import argparse import gzip import hashlib import json import struct from pathlib import Path def resource_path(root, name): parts = name.split("/") if not parts or any(part in ("", ".", "..") or "\\" in part for part in parts): raise ValueError("Chemin de ressource invalide") destination = root.joinpath(*parts).resolve() if root.resolve() not in destination.parents: raise ValueError("Chemin de ressource invalide") return destination def extract(source, destination): raw = gzip.decompress(source.read_bytes()) if raw[:8] != b"WFTTEX1\n": raise ValueError("Format de bibliothèque inconnu") length = struct.unpack_from(" len(body): raise ValueError("Ressource tronquée : " + name) data = body[offset:offset + size] if hashlib.sha256(data).hexdigest() != entry["id"]: raise ValueError("Ressource endommagée : " + name) target = resource_path(destination, name) target.parent.mkdir(parents=True, exist_ok=True) target.write_bytes(data) print(str(len(index["files"])) + " ressources extraites.") def rebuild(source, destination): entries = {} unique = {} body = bytearray() for file in sorted(source.rglob("*")): if not file.is_file() or file.is_symlink(): continue name = file.relative_to(source).as_posix() if not name.startswith("pdftex/"): continue data = file.read_bytes() digest = hashlib.sha256(data).hexdigest() if digest not in unique: unique[digest] = {"offset": len(body), "length": len(data), "id": digest} body.extend(data) entries[name] = unique[digest] if not entries: raise ValueError("Aucune ressource pdftex trouvée") header = json.dumps({"version": 1, "files": entries}, ensure_ascii=False, separators=(",", ":")).encode("utf-8") raw = b"WFTTEX1\n" + struct.pack("