#!/usr/bin/env python3 """ ICAC — Script de renommage assisté de documents PDF. Analyse les PDFs d'un dossier source, extrait la première page, propose un nom selon la convention ICAC, demande confirmation. Convention de nommage : {type}_{annee}_{description}.pdf Exemples : compte_gestion_2022_besseges.pdf pv_conseil_2026_02_24.pdf budget_primitif_2026.pdf bulletin_municipal_2026_01.pdf marche_voirie_2026_colas.pdf Usage : python scripts/rename_docs.py /chemin/vers/dossier python scripts/rename_docs.py ~/Documents/mairie/ """ import sys import shutil import re from pathlib import Path # --- Optional PyMuPDF for text extraction --- try: import fitz # PyMuPDF HAS_FITZ = True except ImportError: HAS_FITZ = False # --- Config --- DST_DIR = Path(__file__).resolve().parent.parent / "data" / "documents" COMMUNE = "besseges" # --- Type detection patterns --- TYPE_PATTERNS = [ (r"compte.*(gestion|administratif)|cg[_\s]|ca[_\s]", "compte_gestion"), (r"budget.*primitif|bp[_\s]", "budget_primitif"), (r"budget.*suppl|bs[_\s]|d[ée]cision.*modif", "budget_supplementaire"), (r"proc[eè]s.?verbal|pv[_\s]|conseil.*municipal", "pv_conseil"), (r"bulletin.*municipal|bm[_\s]", "bulletin_municipal"), (r"d[ée]lib[ée]ration", "deliberation"), (r"march[ée]", "marche"), (r"contrat|convention", "contrat"), (r"rapport.*activit|rapport.*annuel", "rapport"), (r"arr[eê]t[ée]", "arrete"), (r"facture|mandat", "facture"), (r"subvention|detr|dsil|fctva", "subvention"), ] # --- Year extraction --- YEAR_RE = re.compile(r"20[12]\d") # --- Month patterns --- MONTHS_FR = { "janvier": "01", "février": "02", "mars": "03", "avril": "04", "mai": "05", "juin": "06", "juillet": "07", "août": "08", "septembre": "09", "octobre": "10", "novembre": "11", "décembre": "12", "jan": "01", "fev": "02", "fév": "02", "mar": "03", "avr": "04", "jui": "06", "jul": "07", "aou": "08", "aoû": "08", "sep": "09", "oct": "10", "nov": "11", "dec": "12", "déc": "12", } MONTH_RE = re.compile( r"\b(" + "|".join(MONTHS_FR.keys()) + r")\b", re.IGNORECASE ) DATE_RE = re.compile(r"(\d{1,2})[/\-.](\d{1,2})[/\-.](\d{4})") def extract_first_page_text(pdf_path: Path) -> str: """Extract text from first page of PDF.""" if not HAS_FITZ: return f"[PyMuPDF non installé — nom du fichier : {pdf_path.name}]" try: doc = fitz.open(str(pdf_path)) if len(doc) == 0: return "[PDF vide]" text = doc[0].get_text("text").strip() doc.close() return text[:800] if text else "[Page vide]" except Exception as e: return f"[Erreur lecture : {e}]" def detect_type(text: str, filename: str) -> str: """Detect document type from text content and filename.""" combined = (text + " " + filename).lower() for pattern, doc_type in TYPE_PATTERNS: if re.search(pattern, combined): return doc_type return "document" def extract_year(text: str, filename: str) -> str: """Extract year from text or filename.""" # Try filename first m = YEAR_RE.search(filename) if m: return m.group() # Then content m = YEAR_RE.search(text) if m: return m.group() return "" def extract_date_suffix(text: str, filename: str) -> str: """Extract date suffix (month or full date) for PV/bulletins.""" combined = (text + " " + filename).lower() # Try DD/MM/YYYY pattern m = DATE_RE.search(combined) if m: day, month, year = m.groups() return f"{year}_{month.zfill(2)}_{day.zfill(2)}" # Try month name m = MONTH_RE.search(combined) if m: month_num = MONTHS_FR.get(m.group().lower(), "") year = extract_year(text, filename) if year and month_num: return f"{year}_{month_num}" return "" def suggest_name(pdf_path: Path, text: str) -> str: """Suggest a standardized filename.""" filename = pdf_path.name doc_type = detect_type(text, filename) year = extract_year(text, filename) # For PV and bulletins, try to get more specific date if doc_type in ("pv_conseil", "bulletin_municipal", "deliberation"): date_suffix = extract_date_suffix(text, filename) if date_suffix: return f"{doc_type}_{date_suffix}.pdf" if year: return f"{doc_type}_{year}_{COMMUNE}.pdf" return f"{doc_type}_{COMMUNE}.pdf" def main(): if len(sys.argv) < 2: print("Usage : python scripts/rename_docs.py /chemin/vers/dossier") print(" python scripts/rename_docs.py ~/Documents/mairie/") sys.exit(1) src_dir = Path(sys.argv[1]).expanduser().resolve() if not src_dir.is_dir(): print(f"Erreur : {src_dir} n'est pas un dossier valide.") sys.exit(1) DST_DIR.mkdir(parents=True, exist_ok=True) pdfs = sorted(src_dir.glob("*.pdf")) if not pdfs: print(f"Aucun PDF trouvé dans {src_dir}") sys.exit(0) print(f"\n{'=' * 60}") print(f" ICAC — Renommage assisté de documents") print(f" Source : {src_dir}") print(f" Dest. : {DST_DIR}") print(f" PDFs : {len(pdfs)} fichier(s)") if not HAS_FITZ: print(f" ⚠ PyMuPDF non installé — analyse par nom de fichier uniquement") print(f"{'=' * 60}") copied = 0 for pdf in pdfs: text = extract_first_page_text(pdf) suggested = suggest_name(pdf, text) print(f"\n{'─' * 50}") print(f" Fichier : {pdf.name}") print(f" Taille : {pdf.stat().st_size // 1024} Ko") if HAS_FITZ: # Show first 200 chars of extracted text preview = text[:200].replace('\n', ' ').strip() print(f" Extrait : {preview}") print(f" Proposé : \033[1;33m{suggested}\033[0m") user_input = input(" Nouveau nom (Enter = accepter, 's' = skip) : ").strip() if user_input.lower() == 's': print(" → Ignoré.") continue if user_input: new_name = user_input else: new_name = suggested if not new_name.endswith(".pdf"): new_name += ".pdf" dest = DST_DIR / new_name # Avoid overwriting if dest.exists(): print(f" ⚠ {dest.name} existe déjà !") overwrite = input(" Écraser ? (o/N) : ").strip().lower() if overwrite != 'o': print(" → Ignoré.") continue shutil.copy2(pdf, dest) print(f" → Copié : \033[1;32m{dest}\033[0m") copied += 1 print(f"\n{'=' * 60}") print(f" Terminé : {copied}/{len(pdfs)} fichier(s) copiés") print(f" Dossier : {DST_DIR}") print(f"{'=' * 60}\n") if __name__ == "__main__": main()