Spaces:
Running
Running
Update app.py
Browse files
app.py
CHANGED
@@ -11,8 +11,6 @@ import zipfile
|
|
11 |
from pdf2image import convert_from_path
|
12 |
import google.generativeai as genai
|
13 |
import json
|
14 |
-
from typing import Dict, Any, List, Union # Ajout des imports typing nécessaires
|
15 |
-
import logging
|
16 |
|
17 |
from docx import Document
|
18 |
from docx.shared import Pt, RGBColor, Inches
|
@@ -20,6 +18,10 @@ from docx.enum.text import WD_ALIGN_PARAGRAPH
|
|
20 |
from docx.enum.section import WD_SECTION
|
21 |
from docx.oxml import OxmlElement
|
22 |
from docx.oxml.ns import qn
|
|
|
|
|
|
|
|
|
23 |
|
24 |
def authenticate(username, password):
|
25 |
return username == os.getenv("HF_USERNAME") and password == os.getenv("HF_PASSWORD")
|
@@ -183,384 +185,7 @@ def extract_data_with_gemini(text_file_path: str, path_to_data_to_extract: str)
|
|
183 |
except Exception as e:
|
184 |
raise gr.Error(f"Error in Gemini processing: {str(e)}")
|
185 |
|
186 |
-
class RapportGenerator:
|
187 |
-
def __init__(self, json_file: str):
|
188 |
-
self.doc = Document()
|
189 |
-
self.logger = logging.getLogger(__name__)
|
190 |
-
logging.basicConfig(level=logging.INFO)
|
191 |
-
self.json_data = self.load_json(json_file)
|
192 |
-
self._setup_document()
|
193 |
|
194 |
-
def load_json(self, json_file: str) -> Dict:
|
195 |
-
"""Charge le fichier JSON"""
|
196 |
-
try:
|
197 |
-
with open(json_file, 'r', encoding='utf-8') as f:
|
198 |
-
return json.load(f)
|
199 |
-
except Exception as e:
|
200 |
-
self.logger.error(f"Erreur lors de la lecture du JSON: {str(e)}")
|
201 |
-
raise
|
202 |
-
|
203 |
-
def _setup_document(self):
|
204 |
-
"""Configure le document"""
|
205 |
-
sections = self.doc.sections
|
206 |
-
for section in sections:
|
207 |
-
section.bottom_margin = Inches(1)
|
208 |
-
self._create_page_number()
|
209 |
-
|
210 |
-
def _create_page_number(self):
|
211 |
-
"""Crée le numéro de page dans le pied de page"""
|
212 |
-
section = self.doc.sections[0]
|
213 |
-
footer = section.footer
|
214 |
-
paragraph = footer.paragraphs[0]
|
215 |
-
paragraph.alignment = WD_ALIGN_PARAGRAPH.RIGHT
|
216 |
-
|
217 |
-
run = paragraph.add_run()
|
218 |
-
self._create_element(run, 'PAGE')
|
219 |
-
run = paragraph.add_run(" / ")
|
220 |
-
run = paragraph.add_run()
|
221 |
-
self._create_element(run, 'NUMPAGES')
|
222 |
-
|
223 |
-
def _create_element(self, run, text):
|
224 |
-
"""Crée un élément de champ Word"""
|
225 |
-
fldChar1 = OxmlElement('w:fldChar')
|
226 |
-
fldChar1.set(qn('w:fldCharType'), 'begin')
|
227 |
-
run._r.append(fldChar1)
|
228 |
-
|
229 |
-
instrText = OxmlElement('w:instrText')
|
230 |
-
instrText.text = text
|
231 |
-
run._r.append(instrText)
|
232 |
-
|
233 |
-
fldChar2 = OxmlElement('w:fldChar')
|
234 |
-
fldChar2.set(qn('w:fldCharType'), 'end')
|
235 |
-
run._r.append(fldChar2)
|
236 |
-
|
237 |
-
def _add_title(self, text: str, level: int = 1):
|
238 |
-
"""Ajoute un titre avec style et taille appropriés"""
|
239 |
-
heading = self.doc.add_heading(level=level)
|
240 |
-
heading.alignment = WD_ALIGN_PARAGRAPH.LEFT
|
241 |
-
run = heading.add_run(text.strip()) # Ajout de strip()
|
242 |
-
font_sizes = {0: 18, 1: 16, 2: 14, 3: 12}
|
243 |
-
run.font.size = Pt(font_sizes.get(level, 12))
|
244 |
-
|
245 |
-
def _format_key(self, key: str) -> str:
|
246 |
-
"""Formate la clé en supprimant les suffixes _DEMANDEUR et _DEFENDEUR"""
|
247 |
-
key = key.replace("_", " ").title()
|
248 |
-
key = key.replace(" Demandeur", "")
|
249 |
-
key = key.replace(" Defendeur", "")
|
250 |
-
return key
|
251 |
-
|
252 |
-
def _get_safe_value(self, data: Dict, key: str, default: str = '') -> str:
|
253 |
-
"""Récupère une valeur en toute sécurité du dictionnaire"""
|
254 |
-
if isinstance(data, dict):
|
255 |
-
return str(data.get(key, default))
|
256 |
-
return default
|
257 |
-
|
258 |
-
def _format_person_info(self, data: Dict, is_lawyer: bool = False) -> Dict[str, str]:
|
259 |
-
"""Formate les informations d'une personne"""
|
260 |
-
info = {}
|
261 |
-
|
262 |
-
if not isinstance(data, dict):
|
263 |
-
return info
|
264 |
-
|
265 |
-
if is_lawyer:
|
266 |
-
# Informations de l'avocat
|
267 |
-
civilite = self._get_safe_value(data, 'CIVILITE_AVOCAT_DEMANDEUR') or self._get_safe_value(data, 'CIVILITE_AVOCAT_DEFENDEUR')
|
268 |
-
prenom = self._get_safe_value(data, 'PRENOM_AVOCAT_DEMANDEUR') or self._get_safe_value(data, 'PRENOM_AVOCAT_DEFENDEUR')
|
269 |
-
nom = self._get_safe_value(data, 'NOM_AVOCAT_DEMANDEUR') or self._get_safe_value(data, 'NOM_AVOCAT_DEFENDEUR')
|
270 |
-
|
271 |
-
info['Nom'] = f"{civilite} {prenom} {nom}".strip()
|
272 |
-
info['Barreau'] = self._get_safe_value(data, 'BARREAU_AVOCAT_DEMANDEUR') or self._get_safe_value(data, 'BARREAU_AVOCAT_DEFENDEUR')
|
273 |
-
info['Bureau d\'avocats'] = self._get_safe_value(data, 'BUREAU_AVOCAT_DEMANDEUR') or self._get_safe_value(data, 'BUREAU_AVOCAT_DEFENDEUR')
|
274 |
-
|
275 |
-
adresse = data.get('ADRESSE_AVOCAT_DEMANDEUR', {}) if isinstance(data.get('ADRESSE_AVOCAT_DEMANDEUR'), dict) else {}
|
276 |
-
if not adresse:
|
277 |
-
adresse = data.get('ADRESSE_AVOCAT_DEFENDEUR', {}) if isinstance(data.get('ADRESSE_AVOCAT_DEFENDEUR'), dict) else {}
|
278 |
-
|
279 |
-
info['Adresse Complete'] = self._get_safe_value(adresse, 'ADRESSE_COMPLETE_AVOCAT_DEMANDEUR') or self._get_safe_value(adresse, 'ADRESSE_COMPLETE_AVOCAT_DEFENDEUR')
|
280 |
-
info['Email'] = self._get_safe_value(data, 'EMAIL_AVOCAT_DEMANDEUR') or self._get_safe_value(data, 'EMAIL_AVOCAT_DEFENDEUR')
|
281 |
-
info['Telephone'] = self._get_safe_value(data, 'TELEPHONE_AVOCAT_DEMANDEUR') or self._get_safe_value(data, 'TELEPHONE_AVOCAT_DEFENDEUR')
|
282 |
-
else:
|
283 |
-
# Informations de la partie
|
284 |
-
civilite = self._get_safe_value(data, 'CIVILITE_DEMANDEUR') or self._get_safe_value(data, 'CIVILITE_DEFENDEUR')
|
285 |
-
prenom = self._get_safe_value(data, 'PRENOM_DEMANDEUR') or self._get_safe_value(data, 'PRENOM_DEFENDEUR')
|
286 |
-
nom = self._get_safe_value(data, 'NOM_DEMANDEUR') or self._get_safe_value(data, 'NOM_DEFENDEUR')
|
287 |
-
|
288 |
-
info['Nom'] = f"{civilite} {prenom} {nom}".strip()
|
289 |
-
info['Organisme'] = self._get_safe_value(data, 'ORGANISME_DEMANDEUR') or self._get_safe_value(data, 'ORGANISME_DEFENDEUR')
|
290 |
-
|
291 |
-
adresse = data.get('ADRESSE_DEMANDEUR', {}) if isinstance(data.get('ADRESSE_DEMANDEUR'), dict) else {}
|
292 |
-
if not adresse:
|
293 |
-
adresse = data.get('ADRESSE_DEFENDEUR', {}) if isinstance(data.get('ADRESSE_DEFENDEUR'), dict) else {}
|
294 |
-
|
295 |
-
info['Adresse Complete'] = self._get_safe_value(adresse, 'ADRESSE_COMPLETE_DEMANDEUR') or self._get_safe_value(adresse, 'ADRESSE_COMPLETE_DEFENDEUR')
|
296 |
-
info['Email'] = self._get_safe_value(data, 'EMAIL_DEMANDEUR') or self._get_safe_value(data, 'EMAIL_DEFENDEUR')
|
297 |
-
info['Telephone'] = self._get_safe_value(data, 'TELEPHONE_DEMANDEUR') or self._get_safe_value(data, 'TELEPHONE_DEFENDEUR')
|
298 |
-
|
299 |
-
return {k: v for k, v in info.items() if v and v != "Non spécifié"}
|
300 |
-
|
301 |
-
def _process_special_fields(self, key: str, value: Any) -> bool:
|
302 |
-
"""Traite les champs spéciaux (HISTORIQUE et MISSION)"""
|
303 |
-
if key in ["HISTORIQUE", "MISSION"]:
|
304 |
-
self._add_title(key.upper().strip(), 1)
|
305 |
-
|
306 |
-
# Vérifier si la valeur est une liste
|
307 |
-
if isinstance(value, list):
|
308 |
-
for item in value:
|
309 |
-
p = self.doc.add_paragraph()
|
310 |
-
p.add_run(str(item).strip())
|
311 |
-
else:
|
312 |
-
# Si c'est une chaîne simple ou autre type
|
313 |
-
p = self.doc.add_paragraph()
|
314 |
-
p.add_run(str(value).strip())
|
315 |
-
|
316 |
-
# Ajouter un espace après HISTORIQUE seulement
|
317 |
-
if key == "HISTORIQUE":
|
318 |
-
self.doc.add_paragraph()
|
319 |
-
|
320 |
-
return True
|
321 |
-
return False
|
322 |
-
|
323 |
-
def _process_parties(self, title: str, items: Union[str, List], level: int):
|
324 |
-
"""Traite les parties (demandeurs/défendeurs)"""
|
325 |
-
self._add_title(title, level)
|
326 |
-
|
327 |
-
# Si aucune partie n'est spécifiée
|
328 |
-
if isinstance(items, str) and items == "Non spécifié":
|
329 |
-
p = self.doc.add_paragraph(style='List Bullet')
|
330 |
-
p.add_run("Non spécifié")
|
331 |
-
return
|
332 |
-
|
333 |
-
# Traitement de chaque partie
|
334 |
-
if isinstance(items, list):
|
335 |
-
for idx, item in enumerate(items, 1):
|
336 |
-
if not isinstance(item, dict):
|
337 |
-
continue
|
338 |
-
|
339 |
-
# Sous-titre pour chaque partie
|
340 |
-
partie_title = f"{title.rstrip('S')} {idx}"
|
341 |
-
self._add_title(partie_title, level + 1)
|
342 |
-
|
343 |
-
# Formatage du nom complet
|
344 |
-
civilite = self._get_safe_value(item, f'CIVILITE_{title.rstrip("S")}', '')
|
345 |
-
prenom = self._get_safe_value(item, f'PRENOM_{title.rstrip("S")}', '')
|
346 |
-
nom = self._get_safe_value(item, f'NOM_{title.rstrip("S")}', '')
|
347 |
-
nom_complet = f"{civilite} {prenom} {nom}".strip()
|
348 |
-
|
349 |
-
# Ajout du nom
|
350 |
-
if nom_complet and nom_complet != "Non spécifié":
|
351 |
-
p = self.doc.add_paragraph(style='List Bullet')
|
352 |
-
p.add_run("Nom: ").bold = True
|
353 |
-
p.add_run(nom_complet)
|
354 |
-
|
355 |
-
# Récupération de l'adresse
|
356 |
-
adresse = item.get(f'ADRESSE_{title.rstrip("S")}', {})
|
357 |
-
if isinstance(adresse, dict):
|
358 |
-
adresse_complete = adresse.get(f'ADRESSE_COMPLETE_{title.rstrip("S")}', '')
|
359 |
-
if adresse_complete and adresse_complete != "Non spécifié":
|
360 |
-
p = self.doc.add_paragraph(style='List Bullet')
|
361 |
-
p.add_run("Adresse Complete: ").bold = True
|
362 |
-
p.add_run(adresse_complete)
|
363 |
-
|
364 |
-
# Ajout du téléphone
|
365 |
-
telephone = self._get_safe_value(item, f'TELEPHONE_{title.rstrip("S")}', '')
|
366 |
-
if telephone and telephone != "Non spécifié":
|
367 |
-
p = self.doc.add_paragraph(style='List Bullet')
|
368 |
-
p.add_run("Téléphone: ").bold = True
|
369 |
-
p.add_run(telephone)
|
370 |
-
|
371 |
-
# Ajout de l'email
|
372 |
-
email = self._get_safe_value(item, f'EMAIL_{title.rstrip("S")}', '')
|
373 |
-
if email and email != "Non spécifié":
|
374 |
-
p = self.doc.add_paragraph(style='List Bullet')
|
375 |
-
p.add_run("Email: ").bold = True
|
376 |
-
p.add_run(email)
|
377 |
-
|
378 |
-
# Traitement des avocats
|
379 |
-
avocat_key = f'AVOCAT_{title.rstrip("S")}'
|
380 |
-
avocats = item.get(avocat_key, [])
|
381 |
-
|
382 |
-
if isinstance(avocats, list) and avocats and avocats != "Non spécifié":
|
383 |
-
for avocat_idx, avocat in enumerate(avocats, 1):
|
384 |
-
if isinstance(avocat, dict):
|
385 |
-
# Sous-titre pour chaque avocat
|
386 |
-
self._add_title(f"Avocat {avocat_idx}", level + 2)
|
387 |
-
|
388 |
-
# Nom de l'avocat
|
389 |
-
civilite_avocat = self._get_safe_value(avocat, f'CIVILITE_AVOCAT_{title.rstrip("S")}', '')
|
390 |
-
prenom_avocat = self._get_safe_value(avocat, f'PRENOM_AVOCAT_{title.rstrip("S")}', '')
|
391 |
-
nom_avocat = self._get_safe_value(avocat, f'NOM_AVOCAT_{title.rstrip("S")}', '')
|
392 |
-
nom_complet_avocat = f"{civilite_avocat} {prenom_avocat} {nom_avocat}".strip()
|
393 |
-
|
394 |
-
if nom_complet_avocat and nom_complet_avocat != "Non spécifié":
|
395 |
-
p = self.doc.add_paragraph(style='List Bullet')
|
396 |
-
p.add_run("Nom: ").bold = True
|
397 |
-
p.add_run(nom_complet_avocat)
|
398 |
-
|
399 |
-
# Barreau
|
400 |
-
barreau = self._get_safe_value(avocat, f'BARREAU_AVOCAT_{title.rstrip("S")}', '')
|
401 |
-
if barreau and barreau != "Non spécifié":
|
402 |
-
p = self.doc.add_paragraph(style='List Bullet')
|
403 |
-
p.add_run("Barreau: ").bold = True
|
404 |
-
p.add_run(barreau)
|
405 |
-
|
406 |
-
# Bureau d'avocats
|
407 |
-
bureau = self._get_safe_value(avocat, f'BUREAU_AVOCAT_{title.rstrip("S")}', '')
|
408 |
-
if bureau and bureau != "Non spécifié":
|
409 |
-
p = self.doc.add_paragraph(style='List Bullet')
|
410 |
-
p.add_run("Bureau d'avocats: ").bold = True
|
411 |
-
p.add_run(bureau)
|
412 |
-
|
413 |
-
# Adresse de l'avocat
|
414 |
-
adresse_avocat = avocat.get(f'ADRESSE_AVOCAT_{title.rstrip("S")}', {})
|
415 |
-
if isinstance(adresse_avocat, dict):
|
416 |
-
adresse_complete_avocat = adresse_avocat.get(f'ADRESSE_COMPLETE_AVOCAT_{title.rstrip("S")}', '')
|
417 |
-
if adresse_complete_avocat and adresse_complete_avocat != "Non spécifié":
|
418 |
-
p = self.doc.add_paragraph(style='List Bullet')
|
419 |
-
p.add_run("Adresse Complete: ").bold = True
|
420 |
-
p.add_run(adresse_complete_avocat)
|
421 |
-
|
422 |
-
# Téléphone de l'avocat
|
423 |
-
tel_avocat = self._get_safe_value(avocat, f'TELEPHONE_AVOCAT_{title.rstrip("S")}', '')
|
424 |
-
if tel_avocat and tel_avocat != "Non spécifié":
|
425 |
-
p = self.doc.add_paragraph(style='List Bullet')
|
426 |
-
p.add_run("Téléphone: ").bold = True
|
427 |
-
p.add_run(tel_avocat)
|
428 |
-
|
429 |
-
# Email de l'avocat
|
430 |
-
email_avocat = self._get_safe_value(avocat, f'EMAIL_AVOCAT_{title.rstrip("S")}', '')
|
431 |
-
if email_avocat and email_avocat != "Non spécifié":
|
432 |
-
p = self.doc.add_paragraph(style='List Bullet')
|
433 |
-
p.add_run("Email: ").bold = True
|
434 |
-
p.add_run(email_avocat)
|
435 |
-
|
436 |
-
# Ajouter un espace entre chaque partie
|
437 |
-
if idx < len(items):
|
438 |
-
self.doc.add_paragraph()
|
439 |
-
|
440 |
-
def _add_table_of_contents(self):
|
441 |
-
"""Ajoute une table des matières au document"""
|
442 |
-
paragraph = self.doc.add_paragraph()
|
443 |
-
run = paragraph.add_run("TABLE DES MATIÈRES")
|
444 |
-
run.bold = True
|
445 |
-
run.font.size = Pt(14)
|
446 |
-
|
447 |
-
self.doc.add_paragraph()
|
448 |
-
paragraph = self.doc.add_paragraph()
|
449 |
-
|
450 |
-
# Ajouter le champ TOC
|
451 |
-
run = paragraph.add_run()
|
452 |
-
fldChar1 = OxmlElement('w:fldChar')
|
453 |
-
fldChar1.set(qn('w:fldCharType'), 'begin')
|
454 |
-
run._r.append(fldChar1)
|
455 |
-
|
456 |
-
instrText = OxmlElement('w:instrText')
|
457 |
-
instrText.set(qn('xml:space'), 'preserve')
|
458 |
-
instrText.text = 'TOC \\o "1-3" \\h \\z \\u'
|
459 |
-
run._r.append(instrText)
|
460 |
-
|
461 |
-
fldChar2 = OxmlElement('w:fldChar')
|
462 |
-
fldChar2.set(qn('w:fldCharType'), 'separate')
|
463 |
-
run._r.append(fldChar2)
|
464 |
-
|
465 |
-
# Ajouter un espace pour la mise à jour
|
466 |
-
run = paragraph.add_run()
|
467 |
-
|
468 |
-
fldChar4 = OxmlElement('w:fldChar')
|
469 |
-
fldChar4.set(qn('w:fldCharType'), 'end')
|
470 |
-
run._r.append(fldChar4)
|
471 |
-
|
472 |
-
self.doc.add_page_break()
|
473 |
-
|
474 |
-
def _format_person_info_simple(self, data: Dict, prefix: str) -> str:
|
475 |
-
"""Formate les informations d'une personne (magistrat ou greffier)"""
|
476 |
-
prenom = data.get(f'PRENOM_{prefix}', '').strip()
|
477 |
-
nom = data.get(f'NOM_{prefix}', '').strip()
|
478 |
-
titre = data.get(f'TITRE_{prefix}', '').strip()
|
479 |
-
return f"{prenom} {nom}, {titre}".strip()
|
480 |
-
|
481 |
-
def _format_monetary_value(self, data: Dict, prefix: str) -> str:
|
482 |
-
"""Formate les valeurs monétaires"""
|
483 |
-
valeur = str(data.get(f'VALEUR_{prefix}', 0)).strip()
|
484 |
-
monnaie = data.get(f'MONNAIE_{prefix}', '€').strip()
|
485 |
-
return f"{valeur} {monnaie}".strip()
|
486 |
-
|
487 |
-
def generate_report(self):
|
488 |
-
"""Génère le rapport complet"""
|
489 |
-
try:
|
490 |
-
# Titre principal
|
491 |
-
title = self.doc.add_heading("Données extraites d'une ordonnance de référé", 0)
|
492 |
-
title.alignment = WD_ALIGN_PARAGRAPH.CENTER
|
493 |
-
|
494 |
-
# Informations générales
|
495 |
-
self._add_title("INFORMATIONS GÉNÉRALES", 1)
|
496 |
-
if isinstance(self.json_data.get("INFORMATIONS_GENERALES"), dict):
|
497 |
-
info_gen = self.json_data["INFORMATIONS_GENERALES"]
|
498 |
-
|
499 |
-
# Liste des champs à traiter
|
500 |
-
fields_to_process = [
|
501 |
-
# Champs simples
|
502 |
-
("DATE_ORDONNANCE_REFERE", "Date Ordonnance Refere"),
|
503 |
-
("VILLE_TRIBUNAL", "Ville Tribunal"),
|
504 |
-
("NUMERO_RG", "Numero RG"),
|
505 |
-
|
506 |
-
# Champs complexes avec formatage spécial
|
507 |
-
("MAGISTRAT_ORDONNANCE_REFERE", "Magistrat Ordonnance Refere",
|
508 |
-
lambda x: self._format_person_info_simple(x, "MAGISTRAT_ORDONNANCE_REFERE")),
|
509 |
-
|
510 |
-
("GREFFIER_ORDONNANCE_REFERE", "Greffier Ordonnance Refere",
|
511 |
-
lambda x: self._format_person_info_simple(x, "GREFFIER_ORDONNANCE_REFERE")),
|
512 |
-
|
513 |
-
("CONSIGNATION", "Consignation",
|
514 |
-
lambda x: self._format_monetary_value(x, "CONSIGNATION")),
|
515 |
-
|
516 |
-
("SOMME_A_CONSIGER", "Somme A Consiger",
|
517 |
-
lambda x: self._format_monetary_value(x, "SOMME_A_CONSIGER"))
|
518 |
-
]
|
519 |
-
|
520 |
-
# Traitement de chaque champ
|
521 |
-
for field in fields_to_process:
|
522 |
-
key = field[0]
|
523 |
-
display_name = field[1]
|
524 |
-
|
525 |
-
if key in info_gen:
|
526 |
-
p = self.doc.add_paragraph(style='List Bullet')
|
527 |
-
p.add_run(f"{display_name}: ").bold = True
|
528 |
-
|
529 |
-
# Si c'est un champ avec formatage spécial (longueur du tuple = 3)
|
530 |
-
if len(field) == 3:
|
531 |
-
formatter = field[2]
|
532 |
-
value = formatter(info_gen[key])
|
533 |
-
else:
|
534 |
-
value = str(info_gen[key])
|
535 |
-
|
536 |
-
p.add_run(value)
|
537 |
-
|
538 |
-
# Traitement des demandeurs
|
539 |
-
if "DEMANDEURS" in self.json_data:
|
540 |
-
self._process_parties("DEMANDEURS", self.json_data["DEMANDEURS"], 1)
|
541 |
-
|
542 |
-
# Traitement des défendeurs
|
543 |
-
if "DEFENDEURS" in self.json_data:
|
544 |
-
self._process_parties("DEFENDEURS", self.json_data["DEFENDEURS"], 1)
|
545 |
-
|
546 |
-
# Historique
|
547 |
-
if "HISTORIQUE" in self.json_data:
|
548 |
-
self._process_special_fields("HISTORIQUE", self.json_data["HISTORIQUE"])
|
549 |
-
|
550 |
-
# Mission
|
551 |
-
if "MISSION" in self.json_data:
|
552 |
-
self._process_special_fields("MISSION", self.json_data["MISSION"])
|
553 |
-
|
554 |
-
# Sauvegarde du document
|
555 |
-
output_file = "rapport_extraction.docx"
|
556 |
-
self.doc.save(output_file)
|
557 |
-
self.logger.info(f"Rapport généré avec succès: {output_file}")
|
558 |
-
|
559 |
-
return output_file
|
560 |
-
|
561 |
-
except Exception as e:
|
562 |
-
self.logger.error(f"Erreur lors de la génération du rapport: {str(e)}")
|
563 |
-
raise
|
564 |
|
565 |
# Main Processing Function
|
566 |
def process_pdf(pdf_file):
|
|
|
11 |
from pdf2image import convert_from_path
|
12 |
import google.generativeai as genai
|
13 |
import json
|
|
|
|
|
14 |
|
15 |
from docx import Document
|
16 |
from docx.shared import Pt, RGBColor, Inches
|
|
|
18 |
from docx.enum.section import WD_SECTION
|
19 |
from docx.oxml import OxmlElement
|
20 |
from docx.oxml.ns import qn
|
21 |
+
from typing import Dict, Any, List, Union # Ajout des imports typing nécessaires
|
22 |
+
import logging
|
23 |
+
|
24 |
+
from helpers.utils import report-generator
|
25 |
|
26 |
def authenticate(username, password):
|
27 |
return username == os.getenv("HF_USERNAME") and password == os.getenv("HF_PASSWORD")
|
|
|
185 |
except Exception as e:
|
186 |
raise gr.Error(f"Error in Gemini processing: {str(e)}")
|
187 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
188 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
189 |
|
190 |
# Main Processing Function
|
191 |
def process_pdf(pdf_file):
|