keys()))
print i, int(t), round(i / t, 2)
# if i >= 5000:
# break
def dump(self):
# Simulate a file with StringIO
out = open('vector.dump.txt', 'wb')
for i, (_id, projections) in enumerate(self.projected.items()):
if not i % 100:
print i, len(self.projected.keys())
if not projections:
continue
features = [
(self.tfidf.word_by_id[wid], round(weight, 4))
for wid, weight in self.vector_by_category[_id].most_common()
if round(weight, 4)
]
record = (
_id,
self.concept_by_id[_id],
features
)
out.write('%s\n' % str(record))
out.close()
def __call__(self, category):
self.pulling = set([])
return self.__pull(None, 0, category, dict([]))
def __get_parents(self, _id):
parents = []
name = self.concept_by_id[_id]
if (
not self.observed_category[name] or
not self.observed_category[_id] or
not self.has_parents[_id]
):
return []
else:
for i in self.parents_by_category[_id]:
if not self.observed_category[i]:
continue
_name = self.concept_by_id[i]
parents.append(_name)
return set(parents) - self.pulling
def __pull(self, vector, depth, category, tree):
_id = self.id_by_concept[category]
if not self.pulling:
# print
# print
# print category, _id
# print [self.term_by_id[x] for x in self.contributors_by_category[_id]]
# print self.vector_by_category[_id].most_common(20)
vector = self.vector_by_category[_id]
if not self.observed_category[category]:
return dict([])
parents = self.__get_parents(_id)
if not parents or depth >= self.max_depth:
tree[category] = dict([])
else:
subtree = dict([])
self.pulling.update(parents)
for parent in parents:
subtree = self.__pull(vector, depth + 1, parent, subtree)
tree[category] = subtree
self.__project(vector, tree)
return tree
def __project(self, vector, tree):
if not tree.keys():
return
else:
for key, subtree in tree.items():
_id = self.id_by_concept[key]
self.projected[_id] += 1
self.__add2vec(vector, _id)
self.__project(vector, subtree)
def __add2vec(self, vector, _id):
# for w, weight in vector.items():
# __id = self.tfidf.id_by_word[w]
for __id, weight in vector.items():
self.vector_by_category[_id][__id] += weight
def load(self):
self.__load_terms()
self.__load_categories()
self.__load_assignments()
def __load_categories(self):
for concept, _id in from_csv(self.path_categories):
_id = int(_id)
self.id_by_concept[concept] = _id
self.concept_by_id[_id] = concept
self.observed_category[_id] = True
self.observed_category[concept] = True
# print concept, _id, len(self.id_by_concept.keys())
# exit()
def <|fim_middle|>(self):
for term, _id in from_csv(self.path_terms):
_id = int(_id)
self.term_by_id[_id] = term
self.id_by_term[term] = _id
if not term.startswith('Category:'):
continue
self.term_is_category[term] = True
self.term_is_category[_id] = True
def __load_assignments(self):
for row in from_csv(self.path_categories_by_concept):
ints = [int(field) for field in row]
term_id = ints[0]
term = self.term_by_id[term_id]
if self.term_is_category[term_id] and \
self.observed_category[term]:
term = self.term_by_id[term_id]
cat_id = self.id_by_concept[term]
assignments = [i for i in ints[1:] if self.observed_category[i]]
self.parents_by_category[cat_id] = assignments
self.has_parents[cat_id] = True
else:
vector = self.tfidf.content(term_id)
assignments = [i for i in ints[1:] if self.observed_category[i]]
self.parents_by_concept[term_id] = assignments
for a_id in assignments:
for w, weight in vector:
if self.tfidf.df[w] < self.min_df:
continue
#print term, term_id, self.concept_by_id[a_id], w, self.vector_by_category[a_id][w], '\t+%f' % weight
self.vector_by_category[a_id][w] += weight
self.contributors_by_category[a_id].update([term_id])
if __name__ == '__main__':
import random
from random import shuffle as randomize
tfidf = TfIdfMatrix()
tfidf.load_features('bkp.big.out/vector.term.csv')
tfidf.load_distribution('bkp.big.out/vector.index.csv')
# tfidf.load_features('vector.term.csv')
# tfidf.load_distribution('vector.index.csv')
ctree = CategoryTree(
'bkp.big.out/category.index.csv',
'bkp.big.out/term.csv',
'bkp.big.out/category.csv',
# 'category.index.csv',
# 'term.csv',
# 'category.csv',
tfidf,
max_depth=1
)
ctree.load()
ctree.build()
ctree.dump()
<｜fim▁end｜> | __load_terms |
<|file_name|>CategoryTree.py<|end_file_name|><｜fim▁begin｜>import json
from collections import (
Counter,
defaultdict as deft
)
from copy import deepcopy as cp
# from cPickle import (
# dump as to_pickle,
# load as from_pickle
# )
from StringIO import StringIO
from TfIdfMatrix import TfIdfMatrix
from Tools import from_csv
class CategoryTree:
def __init__(self, categories_by_concept, terms,
categories, tfidf, max_depth=5, min_df=20
):
self.min_df = min_df
self.path_categories_by_concept = categories_by_concept
self.path_categories = categories
self.path_terms = terms
self.max_depth = max_depth
self.observed_category = deft(bool)
self.id_by_concept = dict([])
self.concept_by_id = dict([])
self.term_is_category = deft(bool)
self.parents_by_category = dict([])
self.parents_by_concept = deft(list)
self.id_by_term = dict([])
self.term_by_id = dict([])
self.has_parents = deft(bool)
self.tfidf = tfidf
self.pulling = set([])
self.vector_by_category = deft(Counter)
self.contributors_by_category = deft(set)
self.projected = Counter()
def build(self):
for i, c in enumerate(self.concept_by_id.values()):
self(c)
if not i % 100:
t = float(len(self.concept_by_id.keys()))
print i, int(t), round(i / t, 2)
# if i >= 5000:
# break
def dump(self):
# Simulate a file with StringIO
out = open('vector.dump.txt', 'wb')
for i, (_id, projections) in enumerate(self.projected.items()):
if not i % 100:
print i, len(self.projected.keys())
if not projections:
continue
features = [
(self.tfidf.word_by_id[wid], round(weight, 4))
for wid, weight in self.vector_by_category[_id].most_common()
if round(weight, 4)
]
record = (
_id,
self.concept_by_id[_id],
features
)
out.write('%s\n' % str(record))
out.close()
def __call__(self, category):
self.pulling = set([])
return self.__pull(None, 0, category, dict([]))
def __get_parents(self, _id):
parents = []
name = self.concept_by_id[_id]
if (
not self.observed_category[name] or
not self.observed_category[_id] or
not self.has_parents[_id]
):
return []
else:
for i in self.parents_by_category[_id]:
if not self.observed_category[i]:
continue
_name = self.concept_by_id[i]
parents.append(_name)
return set(parents) - self.pulling
def __pull(self, vector, depth, category, tree):
_id = self.id_by_concept[category]
if not self.pulling:
# print
# print
# print category, _id
# print [self.term_by_id[x] for x in self.contributors_by_category[_id]]
# print self.vector_by_category[_id].most_common(20)
vector = self.vector_by_category[_id]
if not self.observed_category[category]:
return dict([])
parents = self.__get_parents(_id)
if not parents or depth >= self.max_depth:
tree[category] = dict([])
else:
subtree = dict([])
self.pulling.update(parents)
for parent in parents:
subtree = self.__pull(vector, depth + 1, parent, subtree)
tree[category] = subtree
self.__project(vector, tree)
return tree
def __project(self, vector, tree):
if not tree.keys():
return
else:
for key, subtree in tree.items():
_id = self.id_by_concept[key]
self.projected[_id] += 1
self.__add2vec(vector, _id)
self.__project(vector, subtree)
def __add2vec(self, vector, _id):
# for w, weight in vector.items():
# __id = self.tfidf.id_by_word[w]
for __id, weight in vector.items():
self.vector_by_category[_id][__id] += weight
def load(self):
self.__load_terms()
self.__load_categories()
self.__load_assignments()
def __load_categories(self):
for concept, _id in from_csv(self.path_categories):
_id = int(_id)
self.id_by_concept[concept] = _id
self.concept_by_id[_id] = concept
self.observed_category[_id] = True
self.observed_category[concept] = True
# print concept, _id, len(self.id_by_concept.keys())
# exit()
def __load_terms(self):
for term, _id in from_csv(self.path_terms):
_id = int(_id)
self.term_by_id[_id] = term
self.id_by_term[term] = _id
if not term.startswith('Category:'):
continue
self.term_is_category[term] = True
self.term_is_category[_id] = True
def <|fim_middle|>(self):
for row in from_csv(self.path_categories_by_concept):
ints = [int(field) for field in row]
term_id = ints[0]
term = self.term_by_id[term_id]
if self.term_is_category[term_id] and \
self.observed_category[term]:
term = self.term_by_id[term_id]
cat_id = self.id_by_concept[term]
assignments = [i for i in ints[1:] if self.observed_category[i]]
self.parents_by_category[cat_id] = assignments
self.has_parents[cat_id] = True
else:
vector = self.tfidf.content(term_id)
assignments = [i for i in ints[1:] if self.observed_category[i]]
self.parents_by_concept[term_id] = assignments
for a_id in assignments:
for w, weight in vector:
if self.tfidf.df[w] < self.min_df:
continue
#print term, term_id, self.concept_by_id[a_id], w, self.vector_by_category[a_id][w], '\t+%f' % weight
self.vector_by_category[a_id][w] += weight
self.contributors_by_category[a_id].update([term_id])
if __name__ == '__main__':
import random
from random import shuffle as randomize
tfidf = TfIdfMatrix()
tfidf.load_features('bkp.big.out/vector.term.csv')
tfidf.load_distribution('bkp.big.out/vector.index.csv')
# tfidf.load_features('vector.term.csv')
# tfidf.load_distribution('vector.index.csv')
ctree = CategoryTree(
'bkp.big.out/category.index.csv',
'bkp.big.out/term.csv',
'bkp.big.out/category.csv',
# 'category.index.csv',
# 'term.csv',
# 'category.csv',
tfidf,
max_depth=1
)
ctree.load()
ctree.build()
ctree.dump()
<｜fim▁end｜> | __load_assignments |
<|file_name|>console.py<|end_file_name|><｜fim▁begin｜>#!/usr/bin/env python
# -*- mode: python; sh-basic-offset: 4; indent-tabs-mode: nil; coding: utf-8 -*-
# vim: tabstop=4 softtabstop=4 expandtab shiftwidth=4 fileencoding=utf-8
#
# Shell command
# Copyright 2010, Jeremy Grosser <[email protected]>
import argparse
import os
import sys
import clusto
from clusto import script_helper
class Console(script_helper.Script):
'''
Use clusto's hardware port mappings to console to a remote server
using the serial console.
'''
def __init__(self):
script_helper.Script.__init__(self)
def _add_arguments(self, parser):
user = os.environ.get('USER')
parser.add_argument('--user', '-u', default=user,
help='SSH User (you can also set this in clusto.conf too'
'in console.user: --user > clusto.conf:console.user > "%s")' % user)
parser.add_argument('server', nargs=1,
help='Object to console to (IP or name)')
def add_subparser(self, subparsers):
parser = self._setup_subparser(subparsers)
self._add_arguments(parser)
def run(self, args):<｜fim▁hole｜> except Exception as e:
self.debug(e)
self.error('No object like "%s" was found' % args.server)
return 1
server = server[0]
if not hasattr(server, 'console'):
self.error('The object %s lacks a console method' % server.name)
return 2
user = os.environ.get('USER')
if args.user:
self.debug('Grabbing user from parameter')
user = args.user
else:
self.debug('Grabbing user from config file or default')
user = self.get_conf('console.user', user)
self.debug('User is "%s"' % user)
return(server.console(ssh_user=user))
def main():
console, args = script_helper.init_arguments(Console)
return(console.run(args))
if __name__ == '__main__':
sys.exit(main())<｜fim▁end｜> | try:
server = clusto.get(args.server[0])
if not server:
raise LookupError('Object "%s" does not exist' % args.server) |
<|file_name|>console.py<|end_file_name|><｜fim▁begin｜>#!/usr/bin/env python
# -*- mode: python; sh-basic-offset: 4; indent-tabs-mode: nil; coding: utf-8 -*-
# vim: tabstop=4 softtabstop=4 expandtab shiftwidth=4 fileencoding=utf-8
#
# Shell command
# Copyright 2010, Jeremy Grosser <[email protected]>
import argparse
import os
import sys
import clusto
from clusto import script_helper
class Console(script_helper.Script):
<|fim_middle|>
def main():
console, args = script_helper.init_arguments(Console)
return(console.run(args))
if __name__ == '__main__':
sys.exit(main())
<｜fim▁end｜> | '''
Use clusto's hardware port mappings to console to a remote server
using the serial console.
'''
def __init__(self):
script_helper.Script.__init__(self)
def _add_arguments(self, parser):
user = os.environ.get('USER')
parser.add_argument('--user', '-u', default=user,
help='SSH User (you can also set this in clusto.conf too'
'in console.user: --user > clusto.conf:console.user > "%s")' % user)
parser.add_argument('server', nargs=1,
help='Object to console to (IP or name)')
def add_subparser(self, subparsers):
parser = self._setup_subparser(subparsers)
self._add_arguments(parser)
def run(self, args):
try:
server = clusto.get(args.server[0])
if not server:
raise LookupError('Object "%s" does not exist' % args.server)
except Exception as e:
self.debug(e)
self.error('No object like "%s" was found' % args.server)
return 1
server = server[0]
if not hasattr(server, 'console'):
self.error('The object %s lacks a console method' % server.name)
return 2
user = os.environ.get('USER')
if args.user:
self.debug('Grabbing user from parameter')
user = args.user
else:
self.debug('Grabbing user from config file or default')
user = self.get_conf('console.user', user)
self.debug('User is "%s"' % user)
return(server.console(ssh_user=user)) |
<|file_name|>console.py<|end_file_name|><｜fim▁begin｜>#!/usr/bin/env python
# -*- mode: python; sh-basic-offset: 4; indent-tabs-mode: nil; coding: utf-8 -*-
# vim: tabstop=4 softtabstop=4 expandtab shiftwidth=4 fileencoding=utf-8
#
# Shell command
# Copyright 2010, Jeremy Grosser <[email protected]>
import argparse
import os
import sys
import clusto
from clusto import script_helper
class Console(script_helper.Script):
'''
Use clusto's hardware port mappings to console to a remote server
using the serial console.
'''
def __init__(self):
<|fim_middle|>
def _add_arguments(self, parser):
user = os.environ.get('USER')
parser.add_argument('--user', '-u', default=user,
help='SSH User (you can also set this in clusto.conf too'
'in console.user: --user > clusto.conf:console.user > "%s")' % user)
parser.add_argument('server', nargs=1,
help='Object to console to (IP or name)')
def add_subparser(self, subparsers):
parser = self._setup_subparser(subparsers)
self._add_arguments(parser)
def run(self, args):
try:
server = clusto.get(args.server[0])
if not server:
raise LookupError('Object "%s" does not exist' % args.server)
except Exception as e:
self.debug(e)
self.error('No object like "%s" was found' % args.server)
return 1
server = server[0]
if not hasattr(server, 'console'):
self.error('The object %s lacks a console method' % server.name)
return 2
user = os.environ.get('USER')
if args.user:
self.debug('Grabbing user from parameter')
user = args.user
else:
self.debug('Grabbing user from config file or default')
user = self.get_conf('console.user', user)
self.debug('User is "%s"' % user)
return(server.console(ssh_user=user))
def main():
console, args = script_helper.init_arguments(Console)
return(console.run(args))
if __name__ == '__main__':
sys.exit(main())
<｜fim▁end｜> | script_helper.Script.__init__(self) |
<|file_name|>console.py<|end_file_name|><｜fim▁begin｜>#!/usr/bin/env python
# -*- mode: python; sh-basic-offset: 4; indent-tabs-mode: nil; coding: utf-8 -*-
# vim: tabstop=4 softtabstop=4 expandtab shiftwidth=4 fileencoding=utf-8
#
# Shell command
# Copyright 2010, Jeremy Grosser <[email protected]>
import argparse
import os
import sys
import clusto
from clusto import script_helper
class Console(script_helper.Script):
'''
Use clusto's hardware port mappings to console to a remote server
using the serial console.
'''
def __init__(self):
script_helper.Script.__init__(self)
def _add_arguments(self, parser):
<|fim_middle|>
def add_subparser(self, subparsers):
parser = self._setup_subparser(subparsers)
self._add_arguments(parser)
def run(self, args):
try:
server = clusto.get(args.server[0])
if not server:
raise LookupError('Object "%s" does not exist' % args.server)
except Exception as e:
self.debug(e)
self.error('No object like "%s" was found' % args.server)
return 1
server = server[0]
if not hasattr(server, 'console'):
self.error('The object %s lacks a console method' % server.name)
return 2
user = os.environ.get('USER')
if args.user:
self.debug('Grabbing user from parameter')
user = args.user
else:
self.debug('Grabbing user from config file or default')
user = self.get_conf('console.user', user)
self.debug('User is "%s"' % user)
return(server.console(ssh_user=user))
def main():
console, args = script_helper.init_arguments(Console)
return(console.run(args))
if __name__ == '__main__':
sys.exit(main())
<｜fim▁end｜> | user = os.environ.get('USER')
parser.add_argument('--user', '-u', default=user,
help='SSH User (you can also set this in clusto.conf too'
'in console.user: --user > clusto.conf:console.user > "%s")' % user)
parser.add_argument('server', nargs=1,
help='Object to console to (IP or name)') |
<|file_name|>console.py<|end_file_name|><｜fim▁begin｜>#!/usr/bin/env python
# -*- mode: python; sh-basic-offset: 4; indent-tabs-mode: nil; coding: utf-8 -*-
# vim: tabstop=4 softtabstop=4 expandtab shiftwidth=4 fileencoding=utf-8
#
# Shell command
# Copyright 2010, Jeremy Grosser <[email protected]>
import argparse
import os
import sys
import clusto
from clusto import script_helper
class Console(script_helper.Script):
'''
Use clusto's hardware port mappings to console to a remote server
using the serial console.
'''
def __init__(self):
script_helper.Script.__init__(self)
def _add_arguments(self, parser):
user = os.environ.get('USER')
parser.add_argument('--user', '-u', default=user,
help='SSH User (you can also set this in clusto.conf too'
'in console.user: --user > clusto.conf:console.user > "%s")' % user)
parser.add_argument('server', nargs=1,
help='Object to console to (IP or name)')
def add_subparser(self, subparsers):
<|fim_middle|>
def run(self, args):
try:
server = clusto.get(args.server[0])
if not server:
raise LookupError('Object "%s" does not exist' % args.server)
except Exception as e:
self.debug(e)
self.error('No object like "%s" was found' % args.server)
return 1
server = server[0]
if not hasattr(server, 'console'):
self.error('The object %s lacks a console method' % server.name)
return 2
user = os.environ.get('USER')
if args.user:
self.debug('Grabbing user from parameter')
user = args.user
else:
self.debug('Grabbing user from config file or default')
user = self.get_conf('console.user', user)
self.debug('User is "%s"' % user)
return(server.console(ssh_user=user))
def main():
console, args = script_helper.init_arguments(Console)
return(console.run(args))
if __name__ == '__main__':
sys.exit(main())
<｜fim▁end｜> | parser = self._setup_subparser(subparsers)
self._add_arguments(parser) |
<|file_name|>console.py<|end_file_name|><｜fim▁begin｜>#!/usr/bin/env python
# -*- mode: python; sh-basic-offset: 4; indent-tabs-mode: nil; coding: utf-8 -*-
# vim: tabstop=4 softtabstop=4 expandtab shiftwidth=4 fileencoding=utf-8
#
# Shell command
# Copyright 2010, Jeremy Grosser <[email protected]>
import argparse
import os
import sys
import clusto
from clusto import script_helper
class Console(script_helper.Script):
'''
Use clusto's hardware port mappings to console to a remote server
using the serial console.
'''
def __init__(self):
script_helper.Script.__init__(self)
def _add_arguments(self, parser):
user = os.environ.get('USER')
parser.add_argument('--user', '-u', default=user,
help='SSH User (you can also set this in clusto.conf too'
'in console.user: --user > clusto.conf:console.user > "%s")' % user)
parser.add_argument('server', nargs=1,
help='Object to console to (IP or name)')
def add_subparser(self, subparsers):
parser = self._setup_subparser(subparsers)
self._add_arguments(parser)
def run(self, args):
<|fim_middle|>
def main():
console, args = script_helper.init_arguments(Console)
return(console.run(args))
if __name__ == '__main__':
sys.exit(main())
<｜fim▁end｜> | try:
server = clusto.get(args.server[0])
if not server:
raise LookupError('Object "%s" does not exist' % args.server)
except Exception as e:
self.debug(e)
self.error('No object like "%s" was found' % args.server)
return 1
server = server[0]
if not hasattr(server, 'console'):
self.error('The object %s lacks a console method' % server.name)
return 2
user = os.environ.get('USER')
if args.user:
self.debug('Grabbing user from parameter')
user = args.user
else:
self.debug('Grabbing user from config file or default')
user = self.get_conf('console.user', user)
self.debug('User is "%s"' % user)
return(server.console(ssh_user=user)) |
<|file_name|>console.py<|end_file_name|><｜fim▁begin｜>#!/usr/bin/env python
# -*- mode: python; sh-basic-offset: 4; indent-tabs-mode: nil; coding: utf-8 -*-
# vim: tabstop=4 softtabstop=4 expandtab shiftwidth=4 fileencoding=utf-8
#
# Shell command
# Copyright 2010, Jeremy Grosser <[email protected]>
import argparse
import os
import sys
import clusto
from clusto import script_helper
class Console(script_helper.Script):
'''
Use clusto's hardware port mappings to console to a remote server
using the serial console.
'''
def __init__(self):
script_helper.Script.__init__(self)
def _add_arguments(self, parser):
user = os.environ.get('USER')
parser.add_argument('--user', '-u', default=user,
help='SSH User (you can also set this in clusto.conf too'
'in console.user: --user > clusto.conf:console.user > "%s")' % user)
parser.add_argument('server', nargs=1,
help='Object to console to (IP or name)')
def add_subparser(self, subparsers):
parser = self._setup_subparser(subparsers)
self._add_arguments(parser)
def run(self, args):
try:
server = clusto.get(args.server[0])
if not server:
raise LookupError('Object "%s" does not exist' % args.server)
except Exception as e:
self.debug(e)
self.error('No object like "%s" was found' % args.server)
return 1
server = server[0]
if not hasattr(server, 'console'):
self.error('The object %s lacks a console method' % server.name)
return 2
user = os.environ.get('USER')
if args.user:
self.debug('Grabbing user from parameter')
user = args.user
else:
self.debug('Grabbing user from config file or default')
user = self.get_conf('console.user', user)
self.debug('User is "%s"' % user)
return(server.console(ssh_user=user))
def main():
<|fim_middle|>
if __name__ == '__main__':
sys.exit(main())
<｜fim▁end｜> | console, args = script_helper.init_arguments(Console)
return(console.run(args)) |
<|file_name|>console.py<|end_file_name|><｜fim▁begin｜>#!/usr/bin/env python
# -*- mode: python; sh-basic-offset: 4; indent-tabs-mode: nil; coding: utf-8 -*-
# vim: tabstop=4 softtabstop=4 expandtab shiftwidth=4 fileencoding=utf-8
#
# Shell command
# Copyright 2010, Jeremy Grosser <[email protected]>
import argparse
import os
import sys
import clusto
from clusto import script_helper
class Console(script_helper.Script):
'''
Use clusto's hardware port mappings to console to a remote server
using the serial console.
'''
def __init__(self):
script_helper.Script.__init__(self)
def _add_arguments(self, parser):
user = os.environ.get('USER')
parser.add_argument('--user', '-u', default=user,
help='SSH User (you can also set this in clusto.conf too'
'in console.user: --user > clusto.conf:console.user > "%s")' % user)
parser.add_argument('server', nargs=1,
help='Object to console to (IP or name)')
def add_subparser(self, subparsers):
parser = self._setup_subparser(subparsers)
self._add_arguments(parser)
def run(self, args):
try:
server = clusto.get(args.server[0])
if not server:
<|fim_middle|>
except Exception as e:
self.debug(e)
self.error('No object like "%s" was found' % args.server)
return 1
server = server[0]
if not hasattr(server, 'console'):
self.error('The object %s lacks a console method' % server.name)
return 2
user = os.environ.get('USER')
if args.user:
self.debug('Grabbing user from parameter')
user = args.user
else:
self.debug('Grabbing user from config file or default')
user = self.get_conf('console.user', user)
self.debug('User is "%s"' % user)
return(server.console(ssh_user=user))
def main():
console, args = script_helper.init_arguments(Console)
return(console.run(args))
if __name__ == '__main__':
sys.exit(main())
<｜fim▁end｜> | raise LookupError('Object "%s" does not exist' % args.server) |
<|file_name|>console.py<|end_file_name|><｜fim▁begin｜>#!/usr/bin/env python
# -*- mode: python; sh-basic-offset: 4; indent-tabs-mode: nil; coding: utf-8 -*-
# vim: tabstop=4 softtabstop=4 expandtab shiftwidth=4 fileencoding=utf-8
#
# Shell command
# Copyright 2010, Jeremy Grosser <[email protected]>
import argparse
import os
import sys
import clusto
from clusto import script_helper
class Console(script_helper.Script):
'''
Use clusto's hardware port mappings to console to a remote server
using the serial console.
'''
def __init__(self):
script_helper.Script.__init__(self)
def _add_arguments(self, parser):
user = os.environ.get('USER')
parser.add_argument('--user', '-u', default=user,
help='SSH User (you can also set this in clusto.conf too'
'in console.user: --user > clusto.conf:console.user > "%s")' % user)
parser.add_argument('server', nargs=1,
help='Object to console to (IP or name)')
def add_subparser(self, subparsers):
parser = self._setup_subparser(subparsers)
self._add_arguments(parser)
def run(self, args):
try:
server = clusto.get(args.server[0])
if not server:
raise LookupError('Object "%s" does not exist' % args.server)
except Exception as e:
self.debug(e)
self.error('No object like "%s" was found' % args.server)
return 1
server = server[0]
if not hasattr(server, 'console'):
<|fim_middle|>
user = os.environ.get('USER')
if args.user:
self.debug('Grabbing user from parameter')
user = args.user
else:
self.debug('Grabbing user from config file or default')
user = self.get_conf('console.user', user)
self.debug('User is "%s"' % user)
return(server.console(ssh_user=user))
def main():
console, args = script_helper.init_arguments(Console)
return(console.run(args))
if __name__ == '__main__':
sys.exit(main())
<｜fim▁end｜> | self.error('The object %s lacks a console method' % server.name)
return 2 |
<|file_name|>console.py<|end_file_name|><｜fim▁begin｜>#!/usr/bin/env python
# -*- mode: python; sh-basic-offset: 4; indent-tabs-mode: nil; coding: utf-8 -*-
# vim: tabstop=4 softtabstop=4 expandtab shiftwidth=4 fileencoding=utf-8
#
# Shell command
# Copyright 2010, Jeremy Grosser <[email protected]>
import argparse
import os
import sys
import clusto
from clusto import script_helper
class Console(script_helper.Script):
'''
Use clusto's hardware port mappings to console to a remote server
using the serial console.
'''
def __init__(self):
script_helper.Script.__init__(self)
def _add_arguments(self, parser):
user = os.environ.get('USER')
parser.add_argument('--user', '-u', default=user,
help='SSH User (you can also set this in clusto.conf too'
'in console.user: --user > clusto.conf:console.user > "%s")' % user)
parser.add_argument('server', nargs=1,
help='Object to console to (IP or name)')
def add_subparser(self, subparsers):
parser = self._setup_subparser(subparsers)
self._add_arguments(parser)
def run(self, args):
try:
server = clusto.get(args.server[0])
if not server:
raise LookupError('Object "%s" does not exist' % args.server)
except Exception as e:
self.debug(e)
self.error('No object like "%s" was found' % args.server)
return 1
server = server[0]
if not hasattr(server, 'console'):
self.error('The object %s lacks a console method' % server.name)
return 2
user = os.environ.get('USER')
if args.user:
<|fim_middle|>
else:
self.debug('Grabbing user from config file or default')
user = self.get_conf('console.user', user)
self.debug('User is "%s"' % user)
return(server.console(ssh_user=user))
def main():
console, args = script_helper.init_arguments(Console)
return(console.run(args))
if __name__ == '__main__':
sys.exit(main())
<｜fim▁end｜> | self.debug('Grabbing user from parameter')
user = args.user |
<|file_name|>console.py<|end_file_name|><｜fim▁begin｜>#!/usr/bin/env python
# -*- mode: python; sh-basic-offset: 4; indent-tabs-mode: nil; coding: utf-8 -*-
# vim: tabstop=4 softtabstop=4 expandtab shiftwidth=4 fileencoding=utf-8
#
# Shell command
# Copyright 2010, Jeremy Grosser <[email protected]>
import argparse
import os
import sys
import clusto
from clusto import script_helper
class Console(script_helper.Script):
'''
Use clusto's hardware port mappings to console to a remote server
using the serial console.
'''
def __init__(self):
script_helper.Script.__init__(self)
def _add_arguments(self, parser):
user = os.environ.get('USER')
parser.add_argument('--user', '-u', default=user,
help='SSH User (you can also set this in clusto.conf too'
'in console.user: --user > clusto.conf:console.user > "%s")' % user)
parser.add_argument('server', nargs=1,
help='Object to console to (IP or name)')
def add_subparser(self, subparsers):
parser = self._setup_subparser(subparsers)
self._add_arguments(parser)
def run(self, args):
try:
server = clusto.get(args.server[0])
if not server:
raise LookupError('Object "%s" does not exist' % args.server)
except Exception as e:
self.debug(e)
self.error('No object like "%s" was found' % args.server)
return 1
server = server[0]
if not hasattr(server, 'console'):
self.error('The object %s lacks a console method' % server.name)
return 2
user = os.environ.get('USER')
if args.user:
self.debug('Grabbing user from parameter')
user = args.user
else:
<|fim_middle|>
self.debug('User is "%s"' % user)
return(server.console(ssh_user=user))
def main():
console, args = script_helper.init_arguments(Console)
return(console.run(args))
if __name__ == '__main__':
sys.exit(main())
<｜fim▁end｜> | self.debug('Grabbing user from config file or default')
user = self.get_conf('console.user', user) |
<|file_name|>console.py<|end_file_name|><｜fim▁begin｜>#!/usr/bin/env python
# -*- mode: python; sh-basic-offset: 4; indent-tabs-mode: nil; coding: utf-8 -*-
# vim: tabstop=4 softtabstop=4 expandtab shiftwidth=4 fileencoding=utf-8
#
# Shell command
# Copyright 2010, Jeremy Grosser <[email protected]>
import argparse
import os
import sys
import clusto
from clusto import script_helper
class Console(script_helper.Script):
'''
Use clusto's hardware port mappings to console to a remote server
using the serial console.
'''
def __init__(self):
script_helper.Script.__init__(self)
def _add_arguments(self, parser):
user = os.environ.get('USER')
parser.add_argument('--user', '-u', default=user,
help='SSH User (you can also set this in clusto.conf too'
'in console.user: --user > clusto.conf:console.user > "%s")' % user)
parser.add_argument('server', nargs=1,
help='Object to console to (IP or name)')
def add_subparser(self, subparsers):
parser = self._setup_subparser(subparsers)
self._add_arguments(parser)
def run(self, args):
try:
server = clusto.get(args.server[0])
if not server:
raise LookupError('Object "%s" does not exist' % args.server)
except Exception as e:
self.debug(e)
self.error('No object like "%s" was found' % args.server)
return 1
server = server[0]
if not hasattr(server, 'console'):
self.error('The object %s lacks a console method' % server.name)
return 2
user = os.environ.get('USER')
if args.user:
self.debug('Grabbing user from parameter')
user = args.user
else:
self.debug('Grabbing user from config file or default')
user = self.get_conf('console.user', user)
self.debug('User is "%s"' % user)
return(server.console(ssh_user=user))
def main():
console, args = script_helper.init_arguments(Console)
return(console.run(args))
if __name__ == '__main__':
<|fim_middle|>
<｜fim▁end｜> | sys.exit(main()) |
<|file_name|>console.py<|end_file_name|><｜fim▁begin｜>#!/usr/bin/env python
# -*- mode: python; sh-basic-offset: 4; indent-tabs-mode: nil; coding: utf-8 -*-
# vim: tabstop=4 softtabstop=4 expandtab shiftwidth=4 fileencoding=utf-8
#
# Shell command
# Copyright 2010, Jeremy Grosser <[email protected]>
import argparse
import os
import sys
import clusto
from clusto import script_helper
class Console(script_helper.Script):
'''
Use clusto's hardware port mappings to console to a remote server
using the serial console.
'''
def <|fim_middle|>(self):
script_helper.Script.__init__(self)
def _add_arguments(self, parser):
user = os.environ.get('USER')
parser.add_argument('--user', '-u', default=user,
help='SSH User (you can also set this in clusto.conf too'
'in console.user: --user > clusto.conf:console.user > "%s")' % user)
parser.add_argument('server', nargs=1,
help='Object to console to (IP or name)')
def add_subparser(self, subparsers):
parser = self._setup_subparser(subparsers)
self._add_arguments(parser)
def run(self, args):
try:
server = clusto.get(args.server[0])
if not server:
raise LookupError('Object "%s" does not exist' % args.server)
except Exception as e:
self.debug(e)
self.error('No object like "%s" was found' % args.server)
return 1
server = server[0]
if not hasattr(server, 'console'):
self.error('The object %s lacks a console method' % server.name)
return 2
user = os.environ.get('USER')
if args.user:
self.debug('Grabbing user from parameter')
user = args.user
else:
self.debug('Grabbing user from config file or default')
user = self.get_conf('console.user', user)
self.debug('User is "%s"' % user)
return(server.console(ssh_user=user))
def main():
console, args = script_helper.init_arguments(Console)
return(console.run(args))
if __name__ == '__main__':
sys.exit(main())
<｜fim▁end｜> | __init__ |
<|file_name|>console.py<|end_file_name|><｜fim▁begin｜>#!/usr/bin/env python
# -*- mode: python; sh-basic-offset: 4; indent-tabs-mode: nil; coding: utf-8 -*-
# vim: tabstop=4 softtabstop=4 expandtab shiftwidth=4 fileencoding=utf-8
#
# Shell command
# Copyright 2010, Jeremy Grosser <[email protected]>
import argparse
import os
import sys
import clusto
from clusto import script_helper
class Console(script_helper.Script):
'''
Use clusto's hardware port mappings to console to a remote server
using the serial console.
'''
def __init__(self):
script_helper.Script.__init__(self)
def <|fim_middle|>(self, parser):
user = os.environ.get('USER')
parser.add_argument('--user', '-u', default=user,
help='SSH User (you can also set this in clusto.conf too'
'in console.user: --user > clusto.conf:console.user > "%s")' % user)
parser.add_argument('server', nargs=1,
help='Object to console to (IP or name)')
def add_subparser(self, subparsers):
parser = self._setup_subparser(subparsers)
self._add_arguments(parser)
def run(self, args):
try:
server = clusto.get(args.server[0])
if not server:
raise LookupError('Object "%s" does not exist' % args.server)
except Exception as e:
self.debug(e)
self.error('No object like "%s" was found' % args.server)
return 1
server = server[0]
if not hasattr(server, 'console'):
self.error('The object %s lacks a console method' % server.name)
return 2
user = os.environ.get('USER')
if args.user:
self.debug('Grabbing user from parameter')
user = args.user
else:
self.debug('Grabbing user from config file or default')
user = self.get_conf('console.user', user)
self.debug('User is "%s"' % user)
return(server.console(ssh_user=user))
def main():
console, args = script_helper.init_arguments(Console)
return(console.run(args))
if __name__ == '__main__':
sys.exit(main())
<｜fim▁end｜> | _add_arguments |
<|file_name|>console.py<|end_file_name|><｜fim▁begin｜>#!/usr/bin/env python
# -*- mode: python; sh-basic-offset: 4; indent-tabs-mode: nil; coding: utf-8 -*-
# vim: tabstop=4 softtabstop=4 expandtab shiftwidth=4 fileencoding=utf-8
#
# Shell command
# Copyright 2010, Jeremy Grosser <[email protected]>
import argparse
import os
import sys
import clusto
from clusto import script_helper
class Console(script_helper.Script):
'''
Use clusto's hardware port mappings to console to a remote server
using the serial console.
'''
def __init__(self):
script_helper.Script.__init__(self)
def _add_arguments(self, parser):
user = os.environ.get('USER')
parser.add_argument('--user', '-u', default=user,
help='SSH User (you can also set this in clusto.conf too'
'in console.user: --user > clusto.conf:console.user > "%s")' % user)
parser.add_argument('server', nargs=1,
help='Object to console to (IP or name)')
def <|fim_middle|>(self, subparsers):
parser = self._setup_subparser(subparsers)
self._add_arguments(parser)
def run(self, args):
try:
server = clusto.get(args.server[0])
if not server:
raise LookupError('Object "%s" does not exist' % args.server)
except Exception as e:
self.debug(e)
self.error('No object like "%s" was found' % args.server)
return 1
server = server[0]
if not hasattr(server, 'console'):
self.error('The object %s lacks a console method' % server.name)
return 2
user = os.environ.get('USER')
if args.user:
self.debug('Grabbing user from parameter')
user = args.user
else:
self.debug('Grabbing user from config file or default')
user = self.get_conf('console.user', user)
self.debug('User is "%s"' % user)
return(server.console(ssh_user=user))
def main():
console, args = script_helper.init_arguments(Console)
return(console.run(args))
if __name__ == '__main__':
sys.exit(main())
<｜fim▁end｜> | add_subparser |
<|file_name|>console.py<|end_file_name|><｜fim▁begin｜>#!/usr/bin/env python
# -*- mode: python; sh-basic-offset: 4; indent-tabs-mode: nil; coding: utf-8 -*-
# vim: tabstop=4 softtabstop=4 expandtab shiftwidth=4 fileencoding=utf-8
#
# Shell command
# Copyright 2010, Jeremy Grosser <[email protected]>
import argparse
import os
import sys
import clusto
from clusto import script_helper
class Console(script_helper.Script):
'''
Use clusto's hardware port mappings to console to a remote server
using the serial console.
'''
def __init__(self):
script_helper.Script.__init__(self)
def _add_arguments(self, parser):
user = os.environ.get('USER')
parser.add_argument('--user', '-u', default=user,
help='SSH User (you can also set this in clusto.conf too'
'in console.user: --user > clusto.conf:console.user > "%s")' % user)
parser.add_argument('server', nargs=1,
help='Object to console to (IP or name)')
def add_subparser(self, subparsers):
parser = self._setup_subparser(subparsers)
self._add_arguments(parser)
def <|fim_middle|>(self, args):
try:
server = clusto.get(args.server[0])
if not server:
raise LookupError('Object "%s" does not exist' % args.server)
except Exception as e:
self.debug(e)
self.error('No object like "%s" was found' % args.server)
return 1
server = server[0]
if not hasattr(server, 'console'):
self.error('The object %s lacks a console method' % server.name)
return 2
user = os.environ.get('USER')
if args.user:
self.debug('Grabbing user from parameter')
user = args.user
else:
self.debug('Grabbing user from config file or default')
user = self.get_conf('console.user', user)
self.debug('User is "%s"' % user)
return(server.console(ssh_user=user))
def main():
console, args = script_helper.init_arguments(Console)
return(console.run(args))
if __name__ == '__main__':
sys.exit(main())
<｜fim▁end｜> | run |
<|file_name|>console.py<|end_file_name|><｜fim▁begin｜>#!/usr/bin/env python
# -*- mode: python; sh-basic-offset: 4; indent-tabs-mode: nil; coding: utf-8 -*-
# vim: tabstop=4 softtabstop=4 expandtab shiftwidth=4 fileencoding=utf-8
#
# Shell command
# Copyright 2010, Jeremy Grosser <[email protected]>
import argparse
import os
import sys
import clusto
from clusto import script_helper
class Console(script_helper.Script):
'''
Use clusto's hardware port mappings to console to a remote server
using the serial console.
'''
def __init__(self):
script_helper.Script.__init__(self)
def _add_arguments(self, parser):
user = os.environ.get('USER')
parser.add_argument('--user', '-u', default=user,
help='SSH User (you can also set this in clusto.conf too'
'in console.user: --user > clusto.conf:console.user > "%s")' % user)
parser.add_argument('server', nargs=1,
help='Object to console to (IP or name)')
def add_subparser(self, subparsers):
parser = self._setup_subparser(subparsers)
self._add_arguments(parser)
def run(self, args):
try:
server = clusto.get(args.server[0])
if not server:
raise LookupError('Object "%s" does not exist' % args.server)
except Exception as e:
self.debug(e)
self.error('No object like "%s" was found' % args.server)
return 1
server = server[0]
if not hasattr(server, 'console'):
self.error('The object %s lacks a console method' % server.name)
return 2
user = os.environ.get('USER')
if args.user:
self.debug('Grabbing user from parameter')
user = args.user
else:
self.debug('Grabbing user from config file or default')
user = self.get_conf('console.user', user)
self.debug('User is "%s"' % user)
return(server.console(ssh_user=user))
def <|fim_middle|>():
console, args = script_helper.init_arguments(Console)
return(console.run(args))
if __name__ == '__main__':
sys.exit(main())
<｜fim▁end｜> | main |
<|file_name|>example.py<|end_file_name|><｜fim▁begin｜>from Robinhood import Robinhood
#Setup
my_trader = Robinhood(username="YOUR_USERNAME", password="YOUR_PASSWORD");
#Get stock information
#Note: Sometimes more than one instrument may be returned for a given stock symbol
stock_instrument = my_trader.instruments("GEVO")[0]
#Get a stock's quote
my_trader.print_quote("AAPL")
#Prompt for a symbol
my_trader.print_quote();
#Print multiple symbols
my_trader.print_quotes(stocks=["BBRY", "FB", "MSFT"])<｜fim▁hole｜>#View all data for a given stock ie. Ask price and size, bid price and size, previous close, adjusted previous close, etc.
quote_info = my_trader.quote_data("GEVO")
print(quote_info);
#Place a buy order (uses market bid price)
buy_order = my_trader.place_buy_order(stock_instrument, 1)
#Place a sell order
sell_order = my_trader.place_sell_order(stock_instrument, 1)<｜fim▁end｜> | |
<|file_name|>sessions_api.py<|end_file_name|><｜fim▁begin｜>import requests
from flask import session, Blueprint, redirect
from flask import request
from grano import authz
from grano.lib.exc import BadRequest
from grano.lib.serialisation import jsonify
from grano.views.cache import validate_cache
from grano.core import db, url_for, app
from grano.providers import github, twitter, facebook
from grano.model import Account
from grano.logic import accounts
blueprint = Blueprint('sessions_api', __name__)
@blueprint.route('/api/1/sessions', methods=['GET'])
def status():
permissions = {}
if authz.logged_in():
for permission in request.account.permissions:
permissions[permission.project.slug] = {
'reader': permission.reader,
'editor': permission.editor,
'admin': permission.admin
}
keys = {
'p': repr(permissions),
'i': request.account.id if authz.logged_in() else None
}
validate_cache(keys=keys)
return jsonify({
'logged_in': authz.logged_in(),
'api_key': request.account.api_key if authz.logged_in() else None,
'account': request.account if request.account else None,
'permissions': permissions
})
def provider_not_enabled(name):
return jsonify({
'status': 501,
'name': 'Provider not configured: %s' % name,
'message': 'There are no OAuth credentials given for %s' % name,
}, status=501)
@blueprint.route('/api/1/sessions/logout', methods=['GET'])
def logout():
#authz.require(authz.logged_in())
session.clear()
return redirect(request.args.get('next_url', '/'))
@blueprint.route('/api/1/sessions/login/github', methods=['GET'])
def github_login():
if not app.config.get('GITHUB_CLIENT_ID'):
return provider_not_enabled('github')
callback=url_for('sessions_api.github_authorized')
session.clear()
if not request.args.get('next_url'):
raise BadRequest("No 'next_url' is specified.")
session['next_url'] = request.args.get('next_url')
return github.authorize(callback=callback)
@blueprint.route('/api/1/sessions/callback/github', methods=['GET'])
@github.authorized_handler
def github_authorized(resp):
next_url = session.get('next_url', '/')
if resp is None or not 'access_token' in resp:
return redirect(next_url)
access_token = resp['access_token']
session['access_token'] = access_token, ''
res = requests.get('https://api.github.com/user?access_token=%s' % access_token,
verify=False)
data = res.json()
account = Account.by_github_id(data.get('id'))
data_ = {
'full_name': data.get('name'),
'login': data.get('login'),
'email': data.get('email'),
'github_id': data.get('id')
}
account = accounts.save(data_, account=account)
db.session.commit()
session['id'] = account.id
return redirect(next_url)
@blueprint.route('/api/1/sessions/login/twitter', methods=['GET'])
def twitter_login():
if not app.config.get('TWITTER_API_KEY'):
return provider_not_enabled('twitter')
callback=url_for('sessions_api.twitter_authorized')
session.clear()
if not request.args.get('next_url'):
raise BadRequest("No 'next_url' is specified.")
session['next_url'] = request.args.get('next_url')
return twitter.authorize(callback=callback)
<｜fim▁hole｜>@blueprint.route('/api/1/sessions/callback/twitter', methods=['GET'])
@twitter.authorized_handler
def twitter_authorized(resp):
next_url = session.get('next_url', '/')
if resp is None or not 'oauth_token' in resp:
return redirect(next_url)
session['twitter_token'] = (resp['oauth_token'],
resp['oauth_token_secret'])
res = twitter.get('users/show.json?user_id=%s' % resp.get('user_id'))
account = Account.by_twitter_id(res.data.get('id'))
data_ = {
'full_name': res.data.get('name'),
'login': res.data.get('screen_name'),
'twitter_id': res.data.get('id')
}
account = accounts.save(data_, account=account)
db.session.commit()
session['id'] = account.id
return redirect(next_url)
@blueprint.route('/api/1/sessions/login/facebook', methods=['GET'])
def facebook_login():
if not app.config.get('FACEBOOK_APP_ID'):
return provider_not_enabled('facebook')
callback=url_for('sessions_api.facebook_authorized')
session.clear()
if not request.args.get('next_url'):
raise BadRequest("No 'next_url' is specified.")
session['next_url'] = request.args.get('next_url')
return facebook.authorize(callback=callback)
@blueprint.route('/api/1/sessions/callback/facebook', methods=['GET'])
@facebook.authorized_handler
def facebook_authorized(resp):
next_url = session.get('next_url', '/')
if resp is None or not 'access_token' in resp:
return redirect(next_url)
session['facebook_token'] = (resp.get('access_token'), '')
data = facebook.get('/me').data
account = Account.by_facebook_id(data.get('id'))
data_ = {
'full_name': data.get('name'),
'login': data.get('username'),
'email': data.get('email'),
'facebook_id': data.get('id')
}
account = accounts.save(data_, account=account)
db.session.commit()
session['id'] = account.id
return redirect(next_url)<｜fim▁end｜> | |
<|file_name|>sessions_api.py<|end_file_name|><｜fim▁begin｜>import requests
from flask import session, Blueprint, redirect
from flask import request
from grano import authz
from grano.lib.exc import BadRequest
from grano.lib.serialisation import jsonify
from grano.views.cache import validate_cache
from grano.core import db, url_for, app
from grano.providers import github, twitter, facebook
from grano.model import Account
from grano.logic import accounts
blueprint = Blueprint('sessions_api', __name__)
@blueprint.route('/api/1/sessions', methods=['GET'])
def status():
<|fim_middle|>
def provider_not_enabled(name):
return jsonify({
'status': 501,
'name': 'Provider not configured: %s' % name,
'message': 'There are no OAuth credentials given for %s' % name,
}, status=501)
@blueprint.route('/api/1/sessions/logout', methods=['GET'])
def logout():
#authz.require(authz.logged_in())
session.clear()
return redirect(request.args.get('next_url', '/'))
@blueprint.route('/api/1/sessions/login/github', methods=['GET'])
def github_login():
if not app.config.get('GITHUB_CLIENT_ID'):
return provider_not_enabled('github')
callback=url_for('sessions_api.github_authorized')
session.clear()
if not request.args.get('next_url'):
raise BadRequest("No 'next_url' is specified.")
session['next_url'] = request.args.get('next_url')
return github.authorize(callback=callback)
@blueprint.route('/api/1/sessions/callback/github', methods=['GET'])
@github.authorized_handler
def github_authorized(resp):
next_url = session.get('next_url', '/')
if resp is None or not 'access_token' in resp:
return redirect(next_url)
access_token = resp['access_token']
session['access_token'] = access_token, ''
res = requests.get('https://api.github.com/user?access_token=%s' % access_token,
verify=False)
data = res.json()
account = Account.by_github_id(data.get('id'))
data_ = {
'full_name': data.get('name'),
'login': data.get('login'),
'email': data.get('email'),
'github_id': data.get('id')
}
account = accounts.save(data_, account=account)
db.session.commit()
session['id'] = account.id
return redirect(next_url)
@blueprint.route('/api/1/sessions/login/twitter', methods=['GET'])
def twitter_login():
if not app.config.get('TWITTER_API_KEY'):
return provider_not_enabled('twitter')
callback=url_for('sessions_api.twitter_authorized')
session.clear()
if not request.args.get('next_url'):
raise BadRequest("No 'next_url' is specified.")
session['next_url'] = request.args.get('next_url')
return twitter.authorize(callback=callback)
@blueprint.route('/api/1/sessions/callback/twitter', methods=['GET'])
@twitter.authorized_handler
def twitter_authorized(resp):
next_url = session.get('next_url', '/')
if resp is None or not 'oauth_token' in resp:
return redirect(next_url)
session['twitter_token'] = (resp['oauth_token'],
resp['oauth_token_secret'])
res = twitter.get('users/show.json?user_id=%s' % resp.get('user_id'))
account = Account.by_twitter_id(res.data.get('id'))
data_ = {
'full_name': res.data.get('name'),
'login': res.data.get('screen_name'),
'twitter_id': res.data.get('id')
}
account = accounts.save(data_, account=account)
db.session.commit()
session['id'] = account.id
return redirect(next_url)
@blueprint.route('/api/1/sessions/login/facebook', methods=['GET'])
def facebook_login():
if not app.config.get('FACEBOOK_APP_ID'):
return provider_not_enabled('facebook')
callback=url_for('sessions_api.facebook_authorized')
session.clear()
if not request.args.get('next_url'):
raise BadRequest("No 'next_url' is specified.")
session['next_url'] = request.args.get('next_url')
return facebook.authorize(callback=callback)
@blueprint.route('/api/1/sessions/callback/facebook', methods=['GET'])
@facebook.authorized_handler
def facebook_authorized(resp):
next_url = session.get('next_url', '/')
if resp is None or not 'access_token' in resp:
return redirect(next_url)
session['facebook_token'] = (resp.get('access_token'), '')
data = facebook.get('/me').data
account = Account.by_facebook_id(data.get('id'))
data_ = {
'full_name': data.get('name'),
'login': data.get('username'),
'email': data.get('email'),
'facebook_id': data.get('id')
}
account = accounts.save(data_, account=account)
db.session.commit()
session['id'] = account.id
return redirect(next_url)
<｜fim▁end｜> | permissions = {}
if authz.logged_in():
for permission in request.account.permissions:
permissions[permission.project.slug] = {
'reader': permission.reader,
'editor': permission.editor,
'admin': permission.admin
}
keys = {
'p': repr(permissions),
'i': request.account.id if authz.logged_in() else None
}
validate_cache(keys=keys)
return jsonify({
'logged_in': authz.logged_in(),
'api_key': request.account.api_key if authz.logged_in() else None,
'account': request.account if request.account else None,
'permissions': permissions
}) |
<|file_name|>sessions_api.py<|end_file_name|><｜fim▁begin｜>import requests
from flask import session, Blueprint, redirect
from flask import request
from grano import authz
from grano.lib.exc import BadRequest
from grano.lib.serialisation import jsonify
from grano.views.cache import validate_cache
from grano.core import db, url_for, app
from grano.providers import github, twitter, facebook
from grano.model import Account
from grano.logic import accounts
blueprint = Blueprint('sessions_api', __name__)
@blueprint.route('/api/1/sessions', methods=['GET'])
def status():
permissions = {}
if authz.logged_in():
for permission in request.account.permissions:
permissions[permission.project.slug] = {
'reader': permission.reader,
'editor': permission.editor,
'admin': permission.admin
}
keys = {
'p': repr(permissions),
'i': request.account.id if authz.logged_in() else None
}
validate_cache(keys=keys)
return jsonify({
'logged_in': authz.logged_in(),
'api_key': request.account.api_key if authz.logged_in() else None,
'account': request.account if request.account else None,
'permissions': permissions
})
def provider_not_enabled(name):
<|fim_middle|>
@blueprint.route('/api/1/sessions/logout', methods=['GET'])
def logout():
#authz.require(authz.logged_in())
session.clear()
return redirect(request.args.get('next_url', '/'))
@blueprint.route('/api/1/sessions/login/github', methods=['GET'])
def github_login():
if not app.config.get('GITHUB_CLIENT_ID'):
return provider_not_enabled('github')
callback=url_for('sessions_api.github_authorized')
session.clear()
if not request.args.get('next_url'):
raise BadRequest("No 'next_url' is specified.")
session['next_url'] = request.args.get('next_url')
return github.authorize(callback=callback)
@blueprint.route('/api/1/sessions/callback/github', methods=['GET'])
@github.authorized_handler
def github_authorized(resp):
next_url = session.get('next_url', '/')
if resp is None or not 'access_token' in resp:
return redirect(next_url)
access_token = resp['access_token']
session['access_token'] = access_token, ''
res = requests.get('https://api.github.com/user?access_token=%s' % access_token,
verify=False)
data = res.json()
account = Account.by_github_id(data.get('id'))
data_ = {
'full_name': data.get('name'),
'login': data.get('login'),
'email': data.get('email'),
'github_id': data.get('id')
}
account = accounts.save(data_, account=account)
db.session.commit()
session['id'] = account.id
return redirect(next_url)
@blueprint.route('/api/1/sessions/login/twitter', methods=['GET'])
def twitter_login():
if not app.config.get('TWITTER_API_KEY'):
return provider_not_enabled('twitter')
callback=url_for('sessions_api.twitter_authorized')
session.clear()
if not request.args.get('next_url'):
raise BadRequest("No 'next_url' is specified.")
session['next_url'] = request.args.get('next_url')
return twitter.authorize(callback=callback)
@blueprint.route('/api/1/sessions/callback/twitter', methods=['GET'])
@twitter.authorized_handler
def twitter_authorized(resp):
next_url = session.get('next_url', '/')
if resp is None or not 'oauth_token' in resp:
return redirect(next_url)
session['twitter_token'] = (resp['oauth_token'],
resp['oauth_token_secret'])
res = twitter.get('users/show.json?user_id=%s' % resp.get('user_id'))
account = Account.by_twitter_id(res.data.get('id'))
data_ = {
'full_name': res.data.get('name'),
'login': res.data.get('screen_name'),
'twitter_id': res.data.get('id')
}
account = accounts.save(data_, account=account)
db.session.commit()
session['id'] = account.id
return redirect(next_url)
@blueprint.route('/api/1/sessions/login/facebook', methods=['GET'])
def facebook_login():
if not app.config.get('FACEBOOK_APP_ID'):
return provider_not_enabled('facebook')
callback=url_for('sessions_api.facebook_authorized')
session.clear()
if not request.args.get('next_url'):
raise BadRequest("No 'next_url' is specified.")
session['next_url'] = request.args.get('next_url')
return facebook.authorize(callback=callback)
@blueprint.route('/api/1/sessions/callback/facebook', methods=['GET'])
@facebook.authorized_handler
def facebook_authorized(resp):
next_url = session.get('next_url', '/')
if resp is None or not 'access_token' in resp:
return redirect(next_url)
session['facebook_token'] = (resp.get('access_token'), '')
data = facebook.get('/me').data
account = Account.by_facebook_id(data.get('id'))
data_ = {
'full_name': data.get('name'),
'login': data.get('username'),
'email': data.get('email'),
'facebook_id': data.get('id')
}
account = accounts.save(data_, account=account)
db.session.commit()
session['id'] = account.id
return redirect(next_url)
<｜fim▁end｜> | return jsonify({
'status': 501,
'name': 'Provider not configured: %s' % name,
'message': 'There are no OAuth credentials given for %s' % name,
}, status=501) |
<|file_name|>sessions_api.py<|end_file_name|><｜fim▁begin｜>import requests
from flask import session, Blueprint, redirect
from flask import request
from grano import authz
from grano.lib.exc import BadRequest
from grano.lib.serialisation import jsonify
from grano.views.cache import validate_cache
from grano.core import db, url_for, app
from grano.providers import github, twitter, facebook
from grano.model import Account
from grano.logic import accounts
blueprint = Blueprint('sessions_api', __name__)
@blueprint.route('/api/1/sessions', methods=['GET'])
def status():
permissions = {}
if authz.logged_in():
for permission in request.account.permissions:
permissions[permission.project.slug] = {
'reader': permission.reader,
'editor': permission.editor,
'admin': permission.admin
}
keys = {
'p': repr(permissions),
'i': request.account.id if authz.logged_in() else None
}
validate_cache(keys=keys)
return jsonify({
'logged_in': authz.logged_in(),
'api_key': request.account.api_key if authz.logged_in() else None,
'account': request.account if request.account else None,
'permissions': permissions
})
def provider_not_enabled(name):
return jsonify({
'status': 501,
'name': 'Provider not configured: %s' % name,
'message': 'There are no OAuth credentials given for %s' % name,
}, status=501)
@blueprint.route('/api/1/sessions/logout', methods=['GET'])
def logout():
#authz.require(authz.logged_in())
<|fim_middle|>
@blueprint.route('/api/1/sessions/login/github', methods=['GET'])
def github_login():
if not app.config.get('GITHUB_CLIENT_ID'):
return provider_not_enabled('github')
callback=url_for('sessions_api.github_authorized')
session.clear()
if not request.args.get('next_url'):
raise BadRequest("No 'next_url' is specified.")
session['next_url'] = request.args.get('next_url')
return github.authorize(callback=callback)
@blueprint.route('/api/1/sessions/callback/github', methods=['GET'])
@github.authorized_handler
def github_authorized(resp):
next_url = session.get('next_url', '/')
if resp is None or not 'access_token' in resp:
return redirect(next_url)
access_token = resp['access_token']
session['access_token'] = access_token, ''
res = requests.get('https://api.github.com/user?access_token=%s' % access_token,
verify=False)
data = res.json()
account = Account.by_github_id(data.get('id'))
data_ = {
'full_name': data.get('name'),
'login': data.get('login'),
'email': data.get('email'),
'github_id': data.get('id')
}
account = accounts.save(data_, account=account)
db.session.commit()
session['id'] = account.id
return redirect(next_url)
@blueprint.route('/api/1/sessions/login/twitter', methods=['GET'])
def twitter_login():
if not app.config.get('TWITTER_API_KEY'):
return provider_not_enabled('twitter')
callback=url_for('sessions_api.twitter_authorized')
session.clear()
if not request.args.get('next_url'):
raise BadRequest("No 'next_url' is specified.")
session['next_url'] = request.args.get('next_url')
return twitter.authorize(callback=callback)
@blueprint.route('/api/1/sessions/callback/twitter', methods=['GET'])
@twitter.authorized_handler
def twitter_authorized(resp):
next_url = session.get('next_url', '/')
if resp is None or not 'oauth_token' in resp:
return redirect(next_url)
session['twitter_token'] = (resp['oauth_token'],
resp['oauth_token_secret'])
res = twitter.get('users/show.json?user_id=%s' % resp.get('user_id'))
account = Account.by_twitter_id(res.data.get('id'))
data_ = {
'full_name': res.data.get('name'),
'login': res.data.get('screen_name'),
'twitter_id': res.data.get('id')
}
account = accounts.save(data_, account=account)
db.session.commit()
session['id'] = account.id
return redirect(next_url)
@blueprint.route('/api/1/sessions/login/facebook', methods=['GET'])
def facebook_login():
if not app.config.get('FACEBOOK_APP_ID'):
return provider_not_enabled('facebook')
callback=url_for('sessions_api.facebook_authorized')
session.clear()
if not request.args.get('next_url'):
raise BadRequest("No 'next_url' is specified.")
session['next_url'] = request.args.get('next_url')
return facebook.authorize(callback=callback)
@blueprint.route('/api/1/sessions/callback/facebook', methods=['GET'])
@facebook.authorized_handler
def facebook_authorized(resp):
next_url = session.get('next_url', '/')
if resp is None or not 'access_token' in resp:
return redirect(next_url)
session['facebook_token'] = (resp.get('access_token'), '')
data = facebook.get('/me').data
account = Account.by_facebook_id(data.get('id'))
data_ = {
'full_name': data.get('name'),
'login': data.get('username'),
'email': data.get('email'),
'facebook_id': data.get('id')
}
account = accounts.save(data_, account=account)
db.session.commit()
session['id'] = account.id
return redirect(next_url)
<｜fim▁end｜> | session.clear()
return redirect(request.args.get('next_url', '/')) |
<|file_name|>sessions_api.py<|end_file_name|><｜fim▁begin｜>import requests
from flask import session, Blueprint, redirect
from flask import request
from grano import authz
from grano.lib.exc import BadRequest
from grano.lib.serialisation import jsonify
from grano.views.cache import validate_cache
from grano.core import db, url_for, app
from grano.providers import github, twitter, facebook
from grano.model import Account
from grano.logic import accounts
blueprint = Blueprint('sessions_api', __name__)
@blueprint.route('/api/1/sessions', methods=['GET'])
def status():
permissions = {}
if authz.logged_in():
for permission in request.account.permissions:
permissions[permission.project.slug] = {
'reader': permission.reader,
'editor': permission.editor,
'admin': permission.admin
}
keys = {
'p': repr(permissions),
'i': request.account.id if authz.logged_in() else None
}
validate_cache(keys=keys)
return jsonify({
'logged_in': authz.logged_in(),
'api_key': request.account.api_key if authz.logged_in() else None,
'account': request.account if request.account else None,
'permissions': permissions
})
def provider_not_enabled(name):
return jsonify({
'status': 501,
'name': 'Provider not configured: %s' % name,
'message': 'There are no OAuth credentials given for %s' % name,
}, status=501)
@blueprint.route('/api/1/sessions/logout', methods=['GET'])
def logout():
#authz.require(authz.logged_in())
session.clear()
return redirect(request.args.get('next_url', '/'))
@blueprint.route('/api/1/sessions/login/github', methods=['GET'])
def github_login():
<|fim_middle|>
@blueprint.route('/api/1/sessions/callback/github', methods=['GET'])
@github.authorized_handler
def github_authorized(resp):
next_url = session.get('next_url', '/')
if resp is None or not 'access_token' in resp:
return redirect(next_url)
access_token = resp['access_token']
session['access_token'] = access_token, ''
res = requests.get('https://api.github.com/user?access_token=%s' % access_token,
verify=False)
data = res.json()
account = Account.by_github_id(data.get('id'))
data_ = {
'full_name': data.get('name'),
'login': data.get('login'),
'email': data.get('email'),
'github_id': data.get('id')
}
account = accounts.save(data_, account=account)
db.session.commit()
session['id'] = account.id
return redirect(next_url)
@blueprint.route('/api/1/sessions/login/twitter', methods=['GET'])
def twitter_login():
if not app.config.get('TWITTER_API_KEY'):
return provider_not_enabled('twitter')
callback=url_for('sessions_api.twitter_authorized')
session.clear()
if not request.args.get('next_url'):
raise BadRequest("No 'next_url' is specified.")
session['next_url'] = request.args.get('next_url')
return twitter.authorize(callback=callback)
@blueprint.route('/api/1/sessions/callback/twitter', methods=['GET'])
@twitter.authorized_handler
def twitter_authorized(resp):
next_url = session.get('next_url', '/')
if resp is None or not 'oauth_token' in resp:
return redirect(next_url)
session['twitter_token'] = (resp['oauth_token'],
resp['oauth_token_secret'])
res = twitter.get('users/show.json?user_id=%s' % resp.get('user_id'))
account = Account.by_twitter_id(res.data.get('id'))
data_ = {
'full_name': res.data.get('name'),
'login': res.data.get('screen_name'),
'twitter_id': res.data.get('id')
}
account = accounts.save(data_, account=account)
db.session.commit()
session['id'] = account.id
return redirect(next_url)
@blueprint.route('/api/1/sessions/login/facebook', methods=['GET'])
def facebook_login():
if not app.config.get('FACEBOOK_APP_ID'):
return provider_not_enabled('facebook')
callback=url_for('sessions_api.facebook_authorized')
session.clear()
if not request.args.get('next_url'):
raise BadRequest("No 'next_url' is specified.")
session['next_url'] = request.args.get('next_url')
return facebook.authorize(callback=callback)
@blueprint.route('/api/1/sessions/callback/facebook', methods=['GET'])
@facebook.authorized_handler
def facebook_authorized(resp):
next_url = session.get('next_url', '/')
if resp is None or not 'access_token' in resp:
return redirect(next_url)
session['facebook_token'] = (resp.get('access_token'), '')
data = facebook.get('/me').data
account = Account.by_facebook_id(data.get('id'))
data_ = {
'full_name': data.get('name'),
'login': data.get('username'),
'email': data.get('email'),
'facebook_id': data.get('id')
}
account = accounts.save(data_, account=account)
db.session.commit()
session['id'] = account.id
return redirect(next_url)
<｜fim▁end｜> | if not app.config.get('GITHUB_CLIENT_ID'):
return provider_not_enabled('github')
callback=url_for('sessions_api.github_authorized')
session.clear()
if not request.args.get('next_url'):
raise BadRequest("No 'next_url' is specified.")
session['next_url'] = request.args.get('next_url')
return github.authorize(callback=callback) |
<|file_name|>sessions_api.py<|end_file_name|><｜fim▁begin｜>import requests
from flask import session, Blueprint, redirect
from flask import request
from grano import authz
from grano.lib.exc import BadRequest
from grano.lib.serialisation import jsonify
from grano.views.cache import validate_cache
from grano.core import db, url_for, app
from grano.providers import github, twitter, facebook
from grano.model import Account
from grano.logic import accounts
blueprint = Blueprint('sessions_api', __name__)
@blueprint.route('/api/1/sessions', methods=['GET'])
def status():
permissions = {}
if authz.logged_in():
for permission in request.account.permissions:
permissions[permission.project.slug] = {
'reader': permission.reader,
'editor': permission.editor,
'admin': permission.admin
}
keys = {
'p': repr(permissions),
'i': request.account.id if authz.logged_in() else None
}
validate_cache(keys=keys)
return jsonify({
'logged_in': authz.logged_in(),
'api_key': request.account.api_key if authz.logged_in() else None,
'account': request.account if request.account else None,
'permissions': permissions
})
def provider_not_enabled(name):
return jsonify({
'status': 501,
'name': 'Provider not configured: %s' % name,
'message': 'There are no OAuth credentials given for %s' % name,
}, status=501)
@blueprint.route('/api/1/sessions/logout', methods=['GET'])
def logout():
#authz.require(authz.logged_in())
session.clear()
return redirect(request.args.get('next_url', '/'))
@blueprint.route('/api/1/sessions/login/github', methods=['GET'])
def github_login():
if not app.config.get('GITHUB_CLIENT_ID'):
return provider_not_enabled('github')
callback=url_for('sessions_api.github_authorized')
session.clear()
if not request.args.get('next_url'):
raise BadRequest("No 'next_url' is specified.")
session['next_url'] = request.args.get('next_url')
return github.authorize(callback=callback)
@blueprint.route('/api/1/sessions/callback/github', methods=['GET'])
@github.authorized_handler
def github_authorized(resp):
<|fim_middle|>
@blueprint.route('/api/1/sessions/login/twitter', methods=['GET'])
def twitter_login():
if not app.config.get('TWITTER_API_KEY'):
return provider_not_enabled('twitter')
callback=url_for('sessions_api.twitter_authorized')
session.clear()
if not request.args.get('next_url'):
raise BadRequest("No 'next_url' is specified.")
session['next_url'] = request.args.get('next_url')
return twitter.authorize(callback=callback)
@blueprint.route('/api/1/sessions/callback/twitter', methods=['GET'])
@twitter.authorized_handler
def twitter_authorized(resp):
next_url = session.get('next_url', '/')
if resp is None or not 'oauth_token' in resp:
return redirect(next_url)
session['twitter_token'] = (resp['oauth_token'],
resp['oauth_token_secret'])
res = twitter.get('users/show.json?user_id=%s' % resp.get('user_id'))
account = Account.by_twitter_id(res.data.get('id'))
data_ = {
'full_name': res.data.get('name'),
'login': res.data.get('screen_name'),
'twitter_id': res.data.get('id')
}
account = accounts.save(data_, account=account)
db.session.commit()
session['id'] = account.id
return redirect(next_url)
@blueprint.route('/api/1/sessions/login/facebook', methods=['GET'])
def facebook_login():
if not app.config.get('FACEBOOK_APP_ID'):
return provider_not_enabled('facebook')
callback=url_for('sessions_api.facebook_authorized')
session.clear()
if not request.args.get('next_url'):
raise BadRequest("No 'next_url' is specified.")
session['next_url'] = request.args.get('next_url')
return facebook.authorize(callback=callback)
@blueprint.route('/api/1/sessions/callback/facebook', methods=['GET'])
@facebook.authorized_handler
def facebook_authorized(resp):
next_url = session.get('next_url', '/')
if resp is None or not 'access_token' in resp:
return redirect(next_url)
session['facebook_token'] = (resp.get('access_token'), '')
data = facebook.get('/me').data
account = Account.by_facebook_id(data.get('id'))
data_ = {
'full_name': data.get('name'),
'login': data.get('username'),
'email': data.get('email'),
'facebook_id': data.get('id')
}
account = accounts.save(data_, account=account)
db.session.commit()
session['id'] = account.id
return redirect(next_url)
<｜fim▁end｜> | next_url = session.get('next_url', '/')
if resp is None or not 'access_token' in resp:
return redirect(next_url)
access_token = resp['access_token']
session['access_token'] = access_token, ''
res = requests.get('https://api.github.com/user?access_token=%s' % access_token,
verify=False)
data = res.json()
account = Account.by_github_id(data.get('id'))
data_ = {
'full_name': data.get('name'),
'login': data.get('login'),
'email': data.get('email'),
'github_id': data.get('id')
}
account = accounts.save(data_, account=account)
db.session.commit()
session['id'] = account.id
return redirect(next_url) |
<|file_name|>sessions_api.py<|end_file_name|><｜fim▁begin｜>import requests
from flask import session, Blueprint, redirect
from flask import request
from grano import authz
from grano.lib.exc import BadRequest
from grano.lib.serialisation import jsonify
from grano.views.cache import validate_cache
from grano.core import db, url_for, app
from grano.providers import github, twitter, facebook
from grano.model import Account
from grano.logic import accounts
blueprint = Blueprint('sessions_api', __name__)
@blueprint.route('/api/1/sessions', methods=['GET'])
def status():
permissions = {}
if authz.logged_in():
for permission in request.account.permissions:
permissions[permission.project.slug] = {
'reader': permission.reader,
'editor': permission.editor,
'admin': permission.admin
}
keys = {
'p': repr(permissions),
'i': request.account.id if authz.logged_in() else None
}
validate_cache(keys=keys)
return jsonify({
'logged_in': authz.logged_in(),
'api_key': request.account.api_key if authz.logged_in() else None,
'account': request.account if request.account else None,
'permissions': permissions
})
def provider_not_enabled(name):
return jsonify({
'status': 501,
'name': 'Provider not configured: %s' % name,
'message': 'There are no OAuth credentials given for %s' % name,
}, status=501)
@blueprint.route('/api/1/sessions/logout', methods=['GET'])
def logout():
#authz.require(authz.logged_in())
session.clear()
return redirect(request.args.get('next_url', '/'))
@blueprint.route('/api/1/sessions/login/github', methods=['GET'])
def github_login():
if not app.config.get('GITHUB_CLIENT_ID'):
return provider_not_enabled('github')
callback=url_for('sessions_api.github_authorized')
session.clear()
if not request.args.get('next_url'):
raise BadRequest("No 'next_url' is specified.")
session['next_url'] = request.args.get('next_url')
return github.authorize(callback=callback)
@blueprint.route('/api/1/sessions/callback/github', methods=['GET'])
@github.authorized_handler
def github_authorized(resp):
next_url = session.get('next_url', '/')
if resp is None or not 'access_token' in resp:
return redirect(next_url)
access_token = resp['access_token']
session['access_token'] = access_token, ''
res = requests.get('https://api.github.com/user?access_token=%s' % access_token,
verify=False)
data = res.json()
account = Account.by_github_id(data.get('id'))
data_ = {
'full_name': data.get('name'),
'login': data.get('login'),
'email': data.get('email'),
'github_id': data.get('id')
}
account = accounts.save(data_, account=account)
db.session.commit()
session['id'] = account.id
return redirect(next_url)
@blueprint.route('/api/1/sessions/login/twitter', methods=['GET'])
def twitter_login():
<|fim_middle|>
@blueprint.route('/api/1/sessions/callback/twitter', methods=['GET'])
@twitter.authorized_handler
def twitter_authorized(resp):
next_url = session.get('next_url', '/')
if resp is None or not 'oauth_token' in resp:
return redirect(next_url)
session['twitter_token'] = (resp['oauth_token'],
resp['oauth_token_secret'])
res = twitter.get('users/show.json?user_id=%s' % resp.get('user_id'))
account = Account.by_twitter_id(res.data.get('id'))
data_ = {
'full_name': res.data.get('name'),
'login': res.data.get('screen_name'),
'twitter_id': res.data.get('id')
}
account = accounts.save(data_, account=account)
db.session.commit()
session['id'] = account.id
return redirect(next_url)
@blueprint.route('/api/1/sessions/login/facebook', methods=['GET'])
def facebook_login():
if not app.config.get('FACEBOOK_APP_ID'):
return provider_not_enabled('facebook')
callback=url_for('sessions_api.facebook_authorized')
session.clear()
if not request.args.get('next_url'):
raise BadRequest("No 'next_url' is specified.")
session['next_url'] = request.args.get('next_url')
return facebook.authorize(callback=callback)
@blueprint.route('/api/1/sessions/callback/facebook', methods=['GET'])
@facebook.authorized_handler
def facebook_authorized(resp):
next_url = session.get('next_url', '/')
if resp is None or not 'access_token' in resp:
return redirect(next_url)
session['facebook_token'] = (resp.get('access_token'), '')
data = facebook.get('/me').data
account = Account.by_facebook_id(data.get('id'))
data_ = {
'full_name': data.get('name'),
'login': data.get('username'),
'email': data.get('email'),
'facebook_id': data.get('id')
}
account = accounts.save(data_, account=account)
db.session.commit()
session['id'] = account.id
return redirect(next_url)
<｜fim▁end｜> | if not app.config.get('TWITTER_API_KEY'):
return provider_not_enabled('twitter')
callback=url_for('sessions_api.twitter_authorized')
session.clear()
if not request.args.get('next_url'):
raise BadRequest("No 'next_url' is specified.")
session['next_url'] = request.args.get('next_url')
return twitter.authorize(callback=callback) |
<|file_name|>sessions_api.py<|end_file_name|><｜fim▁begin｜>import requests
from flask import session, Blueprint, redirect
from flask import request
from grano import authz
from grano.lib.exc import BadRequest
from grano.lib.serialisation import jsonify
from grano.views.cache import validate_cache
from grano.core import db, url_for, app
from grano.providers import github, twitter, facebook
from grano.model import Account
from grano.logic import accounts
blueprint = Blueprint('sessions_api', __name__)
@blueprint.route('/api/1/sessions', methods=['GET'])
def status():
permissions = {}
if authz.logged_in():
for permission in request.account.permissions:
permissions[permission.project.slug] = {
'reader': permission.reader,
'editor': permission.editor,
'admin': permission.admin
}
keys = {
'p': repr(permissions),
'i': request.account.id if authz.logged_in() else None
}
validate_cache(keys=keys)
return jsonify({
'logged_in': authz.logged_in(),
'api_key': request.account.api_key if authz.logged_in() else None,
'account': request.account if request.account else None,
'permissions': permissions
})
def provider_not_enabled(name):
return jsonify({
'status': 501,
'name': 'Provider not configured: %s' % name,
'message': 'There are no OAuth credentials given for %s' % name,
}, status=501)
@blueprint.route('/api/1/sessions/logout', methods=['GET'])
def logout():
#authz.require(authz.logged_in())
session.clear()
return redirect(request.args.get('next_url', '/'))
@blueprint.route('/api/1/sessions/login/github', methods=['GET'])
def github_login():
if not app.config.get('GITHUB_CLIENT_ID'):
return provider_not_enabled('github')
callback=url_for('sessions_api.github_authorized')
session.clear()
if not request.args.get('next_url'):
raise BadRequest("No 'next_url' is specified.")
session['next_url'] = request.args.get('next_url')
return github.authorize(callback=callback)
@blueprint.route('/api/1/sessions/callback/github', methods=['GET'])
@github.authorized_handler
def github_authorized(resp):
next_url = session.get('next_url', '/')
if resp is None or not 'access_token' in resp:
return redirect(next_url)
access_token = resp['access_token']
session['access_token'] = access_token, ''
res = requests.get('https://api.github.com/user?access_token=%s' % access_token,
verify=False)
data = res.json()
account = Account.by_github_id(data.get('id'))
data_ = {
'full_name': data.get('name'),
'login': data.get('login'),
'email': data.get('email'),
'github_id': data.get('id')
}
account = accounts.save(data_, account=account)
db.session.commit()
session['id'] = account.id
return redirect(next_url)
@blueprint.route('/api/1/sessions/login/twitter', methods=['GET'])
def twitter_login():
if not app.config.get('TWITTER_API_KEY'):
return provider_not_enabled('twitter')
callback=url_for('sessions_api.twitter_authorized')
session.clear()
if not request.args.get('next_url'):
raise BadRequest("No 'next_url' is specified.")
session['next_url'] = request.args.get('next_url')
return twitter.authorize(callback=callback)
@blueprint.route('/api/1/sessions/callback/twitter', methods=['GET'])
@twitter.authorized_handler
def twitter_authorized(resp):
<|fim_middle|>
@blueprint.route('/api/1/sessions/login/facebook', methods=['GET'])
def facebook_login():
if not app.config.get('FACEBOOK_APP_ID'):
return provider_not_enabled('facebook')
callback=url_for('sessions_api.facebook_authorized')
session.clear()
if not request.args.get('next_url'):
raise BadRequest("No 'next_url' is specified.")
session['next_url'] = request.args.get('next_url')
return facebook.authorize(callback=callback)
@blueprint.route('/api/1/sessions/callback/facebook', methods=['GET'])
@facebook.authorized_handler
def facebook_authorized(resp):
next_url = session.get('next_url', '/')
if resp is None or not 'access_token' in resp:
return redirect(next_url)
session['facebook_token'] = (resp.get('access_token'), '')
data = facebook.get('/me').data
account = Account.by_facebook_id(data.get('id'))
data_ = {
'full_name': data.get('name'),
'login': data.get('username'),
'email': data.get('email'),
'facebook_id': data.get('id')
}
account = accounts.save(data_, account=account)
db.session.commit()
session['id'] = account.id
return redirect(next_url)
<｜fim▁end｜> | next_url = session.get('next_url', '/')
if resp is None or not 'oauth_token' in resp:
return redirect(next_url)
session['twitter_token'] = (resp['oauth_token'],
resp['oauth_token_secret'])
res = twitter.get('users/show.json?user_id=%s' % resp.get('user_id'))
account = Account.by_twitter_id(res.data.get('id'))
data_ = {
'full_name': res.data.get('name'),
'login': res.data.get('screen_name'),
'twitter_id': res.data.get('id')
}
account = accounts.save(data_, account=account)
db.session.commit()
session['id'] = account.id
return redirect(next_url) |
<|file_name|>sessions_api.py<|end_file_name|><｜fim▁begin｜>import requests
from flask import session, Blueprint, redirect
from flask import request
from grano import authz
from grano.lib.exc import BadRequest
from grano.lib.serialisation import jsonify
from grano.views.cache import validate_cache
from grano.core import db, url_for, app
from grano.providers import github, twitter, facebook
from grano.model import Account
from grano.logic import accounts
blueprint = Blueprint('sessions_api', __name__)
@blueprint.route('/api/1/sessions', methods=['GET'])
def status():
permissions = {}
if authz.logged_in():
for permission in request.account.permissions:
permissions[permission.project.slug] = {
'reader': permission.reader,
'editor': permission.editor,
'admin': permission.admin
}
keys = {
'p': repr(permissions),
'i': request.account.id if authz.logged_in() else None
}
validate_cache(keys=keys)
return jsonify({
'logged_in': authz.logged_in(),
'api_key': request.account.api_key if authz.logged_in() else None,
'account': request.account if request.account else None,
'permissions': permissions
})
def provider_not_enabled(name):
return jsonify({
'status': 501,
'name': 'Provider not configured: %s' % name,
'message': 'There are no OAuth credentials given for %s' % name,
}, status=501)
@blueprint.route('/api/1/sessions/logout', methods=['GET'])
def logout():
#authz.require(authz.logged_in())
session.clear()
return redirect(request.args.get('next_url', '/'))
@blueprint.route('/api/1/sessions/login/github', methods=['GET'])
def github_login():
if not app.config.get('GITHUB_CLIENT_ID'):
return provider_not_enabled('github')
callback=url_for('sessions_api.github_authorized')
session.clear()
if not request.args.get('next_url'):
raise BadRequest("No 'next_url' is specified.")
session['next_url'] = request.args.get('next_url')
return github.authorize(callback=callback)
@blueprint.route('/api/1/sessions/callback/github', methods=['GET'])
@github.authorized_handler
def github_authorized(resp):
next_url = session.get('next_url', '/')
if resp is None or not 'access_token' in resp:
return redirect(next_url)
access_token = resp['access_token']
session['access_token'] = access_token, ''
res = requests.get('https://api.github.com/user?access_token=%s' % access_token,
verify=False)
data = res.json()
account = Account.by_github_id(data.get('id'))
data_ = {
'full_name': data.get('name'),
'login': data.get('login'),
'email': data.get('email'),
'github_id': data.get('id')
}
account = accounts.save(data_, account=account)
db.session.commit()
session['id'] = account.id
return redirect(next_url)
@blueprint.route('/api/1/sessions/login/twitter', methods=['GET'])
def twitter_login():
if not app.config.get('TWITTER_API_KEY'):
return provider_not_enabled('twitter')
callback=url_for('sessions_api.twitter_authorized')
session.clear()
if not request.args.get('next_url'):
raise BadRequest("No 'next_url' is specified.")
session['next_url'] = request.args.get('next_url')
return twitter.authorize(callback=callback)
@blueprint.route('/api/1/sessions/callback/twitter', methods=['GET'])
@twitter.authorized_handler
def twitter_authorized(resp):
next_url = session.get('next_url', '/')
if resp is None or not 'oauth_token' in resp:
return redirect(next_url)
session['twitter_token'] = (resp['oauth_token'],
resp['oauth_token_secret'])
res = twitter.get('users/show.json?user_id=%s' % resp.get('user_id'))
account = Account.by_twitter_id(res.data.get('id'))
data_ = {
'full_name': res.data.get('name'),
'login': res.data.get('screen_name'),
'twitter_id': res.data.get('id')
}
account = accounts.save(data_, account=account)
db.session.commit()
session['id'] = account.id
return redirect(next_url)
@blueprint.route('/api/1/sessions/login/facebook', methods=['GET'])
def facebook_login():
<|fim_middle|>
@blueprint.route('/api/1/sessions/callback/facebook', methods=['GET'])
@facebook.authorized_handler
def facebook_authorized(resp):
next_url = session.get('next_url', '/')
if resp is None or not 'access_token' in resp:
return redirect(next_url)
session['facebook_token'] = (resp.get('access_token'), '')
data = facebook.get('/me').data
account = Account.by_facebook_id(data.get('id'))
data_ = {
'full_name': data.get('name'),
'login': data.get('username'),
'email': data.get('email'),
'facebook_id': data.get('id')
}
account = accounts.save(data_, account=account)
db.session.commit()
session['id'] = account.id
return redirect(next_url)
<｜fim▁end｜> | if not app.config.get('FACEBOOK_APP_ID'):
return provider_not_enabled('facebook')
callback=url_for('sessions_api.facebook_authorized')
session.clear()
if not request.args.get('next_url'):
raise BadRequest("No 'next_url' is specified.")
session['next_url'] = request.args.get('next_url')
return facebook.authorize(callback=callback) |
<|file_name|>sessions_api.py<|end_file_name|><｜fim▁begin｜>import requests
from flask import session, Blueprint, redirect
from flask import request
from grano import authz
from grano.lib.exc import BadRequest
from grano.lib.serialisation import jsonify
from grano.views.cache import validate_cache
from grano.core import db, url_for, app
from grano.providers import github, twitter, facebook
from grano.model import Account
from grano.logic import accounts
blueprint = Blueprint('sessions_api', __name__)
@blueprint.route('/api/1/sessions', methods=['GET'])
def status():
permissions = {}
if authz.logged_in():
for permission in request.account.permissions:
permissions[permission.project.slug] = {
'reader': permission.reader,
'editor': permission.editor,
'admin': permission.admin
}
keys = {
'p': repr(permissions),
'i': request.account.id if authz.logged_in() else None
}
validate_cache(keys=keys)
return jsonify({
'logged_in': authz.logged_in(),
'api_key': request.account.api_key if authz.logged_in() else None,
'account': request.account if request.account else None,
'permissions': permissions
})
def provider_not_enabled(name):
return jsonify({
'status': 501,
'name': 'Provider not configured: %s' % name,
'message': 'There are no OAuth credentials given for %s' % name,
}, status=501)
@blueprint.route('/api/1/sessions/logout', methods=['GET'])
def logout():
#authz.require(authz.logged_in())
session.clear()
return redirect(request.args.get('next_url', '/'))
@blueprint.route('/api/1/sessions/login/github', methods=['GET'])
def github_login():
if not app.config.get('GITHUB_CLIENT_ID'):
return provider_not_enabled('github')
callback=url_for('sessions_api.github_authorized')
session.clear()
if not request.args.get('next_url'):
raise BadRequest("No 'next_url' is specified.")
session['next_url'] = request.args.get('next_url')
return github.authorize(callback=callback)
@blueprint.route('/api/1/sessions/callback/github', methods=['GET'])
@github.authorized_handler
def github_authorized(resp):
next_url = session.get('next_url', '/')
if resp is None or not 'access_token' in resp:
return redirect(next_url)
access_token = resp['access_token']
session['access_token'] = access_token, ''
res = requests.get('https://api.github.com/user?access_token=%s' % access_token,
verify=False)
data = res.json()
account = Account.by_github_id(data.get('id'))
data_ = {
'full_name': data.get('name'),
'login': data.get('login'),
'email': data.get('email'),
'github_id': data.get('id')
}
account = accounts.save(data_, account=account)
db.session.commit()
session['id'] = account.id
return redirect(next_url)
@blueprint.route('/api/1/sessions/login/twitter', methods=['GET'])
def twitter_login():
if not app.config.get('TWITTER_API_KEY'):
return provider_not_enabled('twitter')
callback=url_for('sessions_api.twitter_authorized')
session.clear()
if not request.args.get('next_url'):
raise BadRequest("No 'next_url' is specified.")
session['next_url'] = request.args.get('next_url')
return twitter.authorize(callback=callback)
@blueprint.route('/api/1/sessions/callback/twitter', methods=['GET'])
@twitter.authorized_handler
def twitter_authorized(resp):
next_url = session.get('next_url', '/')
if resp is None or not 'oauth_token' in resp:
return redirect(next_url)
session['twitter_token'] = (resp['oauth_token'],
resp['oauth_token_secret'])
res = twitter.get('users/show.json?user_id=%s' % resp.get('user_id'))
account = Account.by_twitter_id(res.data.get('id'))
data_ = {
'full_name': res.data.get('name'),
'login': res.data.get('screen_name'),
'twitter_id': res.data.get('id')
}
account = accounts.save(data_, account=account)
db.session.commit()
session['id'] = account.id
return redirect(next_url)
@blueprint.route('/api/1/sessions/login/facebook', methods=['GET'])
def facebook_login():
if not app.config.get('FACEBOOK_APP_ID'):
return provider_not_enabled('facebook')
callback=url_for('sessions_api.facebook_authorized')
session.clear()
if not request.args.get('next_url'):
raise BadRequest("No 'next_url' is specified.")
session['next_url'] = request.args.get('next_url')
return facebook.authorize(callback=callback)
@blueprint.route('/api/1/sessions/callback/facebook', methods=['GET'])
@facebook.authorized_handler
def facebook_authorized(resp):
<|fim_middle|>
<｜fim▁end｜> | next_url = session.get('next_url', '/')
if resp is None or not 'access_token' in resp:
return redirect(next_url)
session['facebook_token'] = (resp.get('access_token'), '')
data = facebook.get('/me').data
account = Account.by_facebook_id(data.get('id'))
data_ = {
'full_name': data.get('name'),
'login': data.get('username'),
'email': data.get('email'),
'facebook_id': data.get('id')
}
account = accounts.save(data_, account=account)
db.session.commit()
session['id'] = account.id
return redirect(next_url) |
<|file_name|>sessions_api.py<|end_file_name|><｜fim▁begin｜>import requests
from flask import session, Blueprint, redirect
from flask import request
from grano import authz
from grano.lib.exc import BadRequest
from grano.lib.serialisation import jsonify
from grano.views.cache import validate_cache
from grano.core import db, url_for, app
from grano.providers import github, twitter, facebook
from grano.model import Account
from grano.logic import accounts
blueprint = Blueprint('sessions_api', __name__)
@blueprint.route('/api/1/sessions', methods=['GET'])
def status():
permissions = {}
if authz.logged_in():
<|fim_middle|>
keys = {
'p': repr(permissions),
'i': request.account.id if authz.logged_in() else None
}
validate_cache(keys=keys)
return jsonify({
'logged_in': authz.logged_in(),
'api_key': request.account.api_key if authz.logged_in() else None,
'account': request.account if request.account else None,
'permissions': permissions
})
def provider_not_enabled(name):
return jsonify({
'status': 501,
'name': 'Provider not configured: %s' % name,
'message': 'There are no OAuth credentials given for %s' % name,
}, status=501)
@blueprint.route('/api/1/sessions/logout', methods=['GET'])
def logout():
#authz.require(authz.logged_in())
session.clear()
return redirect(request.args.get('next_url', '/'))
@blueprint.route('/api/1/sessions/login/github', methods=['GET'])
def github_login():
if not app.config.get('GITHUB_CLIENT_ID'):
return provider_not_enabled('github')
callback=url_for('sessions_api.github_authorized')
session.clear()
if not request.args.get('next_url'):
raise BadRequest("No 'next_url' is specified.")
session['next_url'] = request.args.get('next_url')
return github.authorize(callback=callback)
@blueprint.route('/api/1/sessions/callback/github', methods=['GET'])
@github.authorized_handler
def github_authorized(resp):
next_url = session.get('next_url', '/')
if resp is None or not 'access_token' in resp:
return redirect(next_url)
access_token = resp['access_token']
session['access_token'] = access_token, ''
res = requests.get('https://api.github.com/user?access_token=%s' % access_token,
verify=False)
data = res.json()
account = Account.by_github_id(data.get('id'))
data_ = {
'full_name': data.get('name'),
'login': data.get('login'),
'email': data.get('email'),
'github_id': data.get('id')
}
account = accounts.save(data_, account=account)
db.session.commit()
session['id'] = account.id
return redirect(next_url)
@blueprint.route('/api/1/sessions/login/twitter', methods=['GET'])
def twitter_login():
if not app.config.get('TWITTER_API_KEY'):
return provider_not_enabled('twitter')
callback=url_for('sessions_api.twitter_authorized')
session.clear()
if not request.args.get('next_url'):
raise BadRequest("No 'next_url' is specified.")
session['next_url'] = request.args.get('next_url')
return twitter.authorize(callback=callback)
@blueprint.route('/api/1/sessions/callback/twitter', methods=['GET'])
@twitter.authorized_handler
def twitter_authorized(resp):
next_url = session.get('next_url', '/')
if resp is None or not 'oauth_token' in resp:
return redirect(next_url)
session['twitter_token'] = (resp['oauth_token'],
resp['oauth_token_secret'])
res = twitter.get('users/show.json?user_id=%s' % resp.get('user_id'))
account = Account.by_twitter_id(res.data.get('id'))
data_ = {
'full_name': res.data.get('name'),
'login': res.data.get('screen_name'),
'twitter_id': res.data.get('id')
}
account = accounts.save(data_, account=account)
db.session.commit()
session['id'] = account.id
return redirect(next_url)
@blueprint.route('/api/1/sessions/login/facebook', methods=['GET'])
def facebook_login():
if not app.config.get('FACEBOOK_APP_ID'):
return provider_not_enabled('facebook')
callback=url_for('sessions_api.facebook_authorized')
session.clear()
if not request.args.get('next_url'):
raise BadRequest("No 'next_url' is specified.")
session['next_url'] = request.args.get('next_url')
return facebook.authorize(callback=callback)
@blueprint.route('/api/1/sessions/callback/facebook', methods=['GET'])
@facebook.authorized_handler
def facebook_authorized(resp):
next_url = session.get('next_url', '/')
if resp is None or not 'access_token' in resp:
return redirect(next_url)
session['facebook_token'] = (resp.get('access_token'), '')
data = facebook.get('/me').data
account = Account.by_facebook_id(data.get('id'))
data_ = {
'full_name': data.get('name'),
'login': data.get('username'),
'email': data.get('email'),
'facebook_id': data.get('id')
}
account = accounts.save(data_, account=account)
db.session.commit()
session['id'] = account.id
return redirect(next_url)
<｜fim▁end｜> | for permission in request.account.permissions:
permissions[permission.project.slug] = {
'reader': permission.reader,
'editor': permission.editor,
'admin': permission.admin
} |
<|file_name|>sessions_api.py<|end_file_name|><｜fim▁begin｜>import requests
from flask import session, Blueprint, redirect
from flask import request
from grano import authz
from grano.lib.exc import BadRequest
from grano.lib.serialisation import jsonify
from grano.views.cache import validate_cache
from grano.core import db, url_for, app
from grano.providers import github, twitter, facebook
from grano.model import Account
from grano.logic import accounts
blueprint = Blueprint('sessions_api', __name__)
@blueprint.route('/api/1/sessions', methods=['GET'])
def status():
permissions = {}
if authz.logged_in():
for permission in request.account.permissions:
permissions[permission.project.slug] = {
'reader': permission.reader,
'editor': permission.editor,
'admin': permission.admin
}
keys = {
'p': repr(permissions),
'i': request.account.id if authz.logged_in() else None
}
validate_cache(keys=keys)
return jsonify({
'logged_in': authz.logged_in(),
'api_key': request.account.api_key if authz.logged_in() else None,
'account': request.account if request.account else None,
'permissions': permissions
})
def provider_not_enabled(name):
return jsonify({
'status': 501,
'name': 'Provider not configured: %s' % name,
'message': 'There are no OAuth credentials given for %s' % name,
}, status=501)
@blueprint.route('/api/1/sessions/logout', methods=['GET'])
def logout():
#authz.require(authz.logged_in())
session.clear()
return redirect(request.args.get('next_url', '/'))
@blueprint.route('/api/1/sessions/login/github', methods=['GET'])
def github_login():
if not app.config.get('GITHUB_CLIENT_ID'):
<|fim_middle|>
callback=url_for('sessions_api.github_authorized')
session.clear()
if not request.args.get('next_url'):
raise BadRequest("No 'next_url' is specified.")
session['next_url'] = request.args.get('next_url')
return github.authorize(callback=callback)
@blueprint.route('/api/1/sessions/callback/github', methods=['GET'])
@github.authorized_handler
def github_authorized(resp):
next_url = session.get('next_url', '/')
if resp is None or not 'access_token' in resp:
return redirect(next_url)
access_token = resp['access_token']
session['access_token'] = access_token, ''
res = requests.get('https://api.github.com/user?access_token=%s' % access_token,
verify=False)
data = res.json()
account = Account.by_github_id(data.get('id'))
data_ = {
'full_name': data.get('name'),
'login': data.get('login'),
'email': data.get('email'),
'github_id': data.get('id')
}
account = accounts.save(data_, account=account)
db.session.commit()
session['id'] = account.id
return redirect(next_url)
@blueprint.route('/api/1/sessions/login/twitter', methods=['GET'])
def twitter_login():
if not app.config.get('TWITTER_API_KEY'):
return provider_not_enabled('twitter')
callback=url_for('sessions_api.twitter_authorized')
session.clear()
if not request.args.get('next_url'):
raise BadRequest("No 'next_url' is specified.")
session['next_url'] = request.args.get('next_url')
return twitter.authorize(callback=callback)
@blueprint.route('/api/1/sessions/callback/twitter', methods=['GET'])
@twitter.authorized_handler
def twitter_authorized(resp):
next_url = session.get('next_url', '/')
if resp is None or not 'oauth_token' in resp:
return redirect(next_url)
session['twitter_token'] = (resp['oauth_token'],
resp['oauth_token_secret'])
res = twitter.get('users/show.json?user_id=%s' % resp.get('user_id'))
account = Account.by_twitter_id(res.data.get('id'))
data_ = {
'full_name': res.data.get('name'),
'login': res.data.get('screen_name'),
'twitter_id': res.data.get('id')
}
account = accounts.save(data_, account=account)
db.session.commit()
session['id'] = account.id
return redirect(next_url)
@blueprint.route('/api/1/sessions/login/facebook', methods=['GET'])
def facebook_login():
if not app.config.get('FACEBOOK_APP_ID'):
return provider_not_enabled('facebook')
callback=url_for('sessions_api.facebook_authorized')
session.clear()
if not request.args.get('next_url'):
raise BadRequest("No 'next_url' is specified.")
session['next_url'] = request.args.get('next_url')
return facebook.authorize(callback=callback)
@blueprint.route('/api/1/sessions/callback/facebook', methods=['GET'])
@facebook.authorized_handler
def facebook_authorized(resp):
next_url = session.get('next_url', '/')
if resp is None or not 'access_token' in resp:
return redirect(next_url)
session['facebook_token'] = (resp.get('access_token'), '')
data = facebook.get('/me').data
account = Account.by_facebook_id(data.get('id'))
data_ = {
'full_name': data.get('name'),
'login': data.get('username'),
'email': data.get('email'),
'facebook_id': data.get('id')
}
account = accounts.save(data_, account=account)
db.session.commit()
session['id'] = account.id
return redirect(next_url)
<｜fim▁end｜> | return provider_not_enabled('github') |
<|file_name|>sessions_api.py<|end_file_name|><｜fim▁begin｜>import requests
from flask import session, Blueprint, redirect
from flask import request
from grano import authz
from grano.lib.exc import BadRequest
from grano.lib.serialisation import jsonify
from grano.views.cache import validate_cache
from grano.core import db, url_for, app
from grano.providers import github, twitter, facebook
from grano.model import Account
from grano.logic import accounts
blueprint = Blueprint('sessions_api', __name__)
@blueprint.route('/api/1/sessions', methods=['GET'])
def status():
permissions = {}
if authz.logged_in():
for permission in request.account.permissions:
permissions[permission.project.slug] = {
'reader': permission.reader,
'editor': permission.editor,
'admin': permission.admin
}
keys = {
'p': repr(permissions),
'i': request.account.id if authz.logged_in() else None
}
validate_cache(keys=keys)
return jsonify({
'logged_in': authz.logged_in(),
'api_key': request.account.api_key if authz.logged_in() else None,
'account': request.account if request.account else None,
'permissions': permissions
})
def provider_not_enabled(name):
return jsonify({
'status': 501,
'name': 'Provider not configured: %s' % name,
'message': 'There are no OAuth credentials given for %s' % name,
}, status=501)
@blueprint.route('/api/1/sessions/logout', methods=['GET'])
def logout():
#authz.require(authz.logged_in())
session.clear()
return redirect(request.args.get('next_url', '/'))
@blueprint.route('/api/1/sessions/login/github', methods=['GET'])
def github_login():
if not app.config.get('GITHUB_CLIENT_ID'):
return provider_not_enabled('github')
callback=url_for('sessions_api.github_authorized')
session.clear()
if not request.args.get('next_url'):
<|fim_middle|>
session['next_url'] = request.args.get('next_url')
return github.authorize(callback=callback)
@blueprint.route('/api/1/sessions/callback/github', methods=['GET'])
@github.authorized_handler
def github_authorized(resp):
next_url = session.get('next_url', '/')
if resp is None or not 'access_token' in resp:
return redirect(next_url)
access_token = resp['access_token']
session['access_token'] = access_token, ''
res = requests.get('https://api.github.com/user?access_token=%s' % access_token,
verify=False)
data = res.json()
account = Account.by_github_id(data.get('id'))
data_ = {
'full_name': data.get('name'),
'login': data.get('login'),
'email': data.get('email'),
'github_id': data.get('id')
}
account = accounts.save(data_, account=account)
db.session.commit()
session['id'] = account.id
return redirect(next_url)
@blueprint.route('/api/1/sessions/login/twitter', methods=['GET'])
def twitter_login():
if not app.config.get('TWITTER_API_KEY'):
return provider_not_enabled('twitter')
callback=url_for('sessions_api.twitter_authorized')
session.clear()
if not request.args.get('next_url'):
raise BadRequest("No 'next_url' is specified.")
session['next_url'] = request.args.get('next_url')
return twitter.authorize(callback=callback)
@blueprint.route('/api/1/sessions/callback/twitter', methods=['GET'])
@twitter.authorized_handler
def twitter_authorized(resp):
next_url = session.get('next_url', '/')
if resp is None or not 'oauth_token' in resp:
return redirect(next_url)
session['twitter_token'] = (resp['oauth_token'],
resp['oauth_token_secret'])
res = twitter.get('users/show.json?user_id=%s' % resp.get('user_id'))
account = Account.by_twitter_id(res.data.get('id'))
data_ = {
'full_name': res.data.get('name'),
'login': res.data.get('screen_name'),
'twitter_id': res.data.get('id')
}
account = accounts.save(data_, account=account)
db.session.commit()
session['id'] = account.id
return redirect(next_url)
@blueprint.route('/api/1/sessions/login/facebook', methods=['GET'])
def facebook_login():
if not app.config.get('FACEBOOK_APP_ID'):
return provider_not_enabled('facebook')
callback=url_for('sessions_api.facebook_authorized')
session.clear()
if not request.args.get('next_url'):
raise BadRequest("No 'next_url' is specified.")
session['next_url'] = request.args.get('next_url')
return facebook.authorize(callback=callback)
@blueprint.route('/api/1/sessions/callback/facebook', methods=['GET'])
@facebook.authorized_handler
def facebook_authorized(resp):
next_url = session.get('next_url', '/')
if resp is None or not 'access_token' in resp:
return redirect(next_url)
session['facebook_token'] = (resp.get('access_token'), '')
data = facebook.get('/me').data
account = Account.by_facebook_id(data.get('id'))
data_ = {
'full_name': data.get('name'),
'login': data.get('username'),
'email': data.get('email'),
'facebook_id': data.get('id')
}
account = accounts.save(data_, account=account)
db.session.commit()
session['id'] = account.id
return redirect(next_url)
<｜fim▁end｜> | raise BadRequest("No 'next_url' is specified.") |
<|file_name|>sessions_api.py<|end_file_name|><｜fim▁begin｜>import requests
from flask import session, Blueprint, redirect
from flask import request
from grano import authz
from grano.lib.exc import BadRequest
from grano.lib.serialisation import jsonify
from grano.views.cache import validate_cache
from grano.core import db, url_for, app
from grano.providers import github, twitter, facebook
from grano.model import Account
from grano.logic import accounts
blueprint = Blueprint('sessions_api', __name__)
@blueprint.route('/api/1/sessions', methods=['GET'])
def status():
permissions = {}
if authz.logged_in():
for permission in request.account.permissions:
permissions[permission.project.slug] = {
'reader': permission.reader,
'editor': permission.editor,
'admin': permission.admin
}
keys = {
'p': repr(permissions),
'i': request.account.id if authz.logged_in() else None
}
validate_cache(keys=keys)
return jsonify({
'logged_in': authz.logged_in(),
'api_key': request.account.api_key if authz.logged_in() else None,
'account': request.account if request.account else None,
'permissions': permissions
})
def provider_not_enabled(name):
return jsonify({
'status': 501,
'name': 'Provider not configured: %s' % name,
'message': 'There are no OAuth credentials given for %s' % name,
}, status=501)
@blueprint.route('/api/1/sessions/logout', methods=['GET'])
def logout():
#authz.require(authz.logged_in())
session.clear()
return redirect(request.args.get('next_url', '/'))
@blueprint.route('/api/1/sessions/login/github', methods=['GET'])
def github_login():
if not app.config.get('GITHUB_CLIENT_ID'):
return provider_not_enabled('github')
callback=url_for('sessions_api.github_authorized')
session.clear()
if not request.args.get('next_url'):
raise BadRequest("No 'next_url' is specified.")
session['next_url'] = request.args.get('next_url')
return github.authorize(callback=callback)
@blueprint.route('/api/1/sessions/callback/github', methods=['GET'])
@github.authorized_handler
def github_authorized(resp):
next_url = session.get('next_url', '/')
if resp is None or not 'access_token' in resp:
<|fim_middle|>
access_token = resp['access_token']
session['access_token'] = access_token, ''
res = requests.get('https://api.github.com/user?access_token=%s' % access_token,
verify=False)
data = res.json()
account = Account.by_github_id(data.get('id'))
data_ = {
'full_name': data.get('name'),
'login': data.get('login'),
'email': data.get('email'),
'github_id': data.get('id')
}
account = accounts.save(data_, account=account)
db.session.commit()
session['id'] = account.id
return redirect(next_url)
@blueprint.route('/api/1/sessions/login/twitter', methods=['GET'])
def twitter_login():
if not app.config.get('TWITTER_API_KEY'):
return provider_not_enabled('twitter')
callback=url_for('sessions_api.twitter_authorized')
session.clear()
if not request.args.get('next_url'):
raise BadRequest("No 'next_url' is specified.")
session['next_url'] = request.args.get('next_url')
return twitter.authorize(callback=callback)
@blueprint.route('/api/1/sessions/callback/twitter', methods=['GET'])
@twitter.authorized_handler
def twitter_authorized(resp):
next_url = session.get('next_url', '/')
if resp is None or not 'oauth_token' in resp:
return redirect(next_url)
session['twitter_token'] = (resp['oauth_token'],
resp['oauth_token_secret'])
res = twitter.get('users/show.json?user_id=%s' % resp.get('user_id'))
account = Account.by_twitter_id(res.data.get('id'))
data_ = {
'full_name': res.data.get('name'),
'login': res.data.get('screen_name'),
'twitter_id': res.data.get('id')
}
account = accounts.save(data_, account=account)
db.session.commit()
session['id'] = account.id
return redirect(next_url)
@blueprint.route('/api/1/sessions/login/facebook', methods=['GET'])
def facebook_login():
if not app.config.get('FACEBOOK_APP_ID'):
return provider_not_enabled('facebook')
callback=url_for('sessions_api.facebook_authorized')
session.clear()
if not request.args.get('next_url'):
raise BadRequest("No 'next_url' is specified.")
session['next_url'] = request.args.get('next_url')
return facebook.authorize(callback=callback)
@blueprint.route('/api/1/sessions/callback/facebook', methods=['GET'])
@facebook.authorized_handler
def facebook_authorized(resp):
next_url = session.get('next_url', '/')
if resp is None or not 'access_token' in resp:
return redirect(next_url)
session['facebook_token'] = (resp.get('access_token'), '')
data = facebook.get('/me').data
account = Account.by_facebook_id(data.get('id'))
data_ = {
'full_name': data.get('name'),
'login': data.get('username'),
'email': data.get('email'),
'facebook_id': data.get('id')
}
account = accounts.save(data_, account=account)
db.session.commit()
session['id'] = account.id
return redirect(next_url)
<｜fim▁end｜> | return redirect(next_url) |
<|file_name|>sessions_api.py<|end_file_name|><｜fim▁begin｜>import requests
from flask import session, Blueprint, redirect
from flask import request
from grano import authz
from grano.lib.exc import BadRequest
from grano.lib.serialisation import jsonify
from grano.views.cache import validate_cache
from grano.core import db, url_for, app
from grano.providers import github, twitter, facebook
from grano.model import Account
from grano.logic import accounts
blueprint = Blueprint('sessions_api', __name__)
@blueprint.route('/api/1/sessions', methods=['GET'])
def status():
permissions = {}
if authz.logged_in():
for permission in request.account.permissions:
permissions[permission.project.slug] = {
'reader': permission.reader,
'editor': permission.editor,
'admin': permission.admin
}
keys = {
'p': repr(permissions),
'i': request.account.id if authz.logged_in() else None
}
validate_cache(keys=keys)
return jsonify({
'logged_in': authz.logged_in(),
'api_key': request.account.api_key if authz.logged_in() else None,
'account': request.account if request.account else None,
'permissions': permissions
})
def provider_not_enabled(name):
return jsonify({
'status': 501,
'name': 'Provider not configured: %s' % name,
'message': 'There are no OAuth credentials given for %s' % name,
}, status=501)
@blueprint.route('/api/1/sessions/logout', methods=['GET'])
def logout():
#authz.require(authz.logged_in())
session.clear()
return redirect(request.args.get('next_url', '/'))
@blueprint.route('/api/1/sessions/login/github', methods=['GET'])
def github_login():
if not app.config.get('GITHUB_CLIENT_ID'):
return provider_not_enabled('github')
callback=url_for('sessions_api.github_authorized')
session.clear()
if not request.args.get('next_url'):
raise BadRequest("No 'next_url' is specified.")
session['next_url'] = request.args.get('next_url')
return github.authorize(callback=callback)
@blueprint.route('/api/1/sessions/callback/github', methods=['GET'])
@github.authorized_handler
def github_authorized(resp):
next_url = session.get('next_url', '/')
if resp is None or not 'access_token' in resp:
return redirect(next_url)
access_token = resp['access_token']
session['access_token'] = access_token, ''
res = requests.get('https://api.github.com/user?access_token=%s' % access_token,
verify=False)
data = res.json()
account = Account.by_github_id(data.get('id'))
data_ = {
'full_name': data.get('name'),
'login': data.get('login'),
'email': data.get('email'),
'github_id': data.get('id')
}
account = accounts.save(data_, account=account)
db.session.commit()
session['id'] = account.id
return redirect(next_url)
@blueprint.route('/api/1/sessions/login/twitter', methods=['GET'])
def twitter_login():
if not app.config.get('TWITTER_API_KEY'):
<|fim_middle|>
callback=url_for('sessions_api.twitter_authorized')
session.clear()
if not request.args.get('next_url'):
raise BadRequest("No 'next_url' is specified.")
session['next_url'] = request.args.get('next_url')
return twitter.authorize(callback=callback)
@blueprint.route('/api/1/sessions/callback/twitter', methods=['GET'])
@twitter.authorized_handler
def twitter_authorized(resp):
next_url = session.get('next_url', '/')
if resp is None or not 'oauth_token' in resp:
return redirect(next_url)
session['twitter_token'] = (resp['oauth_token'],
resp['oauth_token_secret'])
res = twitter.get('users/show.json?user_id=%s' % resp.get('user_id'))
account = Account.by_twitter_id(res.data.get('id'))
data_ = {
'full_name': res.data.get('name'),
'login': res.data.get('screen_name'),
'twitter_id': res.data.get('id')
}
account = accounts.save(data_, account=account)
db.session.commit()
session['id'] = account.id
return redirect(next_url)
@blueprint.route('/api/1/sessions/login/facebook', methods=['GET'])
def facebook_login():
if not app.config.get('FACEBOOK_APP_ID'):
return provider_not_enabled('facebook')
callback=url_for('sessions_api.facebook_authorized')
session.clear()
if not request.args.get('next_url'):
raise BadRequest("No 'next_url' is specified.")
session['next_url'] = request.args.get('next_url')
return facebook.authorize(callback=callback)
@blueprint.route('/api/1/sessions/callback/facebook', methods=['GET'])
@facebook.authorized_handler
def facebook_authorized(resp):
next_url = session.get('next_url', '/')
if resp is None or not 'access_token' in resp:
return redirect(next_url)
session['facebook_token'] = (resp.get('access_token'), '')
data = facebook.get('/me').data
account = Account.by_facebook_id(data.get('id'))
data_ = {
'full_name': data.get('name'),
'login': data.get('username'),
'email': data.get('email'),
'facebook_id': data.get('id')
}
account = accounts.save(data_, account=account)
db.session.commit()
session['id'] = account.id
return redirect(next_url)
<｜fim▁end｜> | return provider_not_enabled('twitter') |
<|file_name|>sessions_api.py<|end_file_name|><｜fim▁begin｜>import requests
from flask import session, Blueprint, redirect
from flask import request
from grano import authz
from grano.lib.exc import BadRequest
from grano.lib.serialisation import jsonify
from grano.views.cache import validate_cache
from grano.core import db, url_for, app
from grano.providers import github, twitter, facebook
from grano.model import Account
from grano.logic import accounts
blueprint = Blueprint('sessions_api', __name__)
@blueprint.route('/api/1/sessions', methods=['GET'])
def status():
permissions = {}
if authz.logged_in():
for permission in request.account.permissions:
permissions[permission.project.slug] = {
'reader': permission.reader,
'editor': permission.editor,
'admin': permission.admin
}
keys = {
'p': repr(permissions),
'i': request.account.id if authz.logged_in() else None
}
validate_cache(keys=keys)
return jsonify({
'logged_in': authz.logged_in(),
'api_key': request.account.api_key if authz.logged_in() else None,
'account': request.account if request.account else None,
'permissions': permissions
})
def provider_not_enabled(name):
return jsonify({
'status': 501,
'name': 'Provider not configured: %s' % name,
'message': 'There are no OAuth credentials given for %s' % name,
}, status=501)
@blueprint.route('/api/1/sessions/logout', methods=['GET'])
def logout():
#authz.require(authz.logged_in())
session.clear()
return redirect(request.args.get('next_url', '/'))
@blueprint.route('/api/1/sessions/login/github', methods=['GET'])
def github_login():
if not app.config.get('GITHUB_CLIENT_ID'):
return provider_not_enabled('github')
callback=url_for('sessions_api.github_authorized')
session.clear()
if not request.args.get('next_url'):
raise BadRequest("No 'next_url' is specified.")
session['next_url'] = request.args.get('next_url')
return github.authorize(callback=callback)
@blueprint.route('/api/1/sessions/callback/github', methods=['GET'])
@github.authorized_handler
def github_authorized(resp):
next_url = session.get('next_url', '/')
if resp is None or not 'access_token' in resp:
return redirect(next_url)
access_token = resp['access_token']
session['access_token'] = access_token, ''
res = requests.get('https://api.github.com/user?access_token=%s' % access_token,
verify=False)
data = res.json()
account = Account.by_github_id(data.get('id'))
data_ = {
'full_name': data.get('name'),
'login': data.get('login'),
'email': data.get('email'),
'github_id': data.get('id')
}
account = accounts.save(data_, account=account)
db.session.commit()
session['id'] = account.id
return redirect(next_url)
@blueprint.route('/api/1/sessions/login/twitter', methods=['GET'])
def twitter_login():
if not app.config.get('TWITTER_API_KEY'):
return provider_not_enabled('twitter')
callback=url_for('sessions_api.twitter_authorized')
session.clear()
if not request.args.get('next_url'):
<|fim_middle|>
session['next_url'] = request.args.get('next_url')
return twitter.authorize(callback=callback)
@blueprint.route('/api/1/sessions/callback/twitter', methods=['GET'])
@twitter.authorized_handler
def twitter_authorized(resp):
next_url = session.get('next_url', '/')
if resp is None or not 'oauth_token' in resp:
return redirect(next_url)
session['twitter_token'] = (resp['oauth_token'],
resp['oauth_token_secret'])
res = twitter.get('users/show.json?user_id=%s' % resp.get('user_id'))
account = Account.by_twitter_id(res.data.get('id'))
data_ = {
'full_name': res.data.get('name'),
'login': res.data.get('screen_name'),
'twitter_id': res.data.get('id')
}
account = accounts.save(data_, account=account)
db.session.commit()
session['id'] = account.id
return redirect(next_url)
@blueprint.route('/api/1/sessions/login/facebook', methods=['GET'])
def facebook_login():
if not app.config.get('FACEBOOK_APP_ID'):
return provider_not_enabled('facebook')
callback=url_for('sessions_api.facebook_authorized')
session.clear()
if not request.args.get('next_url'):
raise BadRequest("No 'next_url' is specified.")
session['next_url'] = request.args.get('next_url')
return facebook.authorize(callback=callback)
@blueprint.route('/api/1/sessions/callback/facebook', methods=['GET'])
@facebook.authorized_handler
def facebook_authorized(resp):
next_url = session.get('next_url', '/')
if resp is None or not 'access_token' in resp:
return redirect(next_url)
session['facebook_token'] = (resp.get('access_token'), '')
data = facebook.get('/me').data
account = Account.by_facebook_id(data.get('id'))
data_ = {
'full_name': data.get('name'),
'login': data.get('username'),
'email': data.get('email'),
'facebook_id': data.get('id')
}
account = accounts.save(data_, account=account)
db.session.commit()
session['id'] = account.id
return redirect(next_url)
<｜fim▁end｜> | raise BadRequest("No 'next_url' is specified.") |
<|file_name|>sessions_api.py<|end_file_name|><｜fim▁begin｜>import requests
from flask import session, Blueprint, redirect
from flask import request
from grano import authz
from grano.lib.exc import BadRequest
from grano.lib.serialisation import jsonify
from grano.views.cache import validate_cache
from grano.core import db, url_for, app
from grano.providers import github, twitter, facebook
from grano.model import Account
from grano.logic import accounts
blueprint = Blueprint('sessions_api', __name__)
@blueprint.route('/api/1/sessions', methods=['GET'])
def status():
permissions = {}
if authz.logged_in():
for permission in request.account.permissions:
permissions[permission.project.slug] = {
'reader': permission.reader,
'editor': permission.editor,
'admin': permission.admin
}
keys = {
'p': repr(permissions),
'i': request.account.id if authz.logged_in() else None
}
validate_cache(keys=keys)
return jsonify({
'logged_in': authz.logged_in(),
'api_key': request.account.api_key if authz.logged_in() else None,
'account': request.account if request.account else None,
'permissions': permissions
})
def provider_not_enabled(name):
return jsonify({
'status': 501,
'name': 'Provider not configured: %s' % name,
'message': 'There are no OAuth credentials given for %s' % name,
}, status=501)
@blueprint.route('/api/1/sessions/logout', methods=['GET'])
def logout():
#authz.require(authz.logged_in())
session.clear()
return redirect(request.args.get('next_url', '/'))
@blueprint.route('/api/1/sessions/login/github', methods=['GET'])
def github_login():
if not app.config.get('GITHUB_CLIENT_ID'):
return provider_not_enabled('github')
callback=url_for('sessions_api.github_authorized')
session.clear()
if not request.args.get('next_url'):
raise BadRequest("No 'next_url' is specified.")
session['next_url'] = request.args.get('next_url')
return github.authorize(callback=callback)
@blueprint.route('/api/1/sessions/callback/github', methods=['GET'])
@github.authorized_handler
def github_authorized(resp):
next_url = session.get('next_url', '/')
if resp is None or not 'access_token' in resp:
return redirect(next_url)
access_token = resp['access_token']
session['access_token'] = access_token, ''
res = requests.get('https://api.github.com/user?access_token=%s' % access_token,
verify=False)
data = res.json()
account = Account.by_github_id(data.get('id'))
data_ = {
'full_name': data.get('name'),
'login': data.get('login'),
'email': data.get('email'),
'github_id': data.get('id')
}
account = accounts.save(data_, account=account)
db.session.commit()
session['id'] = account.id
return redirect(next_url)
@blueprint.route('/api/1/sessions/login/twitter', methods=['GET'])
def twitter_login():
if not app.config.get('TWITTER_API_KEY'):
return provider_not_enabled('twitter')
callback=url_for('sessions_api.twitter_authorized')
session.clear()
if not request.args.get('next_url'):
raise BadRequest("No 'next_url' is specified.")
session['next_url'] = request.args.get('next_url')
return twitter.authorize(callback=callback)
@blueprint.route('/api/1/sessions/callback/twitter', methods=['GET'])
@twitter.authorized_handler
def twitter_authorized(resp):
next_url = session.get('next_url', '/')
if resp is None or not 'oauth_token' in resp:
<|fim_middle|>
session['twitter_token'] = (resp['oauth_token'],
resp['oauth_token_secret'])
res = twitter.get('users/show.json?user_id=%s' % resp.get('user_id'))
account = Account.by_twitter_id(res.data.get('id'))
data_ = {
'full_name': res.data.get('name'),
'login': res.data.get('screen_name'),
'twitter_id': res.data.get('id')
}
account = accounts.save(data_, account=account)
db.session.commit()
session['id'] = account.id
return redirect(next_url)
@blueprint.route('/api/1/sessions/login/facebook', methods=['GET'])
def facebook_login():
if not app.config.get('FACEBOOK_APP_ID'):
return provider_not_enabled('facebook')
callback=url_for('sessions_api.facebook_authorized')
session.clear()
if not request.args.get('next_url'):
raise BadRequest("No 'next_url' is specified.")
session['next_url'] = request.args.get('next_url')
return facebook.authorize(callback=callback)
@blueprint.route('/api/1/sessions/callback/facebook', methods=['GET'])
@facebook.authorized_handler
def facebook_authorized(resp):
next_url = session.get('next_url', '/')
if resp is None or not 'access_token' in resp:
return redirect(next_url)
session['facebook_token'] = (resp.get('access_token'), '')
data = facebook.get('/me').data
account = Account.by_facebook_id(data.get('id'))
data_ = {
'full_name': data.get('name'),
'login': data.get('username'),
'email': data.get('email'),
'facebook_id': data.get('id')
}
account = accounts.save(data_, account=account)
db.session.commit()
session['id'] = account.id
return redirect(next_url)
<｜fim▁end｜> | return redirect(next_url) |
<|file_name|>sessions_api.py<|end_file_name|><｜fim▁begin｜>import requests
from flask import session, Blueprint, redirect
from flask import request
from grano import authz
from grano.lib.exc import BadRequest
from grano.lib.serialisation import jsonify
from grano.views.cache import validate_cache
from grano.core import db, url_for, app
from grano.providers import github, twitter, facebook
from grano.model import Account
from grano.logic import accounts
blueprint = Blueprint('sessions_api', __name__)
@blueprint.route('/api/1/sessions', methods=['GET'])
def status():
permissions = {}
if authz.logged_in():
for permission in request.account.permissions:
permissions[permission.project.slug] = {
'reader': permission.reader,
'editor': permission.editor,
'admin': permission.admin
}
keys = {
'p': repr(permissions),
'i': request.account.id if authz.logged_in() else None
}
validate_cache(keys=keys)
return jsonify({
'logged_in': authz.logged_in(),
'api_key': request.account.api_key if authz.logged_in() else None,
'account': request.account if request.account else None,
'permissions': permissions
})
def provider_not_enabled(name):
return jsonify({
'status': 501,
'name': 'Provider not configured: %s' % name,
'message': 'There are no OAuth credentials given for %s' % name,
}, status=501)
@blueprint.route('/api/1/sessions/logout', methods=['GET'])
def logout():
#authz.require(authz.logged_in())
session.clear()
return redirect(request.args.get('next_url', '/'))
@blueprint.route('/api/1/sessions/login/github', methods=['GET'])
def github_login():
if not app.config.get('GITHUB_CLIENT_ID'):
return provider_not_enabled('github')
callback=url_for('sessions_api.github_authorized')
session.clear()
if not request.args.get('next_url'):
raise BadRequest("No 'next_url' is specified.")
session['next_url'] = request.args.get('next_url')
return github.authorize(callback=callback)
@blueprint.route('/api/1/sessions/callback/github', methods=['GET'])
@github.authorized_handler
def github_authorized(resp):
next_url = session.get('next_url', '/')
if resp is None or not 'access_token' in resp:
return redirect(next_url)
access_token = resp['access_token']
session['access_token'] = access_token, ''
res = requests.get('https://api.github.com/user?access_token=%s' % access_token,
verify=False)
data = res.json()
account = Account.by_github_id(data.get('id'))
data_ = {
'full_name': data.get('name'),
'login': data.get('login'),
'email': data.get('email'),
'github_id': data.get('id')
}
account = accounts.save(data_, account=account)
db.session.commit()
session['id'] = account.id
return redirect(next_url)
@blueprint.route('/api/1/sessions/login/twitter', methods=['GET'])
def twitter_login():
if not app.config.get('TWITTER_API_KEY'):
return provider_not_enabled('twitter')
callback=url_for('sessions_api.twitter_authorized')
session.clear()
if not request.args.get('next_url'):
raise BadRequest("No 'next_url' is specified.")
session['next_url'] = request.args.get('next_url')
return twitter.authorize(callback=callback)
@blueprint.route('/api/1/sessions/callback/twitter', methods=['GET'])
@twitter.authorized_handler
def twitter_authorized(resp):
next_url = session.get('next_url', '/')
if resp is None or not 'oauth_token' in resp:
return redirect(next_url)
session['twitter_token'] = (resp['oauth_token'],
resp['oauth_token_secret'])
res = twitter.get('users/show.json?user_id=%s' % resp.get('user_id'))
account = Account.by_twitter_id(res.data.get('id'))
data_ = {
'full_name': res.data.get('name'),
'login': res.data.get('screen_name'),
'twitter_id': res.data.get('id')
}
account = accounts.save(data_, account=account)
db.session.commit()
session['id'] = account.id
return redirect(next_url)
@blueprint.route('/api/1/sessions/login/facebook', methods=['GET'])
def facebook_login():
if not app.config.get('FACEBOOK_APP_ID'):
<|fim_middle|>
callback=url_for('sessions_api.facebook_authorized')
session.clear()
if not request.args.get('next_url'):
raise BadRequest("No 'next_url' is specified.")
session['next_url'] = request.args.get('next_url')
return facebook.authorize(callback=callback)
@blueprint.route('/api/1/sessions/callback/facebook', methods=['GET'])
@facebook.authorized_handler
def facebook_authorized(resp):
next_url = session.get('next_url', '/')
if resp is None or not 'access_token' in resp:
return redirect(next_url)
session['facebook_token'] = (resp.get('access_token'), '')
data = facebook.get('/me').data
account = Account.by_facebook_id(data.get('id'))
data_ = {
'full_name': data.get('name'),
'login': data.get('username'),
'email': data.get('email'),
'facebook_id': data.get('id')
}
account = accounts.save(data_, account=account)
db.session.commit()
session['id'] = account.id
return redirect(next_url)
<｜fim▁end｜> | return provider_not_enabled('facebook') |
<|file_name|>sessions_api.py<|end_file_name|><｜fim▁begin｜>import requests
from flask import session, Blueprint, redirect
from flask import request
from grano import authz
from grano.lib.exc import BadRequest
from grano.lib.serialisation import jsonify
from grano.views.cache import validate_cache
from grano.core import db, url_for, app
from grano.providers import github, twitter, facebook
from grano.model import Account
from grano.logic import accounts
blueprint = Blueprint('sessions_api', __name__)
@blueprint.route('/api/1/sessions', methods=['GET'])
def status():
permissions = {}
if authz.logged_in():
for permission in request.account.permissions:
permissions[permission.project.slug] = {
'reader': permission.reader,
'editor': permission.editor,
'admin': permission.admin
}
keys = {
'p': repr(permissions),
'i': request.account.id if authz.logged_in() else None
}
validate_cache(keys=keys)
return jsonify({
'logged_in': authz.logged_in(),
'api_key': request.account.api_key if authz.logged_in() else None,
'account': request.account if request.account else None,
'permissions': permissions
})
def provider_not_enabled(name):
return jsonify({
'status': 501,
'name': 'Provider not configured: %s' % name,
'message': 'There are no OAuth credentials given for %s' % name,
}, status=501)
@blueprint.route('/api/1/sessions/logout', methods=['GET'])
def logout():
#authz.require(authz.logged_in())
session.clear()
return redirect(request.args.get('next_url', '/'))
@blueprint.route('/api/1/sessions/login/github', methods=['GET'])
def github_login():
if not app.config.get('GITHUB_CLIENT_ID'):
return provider_not_enabled('github')
callback=url_for('sessions_api.github_authorized')
session.clear()
if not request.args.get('next_url'):
raise BadRequest("No 'next_url' is specified.")
session['next_url'] = request.args.get('next_url')
return github.authorize(callback=callback)
@blueprint.route('/api/1/sessions/callback/github', methods=['GET'])
@github.authorized_handler
def github_authorized(resp):
next_url = session.get('next_url', '/')
if resp is None or not 'access_token' in resp:
return redirect(next_url)
access_token = resp['access_token']
session['access_token'] = access_token, ''
res = requests.get('https://api.github.com/user?access_token=%s' % access_token,
verify=False)
data = res.json()
account = Account.by_github_id(data.get('id'))
data_ = {
'full_name': data.get('name'),
'login': data.get('login'),
'email': data.get('email'),
'github_id': data.get('id')
}
account = accounts.save(data_, account=account)
db.session.commit()
session['id'] = account.id
return redirect(next_url)
@blueprint.route('/api/1/sessions/login/twitter', methods=['GET'])
def twitter_login():
if not app.config.get('TWITTER_API_KEY'):
return provider_not_enabled('twitter')
callback=url_for('sessions_api.twitter_authorized')
session.clear()
if not request.args.get('next_url'):
raise BadRequest("No 'next_url' is specified.")
session['next_url'] = request.args.get('next_url')
return twitter.authorize(callback=callback)
@blueprint.route('/api/1/sessions/callback/twitter', methods=['GET'])
@twitter.authorized_handler
def twitter_authorized(resp):
next_url = session.get('next_url', '/')
if resp is None or not 'oauth_token' in resp:
return redirect(next_url)
session['twitter_token'] = (resp['oauth_token'],
resp['oauth_token_secret'])
res = twitter.get('users/show.json?user_id=%s' % resp.get('user_id'))
account = Account.by_twitter_id(res.data.get('id'))
data_ = {
'full_name': res.data.get('name'),
'login': res.data.get('screen_name'),
'twitter_id': res.data.get('id')
}
account = accounts.save(data_, account=account)
db.session.commit()
session['id'] = account.id
return redirect(next_url)
@blueprint.route('/api/1/sessions/login/facebook', methods=['GET'])
def facebook_login():
if not app.config.get('FACEBOOK_APP_ID'):
return provider_not_enabled('facebook')
callback=url_for('sessions_api.facebook_authorized')
session.clear()
if not request.args.get('next_url'):
<|fim_middle|>
session['next_url'] = request.args.get('next_url')
return facebook.authorize(callback=callback)
@blueprint.route('/api/1/sessions/callback/facebook', methods=['GET'])
@facebook.authorized_handler
def facebook_authorized(resp):
next_url = session.get('next_url', '/')
if resp is None or not 'access_token' in resp:
return redirect(next_url)
session['facebook_token'] = (resp.get('access_token'), '')
data = facebook.get('/me').data
account = Account.by_facebook_id(data.get('id'))
data_ = {
'full_name': data.get('name'),
'login': data.get('username'),
'email': data.get('email'),
'facebook_id': data.get('id')
}
account = accounts.save(data_, account=account)
db.session.commit()
session['id'] = account.id
return redirect(next_url)
<｜fim▁end｜> | raise BadRequest("No 'next_url' is specified.") |
<|file_name|>sessions_api.py<|end_file_name|><｜fim▁begin｜>import requests
from flask import session, Blueprint, redirect
from flask import request
from grano import authz
from grano.lib.exc import BadRequest
from grano.lib.serialisation import jsonify
from grano.views.cache import validate_cache
from grano.core import db, url_for, app
from grano.providers import github, twitter, facebook
from grano.model import Account
from grano.logic import accounts
blueprint = Blueprint('sessions_api', __name__)
@blueprint.route('/api/1/sessions', methods=['GET'])
def status():
permissions = {}
if authz.logged_in():
for permission in request.account.permissions:
permissions[permission.project.slug] = {
'reader': permission.reader,
'editor': permission.editor,
'admin': permission.admin
}
keys = {
'p': repr(permissions),
'i': request.account.id if authz.logged_in() else None
}
validate_cache(keys=keys)
return jsonify({
'logged_in': authz.logged_in(),
'api_key': request.account.api_key if authz.logged_in() else None,
'account': request.account if request.account else None,
'permissions': permissions
})
def provider_not_enabled(name):
return jsonify({
'status': 501,
'name': 'Provider not configured: %s' % name,
'message': 'There are no OAuth credentials given for %s' % name,
}, status=501)
@blueprint.route('/api/1/sessions/logout', methods=['GET'])
def logout():
#authz.require(authz.logged_in())
session.clear()
return redirect(request.args.get('next_url', '/'))
@blueprint.route('/api/1/sessions/login/github', methods=['GET'])
def github_login():
if not app.config.get('GITHUB_CLIENT_ID'):
return provider_not_enabled('github')
callback=url_for('sessions_api.github_authorized')
session.clear()
if not request.args.get('next_url'):
raise BadRequest("No 'next_url' is specified.")
session['next_url'] = request.args.get('next_url')
return github.authorize(callback=callback)
@blueprint.route('/api/1/sessions/callback/github', methods=['GET'])
@github.authorized_handler
def github_authorized(resp):
next_url = session.get('next_url', '/')
if resp is None or not 'access_token' in resp:
return redirect(next_url)
access_token = resp['access_token']
session['access_token'] = access_token, ''
res = requests.get('https://api.github.com/user?access_token=%s' % access_token,
verify=False)
data = res.json()
account = Account.by_github_id(data.get('id'))
data_ = {
'full_name': data.get('name'),
'login': data.get('login'),
'email': data.get('email'),
'github_id': data.get('id')
}
account = accounts.save(data_, account=account)
db.session.commit()
session['id'] = account.id
return redirect(next_url)
@blueprint.route('/api/1/sessions/login/twitter', methods=['GET'])
def twitter_login():
if not app.config.get('TWITTER_API_KEY'):
return provider_not_enabled('twitter')
callback=url_for('sessions_api.twitter_authorized')
session.clear()
if not request.args.get('next_url'):
raise BadRequest("No 'next_url' is specified.")
session['next_url'] = request.args.get('next_url')
return twitter.authorize(callback=callback)
@blueprint.route('/api/1/sessions/callback/twitter', methods=['GET'])
@twitter.authorized_handler
def twitter_authorized(resp):
next_url = session.get('next_url', '/')
if resp is None or not 'oauth_token' in resp:
return redirect(next_url)
session['twitter_token'] = (resp['oauth_token'],
resp['oauth_token_secret'])
res = twitter.get('users/show.json?user_id=%s' % resp.get('user_id'))
account = Account.by_twitter_id(res.data.get('id'))
data_ = {
'full_name': res.data.get('name'),
'login': res.data.get('screen_name'),
'twitter_id': res.data.get('id')
}
account = accounts.save(data_, account=account)
db.session.commit()
session['id'] = account.id
return redirect(next_url)
@blueprint.route('/api/1/sessions/login/facebook', methods=['GET'])
def facebook_login():
if not app.config.get('FACEBOOK_APP_ID'):
return provider_not_enabled('facebook')
callback=url_for('sessions_api.facebook_authorized')
session.clear()
if not request.args.get('next_url'):
raise BadRequest("No 'next_url' is specified.")
session['next_url'] = request.args.get('next_url')
return facebook.authorize(callback=callback)
@blueprint.route('/api/1/sessions/callback/facebook', methods=['GET'])
@facebook.authorized_handler
def facebook_authorized(resp):
next_url = session.get('next_url', '/')
if resp is None or not 'access_token' in resp:
<|fim_middle|>
session['facebook_token'] = (resp.get('access_token'), '')
data = facebook.get('/me').data
account = Account.by_facebook_id(data.get('id'))
data_ = {
'full_name': data.get('name'),
'login': data.get('username'),
'email': data.get('email'),
'facebook_id': data.get('id')
}
account = accounts.save(data_, account=account)
db.session.commit()
session['id'] = account.id
return redirect(next_url)
<｜fim▁end｜> | return redirect(next_url) |
<|file_name|>sessions_api.py<|end_file_name|><｜fim▁begin｜>import requests
from flask import session, Blueprint, redirect
from flask import request
from grano import authz
from grano.lib.exc import BadRequest
from grano.lib.serialisation import jsonify
from grano.views.cache import validate_cache
from grano.core import db, url_for, app
from grano.providers import github, twitter, facebook
from grano.model import Account
from grano.logic import accounts
blueprint = Blueprint('sessions_api', __name__)
@blueprint.route('/api/1/sessions', methods=['GET'])
def <|fim_middle|>():
permissions = {}
if authz.logged_in():
for permission in request.account.permissions:
permissions[permission.project.slug] = {
'reader': permission.reader,
'editor': permission.editor,
'admin': permission.admin
}
keys = {
'p': repr(permissions),
'i': request.account.id if authz.logged_in() else None
}
validate_cache(keys=keys)
return jsonify({
'logged_in': authz.logged_in(),
'api_key': request.account.api_key if authz.logged_in() else None,
'account': request.account if request.account else None,
'permissions': permissions
})
def provider_not_enabled(name):
return jsonify({
'status': 501,
'name': 'Provider not configured: %s' % name,
'message': 'There are no OAuth credentials given for %s' % name,
}, status=501)
@blueprint.route('/api/1/sessions/logout', methods=['GET'])
def logout():
#authz.require(authz.logged_in())
session.clear()
return redirect(request.args.get('next_url', '/'))
@blueprint.route('/api/1/sessions/login/github', methods=['GET'])
def github_login():
if not app.config.get('GITHUB_CLIENT_ID'):
return provider_not_enabled('github')
callback=url_for('sessions_api.github_authorized')
session.clear()
if not request.args.get('next_url'):
raise BadRequest("No 'next_url' is specified.")
session['next_url'] = request.args.get('next_url')
return github.authorize(callback=callback)
@blueprint.route('/api/1/sessions/callback/github', methods=['GET'])
@github.authorized_handler
def github_authorized(resp):
next_url = session.get('next_url', '/')
if resp is None or not 'access_token' in resp:
return redirect(next_url)
access_token = resp['access_token']
session['access_token'] = access_token, ''
res = requests.get('https://api.github.com/user?access_token=%s' % access_token,
verify=False)
data = res.json()
account = Account.by_github_id(data.get('id'))
data_ = {
'full_name': data.get('name'),
'login': data.get('login'),
'email': data.get('email'),
'github_id': data.get('id')
}
account = accounts.save(data_, account=account)
db.session.commit()
session['id'] = account.id
return redirect(next_url)
@blueprint.route('/api/1/sessions/login/twitter', methods=['GET'])
def twitter_login():
if not app.config.get('TWITTER_API_KEY'):
return provider_not_enabled('twitter')
callback=url_for('sessions_api.twitter_authorized')
session.clear()
if not request.args.get('next_url'):
raise BadRequest("No 'next_url' is specified.")
session['next_url'] = request.args.get('next_url')
return twitter.authorize(callback=callback)
@blueprint.route('/api/1/sessions/callback/twitter', methods=['GET'])
@twitter.authorized_handler
def twitter_authorized(resp):
next_url = session.get('next_url', '/')
if resp is None or not 'oauth_token' in resp:
return redirect(next_url)
session['twitter_token'] = (resp['oauth_token'],
resp['oauth_token_secret'])
res = twitter.get('users/show.json?user_id=%s' % resp.get('user_id'))
account = Account.by_twitter_id(res.data.get('id'))
data_ = {
'full_name': res.data.get('name'),
'login': res.data.get('screen_name'),
'twitter_id': res.data.get('id')
}
account = accounts.save(data_, account=account)
db.session.commit()
session['id'] = account.id
return redirect(next_url)
@blueprint.route('/api/1/sessions/login/facebook', methods=['GET'])
def facebook_login():
if not app.config.get('FACEBOOK_APP_ID'):
return provider_not_enabled('facebook')
callback=url_for('sessions_api.facebook_authorized')
session.clear()
if not request.args.get('next_url'):
raise BadRequest("No 'next_url' is specified.")
session['next_url'] = request.args.get('next_url')
return facebook.authorize(callback=callback)
@blueprint.route('/api/1/sessions/callback/facebook', methods=['GET'])
@facebook.authorized_handler
def facebook_authorized(resp):
next_url = session.get('next_url', '/')
if resp is None or not 'access_token' in resp:
return redirect(next_url)
session['facebook_token'] = (resp.get('access_token'), '')
data = facebook.get('/me').data
account = Account.by_facebook_id(data.get('id'))
data_ = {
'full_name': data.get('name'),
'login': data.get('username'),
'email': data.get('email'),
'facebook_id': data.get('id')
}
account = accounts.save(data_, account=account)
db.session.commit()
session['id'] = account.id
return redirect(next_url)
<｜fim▁end｜> | status |
<|file_name|>sessions_api.py<|end_file_name|><｜fim▁begin｜>import requests
from flask import session, Blueprint, redirect
from flask import request
from grano import authz
from grano.lib.exc import BadRequest
from grano.lib.serialisation import jsonify
from grano.views.cache import validate_cache
from grano.core import db, url_for, app
from grano.providers import github, twitter, facebook
from grano.model import Account
from grano.logic import accounts
blueprint = Blueprint('sessions_api', __name__)
@blueprint.route('/api/1/sessions', methods=['GET'])
def status():
permissions = {}
if authz.logged_in():
for permission in request.account.permissions:
permissions[permission.project.slug] = {
'reader': permission.reader,
'editor': permission.editor,
'admin': permission.admin
}
keys = {
'p': repr(permissions),
'i': request.account.id if authz.logged_in() else None
}
validate_cache(keys=keys)
return jsonify({
'logged_in': authz.logged_in(),
'api_key': request.account.api_key if authz.logged_in() else None,
'account': request.account if request.account else None,
'permissions': permissions
})
def <|fim_middle|>(name):
return jsonify({
'status': 501,
'name': 'Provider not configured: %s' % name,
'message': 'There are no OAuth credentials given for %s' % name,
}, status=501)
@blueprint.route('/api/1/sessions/logout', methods=['GET'])
def logout():
#authz.require(authz.logged_in())
session.clear()
return redirect(request.args.get('next_url', '/'))
@blueprint.route('/api/1/sessions/login/github', methods=['GET'])
def github_login():
if not app.config.get('GITHUB_CLIENT_ID'):
return provider_not_enabled('github')
callback=url_for('sessions_api.github_authorized')
session.clear()
if not request.args.get('next_url'):
raise BadRequest("No 'next_url' is specified.")
session['next_url'] = request.args.get('next_url')
return github.authorize(callback=callback)
@blueprint.route('/api/1/sessions/callback/github', methods=['GET'])
@github.authorized_handler
def github_authorized(resp):
next_url = session.get('next_url', '/')
if resp is None or not 'access_token' in resp:
return redirect(next_url)
access_token = resp['access_token']
session['access_token'] = access_token, ''
res = requests.get('https://api.github.com/user?access_token=%s' % access_token,
verify=False)
data = res.json()
account = Account.by_github_id(data.get('id'))
data_ = {
'full_name': data.get('name'),
'login': data.get('login'),
'email': data.get('email'),
'github_id': data.get('id')
}
account = accounts.save(data_, account=account)
db.session.commit()
session['id'] = account.id
return redirect(next_url)
@blueprint.route('/api/1/sessions/login/twitter', methods=['GET'])
def twitter_login():
if not app.config.get('TWITTER_API_KEY'):
return provider_not_enabled('twitter')
callback=url_for('sessions_api.twitter_authorized')
session.clear()
if not request.args.get('next_url'):
raise BadRequest("No 'next_url' is specified.")
session['next_url'] = request.args.get('next_url')
return twitter.authorize(callback=callback)
@blueprint.route('/api/1/sessions/callback/twitter', methods=['GET'])
@twitter.authorized_handler
def twitter_authorized(resp):
next_url = session.get('next_url', '/')
if resp is None or not 'oauth_token' in resp:
return redirect(next_url)
session['twitter_token'] = (resp['oauth_token'],
resp['oauth_token_secret'])
res = twitter.get('users/show.json?user_id=%s' % resp.get('user_id'))
account = Account.by_twitter_id(res.data.get('id'))
data_ = {
'full_name': res.data.get('name'),
'login': res.data.get('screen_name'),
'twitter_id': res.data.get('id')
}
account = accounts.save(data_, account=account)
db.session.commit()
session['id'] = account.id
return redirect(next_url)
@blueprint.route('/api/1/sessions/login/facebook', methods=['GET'])
def facebook_login():
if not app.config.get('FACEBOOK_APP_ID'):
return provider_not_enabled('facebook')
callback=url_for('sessions_api.facebook_authorized')
session.clear()
if not request.args.get('next_url'):
raise BadRequest("No 'next_url' is specified.")
session['next_url'] = request.args.get('next_url')
return facebook.authorize(callback=callback)
@blueprint.route('/api/1/sessions/callback/facebook', methods=['GET'])
@facebook.authorized_handler
def facebook_authorized(resp):
next_url = session.get('next_url', '/')
if resp is None or not 'access_token' in resp:
return redirect(next_url)
session['facebook_token'] = (resp.get('access_token'), '')
data = facebook.get('/me').data
account = Account.by_facebook_id(data.get('id'))
data_ = {
'full_name': data.get('name'),
'login': data.get('username'),
'email': data.get('email'),
'facebook_id': data.get('id')
}
account = accounts.save(data_, account=account)
db.session.commit()
session['id'] = account.id
return redirect(next_url)
<｜fim▁end｜> | provider_not_enabled |
<|file_name|>sessions_api.py<|end_file_name|><｜fim▁begin｜>import requests
from flask import session, Blueprint, redirect
from flask import request
from grano import authz
from grano.lib.exc import BadRequest
from grano.lib.serialisation import jsonify
from grano.views.cache import validate_cache
from grano.core import db, url_for, app
from grano.providers import github, twitter, facebook
from grano.model import Account
from grano.logic import accounts
blueprint = Blueprint('sessions_api', __name__)
@blueprint.route('/api/1/sessions', methods=['GET'])
def status():
permissions = {}
if authz.logged_in():
for permission in request.account.permissions:
permissions[permission.project.slug] = {
'reader': permission.reader,
'editor': permission.editor,
'admin': permission.admin
}
keys = {
'p': repr(permissions),
'i': request.account.id if authz.logged_in() else None
}
validate_cache(keys=keys)
return jsonify({
'logged_in': authz.logged_in(),
'api_key': request.account.api_key if authz.logged_in() else None,
'account': request.account if request.account else None,
'permissions': permissions
})
def provider_not_enabled(name):
return jsonify({
'status': 501,
'name': 'Provider not configured: %s' % name,
'message': 'There are no OAuth credentials given for %s' % name,
}, status=501)
@blueprint.route('/api/1/sessions/logout', methods=['GET'])
def <|fim_middle|>():
#authz.require(authz.logged_in())
session.clear()
return redirect(request.args.get('next_url', '/'))
@blueprint.route('/api/1/sessions/login/github', methods=['GET'])
def github_login():
if not app.config.get('GITHUB_CLIENT_ID'):
return provider_not_enabled('github')
callback=url_for('sessions_api.github_authorized')
session.clear()
if not request.args.get('next_url'):
raise BadRequest("No 'next_url' is specified.")
session['next_url'] = request.args.get('next_url')
return github.authorize(callback=callback)
@blueprint.route('/api/1/sessions/callback/github', methods=['GET'])
@github.authorized_handler
def github_authorized(resp):
next_url = session.get('next_url', '/')
if resp is None or not 'access_token' in resp:
return redirect(next_url)
access_token = resp['access_token']
session['access_token'] = access_token, ''
res = requests.get('https://api.github.com/user?access_token=%s' % access_token,
verify=False)
data = res.json()
account = Account.by_github_id(data.get('id'))
data_ = {
'full_name': data.get('name'),
'login': data.get('login'),
'email': data.get('email'),
'github_id': data.get('id')
}
account = accounts.save(data_, account=account)
db.session.commit()
session['id'] = account.id
return redirect(next_url)
@blueprint.route('/api/1/sessions/login/twitter', methods=['GET'])
def twitter_login():
if not app.config.get('TWITTER_API_KEY'):
return provider_not_enabled('twitter')
callback=url_for('sessions_api.twitter_authorized')
session.clear()
if not request.args.get('next_url'):
raise BadRequest("No 'next_url' is specified.")
session['next_url'] = request.args.get('next_url')
return twitter.authorize(callback=callback)
@blueprint.route('/api/1/sessions/callback/twitter', methods=['GET'])
@twitter.authorized_handler
def twitter_authorized(resp):
next_url = session.get('next_url', '/')
if resp is None or not 'oauth_token' in resp:
return redirect(next_url)
session['twitter_token'] = (resp['oauth_token'],
resp['oauth_token_secret'])
res = twitter.get('users/show.json?user_id=%s' % resp.get('user_id'))
account = Account.by_twitter_id(res.data.get('id'))
data_ = {
'full_name': res.data.get('name'),
'login': res.data.get('screen_name'),
'twitter_id': res.data.get('id')
}
account = accounts.save(data_, account=account)
db.session.commit()
session['id'] = account.id
return redirect(next_url)
@blueprint.route('/api/1/sessions/login/facebook', methods=['GET'])
def facebook_login():
if not app.config.get('FACEBOOK_APP_ID'):
return provider_not_enabled('facebook')
callback=url_for('sessions_api.facebook_authorized')
session.clear()
if not request.args.get('next_url'):
raise BadRequest("No 'next_url' is specified.")
session['next_url'] = request.args.get('next_url')
return facebook.authorize(callback=callback)
@blueprint.route('/api/1/sessions/callback/facebook', methods=['GET'])
@facebook.authorized_handler
def facebook_authorized(resp):
next_url = session.get('next_url', '/')
if resp is None or not 'access_token' in resp:
return redirect(next_url)
session['facebook_token'] = (resp.get('access_token'), '')
data = facebook.get('/me').data
account = Account.by_facebook_id(data.get('id'))
data_ = {
'full_name': data.get('name'),
'login': data.get('username'),
'email': data.get('email'),
'facebook_id': data.get('id')
}
account = accounts.save(data_, account=account)
db.session.commit()
session['id'] = account.id
return redirect(next_url)
<｜fim▁end｜> | logout |
<|file_name|>sessions_api.py<|end_file_name|><｜fim▁begin｜>import requests
from flask import session, Blueprint, redirect
from flask import request
from grano import authz
from grano.lib.exc import BadRequest
from grano.lib.serialisation import jsonify
from grano.views.cache import validate_cache
from grano.core import db, url_for, app
from grano.providers import github, twitter, facebook
from grano.model import Account
from grano.logic import accounts
blueprint = Blueprint('sessions_api', __name__)
@blueprint.route('/api/1/sessions', methods=['GET'])
def status():
permissions = {}
if authz.logged_in():
for permission in request.account.permissions:
permissions[permission.project.slug] = {
'reader': permission.reader,
'editor': permission.editor,
'admin': permission.admin
}
keys = {
'p': repr(permissions),
'i': request.account.id if authz.logged_in() else None
}
validate_cache(keys=keys)
return jsonify({
'logged_in': authz.logged_in(),
'api_key': request.account.api_key if authz.logged_in() else None,
'account': request.account if request.account else None,
'permissions': permissions
})
def provider_not_enabled(name):
return jsonify({
'status': 501,
'name': 'Provider not configured: %s' % name,
'message': 'There are no OAuth credentials given for %s' % name,
}, status=501)
@blueprint.route('/api/1/sessions/logout', methods=['GET'])
def logout():
#authz.require(authz.logged_in())
session.clear()
return redirect(request.args.get('next_url', '/'))
@blueprint.route('/api/1/sessions/login/github', methods=['GET'])
def <|fim_middle|>():
if not app.config.get('GITHUB_CLIENT_ID'):
return provider_not_enabled('github')
callback=url_for('sessions_api.github_authorized')
session.clear()
if not request.args.get('next_url'):
raise BadRequest("No 'next_url' is specified.")
session['next_url'] = request.args.get('next_url')
return github.authorize(callback=callback)
@blueprint.route('/api/1/sessions/callback/github', methods=['GET'])
@github.authorized_handler
def github_authorized(resp):
next_url = session.get('next_url', '/')
if resp is None or not 'access_token' in resp:
return redirect(next_url)
access_token = resp['access_token']
session['access_token'] = access_token, ''
res = requests.get('https://api.github.com/user?access_token=%s' % access_token,
verify=False)
data = res.json()
account = Account.by_github_id(data.get('id'))
data_ = {
'full_name': data.get('name'),
'login': data.get('login'),
'email': data.get('email'),
'github_id': data.get('id')
}
account = accounts.save(data_, account=account)
db.session.commit()
session['id'] = account.id
return redirect(next_url)
@blueprint.route('/api/1/sessions/login/twitter', methods=['GET'])
def twitter_login():
if not app.config.get('TWITTER_API_KEY'):
return provider_not_enabled('twitter')
callback=url_for('sessions_api.twitter_authorized')
session.clear()
if not request.args.get('next_url'):
raise BadRequest("No 'next_url' is specified.")
session['next_url'] = request.args.get('next_url')
return twitter.authorize(callback=callback)
@blueprint.route('/api/1/sessions/callback/twitter', methods=['GET'])
@twitter.authorized_handler
def twitter_authorized(resp):
next_url = session.get('next_url', '/')
if resp is None or not 'oauth_token' in resp:
return redirect(next_url)
session['twitter_token'] = (resp['oauth_token'],
resp['oauth_token_secret'])
res = twitter.get('users/show.json?user_id=%s' % resp.get('user_id'))
account = Account.by_twitter_id(res.data.get('id'))
data_ = {
'full_name': res.data.get('name'),
'login': res.data.get('screen_name'),
'twitter_id': res.data.get('id')
}
account = accounts.save(data_, account=account)
db.session.commit()
session['id'] = account.id
return redirect(next_url)
@blueprint.route('/api/1/sessions/login/facebook', methods=['GET'])
def facebook_login():
if not app.config.get('FACEBOOK_APP_ID'):
return provider_not_enabled('facebook')
callback=url_for('sessions_api.facebook_authorized')
session.clear()
if not request.args.get('next_url'):
raise BadRequest("No 'next_url' is specified.")
session['next_url'] = request.args.get('next_url')
return facebook.authorize(callback=callback)
@blueprint.route('/api/1/sessions/callback/facebook', methods=['GET'])
@facebook.authorized_handler
def facebook_authorized(resp):
next_url = session.get('next_url', '/')
if resp is None or not 'access_token' in resp:
return redirect(next_url)
session['facebook_token'] = (resp.get('access_token'), '')
data = facebook.get('/me').data
account = Account.by_facebook_id(data.get('id'))
data_ = {
'full_name': data.get('name'),
'login': data.get('username'),
'email': data.get('email'),
'facebook_id': data.get('id')
}
account = accounts.save(data_, account=account)
db.session.commit()
session['id'] = account.id
return redirect(next_url)
<｜fim▁end｜> | github_login |
<|file_name|>sessions_api.py<|end_file_name|><｜fim▁begin｜>import requests
from flask import session, Blueprint, redirect
from flask import request
from grano import authz
from grano.lib.exc import BadRequest
from grano.lib.serialisation import jsonify
from grano.views.cache import validate_cache
from grano.core import db, url_for, app
from grano.providers import github, twitter, facebook
from grano.model import Account
from grano.logic import accounts
blueprint = Blueprint('sessions_api', __name__)
@blueprint.route('/api/1/sessions', methods=['GET'])
def status():
permissions = {}
if authz.logged_in():
for permission in request.account.permissions:
permissions[permission.project.slug] = {
'reader': permission.reader,
'editor': permission.editor,
'admin': permission.admin
}
keys = {
'p': repr(permissions),
'i': request.account.id if authz.logged_in() else None
}
validate_cache(keys=keys)
return jsonify({
'logged_in': authz.logged_in(),
'api_key': request.account.api_key if authz.logged_in() else None,
'account': request.account if request.account else None,
'permissions': permissions
})
def provider_not_enabled(name):
return jsonify({
'status': 501,
'name': 'Provider not configured: %s' % name,
'message': 'There are no OAuth credentials given for %s' % name,
}, status=501)
@blueprint.route('/api/1/sessions/logout', methods=['GET'])
def logout():
#authz.require(authz.logged_in())
session.clear()
return redirect(request.args.get('next_url', '/'))
@blueprint.route('/api/1/sessions/login/github', methods=['GET'])
def github_login():
if not app.config.get('GITHUB_CLIENT_ID'):
return provider_not_enabled('github')
callback=url_for('sessions_api.github_authorized')
session.clear()
if not request.args.get('next_url'):
raise BadRequest("No 'next_url' is specified.")
session['next_url'] = request.args.get('next_url')
return github.authorize(callback=callback)
@blueprint.route('/api/1/sessions/callback/github', methods=['GET'])
@github.authorized_handler
def <|fim_middle|>(resp):
next_url = session.get('next_url', '/')
if resp is None or not 'access_token' in resp:
return redirect(next_url)
access_token = resp['access_token']
session['access_token'] = access_token, ''
res = requests.get('https://api.github.com/user?access_token=%s' % access_token,
verify=False)
data = res.json()
account = Account.by_github_id(data.get('id'))
data_ = {
'full_name': data.get('name'),
'login': data.get('login'),
'email': data.get('email'),
'github_id': data.get('id')
}
account = accounts.save(data_, account=account)
db.session.commit()
session['id'] = account.id
return redirect(next_url)
@blueprint.route('/api/1/sessions/login/twitter', methods=['GET'])
def twitter_login():
if not app.config.get('TWITTER_API_KEY'):
return provider_not_enabled('twitter')
callback=url_for('sessions_api.twitter_authorized')
session.clear()
if not request.args.get('next_url'):
raise BadRequest("No 'next_url' is specified.")
session['next_url'] = request.args.get('next_url')
return twitter.authorize(callback=callback)
@blueprint.route('/api/1/sessions/callback/twitter', methods=['GET'])
@twitter.authorized_handler
def twitter_authorized(resp):
next_url = session.get('next_url', '/')
if resp is None or not 'oauth_token' in resp:
return redirect(next_url)
session['twitter_token'] = (resp['oauth_token'],
resp['oauth_token_secret'])
res = twitter.get('users/show.json?user_id=%s' % resp.get('user_id'))
account = Account.by_twitter_id(res.data.get('id'))
data_ = {
'full_name': res.data.get('name'),
'login': res.data.get('screen_name'),
'twitter_id': res.data.get('id')
}
account = accounts.save(data_, account=account)
db.session.commit()
session['id'] = account.id
return redirect(next_url)
@blueprint.route('/api/1/sessions/login/facebook', methods=['GET'])
def facebook_login():
if not app.config.get('FACEBOOK_APP_ID'):
return provider_not_enabled('facebook')
callback=url_for('sessions_api.facebook_authorized')
session.clear()
if not request.args.get('next_url'):
raise BadRequest("No 'next_url' is specified.")
session['next_url'] = request.args.get('next_url')
return facebook.authorize(callback=callback)
@blueprint.route('/api/1/sessions/callback/facebook', methods=['GET'])
@facebook.authorized_handler
def facebook_authorized(resp):
next_url = session.get('next_url', '/')
if resp is None or not 'access_token' in resp:
return redirect(next_url)
session['facebook_token'] = (resp.get('access_token'), '')
data = facebook.get('/me').data
account = Account.by_facebook_id(data.get('id'))
data_ = {
'full_name': data.get('name'),
'login': data.get('username'),
'email': data.get('email'),
'facebook_id': data.get('id')
}
account = accounts.save(data_, account=account)
db.session.commit()
session['id'] = account.id
return redirect(next_url)
<｜fim▁end｜> | github_authorized |
<|file_name|>sessions_api.py<|end_file_name|><｜fim▁begin｜>import requests
from flask import session, Blueprint, redirect
from flask import request
from grano import authz
from grano.lib.exc import BadRequest
from grano.lib.serialisation import jsonify
from grano.views.cache import validate_cache
from grano.core import db, url_for, app
from grano.providers import github, twitter, facebook
from grano.model import Account
from grano.logic import accounts
blueprint = Blueprint('sessions_api', __name__)
@blueprint.route('/api/1/sessions', methods=['GET'])
def status():
permissions = {}
if authz.logged_in():
for permission in request.account.permissions:
permissions[permission.project.slug] = {
'reader': permission.reader,
'editor': permission.editor,
'admin': permission.admin
}
keys = {
'p': repr(permissions),
'i': request.account.id if authz.logged_in() else None
}
validate_cache(keys=keys)
return jsonify({
'logged_in': authz.logged_in(),
'api_key': request.account.api_key if authz.logged_in() else None,
'account': request.account if request.account else None,
'permissions': permissions
})
def provider_not_enabled(name):
return jsonify({
'status': 501,
'name': 'Provider not configured: %s' % name,
'message': 'There are no OAuth credentials given for %s' % name,
}, status=501)
@blueprint.route('/api/1/sessions/logout', methods=['GET'])
def logout():
#authz.require(authz.logged_in())
session.clear()
return redirect(request.args.get('next_url', '/'))
@blueprint.route('/api/1/sessions/login/github', methods=['GET'])
def github_login():
if not app.config.get('GITHUB_CLIENT_ID'):
return provider_not_enabled('github')
callback=url_for('sessions_api.github_authorized')
session.clear()
if not request.args.get('next_url'):
raise BadRequest("No 'next_url' is specified.")
session['next_url'] = request.args.get('next_url')
return github.authorize(callback=callback)
@blueprint.route('/api/1/sessions/callback/github', methods=['GET'])
@github.authorized_handler
def github_authorized(resp):
next_url = session.get('next_url', '/')
if resp is None or not 'access_token' in resp:
return redirect(next_url)
access_token = resp['access_token']
session['access_token'] = access_token, ''
res = requests.get('https://api.github.com/user?access_token=%s' % access_token,
verify=False)
data = res.json()
account = Account.by_github_id(data.get('id'))
data_ = {
'full_name': data.get('name'),
'login': data.get('login'),
'email': data.get('email'),
'github_id': data.get('id')
}
account = accounts.save(data_, account=account)
db.session.commit()
session['id'] = account.id
return redirect(next_url)
@blueprint.route('/api/1/sessions/login/twitter', methods=['GET'])
def <|fim_middle|>():
if not app.config.get('TWITTER_API_KEY'):
return provider_not_enabled('twitter')
callback=url_for('sessions_api.twitter_authorized')
session.clear()
if not request.args.get('next_url'):
raise BadRequest("No 'next_url' is specified.")
session['next_url'] = request.args.get('next_url')
return twitter.authorize(callback=callback)
@blueprint.route('/api/1/sessions/callback/twitter', methods=['GET'])
@twitter.authorized_handler
def twitter_authorized(resp):
next_url = session.get('next_url', '/')
if resp is None or not 'oauth_token' in resp:
return redirect(next_url)
session['twitter_token'] = (resp['oauth_token'],
resp['oauth_token_secret'])
res = twitter.get('users/show.json?user_id=%s' % resp.get('user_id'))
account = Account.by_twitter_id(res.data.get('id'))
data_ = {
'full_name': res.data.get('name'),
'login': res.data.get('screen_name'),
'twitter_id': res.data.get('id')
}
account = accounts.save(data_, account=account)
db.session.commit()
session['id'] = account.id
return redirect(next_url)
@blueprint.route('/api/1/sessions/login/facebook', methods=['GET'])
def facebook_login():
if not app.config.get('FACEBOOK_APP_ID'):
return provider_not_enabled('facebook')
callback=url_for('sessions_api.facebook_authorized')
session.clear()
if not request.args.get('next_url'):
raise BadRequest("No 'next_url' is specified.")
session['next_url'] = request.args.get('next_url')
return facebook.authorize(callback=callback)
@blueprint.route('/api/1/sessions/callback/facebook', methods=['GET'])
@facebook.authorized_handler
def facebook_authorized(resp):
next_url = session.get('next_url', '/')
if resp is None or not 'access_token' in resp:
return redirect(next_url)
session['facebook_token'] = (resp.get('access_token'), '')
data = facebook.get('/me').data
account = Account.by_facebook_id(data.get('id'))
data_ = {
'full_name': data.get('name'),
'login': data.get('username'),
'email': data.get('email'),
'facebook_id': data.get('id')
}
account = accounts.save(data_, account=account)
db.session.commit()
session['id'] = account.id
return redirect(next_url)
<｜fim▁end｜> | twitter_login |
<|file_name|>sessions_api.py<|end_file_name|><｜fim▁begin｜>import requests
from flask import session, Blueprint, redirect
from flask import request
from grano import authz
from grano.lib.exc import BadRequest
from grano.lib.serialisation import jsonify
from grano.views.cache import validate_cache
from grano.core import db, url_for, app
from grano.providers import github, twitter, facebook
from grano.model import Account
from grano.logic import accounts
blueprint = Blueprint('sessions_api', __name__)
@blueprint.route('/api/1/sessions', methods=['GET'])
def status():
permissions = {}
if authz.logged_in():
for permission in request.account.permissions:
permissions[permission.project.slug] = {
'reader': permission.reader,
'editor': permission.editor,
'admin': permission.admin
}
keys = {
'p': repr(permissions),
'i': request.account.id if authz.logged_in() else None
}
validate_cache(keys=keys)
return jsonify({
'logged_in': authz.logged_in(),
'api_key': request.account.api_key if authz.logged_in() else None,
'account': request.account if request.account else None,
'permissions': permissions
})
def provider_not_enabled(name):
return jsonify({
'status': 501,
'name': 'Provider not configured: %s' % name,
'message': 'There are no OAuth credentials given for %s' % name,
}, status=501)
@blueprint.route('/api/1/sessions/logout', methods=['GET'])
def logout():
#authz.require(authz.logged_in())
session.clear()
return redirect(request.args.get('next_url', '/'))
@blueprint.route('/api/1/sessions/login/github', methods=['GET'])
def github_login():
if not app.config.get('GITHUB_CLIENT_ID'):
return provider_not_enabled('github')
callback=url_for('sessions_api.github_authorized')
session.clear()
if not request.args.get('next_url'):
raise BadRequest("No 'next_url' is specified.")
session['next_url'] = request.args.get('next_url')
return github.authorize(callback=callback)
@blueprint.route('/api/1/sessions/callback/github', methods=['GET'])
@github.authorized_handler
def github_authorized(resp):
next_url = session.get('next_url', '/')
if resp is None or not 'access_token' in resp:
return redirect(next_url)
access_token = resp['access_token']
session['access_token'] = access_token, ''
res = requests.get('https://api.github.com/user?access_token=%s' % access_token,
verify=False)
data = res.json()
account = Account.by_github_id(data.get('id'))
data_ = {
'full_name': data.get('name'),
'login': data.get('login'),
'email': data.get('email'),
'github_id': data.get('id')
}
account = accounts.save(data_, account=account)
db.session.commit()
session['id'] = account.id
return redirect(next_url)
@blueprint.route('/api/1/sessions/login/twitter', methods=['GET'])
def twitter_login():
if not app.config.get('TWITTER_API_KEY'):
return provider_not_enabled('twitter')
callback=url_for('sessions_api.twitter_authorized')
session.clear()
if not request.args.get('next_url'):
raise BadRequest("No 'next_url' is specified.")
session['next_url'] = request.args.get('next_url')
return twitter.authorize(callback=callback)
@blueprint.route('/api/1/sessions/callback/twitter', methods=['GET'])
@twitter.authorized_handler
def <|fim_middle|>(resp):
next_url = session.get('next_url', '/')
if resp is None or not 'oauth_token' in resp:
return redirect(next_url)
session['twitter_token'] = (resp['oauth_token'],
resp['oauth_token_secret'])
res = twitter.get('users/show.json?user_id=%s' % resp.get('user_id'))
account = Account.by_twitter_id(res.data.get('id'))
data_ = {
'full_name': res.data.get('name'),
'login': res.data.get('screen_name'),
'twitter_id': res.data.get('id')
}
account = accounts.save(data_, account=account)
db.session.commit()
session['id'] = account.id
return redirect(next_url)
@blueprint.route('/api/1/sessions/login/facebook', methods=['GET'])
def facebook_login():
if not app.config.get('FACEBOOK_APP_ID'):
return provider_not_enabled('facebook')
callback=url_for('sessions_api.facebook_authorized')
session.clear()
if not request.args.get('next_url'):
raise BadRequest("No 'next_url' is specified.")
session['next_url'] = request.args.get('next_url')
return facebook.authorize(callback=callback)
@blueprint.route('/api/1/sessions/callback/facebook', methods=['GET'])
@facebook.authorized_handler
def facebook_authorized(resp):
next_url = session.get('next_url', '/')
if resp is None or not 'access_token' in resp:
return redirect(next_url)
session['facebook_token'] = (resp.get('access_token'), '')
data = facebook.get('/me').data
account = Account.by_facebook_id(data.get('id'))
data_ = {
'full_name': data.get('name'),
'login': data.get('username'),
'email': data.get('email'),
'facebook_id': data.get('id')
}
account = accounts.save(data_, account=account)
db.session.commit()
session['id'] = account.id
return redirect(next_url)
<｜fim▁end｜> | twitter_authorized |
<|file_name|>sessions_api.py<|end_file_name|><｜fim▁begin｜>import requests
from flask import session, Blueprint, redirect
from flask import request
from grano import authz
from grano.lib.exc import BadRequest
from grano.lib.serialisation import jsonify
from grano.views.cache import validate_cache
from grano.core import db, url_for, app
from grano.providers import github, twitter, facebook
from grano.model import Account
from grano.logic import accounts
blueprint = Blueprint('sessions_api', __name__)
@blueprint.route('/api/1/sessions', methods=['GET'])
def status():
permissions = {}
if authz.logged_in():
for permission in request.account.permissions:
permissions[permission.project.slug] = {
'reader': permission.reader,
'editor': permission.editor,
'admin': permission.admin
}
keys = {
'p': repr(permissions),
'i': request.account.id if authz.logged_in() else None
}
validate_cache(keys=keys)
return jsonify({
'logged_in': authz.logged_in(),
'api_key': request.account.api_key if authz.logged_in() else None,
'account': request.account if request.account else None,
'permissions': permissions
})
def provider_not_enabled(name):
return jsonify({
'status': 501,
'name': 'Provider not configured: %s' % name,
'message': 'There are no OAuth credentials given for %s' % name,
}, status=501)
@blueprint.route('/api/1/sessions/logout', methods=['GET'])
def logout():
#authz.require(authz.logged_in())
session.clear()
return redirect(request.args.get('next_url', '/'))
@blueprint.route('/api/1/sessions/login/github', methods=['GET'])
def github_login():
if not app.config.get('GITHUB_CLIENT_ID'):
return provider_not_enabled('github')
callback=url_for('sessions_api.github_authorized')
session.clear()
if not request.args.get('next_url'):
raise BadRequest("No 'next_url' is specified.")
session['next_url'] = request.args.get('next_url')
return github.authorize(callback=callback)
@blueprint.route('/api/1/sessions/callback/github', methods=['GET'])
@github.authorized_handler
def github_authorized(resp):
next_url = session.get('next_url', '/')
if resp is None or not 'access_token' in resp:
return redirect(next_url)
access_token = resp['access_token']
session['access_token'] = access_token, ''
res = requests.get('https://api.github.com/user?access_token=%s' % access_token,
verify=False)
data = res.json()
account = Account.by_github_id(data.get('id'))
data_ = {
'full_name': data.get('name'),
'login': data.get('login'),
'email': data.get('email'),
'github_id': data.get('id')
}
account = accounts.save(data_, account=account)
db.session.commit()
session['id'] = account.id
return redirect(next_url)
@blueprint.route('/api/1/sessions/login/twitter', methods=['GET'])
def twitter_login():
if not app.config.get('TWITTER_API_KEY'):
return provider_not_enabled('twitter')
callback=url_for('sessions_api.twitter_authorized')
session.clear()
if not request.args.get('next_url'):
raise BadRequest("No 'next_url' is specified.")
session['next_url'] = request.args.get('next_url')
return twitter.authorize(callback=callback)
@blueprint.route('/api/1/sessions/callback/twitter', methods=['GET'])
@twitter.authorized_handler
def twitter_authorized(resp):
next_url = session.get('next_url', '/')
if resp is None or not 'oauth_token' in resp:
return redirect(next_url)
session['twitter_token'] = (resp['oauth_token'],
resp['oauth_token_secret'])
res = twitter.get('users/show.json?user_id=%s' % resp.get('user_id'))
account = Account.by_twitter_id(res.data.get('id'))
data_ = {
'full_name': res.data.get('name'),
'login': res.data.get('screen_name'),
'twitter_id': res.data.get('id')
}
account = accounts.save(data_, account=account)
db.session.commit()
session['id'] = account.id
return redirect(next_url)
@blueprint.route('/api/1/sessions/login/facebook', methods=['GET'])
def <|fim_middle|>():
if not app.config.get('FACEBOOK_APP_ID'):
return provider_not_enabled('facebook')
callback=url_for('sessions_api.facebook_authorized')
session.clear()
if not request.args.get('next_url'):
raise BadRequest("No 'next_url' is specified.")
session['next_url'] = request.args.get('next_url')
return facebook.authorize(callback=callback)
@blueprint.route('/api/1/sessions/callback/facebook', methods=['GET'])
@facebook.authorized_handler
def facebook_authorized(resp):
next_url = session.get('next_url', '/')
if resp is None or not 'access_token' in resp:
return redirect(next_url)
session['facebook_token'] = (resp.get('access_token'), '')
data = facebook.get('/me').data
account = Account.by_facebook_id(data.get('id'))
data_ = {
'full_name': data.get('name'),
'login': data.get('username'),
'email': data.get('email'),
'facebook_id': data.get('id')
}
account = accounts.save(data_, account=account)
db.session.commit()
session['id'] = account.id
return redirect(next_url)
<｜fim▁end｜> | facebook_login |
<|file_name|>sessions_api.py<|end_file_name|><｜fim▁begin｜>import requests
from flask import session, Blueprint, redirect
from flask import request
from grano import authz
from grano.lib.exc import BadRequest
from grano.lib.serialisation import jsonify
from grano.views.cache import validate_cache
from grano.core import db, url_for, app
from grano.providers import github, twitter, facebook
from grano.model import Account
from grano.logic import accounts
blueprint = Blueprint('sessions_api', __name__)
@blueprint.route('/api/1/sessions', methods=['GET'])
def status():
permissions = {}
if authz.logged_in():
for permission in request.account.permissions:
permissions[permission.project.slug] = {
'reader': permission.reader,
'editor': permission.editor,
'admin': permission.admin
}
keys = {
'p': repr(permissions),
'i': request.account.id if authz.logged_in() else None
}
validate_cache(keys=keys)
return jsonify({
'logged_in': authz.logged_in(),
'api_key': request.account.api_key if authz.logged_in() else None,
'account': request.account if request.account else None,
'permissions': permissions
})
def provider_not_enabled(name):
return jsonify({
'status': 501,
'name': 'Provider not configured: %s' % name,
'message': 'There are no OAuth credentials given for %s' % name,
}, status=501)
@blueprint.route('/api/1/sessions/logout', methods=['GET'])
def logout():
#authz.require(authz.logged_in())
session.clear()
return redirect(request.args.get('next_url', '/'))
@blueprint.route('/api/1/sessions/login/github', methods=['GET'])
def github_login():
if not app.config.get('GITHUB_CLIENT_ID'):
return provider_not_enabled('github')
callback=url_for('sessions_api.github_authorized')
session.clear()
if not request.args.get('next_url'):
raise BadRequest("No 'next_url' is specified.")
session['next_url'] = request.args.get('next_url')
return github.authorize(callback=callback)
@blueprint.route('/api/1/sessions/callback/github', methods=['GET'])
@github.authorized_handler
def github_authorized(resp):
next_url = session.get('next_url', '/')
if resp is None or not 'access_token' in resp:
return redirect(next_url)
access_token = resp['access_token']
session['access_token'] = access_token, ''
res = requests.get('https://api.github.com/user?access_token=%s' % access_token,
verify=False)
data = res.json()
account = Account.by_github_id(data.get('id'))
data_ = {
'full_name': data.get('name'),
'login': data.get('login'),
'email': data.get('email'),
'github_id': data.get('id')
}
account = accounts.save(data_, account=account)
db.session.commit()
session['id'] = account.id
return redirect(next_url)
@blueprint.route('/api/1/sessions/login/twitter', methods=['GET'])
def twitter_login():
if not app.config.get('TWITTER_API_KEY'):
return provider_not_enabled('twitter')
callback=url_for('sessions_api.twitter_authorized')
session.clear()
if not request.args.get('next_url'):
raise BadRequest("No 'next_url' is specified.")
session['next_url'] = request.args.get('next_url')
return twitter.authorize(callback=callback)
@blueprint.route('/api/1/sessions/callback/twitter', methods=['GET'])
@twitter.authorized_handler
def twitter_authorized(resp):
next_url = session.get('next_url', '/')
if resp is None or not 'oauth_token' in resp:
return redirect(next_url)
session['twitter_token'] = (resp['oauth_token'],
resp['oauth_token_secret'])
res = twitter.get('users/show.json?user_id=%s' % resp.get('user_id'))
account = Account.by_twitter_id(res.data.get('id'))
data_ = {
'full_name': res.data.get('name'),
'login': res.data.get('screen_name'),
'twitter_id': res.data.get('id')
}
account = accounts.save(data_, account=account)
db.session.commit()
session['id'] = account.id
return redirect(next_url)
@blueprint.route('/api/1/sessions/login/facebook', methods=['GET'])
def facebook_login():
if not app.config.get('FACEBOOK_APP_ID'):
return provider_not_enabled('facebook')
callback=url_for('sessions_api.facebook_authorized')
session.clear()
if not request.args.get('next_url'):
raise BadRequest("No 'next_url' is specified.")
session['next_url'] = request.args.get('next_url')
return facebook.authorize(callback=callback)
@blueprint.route('/api/1/sessions/callback/facebook', methods=['GET'])
@facebook.authorized_handler
def <|fim_middle|>(resp):
next_url = session.get('next_url', '/')
if resp is None or not 'access_token' in resp:
return redirect(next_url)
session['facebook_token'] = (resp.get('access_token'), '')
data = facebook.get('/me').data
account = Account.by_facebook_id(data.get('id'))
data_ = {
'full_name': data.get('name'),
'login': data.get('username'),
'email': data.get('email'),
'facebook_id': data.get('id')
}
account = accounts.save(data_, account=account)
db.session.commit()
session['id'] = account.id
return redirect(next_url)
<｜fim▁end｜> | facebook_authorized |
<|file_name|>setup_pkgresource.py<|end_file_name|><｜fim▁begin｜>def _setup_pkgresources():
import pkg_resources
import os
import plistlib
pl = plistlib.readPlist(os.path.join(
os.path.dirname(os.getenv('RESOURCEPATH')), "Info.plist"))<｜fim▁hole｜> pkg_resources.set_extraction_path(path)
_setup_pkgresources()<｜fim▁end｜> | appname = pl.get('CFBundleIdentifier')
if appname is None:
appname = pl['CFBundleDisplayName']
path = os.path.expanduser('~/Library/Caches/%s/python-eggs' % (appname,)) |
<|file_name|>setup_pkgresource.py<|end_file_name|><｜fim▁begin｜>def _setup_pkgresources():
<|fim_middle|>
_setup_pkgresources()
<｜fim▁end｜> | import pkg_resources
import os
import plistlib
pl = plistlib.readPlist(os.path.join(
os.path.dirname(os.getenv('RESOURCEPATH')), "Info.plist"))
appname = pl.get('CFBundleIdentifier')
if appname is None:
appname = pl['CFBundleDisplayName']
path = os.path.expanduser('~/Library/Caches/%s/python-eggs' % (appname,))
pkg_resources.set_extraction_path(path) |
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.