# -*- coding: utf-8 -*- """Remplit la colonne Code du CSV mariage à partir des fiches guests Statamic.""" import csv import re from pathlib import Path GUESTS_DIR = Path(__file__).resolve().parent.parent / "content" / "collections" / "guests" CSV_PATH = Path(r"c:\Users\yanni\OneDrive\mariage.csv") def parse_frontmatter(md: str) -> dict: if "---" not in md: return {} parts = md.split("---", 2) if len(parts) < 3: return {} fm = parts[1] d = {} for line in fm.splitlines(): if ":" not in line: continue k, _, rest = line.partition(":") k = k.strip() v = rest.strip() if v.startswith("'") and v.endswith("'"): v = v[1:-1] elif v.startswith('"') and v.endswith('"'): v = v[1:-1] if k in ("code", "first_name", "last_name", "title"): d[k] = v return d def norm(s: str) -> str: return ( s.lower() .replace("é", "e") .replace("è", "e") .replace("ê", "e") .replace("ë", "e") .replace("à", "a") .replace("â", "a") .replace("ô", "o") .replace("ö", "o") .replace("ï", "i") .replace("î", "i") .replace("ù", "u") .replace("û", "u") .replace("ç", "c") .replace("œ", "oe") .strip() ) def load_guests(): by_full = {} by_first_last = {} by_last_first = {} singles = [] # (norm_first, norm_last, code) for p in GUESTS_DIR.glob("*.md"): d = parse_frontmatter(p.read_text(encoding="utf-8")) code = d.get("code") if not code: continue fn = d.get("first_name") or "" ln = d.get("last_name") or "" title = d.get("title") or "" # last name fallback: title if not ln and title: ln = title full = f"{fn} {ln}".strip() if full: by_full[norm(full)] = code if fn and ln: key = (norm(fn), norm(ln)) by_first_last[key] = code by_last_first[(norm(ln), norm(fn))] = code singles.append( { "code": code, "fn": norm(fn), "ln": norm(ln), "fn_raw": fn, "ln_raw": ln, } ) return by_full, by_first_last, singles def split_csv_names(cell: str): """Découpe 'Alice Blumer & Florian' ou 'Stéphanie et Pascal Dubois'.""" s = cell.strip() # retirer parenthèses type (Tarlala Lita) s = re.sub(r"\s*\([^)]*\)\s*", " ", s).strip() parts = re.split(r"\s*(?:&|\+|/)\s*|\s+et\s+", s, flags=re.IGNORECASE) return [p.strip() for p in parts if p.strip()] def tokenize_person(s: str): """Retourne (prénom_tokens, nom_tokens) heuristique.""" words = s.split() if len(words) == 1: return ([norm(words[0])], []) if len(words) == 2: return ([norm(words[0])], [norm(words[1])]) # 3+ mots: dernier = nom de famille, reste = prénoms return ([norm(w) for w in words[:-1]], [norm(words[-1])]) def find_code(person: str, by_full, by_first_last, singles) -> str | None: person = person.strip() nfull = norm(person) if nfull in by_full: return by_full[nfull] words = person.split() if len(words) >= 2: fn, ln = norm(words[0]), norm(words[-1]) if (fn, ln) in by_first_last: return by_first_last[(fn, ln)] # prénom nom avec plusieurs prénoms if len(words) > 2: fn2, ln2 = norm(" ".join(words[:-1])), norm(words[-1]) # essai prénom composé + nom if (fn2, ln2) in by_first_last: return by_first_last[(fn2, ln2)] # une seule unité (prénom seul souvent) if len(words) == 1: n1 = norm(words[0]) matches = [g for g in singles if g["fn"] == n1 and not g["ln"]] if len(matches) == 1: return matches[0]["code"] matches2 = [g for g in singles if g["fn"] == n1] if len(matches2) == 1: return matches2[0]["code"] # dernier recours: match sur nom de famille seul si unique if len(words) >= 2: ln_only = norm(words[-1]) cand = [g for g in singles if g["ln"] == ln_only] if len(cand) == 1: return cand[0]["code"] return None def main(): by_full, by_first_last, singles = load_guests() rows = [] raw = CSV_PATH.read_bytes() for enc in ("utf-8-sig", "utf-8", "cp1252", "latin-1"): try: text = raw.decode(enc) break except UnicodeDecodeError: continue else: text = raw.decode("utf-8", errors="replace") from io import StringIO f = StringIO(text) reader = csv.reader(f, delimiter=";") header = next(reader) for row in reader: rows.append(row) # colonnes attendues: Liste de nom;Table ID;... try: idx_name = header.index("Liste de nom") idx_code = header.index("Code") except ValueError: idx_name = 0 idx_code = 4 # Ajouter Code 2 si pas présent (insérer la cellule dans chaque ligne) if "Code 2" not in header: insert_at = idx_code + 1 header.insert(insert_at, "Code 2") new_rows = [] for row in rows: while len(row) < insert_at: row.append("") new_rows.append(row[:insert_at] + [""] + row[insert_at:]) rows = new_rows idx_code2 = header.index("Code 2") unmatched = [] for row in rows: while len(row) < len(header): row.append("") name_cell = row[idx_name].strip() if not name_cell or name_cell.startswith("#"): row[idx_code] = "" if idx_code2 < len(row): row[idx_code2] = "" continue parts = split_csv_names(name_cell) codes = [] for p in parts: c = find_code(p, by_full, by_first_last, singles) codes.append(c) if all(codes): row[idx_code] = codes[0] row[idx_code2] = codes[1] if len(codes) > 1 else "" elif len(codes) == 1 and codes[0]: row[idx_code] = codes[0] row[idx_code2] = "" else: # fallback: une seule entrée couple_id ? non row[idx_code] = " | ".join(c for c in codes if c) or "" row[idx_code2] = "" if not row[idx_code]: unmatched.append((name_cell, parts, codes)) # Excel Windows: cp1252 ou UTF-8 BOM with CSV_PATH.open("w", encoding="utf-8-sig", newline="") as f: w = csv.writer(f, delimiter=";") w.writerow(header) for row in rows: # pad while len(row) < len(header): row.append("") w.writerow(row[: len(header)]) print("OK, fichier écrit:", CSV_PATH) if unmatched: print("\nNon résolus:") for item in unmatched: print(" ", item) if __name__ == "__main__": main()