Function parse_github_resource_url

scripts/utils/github_utils.py:132–172

Parse GitHub URL and extract owner, repo, branch, and path. Returns a dict with keys: owner, repo, branch, path, type.

(url: str)

def parse_github_resource_url(url: str) -> dict[str, str] | None:
    """
    Parse GitHub URL and extract owner, repo, branch, and path.
    Returns a dict with keys: owner, repo, branch, path, type.
    """
    patterns = {
        # File in repository
        "file": r"https://github\.com/([^/]+)/([^/]+)/(?:blob|raw)/([^/]+)/(.+)",
        # Directory in repository
        "dir": r"https://github\.com/([^/]+)/([^/]+)/tree/([^/]+)/(.+)",
        # Repository root
        "repo": r"https://github\.com/([^/]+)/([^/]+)/?$",
        # Gist
        "gist": r"https://gist\.github\.com/([^/]+)/([^/#]+)",
    }

    for url_type, pattern in patterns.items():
        match = re.match(pattern, url)
        if match:
            if url_type == "gist":
                return {
                    "type": "gist",
                    "owner": match.group(1),
                    "gist_id": match.group(2),
                }
            elif url_type == "repo":
                return {
                    "type": "repo",
                    "owner": match.group(1),
                    "repo": _normalize_repo_name(match.group(2)),
                }
            else:
                return {
                    "type": url_type,
                    "owner": match.group(1),
                    "repo": _normalize_repo_name(match.group(2)),
                    "branch": match.group(3),
                    "path": match.group(4),
                }

    return None