#!/usr/bin/env python3 """Текстовое сравнение двух PDF.""" from __future__ import annotations import argparse import difflib import json import re import sys from pathlib import Path import pdfplumber def page_texts(path: Path) -> list[str]: with pdfplumber.open(str(path)) as pdf: return [(page.extract_text() or "") for page in pdf.pages] def normalize(text: str, collapse: bool) -> str: if collapse: return re.sub(r"\s+", " ", text).strip() return text def main() -> int: parser = argparse.ArgumentParser(description="Сравнить два PDF по тексту") parser.add_argument("first", type=Path, help="Первый PDF") parser.add_argument("second", type=Path, help="Второй PDF") parser.add_argument("--ignore-whitespace", action="store_true", help="Не учитывать пробелы и переносы") parser.add_argument("--diff-lines", type=int, default=40, help="Сколько строк diff показать") parser.add_argument("--json", type=Path, help="Сохранить отчёт в JSON") args = parser.parse_args() first = args.first.expanduser() second = args.second.expanduser() for path in (first, second): if not path.is_file(): print(f"Файл не найден: {path}", file=sys.stderr) return 2 try: left = page_texts(first) right = page_texts(second) except Exception as error: print(f"Не удалось прочитать PDF: {error}", file=sys.stderr) return 1 total = max(len(left), len(right)) rows = [] different = 0 for index in range(total): a = normalize(left[index], args.ignore_whitespace) if index < len(left) else "" b = normalize(right[index], args.ignore_whitespace) if index < len(right) else "" ratio = difflib.SequenceMatcher(None, a, b).ratio() if a != b: different += 1 rows.append({"page": index + 1, "similarity": round(ratio, 4), "same": a == b}) print(f"Страниц: {len(left)} и {len(right)}") print(f"Отличаются страницы: {different} из {total}") for row in rows: if not row["same"]: print(f"стр. {row['page']}: сходство {row['similarity']:.1%}") shown = 0 for index in range(total): a = normalize(left[index], args.ignore_whitespace) if index < len(left) else "" b = normalize(right[index], args.ignore_whitespace) if index < len(right) else "" diff = list(difflib.unified_diff(a.splitlines(), b.splitlines(), fromfile=f"old стр.{index + 1}", tofile=f"new стр.{index + 1}", lineterm="", n=1)) for line in diff: if shown >= args.diff_lines: break print(line) shown += 1 if shown >= args.diff_lines: break if args.json: target = args.json.expanduser() target.parent.mkdir(parents=True, exist_ok=True) payload = {"first": str(first), "second": str(second), "pages": rows} target.write_text(json.dumps(payload, ensure_ascii=False, indent=2) + "\n", encoding="utf-8") print(f"Отчёт сохранён: {target}") return 0 if different == 0 else 1 if __name__ == "__main__": raise SystemExit(main())