Spaces:
Running
Running
# -- coding: utf-8 -- | |
import json | |
import requests | |
import logging | |
from datetime import datetime, timezone, timedelta | |
from urllib.parse import quote, urlencode | |
import requests_oauthlib # For version logging | |
# Assuming you have a sessions.py with create_session | |
# If sessions.py or create_session is not found, it will raise an ImportError, | |
# which is appropriate for a module that depends on it. | |
from sessions import create_session # Make sure this file exists and is correct | |
logging.basicConfig(level=logging.INFO, format='%(asctime)s - %(levelname)s - %(message)s') | |
API_V2_BASE = 'https://api.linkedin.com/v2' | |
API_REST_BASE = "https://api.linkedin.com/rest" | |
LINKEDIN_API_VERSION = "202502" | |
# --- ID to Name Mapping Helper Functions --- | |
def _fetch_linkedin_names(session, url, params, result_key_path, name_key_path, id_key="id"): | |
""" | |
Generic helper to fetch and map IDs to names from a LinkedIn API endpoint. | |
""" | |
mapping = {} | |
request_url_for_logging = url | |
response_obj = None | |
try: | |
logging.debug(f"_fetch_linkedin_names (id: {id(session)}): About to prepare request. Session token: {session.token}, Session authorized: {session.authorized}, Expires at: {session.token.get('expires_at') if session.token else 'N/A'}") | |
logging.debug(f"_fetch_linkedin_names (id: {id(session)}): Session auth object: type={type(session.auth)}, value={session.auth}") | |
req = requests.Request('GET', url, params=params) | |
prepared_req = session.prepare_request(req) | |
request_url_for_logging = prepared_req.url | |
logging.debug(f"Fetching names from URL: {request_url_for_logging}") | |
logging.debug(f"Request headers for _fetch_linkedin_names: {json.dumps(dict(prepared_req.headers), indent=2)}") | |
response_obj = session.send(prepared_req, timeout=30) | |
response_obj.raise_for_status() | |
data = response_obj.json() | |
items = data | |
for key in result_key_path: | |
if isinstance(items, dict): | |
items = items.get(key) | |
if items is None: | |
logging.warning(f"Key '{key}' not found in response from {request_url_for_logging} at path {result_key_path}. Response data: {json.dumps(data, indent=2)}") | |
return mapping | |
else: | |
logging.warning(f"Expected dict to get key '{key}' but got {type(items)} at path {result_key_path} for URL {request_url_for_logging}. Check result_key_path.") | |
return mapping | |
if items is None: | |
logging.warning(f"Items became None after navigating result_key_path for URL {request_url_for_logging}. Path: {result_key_path}") | |
return mapping | |
if isinstance(items, dict): | |
for item_id_str, item_data in items.items(): | |
name = item_data | |
for key_nav in name_key_path: | |
if isinstance(name, dict): | |
name = name.get(key_nav) | |
else: | |
name = None | |
break | |
if name: | |
mapping[item_id_str] = name | |
else: | |
logging.warning(f"No name found for ID {item_id_str} at path {name_key_path} in item: {item_data} from URL {request_url_for_logging}") | |
elif isinstance(items, list): | |
for item in items: | |
item_id_val = item.get(id_key) | |
name = item | |
for key_nav in name_key_path: | |
if isinstance(name, dict): | |
name = name.get(key_nav) | |
else: | |
name = None | |
break | |
if item_id_val is not None and name: | |
mapping[str(item_id_val)] = name | |
else: | |
if item_id_val is None: | |
logging.warning(f"No ID ('{id_key}') found in item: {item} from URL {request_url_for_logging}") | |
if name is None: | |
logging.warning(f"No name found at path {name_key_path} for item with ID '{item_id_val}' in item: {item} from URL {request_url_for_logging}") | |
else: | |
logging.warning(f"Expected list or dict of items at {result_key_path} from URL {request_url_for_logging}, got {type(items)}. Full items: {items}") | |
except requests.exceptions.HTTPError as http_err: | |
status_code = "N/A"; error_text = str(http_err); response_headers = {} | |
if http_err.response is not None: | |
status_code = http_err.response.status_code | |
error_text = http_err.response.text | |
response_headers = dict(http_err.response.headers) | |
logging.error(f"HTTP error in _fetch_linkedin_names from {request_url_for_logging} (Status: {status_code}): {error_text}") | |
logging.error(f"Response Headers: {json.dumps(response_headers, indent=2)}") | |
except requests.exceptions.RequestException as req_err: | |
logging.error(f"Request error in _fetch_linkedin_names from {request_url_for_logging}: {str(req_err)}") | |
if req_err.response is not None: | |
logging.error(f"Associated Response Status: {req_err.response.status_code}, Text: {req_err.response.text}, Headers: {json.dumps(dict(req_err.response.headers), indent=2)}") | |
except json.JSONDecodeError as json_err: | |
response_text = "Not available" | |
if response_obj is not None and hasattr(response_obj, 'text'): response_text = response_obj.text | |
logging.error(f"Error decoding JSON for names from {request_url_for_logging}: {json_err}. Response text: {response_text}") | |
except Exception as e: | |
logging.error(f"Unexpected error in _fetch_linkedin_names from {request_url_for_logging}: {e}", exc_info=True) | |
if response_obj is not None: | |
logging.error(f"Associated Response (if any) Status: {getattr(response_obj, 'status_code', 'N/A')}, Text: {getattr(response_obj, 'text', 'N/A')}") | |
return mapping | |
def get_functions_map(session): | |
url = f"{API_V2_BASE}/functions" | |
params = {'count': 200} | |
logging.info("Fetching all LinkedIn functions.") | |
return _fetch_linkedin_names(session, url, params, ["elements"], ["name", "localized", "en_US"], "id") | |
def get_seniorities_map(session): | |
url = f"{API_V2_BASE}/seniorities" | |
params = {'count': 200} | |
logging.info("Fetching all LinkedIn seniorities.") | |
return _fetch_linkedin_names(session, url, params, ["elements"], ["name", "localized", "en_US"], "id") | |
def get_industries_map(session, industry_urns, version="DEFAULT"): | |
industry_ids = [_parse_urn_to_id(urn) for urn in industry_urns or []] | |
unique_ids = set(filter(None, industry_ids)) | |
if not unique_ids: | |
return {} | |
url = f"{API_V2_BASE}/industryTaxonomyVersions/{version}/industries" | |
params = { 'start': 0, 'count': 500 } | |
request_url_for_logging = url | |
response_obj = None | |
logging.info(f"Fetching all industries (to filter {len(unique_ids)} IDs)") | |
try: | |
logging.debug(f"get_industries_map (id: {id(session)}): About to prepare request. Session token: {session.token}, Session authorized: {session.authorized}, Expires at: {session.token.get('expires_at') if session.token else 'N/A'}") | |
logging.debug(f"get_industries_map (id: {id(session)}): Session auth object: type={type(session.auth)}, value={session.auth}") | |
req = requests.Request('GET', url, params=params) | |
prepared_req = session.prepare_request(req) | |
request_url_for_logging = prepared_req.url | |
logging.debug(f"Requesting all industries from URL: {request_url_for_logging}") | |
logging.debug(f"Request headers for get_industries_map: {json.dumps(dict(prepared_req.headers), indent=2)}") | |
response_obj = session.send(prepared_req, timeout=30) | |
response_obj.raise_for_status() | |
data = response_obj.json() | |
elements = data.get('elements', []) | |
mapping = {} | |
for el in elements: | |
el_id = el.get('id') | |
if el_id and str(el_id) in unique_ids: | |
name = el.get('name', {}).get('localized', {}).get('en_US') | |
if name: | |
mapping[str(el_id)] = name | |
else: | |
logging.warning(f"Industry {el_id} has no en_US name field in element: {el} from URL {request_url_for_logging}") | |
return mapping | |
except requests.exceptions.HTTPError as http_err: | |
status_code = "N/A"; error_text = str(http_err); response_headers = {} | |
if http_err.response is not None: | |
status_code = http_err.response.status_code | |
error_text = http_err.response.text | |
response_headers = dict(http_err.response.headers) | |
logging.error(f"HTTP error fetching all industries from {request_url_for_logging} (Status: {status_code}): {error_text}") | |
logging.error(f"Response Headers: {json.dumps(response_headers, indent=2)}") | |
return {} | |
except requests.exceptions.RequestException as req_err: | |
logging.error(f"Request error fetching all industries from {request_url_for_logging}: {str(req_err)}") | |
if req_err.response is not None: | |
logging.error(f"Associated Response Status: {req_err.response.status_code}, Text: {req_err.response.text}, Headers: {json.dumps(dict(req_err.response.headers), indent=2)}") | |
return {} | |
except json.JSONDecodeError as json_err: | |
response_text = "Not available" | |
if response_obj is not None and hasattr(response_obj, 'text'): response_text = response_obj.text | |
logging.error(f"Error decoding JSON for industries from {request_url_for_logging}: {json_err}. Response text: {response_text}") | |
return {} | |
except Exception as e: | |
logging.error(f"Unexpected error fetching all industries from {request_url_for_logging}: {e}", exc_info=True) | |
if response_obj is not None: | |
logging.error(f"Associated Response (if any) Status: {getattr(response_obj, 'status_code', 'N/A')}, Text: {getattr(response_obj, 'text', 'N/A')}") | |
return {} | |
def get_geo_map(session, geo_urns): | |
if not geo_urns: return {} | |
geo_ids = [_parse_urn_to_id(urn) for urn in geo_urns if urn] | |
unique_ids = list(set(filter(None, geo_ids))) | |
if not unique_ids: return {} | |
MAX_GEO_IDS_PER_CALL = 100 | |
all_geo_mappings = {} | |
for i in range(0, len(unique_ids), MAX_GEO_IDS_PER_CALL): | |
chunk_ids = unique_ids[i:i + MAX_GEO_IDS_PER_CALL] | |
if not chunk_ids: continue | |
ids_param_value = "List(" + ",".join(map(str, chunk_ids)) + ")" | |
locale_param_value = "(language:en,country:US)" | |
url = f"{API_V2_BASE}/geo" | |
geo_params = { 'ids': ids_param_value, 'locale': locale_param_value } | |
logging.info(f"Fetching names for {len(chunk_ids)} geo IDs (chunk {i//MAX_GEO_IDS_PER_CALL + 1})") | |
chunk_mapping = _fetch_linkedin_names(session, url, geo_params, ["results"], ["defaultLocalizedName", "value"]) | |
all_geo_mappings.update(chunk_mapping) | |
return all_geo_mappings | |
def _parse_urn_to_id(urn_string): | |
if not isinstance(urn_string, str): | |
logging.debug(f"Invalid URN type: {type(urn_string)}, value: {urn_string}. Cannot parse ID.") | |
return None | |
try: | |
return urn_string.split(':')[-1] | |
except IndexError: | |
logging.warning(f"Could not parse ID from URN: {urn_string}") | |
return None | |
except Exception as e: | |
logging.error(f"Unexpected error parsing URN {urn_string}: {e}") | |
return None | |
# --- Follower Data Fetching Functions --- | |
def fetch_monthly_follower_gains(session, org_urn, api_rest_base_url): | |
now_utc = datetime.now(timezone.utc) | |
start_of_reporting_period = (now_utc - timedelta(days=365)).replace(day=1, hour=0, minute=0, second=0, microsecond=0) | |
start_ms = int(start_of_reporting_period.timestamp() * 1000) | |
base_url = f"{api_rest_base_url}/organizationalEntityFollowerStatistics" | |
time_intervals_value = f"(timeRange:(start:{start_ms}),timeGranularityType:MONTH)" | |
api_params = { | |
"q": "organizationalEntity", | |
"organizationalEntity": org_urn, | |
"timeIntervals": time_intervals_value | |
} | |
logging.info(f"Preparing to fetch monthly follower gains for {org_urn}.") | |
logging.debug(f"API Parameters for monthly gains: {json.dumps(api_params)}") | |
results = [] | |
request_url_for_logging = "Not constructed" | |
response_obj = None | |
try: | |
logging.debug(f"fetch_monthly_follower_gains (id: {id(session)}): About to prepare request. Session token: {session.token}, Session authorized: {session.authorized}, Expires at: {session.token.get('expires_at') if session.token else 'N/A'}") | |
logging.debug(f"fetch_monthly_follower_gains (id: {id(session)}): Session auth object: type={type(session.auth)}, value={session.auth}") | |
logging.debug(f"fetch_monthly_follower_gains (id: {id(session)}): Auto-refresh URL: {session.auto_refresh_url}, Auto-refresh kwargs: {session.auto_refresh_kwargs}") | |
if session.token and 'access_token' in session.token: | |
logging.debug(f"fetch_monthly_follower_gains: Access token (partial): {str(session.token['access_token'])[:20]}...") | |
else: | |
logging.warning("fetch_monthly_follower_gains: session.token is None or 'access_token' key is missing before prepare_request.") | |
req = requests.Request('GET', base_url, params=api_params) | |
prepared_req = session.prepare_request(req) | |
request_url_for_logging = prepared_req.url | |
logging.info(f"Requesting monthly follower gains from URL: {request_url_for_logging}") | |
logging.debug(f"Request Headers for monthly gains: {json.dumps(dict(prepared_req.headers), indent=2)}") | |
response_obj = session.send(prepared_req, timeout=30) | |
response_obj.raise_for_status() | |
data = response_obj.json() | |
elements = data.get('elements', []) | |
if not elements: | |
logging.info(f"No 'elements' found in monthly follower gains API response for {org_urn} (start_ms {start_ms}). Response data: {json.dumps(data, indent=2)}") | |
for item in elements: | |
time_range = item.get('timeRange', {}) | |
ts = time_range.get('start') | |
if ts is None: | |
logging.warning(f"Skipping item due to missing 'start' timestamp: {item}") | |
continue | |
date_obj = datetime.fromtimestamp(ts / 1000, tz=timezone.utc) | |
date_str = date_obj.strftime('%Y-%m-%d') | |
gains = item.get('followerGains', {}) | |
if gains is None: gains = {} | |
results.append({ | |
'category_name': date_str, | |
'follower_count_organic': gains.get('organicFollowerGain', 0), | |
'follower_count_paid': gains.get('paidFollowerGain', 0), | |
'follower_count_type': 'follower_gains_monthly', | |
'organization_urn': org_urn | |
}) | |
logging.info(f"Fetched {len(results)} monthly follower entries for {org_urn} starting from {start_of_reporting_period.strftime('%Y-%m-%d')}.") | |
except requests.exceptions.HTTPError as http_err: | |
code = "N/A"; text = str(http_err); resp_headers = {} | |
if http_err.response is not None: | |
code = http_err.response.status_code | |
text = http_err.response.text | |
resp_headers = dict(http_err.response.headers) | |
logging.error(f"HTTP error fetching monthly gains for {org_urn}: {code} - {text}") | |
logging.error(f"Request URL was: {request_url_for_logging}") | |
logging.error(f"Response Headers: {json.dumps(resp_headers, indent=2)}") | |
except requests.exceptions.RequestException as req_err: | |
logging.error(f"RequestException fetching monthly gains for {org_urn}: {str(req_err)}") | |
logging.error(f"Request URL was: {request_url_for_logging}") | |
if req_err.response is not None: | |
logging.error(f"Associated Response Status: {req_err.response.status_code}") | |
logging.error(f"Associated Response Text: {req_err.response.text}") | |
logging.error(f"Associated Response Headers: {json.dumps(dict(req_err.response.headers), indent=2)}") | |
except json.JSONDecodeError as json_err: | |
response_text = "Not available" | |
if response_obj is not None and hasattr(response_obj, 'text'): | |
response_text = response_obj.text | |
logging.error(f"Error decoding JSON for monthly follower gains for {org_urn}: {json_err}") | |
logging.error(f"Request URL was: {request_url_for_logging}") | |
logging.error(f"Raw Response Text: {response_text}") | |
except Exception as ex: | |
logging.error(f"An unexpected error occurred while fetching monthly gains for {org_urn}: {str(ex)}", exc_info=True) | |
logging.error(f"Request URL was: {request_url_for_logging}") | |
if response_obj is not None: | |
logging.error(f"Response Status (if available): {getattr(response_obj, 'status_code', 'N/A')}") | |
logging.error(f"Response Text (if available): {getattr(response_obj, 'text', 'N/A')}") | |
logging.error(f"Response Headers (if available): {json.dumps(dict(getattr(response_obj, 'headers', {})), indent=2)}") | |
return results | |
def fetch_follower_demographics(session, org_urn, functions_map, seniorities_map): | |
final_demographics_results = [] | |
base_url = f"{API_REST_BASE}/organizationalEntityFollowerStatistics" | |
params = { 'q': 'organizationalEntity', 'organizationalEntity': org_urn } | |
logging.info(f"Preparing to fetch follower demographics for org URN {org_urn}.") | |
logging.debug(f"API Parameters for demographics: {json.dumps(params)}") | |
request_url_for_logging = "Not constructed" | |
response_obj = None | |
try: | |
logging.debug(f"fetch_follower_demographics (id: {id(session)}): About to prepare request. Session token: {session.token}, Session authorized: {session.authorized}, Expires at: {session.token.get('expires_at') if session.token else 'N/A'}") | |
logging.debug(f"fetch_follower_demographics (id: {id(session)}): Session auth object: type={type(session.auth)}, value={session.auth}") | |
logging.debug(f"fetch_follower_demographics (id: {id(session)}): Auto-refresh URL: {session.auto_refresh_url}, Auto-refresh kwargs: {session.auto_refresh_kwargs}") | |
if session.token and 'access_token' in session.token: | |
logging.debug(f"fetch_follower_demographics: Access token (partial): {str(session.token['access_token'])[:20]}...") | |
else: | |
logging.warning("fetch_follower_demographics: session.token is None or 'access_token' key is missing before prepare_request.") | |
req = requests.Request('GET', base_url, params=params) | |
prepared_req = session.prepare_request(req) | |
request_url_for_logging = prepared_req.url | |
logging.info(f"Requesting follower demographics from URL: {request_url_for_logging}") | |
logging.debug(f"Request Headers for demographics: {json.dumps(dict(prepared_req.headers), indent=2)}") | |
response_obj = session.send(prepared_req, timeout=30) | |
response_obj.raise_for_status() | |
data = response_obj.json() | |
elements = data.get("elements", []) | |
if not elements: | |
logging.warning(f"No elements found in follower demographics response for {org_urn}. Response data: {json.dumps(data, indent=2)}") | |
return [] | |
stat_element = elements[0] if len(elements) > 0 else None | |
if not stat_element: | |
logging.warning(f"Elements list is empty or stat_element is None in demographics response for {org_urn}. Response data: {json.dumps(data, indent=2)}") | |
return [] | |
def _get_entries_for_type(raw_items_list, type_name, id_map, id_field_name_in_item, org_urn_val): | |
current_type_entries = [] | |
if not raw_items_list: | |
logging.debug(f"No raw items for demographic type '{type_name}' for org {org_urn_val}.") | |
return current_type_entries | |
for item in raw_items_list: | |
category_name_val = "Unknown" | |
if type_name == "follower_association": | |
category_name_val = item.get(id_field_name_in_item, f"Unknown {id_field_name_in_item}") | |
else: | |
urn_val = item.get(id_field_name_in_item) | |
entity_id = _parse_urn_to_id(urn_val) | |
if entity_id and id_map: | |
category_name_val = id_map.get(str(entity_id), f"Unknown {type_name.split('_')[-1].capitalize()} (ID: {entity_id})") | |
elif urn_val: | |
category_name_val = f"Unmapped {type_name.split('_')[-1].capitalize()} (URN: {urn_val})" | |
else: | |
category_name_val = f"Missing URN for {type_name.split('_')[-1].capitalize()}" | |
counts = item.get("followerCounts", {}) | |
organic_count = counts.get("organicFollowerCount", 0) | |
paid_count = counts.get("paidFollowerCount", 0) | |
current_type_entries.append({ | |
"category_name": category_name_val, | |
"follower_count_organic": organic_count, | |
"follower_count_paid": paid_count, | |
"follower_count_type": type_name, | |
"organization_urn": org_urn_val | |
}) | |
return current_type_entries | |
industry_urns_to_map = [item.get("industry") for item in stat_element.get("followerCountsByIndustry", []) if item.get("industry")] | |
geo_urns_to_map = [item.get("geo") for item in stat_element.get("followerCountsByGeoCountry", []) if item.get("geo")] | |
live_industries_map = get_industries_map(session, list(set(industry_urns_to_map))) | |
live_geo_map = get_geo_map(session, list(set(geo_urns_to_map))) | |
demographic_configs = [ | |
{"items_key": "followerCountsBySeniority", "type_name": "follower_seniority", "id_map": seniorities_map, "id_field": "seniority", "top_n": 10}, | |
{"items_key": "followerCountsByFunction", "type_name": "follower_function", "id_map": functions_map, "id_field": "function", "top_n": 10}, | |
{"items_key": "followerCountsByIndustry", "type_name": "follower_industry", "id_map": live_industries_map, "id_field": "industry", "top_n": 10}, | |
{"items_key": "followerCountsByGeoCountry", "type_name": "follower_geo", "id_map": live_geo_map, "id_field": "geo", "top_n": 10}, | |
{"items_key": "followerCountsByAssociationType", "type_name": "follower_association", "id_map": {}, "id_field": "associationType", "top_n": None} | |
] | |
for config in demographic_configs: | |
raw_items = stat_element.get(config["items_key"], []) | |
processed_entries = _get_entries_for_type(raw_items, config["type_name"], config["id_map"], config["id_field"], org_urn) | |
if config["top_n"] is not None and processed_entries: | |
for entry in processed_entries: | |
if not isinstance(entry.get("follower_count_organic"), (int, float)): | |
logging.warning(f"Invalid organic follower count for sorting in {config['type_name']}: {entry['follower_count_organic']}. Setting to 0.") | |
entry["follower_count_organic"] = 0 | |
sorted_entries = sorted(processed_entries, key=lambda x: x.get("follower_count_organic", 0), reverse=True) | |
final_demographics_results.extend(sorted_entries[:config["top_n"]]) | |
logging.debug(f"Added top {config['top_n']} for {config['type_name']}. Count: {len(sorted_entries[:config['top_n']])}") | |
else: | |
final_demographics_results.extend(processed_entries) | |
logging.debug(f"Added all for {config['type_name']}. Count: {len(processed_entries)}") | |
logging.info(f"Processed follower demographics for {org_urn}. Total entries from all types: {len(final_demographics_results)}") | |
except requests.exceptions.HTTPError as http_err: | |
code = "N/A"; text = str(http_err); resp_headers = {} | |
if http_err.response is not None: | |
code = http_err.response.status_code | |
text = http_err.response.text | |
resp_headers = dict(http_err.response.headers) | |
logging.error(f"HTTP error fetching follower demographics for {org_urn} (Status: {code}): {text}") | |
logging.error(f"Request URL was: {request_url_for_logging}") | |
logging.error(f"Response Headers: {json.dumps(resp_headers, indent=2)}") | |
except requests.exceptions.RequestException as req_err: | |
logging.error(f"RequestException fetching follower demographics for {org_urn}: {str(req_err)}") | |
logging.error(f"Request URL was: {request_url_for_logging}") | |
if req_err.response is not None: | |
logging.error(f"Associated Response Status: {req_err.response.status_code}") | |
logging.error(f"Associated Response Text: {req_err.response.text}") | |
logging.error(f"Associated Response Headers: {json.dumps(dict(req_err.response.headers), indent=2)}") | |
except json.JSONDecodeError as json_err: | |
response_text = "Not available" | |
if response_obj is not None and hasattr(response_obj, 'text'): | |
response_text = response_obj.text | |
logging.error(f"Error decoding JSON for follower demographics for {org_urn}: {json_err}") | |
logging.error(f"Request URL was: {request_url_for_logging}") | |
logging.error(f"Raw Response Text: {response_text}") | |
except Exception as e: | |
logging.error(f"Unexpected error fetching follower demographics for {org_urn}: {e}", exc_info=True) | |
logging.error(f"Request URL was: {request_url_for_logging}") | |
if response_obj is not None: | |
logging.error(f"Response Status (if available): {getattr(response_obj, 'status_code', 'N/A')}") | |
logging.error(f"Response Text (if available): {getattr(response_obj, 'text', 'N/A')}") | |
logging.error(f"Response Headers (if available): {json.dumps(dict(getattr(response_obj, 'headers', {})), indent=2)}") | |
return final_demographics_results | |
# --- Main Orchestration Function --- | |
def get_linkedin_follower_stats(comm_client_id, community_token, org_urn): | |
logging.info(f"--- Initiating get_linkedin_follower_stats for org: {org_urn} ---") | |
logging.info(f"requests-oauthlib version: {requests_oauthlib.__version__}") # Log version | |
logging.debug(f"Received comm_client_id: {comm_client_id}") | |
logging.debug(f"Received community_token - Type: {type(community_token)}, IsSet: {bool(community_token)}") | |
if isinstance(community_token, str) and len(community_token) > 10: | |
logging.debug(f"Received community_token (partial): {community_token[:10]}...") | |
elif isinstance(community_token, dict): | |
logging.debug(f"Received community_token (dict): { {k: (v[:10] + '...' if k == 'access_token' and isinstance(v, str) and len(v)>10 else v) for k,v in community_token.items()} }") | |
if not all([comm_client_id, community_token, org_urn]): | |
logging.error("Client ID, community_token, or Organization URN is missing or empty.") | |
if not comm_client_id: logging.error("comm_client_id is missing.") | |
if not community_token: logging.error("community_token is missing or empty.") | |
if not org_urn: logging.error("org_urn is missing.") | |
return [] | |
token_dict = community_token if isinstance(community_token, dict) else {'access_token': str(community_token), 'token_type': 'Bearer'} | |
if not token_dict.get('access_token'): | |
logging.error("Failed to construct a valid token_dict: 'access_token' is empty.") | |
logging.debug(f"Problematic token_dict: {token_dict}") | |
return [] | |
logging.debug(f"Constructed token_dict for session: { {k: (v[:10] + '...' if k == 'access_token' and isinstance(v, str) and len(v)>10 else v) for k,v in token_dict.items()} }") | |
session = None | |
try: | |
session = create_session(comm_client_id, token=token_dict) | |
session.headers.update({ | |
"X-Restli-Protocol-Version": "2.0.0", | |
"LinkedIn-Version": LINKEDIN_API_VERSION, | |
"Accept-Language": "en_US" | |
}) | |
logging.info(f"Session (id: {id(session)}) created and headers updated for org {org_urn}.") | |
logging.debug(f"get_linkedin_follower_stats (id: {id(session)}): Session token after creation: {session.token}, Session authorized: {session.authorized}, Expires at: {session.token.get('expires_at') if session.token else 'N/A'}") | |
logging.debug(f"get_linkedin_follower_stats (id: {id(session)}): Session auth object: type={type(session.auth)}, value={session.auth}") | |
logging.debug(f"get_linkedin_follower_stats (id: {id(session)}): Auto-refresh URL: {session.auto_refresh_url}, Auto-refresh kwargs: {session.auto_refresh_kwargs}") | |
if session.token and 'access_token' in session.token: | |
logging.debug(f"get_linkedin_follower_stats: Access token in session (partial): {str(session.token['access_token'])[:20]}...") | |
else: | |
logging.warning("get_linkedin_follower_stats: session.token is None or 'access_token' key is missing after session creation.") | |
except Exception as e: | |
logging.error(f"Failed to create session or update headers for org {org_urn}: {e}", exc_info=True) | |
return [] | |
logging.info(f"Starting follower stats retrieval for org: {org_urn} using session (id: {id(session)})") | |
functions_map = get_functions_map(session) # Pass session | |
seniorities_map = get_seniorities_map(session) # Pass session | |
if not functions_map: logging.warning(f"Functions map is empty for org {org_urn}.") | |
if not seniorities_map: logging.warning(f"Seniorities map is empty for org {org_urn}.") | |
all_follower_data = [] | |
monthly_gains = fetch_monthly_follower_gains(session, org_urn, API_REST_BASE) # Pass session | |
all_follower_data.extend(monthly_gains) | |
demographics = fetch_follower_demographics(session, org_urn, functions_map, seniorities_map) # Pass session | |
all_follower_data.extend(demographics) | |
if not all_follower_data: | |
logging.warning(f"No follower data (gains or demographics) could be compiled for {org_urn}.") | |
else: | |
logging.info(f"Successfully compiled {len(all_follower_data)} total follower stat entries for {org_urn}.") | |
return all_follower_data | |