"""requirements.txt 의 패키지 이름이 실제로 존재하는지, 너무 새롭지 않은지 설치 전에 점검합니다. 사용법: python pkgcheck.py requirements.txt [--index URL] [--recent-days 30] [--min-releases 3] [--strict] 종료 코드: 0 통과, 1 존재하지 않는 이름이 있음(--strict 이면 경고도 1), 2 조회 실패(네트워크 등) 조회만 하며 어떤 패키지도 설치하지 않습니다. """ import argparse import difflib import json import re import sys import urllib.error import urllib.request from datetime import datetime, timezone from pathlib import Path HERE = Path(__file__).resolve().parent def normalize(name): return re.sub(r"[-_.]+", "-", name).lower() def read_names(path): names = [] for line in Path(path).read_text(encoding="utf-8").splitlines(): line = line.split("#", 1)[0].strip() if not line or line.startswith("-"): continue m = re.match(r"^([A-Za-z0-9][A-Za-z0-9._-]*)", line) if m: names.append(m.group(1)) return names def read_list(path): p = Path(path) if not p.exists(): return set() return {normalize(x.split("#", 1)[0].strip()) for x in p.read_text(encoding="utf-8").splitlines() if x.split("#", 1)[0].strip()} def fetch(index, name): """(상태코드, JSON 또는 None). 404 는 (404, None), 그 밖의 오류는 예외.""" url = f"{index.rstrip('/')}/{normalize(name)}/json" try: with urllib.request.urlopen(url, timeout=15) as r: return r.status, json.load(r) except urllib.error.HTTPError as e: if e.code == 404: return 404, None raise def first_release(data): times = [] for files in data.get("releases", {}).values(): for f in files: t = f.get("upload_time_iso_8601") if t: times.append(datetime.fromisoformat(t.replace("Z", "+00:00"))) return min(times) if times else None def similar_to(name, popular, cutoff=0.85): n = normalize(name) if n in popular: return None best = difflib.get_close_matches(n, sorted(popular), n=1, cutoff=cutoff) return best[0] if best else None def main(): ap = argparse.ArgumentParser() ap.add_argument("requirements") ap.add_argument("--index", default="https://pypi.org/pypi") ap.add_argument("--recent-days", type=int, default=30) ap.add_argument("--min-releases", type=int, default=3) ap.add_argument("--popular", default=str(HERE / "popular.txt")) ap.add_argument("--allow", default=str(HERE / "allowlist.txt")) ap.add_argument("--strict", action="store_true") a = ap.parse_args() popular, allow = read_list(a.popular), read_list(a.allow) now = datetime.now(timezone.utc) missing = warned = 0 print(f"{'이름':<30}{'판정':<8}{'최초 릴리스':<13}{'릴리스':>4} 비고") for name in read_names(a.requirements): n = normalize(name) try: code, data = fetch(a.index, name) except (urllib.error.URLError, TimeoutError) as e: print(f"{name:<30}{'조회실패':<8}{'-':<13}{'-':>4} {e}") return 2 notes = [] if code == 404: missing += 1 verdict, first, count = "없음", "-", "-" notes.append("인덱스에 없는 이름") if (s := similar_to(n, popular)): notes.append(f"{s} 와 비슷") else: first_dt = first_release(data) count = len(data.get("releases", {})) first = first_dt.strftime("%Y-%m-%d") if first_dt else "-" if n in allow: verdict = "허용" notes.append("allowlist") else: if first_dt and (now - first_dt).days < a.recent_days: notes.append(f"최초 릴리스 {(now - first_dt).days}일 전") if count < a.min_releases: notes.append(f"릴리스 {count}개") if (s := similar_to(n, popular)): notes.append(f"{s} 와 비슷") verdict = "경고" if notes else "통과" warned += bool(notes) print(f"{name:<30}{verdict:<8}{first:<13}{count!s:>4} {', '.join(notes)}") print(f"\n없음 {missing}건, 경고 {warned}건") return 1 if missing or (a.strict and warned) else 0 if __name__ == "__main__": sys.exit(main())