Spaces:
Paused
Paused
File size: 33,196 Bytes
e60c00e 525d347 bd5de4f bf9a234 67360ee 10eded2 5f9453e 0a1b210 9f4cd2a bf9a234 2c24f0a 234506b 9f4cd2a e60c00e 2c24f0a e60c00e 234506b e34874a 06b2f8b cb6077d 83fcc35 5d37246 1e75abc 5d37246 1e75abc 5d37246 1e75abc 83fcc35 cb6077d 2bef542 62cf3f2 bf9a234 cc1a429 62cf3f2 cc1a429 62cf3f2 cc1a429 62cf3f2 cc1a429 62cf3f2 392b6fc 62cf3f2 21c6284 62cf3f2 27430be 06b2f8b f21f4b0 27430be f21f4b0 10eded2 f21f4b0 2bef542 62cf3f2 27430be 10eded2 2bef542 27430be 2bef542 27430be 2bef542 27430be 2bef542 27430be 2bef542 27430be 2bef542 27430be 2bef542 27430be 2bef542 27430be 2bef542 27430be 2bef542 62cf3f2 7073225 2bef542 7946627 06b2f8b 18e8106 4aef3ed 18e8106 4aef3ed 18e8106 4aef3ed 7946627 234506b 9341c86 7038ced 62cf3f2 4699f5b 392b6fc d212a64 18e8106 d212a64 7eacc0a 4aef3ed 732ae7b 18e8106 7eacc0a 18e8106 d212a64 18e8106 9341c86 d1d0218 18e8106 9341c86 d1d0218 9acc52c 4aef3ed d1d0218 18e8106 67360ee e34874a d212a64 7038ced 174db07 7038ced 62cf3f2 3a85524 234506b 27430be 21c6284 27430be 234506b 27430be 7eacc0a 7e236fd 392b6fc 27430be 234506b 7e236fd 27430be 234506b 27430be 234506b 7e236fd 27430be 7e236fd d555284 234506b dabf044 7e236fd 8aa81e7 7e236fd 234506b dabf044 234506b 8aa81e7 392b6fc d555284 7e236fd dabf044 7e236fd d555284 cc1a429 7e236fd dabf044 d555284 7e236fd d555284 392b6fc 7e236fd d555284 c1813c1 21c6284 7e236fd 21c6284 2c24f0a 21c6284 d555284 2c24f0a 21c6284 2c24f0a 94cf5e8 2c24f0a 94cf5e8 2c24f0a 94cf5e8 2c24f0a 21c6284 2c24f0a 21c6284 2c24f0a 21c6284 94cf5e8 21c6284 2c24f0a d555284 234506b d555284 21c6284 d555284 732ae7b 7e236fd d555284 7e236fd d555284 7e236fd 392b6fc 7e236fd 392b6fc 4699f5b d555284 732ae7b 27430be 732ae7b 7e236fd c1813c1 7e236fd c1813c1 7e236fd c1813c1 7e236fd c1813c1 7e236fd d555284 cc1a429 3a85524 94cf5e8 |
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 285 286 287 288 289 290 291 292 293 294 295 296 297 298 299 300 301 302 303 304 305 306 307 308 309 310 311 312 313 314 315 316 317 318 319 320 321 322 323 324 325 326 327 328 329 330 331 332 333 334 335 336 337 338 339 340 341 342 343 344 345 346 347 348 349 350 351 352 353 354 355 356 357 358 359 360 361 362 363 364 365 366 367 368 369 370 371 372 373 374 375 376 377 378 379 380 381 382 383 384 385 386 387 388 389 390 391 392 393 394 395 396 397 398 399 400 401 402 403 404 405 406 407 408 409 410 411 412 413 414 415 416 417 418 419 420 421 422 423 424 425 426 427 428 429 430 431 432 433 434 435 436 437 438 439 440 441 442 443 444 445 446 447 448 449 450 451 452 453 454 455 456 457 458 459 460 461 462 463 464 465 466 467 468 469 470 471 472 473 474 475 476 477 478 479 480 481 482 483 484 485 486 487 488 489 490 491 492 493 494 495 496 497 498 499 500 501 502 503 504 505 506 507 508 509 510 511 512 513 514 515 516 517 518 519 520 521 522 523 524 525 526 527 528 529 530 531 532 533 534 535 536 537 538 539 540 541 542 543 544 545 546 547 548 549 550 551 552 553 554 555 556 557 558 559 560 561 562 563 564 565 566 567 568 569 570 571 572 573 574 575 576 577 578 579 580 581 582 583 584 585 586 587 588 589 590 591 592 593 594 595 596 597 598 599 600 601 602 603 604 605 606 607 608 609 610 611 612 613 614 615 616 617 618 619 620 621 622 623 624 625 626 627 628 629 630 631 632 633 634 635 636 637 638 639 640 641 642 643 644 645 646 647 648 649 650 651 652 653 654 655 656 657 658 659 660 661 662 663 664 665 666 667 668 669 670 671 672 673 674 675 676 677 678 679 680 681 682 683 684 685 686 687 688 689 690 691 692 693 694 695 696 697 698 699 700 701 702 703 704 705 706 707 708 709 710 711 712 713 714 715 716 717 718 719 720 721 722 723 724 725 726 727 728 729 730 731 732 733 734 735 736 737 738 739 740 741 742 743 744 745 746 747 748 749 750 751 752 753 754 755 756 757 758 759 760 761 |
import base64
import io
import os
import pandas as pd
from docx import Document
from io import BytesIO
import dash
import dash_bootstrap_components as dbc
from dash import html, dcc, Input, Output, State, callback_context, ALL
from dash.dash_table import DataTable
from PyPDF2 import PdfReader
import logging
import threading
import re
import anthropic
import sqlite3
import time
logging.basicConfig(level=logging.INFO, format='%(asctime)s %(levelname)s %(message)s')
ANTHROPIC_KEY = os.environ.get("ANTHROPIC_API_KEY", "")
anthropic_client = anthropic.Anthropic(api_key=ANTHROPIC_KEY)
CLAUDE3_SONNET_MODEL = "claude-3-7-sonnet-20250219"
CLAUDE3_MAX_CONTEXT_TOKENS = 200_000
CLAUDE3_MAX_OUTPUT_TOKENS = 64_000
DB_PATH = "maiko_dash_docs.sqlite"
def init_db():
with sqlite3.connect(DB_PATH) as conn:
c = conn.cursor()
c.execute('''CREATE TABLE IF NOT EXISTS uploaded_docs
(doc_type TEXT PRIMARY KEY, doc_content TEXT, file_name TEXT, last_update TIMESTAMP)''')
c.execute('''CREATE TABLE IF NOT EXISTS generated_docs
(doc_type TEXT PRIMARY KEY, doc_content TEXT, last_update TIMESTAMP)''')
conn.commit()
init_db()
def set_uploaded_doc(doc_type, doc_content, file_name):
with sqlite3.connect(DB_PATH) as conn:
c = conn.cursor()
c.execute('''INSERT OR REPLACE INTO uploaded_docs (doc_type, doc_content, file_name, last_update)
VALUES (?, ?, ?, ?)''', (doc_type, doc_content, file_name, time.time()))
conn.commit()
def get_uploaded_doc(doc_type):
with sqlite3.connect(DB_PATH) as conn:
c = conn.cursor()
c.execute('''SELECT doc_content, file_name FROM uploaded_docs WHERE doc_type=?''', (doc_type,))
row = c.fetchone()
if row:
return row[0], row[1]
return None, None
def set_generated_doc(doc_type, doc_content):
with sqlite3.connect(DB_PATH) as conn:
c = conn.cursor()
c.execute('''INSERT OR REPLACE INTO generated_docs (doc_type, doc_content, last_update)
VALUES (?, ?, ?)''', (doc_type, doc_content, time.time()))
conn.commit()
def get_generated_doc(doc_type):
with sqlite3.connect(DB_PATH) as conn:
c = conn.cursor()
c.execute('''SELECT doc_content FROM generated_docs WHERE doc_type=?''', (doc_type,))
row = c.fetchone()
if row:
return row[0]
return None
spreadsheet_types = ["Shred", "Pink Review", "Red Review", "Gold Review", "Virtual Board", "LOE"]
narrative_types = ["Pink", "Red", "Gold"]
document_types = {
"Shred": "Ignore all other instructions and generate only requirements spreadsheet of the Project Work Statement (PWS) identified by action words like shall, will, perform etc. by pws section, requirement. Do not write as if you're responding to the proposal. Its a spreadsheet to distill the requirements, not microhealth's approach",
"Pink": "Create a highly detailed Pink Team document based on the PWS outline. Your goal is to be compliant and compelling. Focus on describing the approach and how it will be done, the steps, workflow, people, processes and technology based on well known industry standards to accomplish the task. Be sure to demonstrate innovation.",
"Pink Review": "Ignore all other instructions and generate and evaluate compliance of the Pink Team document against the requirements and output only a spreadsheet of non compliant findings by pws number, the goal of that pws section, what made it non compliant and your recommendations for recovery. you must also take into account section L&M of the document which is the evaluation criteria to be sure we address them.",
"Red": "Produce a highly detailed Red Team document based on the Pink Review by pws sections. Your goal is to be compliant and compelling by recovering all the findings in Pink Review. Focus on describing the approach and how it will be done, the steps, workflow, people, processes and technology to accomplish the task. Be sure to refer to research that validates the approach and cite sources with measurable outcomes",
"Red Review": "Ignore all other instructions and generate and evaluate compliance of the Red Team document against the requirements and output a only a spreadsheet of non compliant findings by pws number, the goal of that pws section, what made it non compliant and your recommendations for recovery. you must also take into account section L&M of the document which is the evaluation criteria to be sure we address them",
"Gold": "Create a highly detailed Gold Team document based on the PWS response by pws sections. Your goal is to be compliant and compelling by recovering all the findings in Red Review. Focus on describing the approach and how it will be done, the steps, workflow, people, processes and technology to accomplish the task. Be sure to refer to research that validates the approach and cite sources with measurable outcomes and improve on innovations of the approach",
"Gold Review": "Ignore all other instructions and generate and perform a final compliance review against the requirements and output only a spreadsheet of non compliant findings by pws number, the goal of that pws section, what made it non compliant and your recommendations for recovery. you must also take into account section L&M of the document which is the evaluation criteria to be sure we address them",
"Virtual Board": "Ignore all other instructions and generate and based on the requirements and in particular the evaulation criteria, you will evaluate the proposal as if you were a contracting office and provide section by section evaluation as unsatisfactory, satisfactory, good, very good, excellent and why and produce only spreadsheet",
"LOE": "Ignore all other instructions and generate and generate a Level of Effort (LOE) breakdown and produce only spreadsheet"
}
doc_dependencies = {
"Pink": {"source": ["shred"], "require": True},
"Pink Review": {"source": ["pink", "shred"], "require": True},
"Red": {"source": ["pink_review"], "require": True},
"Red Review": {"source": ["red", "shred"], "require": True},
"Gold": {"source": ["red_review"], "require": True},
"Gold Review": {"source": ["gold", "shred"], "require": True},
"LOE": {"source": ["gold"], "require": True},
"Virtual Board": {"source": ["shred"], "require": True},
}
def extract_markdown_tables(md_text):
tables = []
lines = md_text.split('\n')
in_table = False
table_lines = []
for line in lines:
if re.match(r'^\s*\|.*\|\s*$', line):
in_table = True
table_lines.append(line)
elif in_table and (re.match(r'^\s*\|.*\|\s*$', line) or re.match(r'^\s*$', line)):
table_lines.append(line)
else:
if in_table and table_lines:
tables.append('\n'.join(table_lines))
table_lines = []
in_table = False
if in_table and table_lines:
tables.append('\n'.join(table_lines))
return tables
def markdown_table_to_df(md_table):
lines = [line.strip() for line in md_table.split('\n') if line.strip()]
if len(lines) < 2:
return None
header = [h.strip() for h in lines[0].strip('|').split('|')]
sep_idx = 1
while sep_idx < len(lines) and not re.match(r'^\|\s*:?-+:?\s*(\|\s*:?-+:?\s*)+\|?$', lines[sep_idx]):
sep_idx += 1
data_lines = lines[sep_idx+1:] if sep_idx+1 < len(lines) else []
rows = []
for row in data_lines:
if not row.strip() or not row.strip().startswith('|'):
continue
cells = [c.strip() for c in row.strip('|').split('|')]
if len(cells) < len(header):
cells += [''] * (len(header) - len(cells))
elif len(cells) > len(header):
cells = cells[:len(header)]
rows.append(cells)
df = pd.DataFrame(rows, columns=header)
return df
def markdown_table_preview(md_text):
if not md_text:
logging.warning("No markdown text provided for preview.")
return html.Div("No preview available.")
tables = extract_markdown_tables(md_text)
if not tables:
logging.warning("No tables found in markdown text.")
return html.Div("No tables to preview.")
table_divs = []
for i, table in enumerate(tables):
df = markdown_table_to_df(table)
if df is not None and not df.empty:
table_divs.append(
html.Div([
DataTable(
columns=[{"name": str(col), "id": str(col)} for col in df.columns],
data=df.to_dict('records'),
style_table={'overflowX': 'auto'},
style_cell={'whiteSpace': 'normal', 'height': 'auto', 'textAlign': 'left', 'fontFamily': 'monospace', 'fontSize': '14px', 'maxWidth': '400px', 'minWidth': '80px', 'wordBreak': 'break-word'},
style_header={'fontWeight': 'bold'},
page_size=100,
id={'type': 'datatable-preview', 'index': i}
)
], className="mb-4")
)
else:
logging.warning(f"Table {i} could not be parsed into DataFrame.")
if not table_divs:
return html.Div("No valid tables to preview.")
return html.Div(table_divs)
def markdown_narrative_preview(md_text):
if not md_text:
logging.warning("No markdown text provided for narrative preview.")
return html.Div("No preview available.")
return html.Div(dcc.Markdown(md_text, dangerously_allow_html=True, style={'whiteSpace': 'pre-wrap', 'fontFamily': 'sans-serif'}))
def markdown_tables_to_xlsx(md_text):
tables = extract_markdown_tables(md_text)
output = BytesIO()
with pd.ExcelWriter(output, engine='xlsxwriter') as writer:
for i, table in enumerate(tables):
df = markdown_table_to_df(table)
if df is not None:
sheet_name = f"Table{i+1}"
df.to_excel(writer, sheet_name=sheet_name, index=False)
output.seek(0)
return output
def strip_markdown(text):
text = re.sub(r'(\*\*|__)(.*?)\1', r'\2', text)
text = re.sub(r'(\*|_)(.*?)\1', r'\2', text)
text = re.sub(r'`{1,3}[^`]*`{1,3}', '', text)
text = re.sub(r'^#+ ', '', text, flags=re.MULTILINE)
text = re.sub(r'^> ', '', text, flags=re.MULTILINE)
text = re.sub(r'!\[.*?\]\(.*?\)', '', text)
text = re.sub(r'\[([^\]]+)\]\([^)]+\)', r'\1', text)
text = re.sub(r'^\s*[-*+] ', '', text, flags=re.MULTILINE)
text = re.sub(r'^\s*\d+\.\s+', '', text, flags=re.MULTILINE)
text = text.replace('---', '')
text = text.replace('___', '')
text = text.replace('***', '')
return text.strip()
def process_document(contents, filename):
content_type, content_string = contents.split(',')
decoded = base64.b64decode(content_string)
try:
if filename.lower().endswith('.docx'):
doc = Document(BytesIO(decoded))
text = "\n".join([para.text for para in doc.paragraphs])
return text
elif filename.lower().endswith('.pdf'):
pdf = PdfReader(BytesIO(decoded))
text = ""
for page in pdf.pages:
page_text = page.extract_text()
if page_text:
text += page_text
return text
else:
return f"Unsupported file format: {filename}. Please upload a PDF or DOCX file."
except Exception as e:
logging.error(f"Error processing document: {str(e)}")
return f"Error processing document: {str(e)}"
def file_list_component():
return html.Div(
id='file-list-container',
children=[
html.Div(id='file-list')
]
)
def get_left_col_content():
chat_card = dbc.Card(
dbc.CardBody([
html.H5("Maiko Chat", className="mb-2"),
dcc.Loading(
id="chat-loading",
type="default",
children=[
dcc.Textarea(
id="chat-input",
placeholder="Chat with AI to update document...",
className="mb-2",
style={
'whiteSpace':'pre-wrap',
'width': '100%',
'minHeight': '100px',
'maxHeight': '300px',
'resize': 'vertical',
'overflowY': 'auto'
},
rows=5,
wrap='soft'
),
dbc.Row([
dbc.Col(
dbc.Button("Send", id="btn-send-chat", color="primary", className="mb-3 w-100"),
width=6
),
dbc.Col(
dbc.Button("Clear Chat", id="btn-clear-chat", color="secondary", className="mb-3 w-100"),
width=6
),
], className="g-1 mb-2"),
html.Div(id="chat-output")
]
)
]),
className="mt-4"
)
return [
html.H4("Proposal Writer", className="mt-3 mb-2", style={'marginBottom': '12px'}),
html.Hr(style={'marginTop': '8px', 'marginBottom': '16px'}),
html.Div(
id='doc-type-buttons'
),
chat_card
]
def get_right_col_content(selected_type, store_data):
controls = []
if selected_type == "Shred":
controls.append(
html.Div([
dcc.Upload(
id={'type': 'upload-doc-type', 'subtype': 'shred', 'index': selected_type},
children=html.Div([
'Drag and Drop or ',
html.A('Select Files')
]),
style={
'width': '100%',
'height': '60px',
'lineHeight': '60px',
'borderWidth': '1px',
'borderStyle': 'dashed',
'borderRadius': '5px',
'textAlign': 'center',
'margin': '10px 0'
},
multiple=True
),
file_list_component(),
html.Div(id='uploaded-doc-name-shred')
])
)
elif selected_type in doc_dependencies:
sources = doc_dependencies[selected_type]["source"]
for src in sources:
label = ""
loaded_preview = None
store_var = store_data.get(src)
doc_name_id = f"uploaded-doc-name-{src}"
if src == "shred":
label = "Shred (Requirements)"
loaded_preview = markdown_table_preview(store_var)
elif src == "pink":
label = "Pink Document"
loaded_preview = markdown_narrative_preview(store_var)
elif src == "pink_review":
label = "Pink Review"
loaded_preview = markdown_table_preview(store_var)
elif src == "red":
label = "Red Document"
loaded_preview = markdown_narrative_preview(store_var)
elif src == "red_review":
label = "Red Review"
loaded_preview = markdown_table_preview(store_var)
elif src == "gold":
label = "Gold Document"
loaded_preview = markdown_narrative_preview(store_var)
controls.append(html.Div([
html.Label(f"Upload {label}"),
dcc.Upload(
id={'type': 'upload-doc-type', 'subtype': src, 'index': selected_type},
children=html.Div(['Drag and Drop or ', html.A('Select File')]),
style={
'width': '100%',
'height': '60px',
'lineHeight': '60px',
'borderWidth': '1px',
'borderStyle': 'dashed',
'borderRadius': '5px',
'textAlign': 'center',
'margin': '10px 0'
},
multiple=False
),
dbc.RadioItems(
id={'type': 'radio-doc-source', 'subtype': src, 'index': selected_type},
options=[
{'label': 'Loaded Document', 'value': 'loaded'}
] if store_var else [],
value='loaded' if store_var else None,
inline=True,
className="mb-2"
),
html.Div([
html.Div("Loaded Document Preview:", className="mt-2 mb-1"),
loaded_preview
], style={'marginBottom': '12px'}),
], id={'type': f'doc-type-controls-{src}', 'index': selected_type}))
controls.append(
dbc.Row([
dbc.Col(
dbc.Button("Generate Document", id={'type': 'btn-generate-doc', 'index': selected_type}, color="primary", className="mb-3 w-100"),
width=6
),
dbc.Col(
dbc.Button("Download Document", id="btn-download", color="success", className="mb-3 w-100"),
width=6
),
dcc.Download(id="download-document")
], className="mb-2")
)
controls.append(html.Hr())
controls.append(
dbc.Card(
dbc.CardBody(
dcc.Loading(
id='preview-loading',
type='default',
children=html.Div(
id='document-preview',
style={
'height': '100%',
'overflowY': 'auto',
'overflowX': 'auto',
'padding': '8px'
}
)
)
),
style={
'marginBottom': '10px',
'height': 'calc(100vh - 155px)',
'maxHeight': '820px',
'minHeight': '460px',
'boxSizing': 'border-box'
}
)
)
return dbc.Card(dbc.CardBody(controls), style={'height': 'calc(100vh - 140px)', 'maxHeight': '820px', 'minHeight': '460px', 'boxSizing': 'border-box'})
app = dash.Dash(__name__, external_stylesheets=[dbc.themes.BOOTSTRAP], suppress_callback_exceptions=True)
app.layout = dbc.Container([
dcc.Store(id='selected-doc-type', data="Shred"),
dcc.Store(id='store-shred'),
dcc.Store(id='store-pink'),
dcc.Store(id='store-pink-review'),
dcc.Store(id='store-red'),
dcc.Store(id='store-red-review'),
dcc.Store(id='store-gold'),
dcc.Store(id='store-gold-review'),
dcc.Store(id='store-loe'),
dcc.Store(id='store-virtual-board'),
dcc.Store(id='store-generated-doc'),
dcc.Store(id='main-loader-trigger'),
dbc.Row([
dbc.Col(
html.H2(id='main-title', className="mt-3 mb-2", style={'textAlign': 'center', 'width':'100%'}),
width=12
)
], justify='center'),
dbc.Row([
dbc.Col(
dcc.Loading(
id="main-progress-dot",
type="default",
children=html.Div(id="main-progress-dot-output"),
style={'display': 'flex', 'justifyContent': 'center', 'alignItems': 'center', 'height':'40px'}
),
width=12
)
], justify='center'),
dbc.Row([
dbc.Col(
dbc.Card(
dbc.CardBody(get_left_col_content()),
style={
'height': 'calc(100vh - 140px)',
'maxHeight': '820px',
'minHeight': '460px',
'boxSizing': 'border-box'
}
),
width=3
),
dbc.Col(
html.Div([
html.Div(id='right-col-content')
], style={'height': 'calc(100vh - 140px)', 'maxHeight': '820px', 'minHeight': '460px', 'boxSizing': 'border-box'}),
width=9
)
], align='start')
], fluid=True)
@app.callback(
Output('main-title', 'children'),
Input('selected-doc-type', 'data')
)
def update_main_title(selected_type):
return selected_type
@app.callback(
Output('doc-type-buttons', 'children'),
Input('selected-doc-type', 'data')
)
def render_doc_type_buttons(selected_type):
buttons = []
for doc_type in document_types.keys():
btn_style = {'overflow': 'hidden', 'text-overflow': 'ellipsis', 'white-space': 'nowrap'}
btn_class = "mb-2 w-100 text-left custom-button"
if doc_type == selected_type:
btn_class += " active-doc-type"
buttons.append(
dbc.Button(
doc_type,
id={'type': 'btn-doc-type', 'index': doc_type},
color="link",
className=btn_class,
style=btn_style
)
)
return buttons
@app.callback(
Output('selected-doc-type', 'data'),
[Input({'type': 'btn-doc-type', 'index': ALL}, 'n_clicks')],
[State({'type': 'btn-doc-type', 'index': ALL}, 'id')],
prevent_initial_call=True
)
def update_selected_doc_type(n_clicks_list, btn_ids):
triggered = callback_context.triggered
if not triggered or all(x is None for x in n_clicks_list):
raise dash.exceptions.PreventUpdate
idx = [i for i, x in enumerate(n_clicks_list) if x]
if idx:
selected_type = btn_ids[idx[-1]]['index']
else:
selected_type = "Shred"
logging.info(f"Doc type selected: {selected_type}")
return selected_type
@app.callback(
Output('right-col-content', 'children'),
Input('selected-doc-type', 'data'),
)
def update_right_col(selected_type):
store_data = {}
for store_key in ["shred","pink","pink_review","red","red_review","gold","gold_review","loe","virtual_board"]:
doc_content, _ = get_uploaded_doc(store_key)
store_data[store_key] = doc_content
return get_right_col_content(selected_type, store_data)
@app.callback(
[
Output('store-shred', 'data'),
Output('store-pink', 'data'),
Output('store-pink-review', 'data'),
Output('store-red', 'data'),
Output('store-red-review', 'data'),
Output('store-gold', 'data'),
Output('store-gold-review', 'data'),
Output('store-loe', 'data'),
Output('store-virtual-board', 'data'),
Output('file-list', 'children'),
Output('uploaded-doc-name-shred', 'children'),
Output('document-preview', 'children'),
Output('store-generated-doc', 'data'),
Output('download-document', 'data'),
Output('main-loader-trigger', 'data')
],
[
Input({'type': 'upload-doc-type', 'subtype': ALL, 'index': ALL}, 'contents'),
Input({'type': 'btn-generate-doc', 'index': ALL}, 'n_clicks'),
Input("btn-download", "n_clicks"),
Input('selected-doc-type', 'data')
],
[
State({'type': 'upload-doc-type', 'subtype': ALL, 'index': ALL}, 'filename'),
State('store-shred', 'data'),
State('store-pink', 'data'),
State('store-pink-review', 'data'),
State('store-red', 'data'),
State('store-red-review', 'data'),
State('store-gold', 'data'),
State('store-gold-review', 'data'),
State('store-loe', 'data'),
State('store-virtual-board', 'data'),
State('store-generated-doc', 'data')
],
prevent_initial_call=True
)
def unified_handler(contents_lists, n_clicks_generate_list, n_clicks_download, selected_type,
filenames_lists, shred, pink, pink_review, red, red_review, gold, gold_review, loe, virtual_board, prev_generated):
logging.info(f"unified_handler triggered. Trigger: {callback_context.triggered}")
stores = [shred, pink, pink_review, red, red_review, gold, gold_review, loe, virtual_board]
file_list = []
uploaded_doc_name_shred = dash.no_update
preview = dash.no_update
generated_doc = dash.no_update
download_data = dash.no_update
ctx = callback_context
trigger = ctx.triggered[0]['prop_id'] if ctx.triggered else ""
subtype_map = {
"shred": 0,
"pink": 1,
"pink_review": 2,
"red": 3,
"red_review": 4,
"gold": 5,
"gold_review": 6,
"loe": 7,
"virtual_board": 8
}
loader_trigger = time.time()
# Handle file upload
if "upload-doc-type" in trigger:
for i, (contents, filenames) in enumerate(zip(contents_lists, filenames_lists)):
if contents and filenames:
id_dict = ctx.inputs_list[0][i]['id']
subtype = id_dict['subtype']
idx = subtype_map[subtype]
if isinstance(contents, str):
contents = [contents]
filenames = [filenames]
content = contents[-1]
name = filenames[-1]
file_text = process_document(content, name)
logging.info(f"Processed uploaded file for subtype '{subtype}': {name}")
logging.info(f"Extracted text from uploaded document (first 500 chars): {file_text[:500]}")
set_uploaded_doc(subtype, file_text, name)
stores[idx] = file_text
if subtype == "shred":
file_list = [
dbc.Row([
dbc.Col(
html.Button('×', id={'type': 'remove-file', 'index': name}, style={'marginRight': '5px', 'fontSize': '10px'}),
width=1
),
dbc.Col(
html.Span(name, style={'wordBreak': 'break-all'}),
width=11
)
], id={'type': 'file-row', 'index': name}, align="center", className="mb-1")
]
uploaded_doc_name_shred = name
# Preview logic with fallback
if subtype == "shred":
preview = markdown_table_preview(file_text)
elif subtype in ["pink", "red", "gold"]:
preview = markdown_narrative_preview(file_text)
else:
preview = markdown_table_preview(file_text)
generated_doc = dash.no_update
download_data = dash.no_update
logging.info("Returning after file upload.")
return stores + [file_list, uploaded_doc_name_shred, preview, generated_doc, download_data, loader_trigger]
# Handle document generation
if any(n_clicks_generate_list):
logging.info(f"Generating document for type: {selected_type}")
inputs = {
"shred": stores[0],
"pink": stores[1],
"pink_review": stores[2],
"red": stores[3],
"red_review": stores[4],
"gold": stores[5],
"gold_review": stores[6],
"loe": stores[7],
"virtual_board": stores[8]
}
prompt = document_types.get(selected_type, "")
context = ""
if selected_type in doc_dependencies:
for dep in doc_dependencies[selected_type]["source"]:
context += f"\n\n{dep.upper()}:\n{inputs.get(dep, '')}"
else:
context += f"\n\n{inputs.get('shred','')}"
full_prompt = f"{prompt}\n\n{context}"
logging.info(f"Prompt sent to Claude 3 (first 1000 chars): {full_prompt[:1000]}")
result_holder = {}
def anthropic_thread():
try:
message = anthropic_client.messages.create(
model=CLAUDE3_SONNET_MODEL,
max_tokens=CLAUDE3_MAX_OUTPUT_TOKENS,
system="You are a helpful assistant for proposal writing. Always output in markdown, and use markdown tables for spreadsheets if needed.",
messages=[
{"role": "user", "content": full_prompt}
]
)
result_text = message.content[0].text if hasattr(message, "content") and message.content else str(message)
result_holder['result'] = result_text
logging.info("Claude 3 document generated successfully.")
except Exception as e:
logging.error(f"Error with Claude 3 API: {str(e)}")
result_holder['result'] = f"Error generating document: {str(e)}"
t = threading.Thread(target=anthropic_thread)
t.start()
t.join(timeout=120)
generated_doc = result_holder.get('result', 'Error: No document generated.')
logging.info(f"Claude 3 API returned: {generated_doc[:500]}")
set_generated_doc(selected_type, generated_doc)
# Preview with fallback
if selected_type in spreadsheet_types:
preview = markdown_table_preview(generated_doc)
else:
preview = markdown_narrative_preview(generated_doc)
logging.info("Returning after document generation.")
return stores + [file_list, uploaded_doc_name_shred, preview, generated_doc, dash.no_update, loader_trigger]
# Handle download
if "btn-download" in trigger:
doc = prev_generated
if not doc:
doc = get_generated_doc(selected_type)
if not doc:
if selected_type == "Shred":
doc = stores[0]
elif selected_type == "Pink":
doc = stores[1]
elif selected_type == "Pink Review":
doc = stores[2]
elif selected_type == "Red":
doc = stores[3]
elif selected_type == "Red Review":
doc = stores[4]
elif selected_type == "Gold":
doc = stores[5]
elif selected_type == "Gold Review":
doc = stores[6]
elif selected_type == "LOE":
doc = stores[7]
elif selected_type == "Virtual Board":
doc = stores[8]
if not doc:
logging.warning("No generated document to download.")
preview = html.Div("No document loaded or generated.")
return stores + [file_list, uploaded_doc_name_shred, preview, dash.no_update, dash.no_update, loader_trigger]
if selected_type in spreadsheet_types:
xlsx_io = markdown_tables_to_xlsx(doc)
logging.info("Spreadsheet document prepared for download.")
download_data = dcc.send_bytes(xlsx_io.getvalue(), filename=f"{selected_type}_output.xlsx")
return stores + [file_list, uploaded_doc_name_shred, dash.no_update, dash.no_update, download_data, loader_trigger]
else:
content = doc.encode('utf-8')
logging.info("Narrative document prepared for download.")
download_data = dcc.send_bytes(content, filename=f"{selected_type}_output.md")
return stores + [file_list, uploaded_doc_name_shred, dash.no_update, dash.no_update, download_data, loader_trigger]
# Default preview logic
doc = prev_generated
if not doc:
doc = get_generated_doc(selected_type)
if doc:
if selected_type in spreadsheet_types:
preview = markdown_table_preview(doc)
else:
preview = markdown_narrative_preview(doc)
else:
if selected_type == "Shred" and stores[0]:
preview = markdown_table_preview(stores[0])
elif selected_type in spreadsheet_types:
doc_store = {
"Pink Review": stores[2],
"Red Review": stores[4],
"Gold Review": stores[6],
"Virtual Board": stores[8],
"LOE": stores[7]
}
doc = doc_store.get(selected_type, "")
preview = markdown_table_preview(doc)
elif selected_type in narrative_types:
doc_store = {
"Pink": stores[1],
"Red": stores[3],
"Gold": stores[5]
}
doc = doc_store.get(selected_type, "")
preview = markdown_narrative_preview(doc)
else:
preview = html.Div("No document loaded or generated.")
return stores + [file_list, uploaded_doc_name_shred, preview, dash.no_update, dash.no_update, loader_trigger]
if __name__ == '__main__':
print("Starting the Dash application...")
app.run(debug=True, host='0.0.0.0', port=7860)
print("Dash application has finished running.") |