#!/usr/bin/env python3 """Оценка читаемости текста по индексу Flesch.""" from __future__ import annotations import argparse import json import re from pathlib import Path WORD = re.compile(r"[A-Za-zА-Яа-яЁё]+(?:-[A-Za-zА-Яа-яЁё]+)*") SENTENCE = re.compile(r"[.!?]+") RU_VOWELS = set("аеёиоуыэюя") EN_VOWELS = set("aeiouy") def detect_language(text: str) -> str: cyrillic = len(re.findall(r"[А-Яа-яЁё]", text)) latin = len(re.findall(r"[A-Za-z]", text)) return "ru" if cyrillic >= latin else "en" def syllables(word: str, language: str) -> int: lower = word.lower() if language == "ru": return max(1, sum(char in RU_VOWELS for char in lower)) clean = re.sub(r"[^a-z]", "", lower) groups = re.findall(r"[aeiouy]+", clean) count = len(groups) if clean.endswith("e") and count > 1 and not clean.endswith(("le", "ee")): count -= 1 return max(1, count) def grade(index: float) -> str: if index >= 80: return "лёгкий" if index >= 60: return "средний" if index >= 30: return "сложный" return "очень сложный" def analyze(text: str, language: str) -> dict[str, object]: words = WORD.findall(text) sentences = [part.strip() for part in SENTENCE.split(text) if part.strip()] if not sentences and text.strip(): sentences = [text.strip()] if not words or not sentences: return {} syllable_count = sum(syllables(word, language) for word in words) asl = len(words) / len(sentences) asw = syllable_count / len(words) index = 206.835 - 1.3 * asl - 60.1 * asw long_sentences = [] for sentence in sentences: length = len(WORD.findall(sentence)) if length >= 25: long_sentences.append((length, sentence[:160])) long_sentences.sort(reverse=True) return { "language": language, "words": len(words), "sentences": len(sentences), "syllables": syllable_count, "avg_sentence_words": round(asl, 1), "avg_word_syllables": round(asw, 2), "flesch": round(index, 1), "grade": grade(index), "long_sentences": [{"words": length, "text": text_part} for length, text_part in long_sentences[:5]], } def main() -> int: parser = argparse.ArgumentParser(description="Оценить сложность текста") parser.add_argument("paths", nargs="+", type=Path, help="Файлы или папки") parser.add_argument("--glob", default="*.txt", help="Маска при обходе папки") parser.add_argument("--lang", choices=["auto", "ru", "en"], default="auto", help="Язык текста") parser.add_argument("--json", type=Path, help="Сохранить отчёт в JSON") args = parser.parse_args() reports = [] for path in args.paths: root = path.expanduser() if not root.exists(): print(f"Пропуск: {root} не найден") continue files = [root] if root.is_file() else sorted(item for item in root.rglob(args.glob) if item.is_file()) for source in files: try: text = source.read_text(encoding="utf-8") except (UnicodeDecodeError, OSError) as error: print(f"Пропуск {source}: {error}") continue language = detect_language(text) if args.lang == "auto" else args.lang stats = analyze(text, language) if not stats: continue stats["file"] = str(source) reports.append(stats) print(f"\n{source}") print(f" Язык: {stats['language']}, слов: {stats['words']}, предложений: {stats['sentences']}") print(f" Средняя длина: {stats['avg_sentence_words']} слов, слогов на слово: {stats['avg_word_syllables']}") print(f" Flesch: {stats['flesch']} ({stats['grade']})") for sentence in stats["long_sentences"]: print(f" Длинное предложение ({sentence['words']} слов): {sentence['text']}") if not reports: print("Подходящих файлов не найдено") return 1 if args.json: args.json.expanduser().write_text( json.dumps({"reports": reports}, ensure_ascii=False, indent=2) + "\n", encoding="utf-8", ) print(f"\nJSON: {args.json}") return 0 if __name__ == "__main__": raise SystemExit(main())