repo_fetcher.py

import os
import base64
import requests

API_BASE = "https://api.github.com"

DEPENDENCY_FILENAMES = [
    "requirements.txt", "environment.yml", "environment.yaml",
    "setup.py", "setup.cfg", "pyproject.toml",
    "Pipfile", "Pipfile.lock", "Dockerfile", "Makefile", "CMakeLists.txt",
]


def _get_headers():
    token = os.getenv("GITHUB_TOKEN", "")
    headers = {"Accept": "application/vnd.github.v3+json"}
    if token:
        headers["Authorization"] = f"Bearer {token}"
    return headers


def _github_get(path: str) -> dict | None:
    """统一的 GitHub API GET 请求，失败返回 None"""
    headers = _get_headers()
    url = f"{API_BASE}{path}"
    try:
        resp = requests.get(url, headers=headers, timeout=15)
        if resp.status_code == 404:
            return None
        if resp.status_code in (403, 429):
            print(f"[警告] GitHub API 限速: {path}")
            return None
        resp.raise_for_status()
        return resp.json()
    except Exception as e:
        print(f"[警告] GitHub API 请求失败 ({path}): {e}")
        return None


def fetch_readme(owner: str, name: str) -> str | None:
    """获取仓库 README 文件内容（带缓存，1 小时 TTL）"""
    from cache import github_readme_cache
    cache_key = f"{owner}/{name}"
    cached = github_readme_cache.get(cache_key)
    if cached is not None:
        return cached

    try:
        data = _github_get(f"/repos/{owner}/{name}/readme")
        if not data:
            return None
        content = data.get("content", "")
        result = base64.b64decode(content).decode("utf-8", errors="replace")
        github_readme_cache.set(cache_key, result)
        return result
    except Exception:
        return None


def fetch_dependencies(owner: str, name: str) -> dict[str, str]:
    """获取仓库的依赖文件（带缓存，1 小时 TTL）

    Returns:
        dict: {文件名: 文件内容}，找不到任何依赖文件时返回 {}
    """
    from cache import github_deps_cache
    cache_key = f"{owner}/{name}"
    cached = github_deps_cache.get(cache_key)
    if cached is not None:
        return cached

    result = {}
    try:
        contents = _github_get(f"/repos/{owner}/{name}/contents") or []
        for item in contents:
            fname = item.get("name", "")
            if fname in DEPENDENCY_FILENAMES:
                file_data = _github_get(f"/repos/{owner}/{name}/contents/{fname}")
                if file_data and file_data.get("content"):
                    try:
                        decoded = base64.b64decode(file_data["content"]).decode("utf-8", errors="replace")
                        result[fname] = decoded
                    except Exception:
                        pass
    except Exception:
        pass
    github_deps_cache.set(cache_key, result)
    return result


if __name__ == "__main__":
    # 自测：对 anomalib 仓库调用
    owner, name = "openvinotoolkit", "anomalib"
    print("=== repo_fetcher 自测 ===")

    readme = fetch_readme(owner, name)
    print(f"README 长度: {len(readme) if readme else 0} 字符")
    if readme:
        print(f"  前 200 字符: {readme[:200]}...")

    deps = fetch_dependencies(owner, name)
    print(f"依赖文件: {len(deps)} 个 — {list(deps.keys())}")
    for fname, content in deps.items():
        print(f"  {fname}: {len(content)} 字符")