Function parse_github_url

src/python/lfs_plugins/installer.py:638–695

Parse GitHub URL into (owner, repo, branch). Supports: - https://github.com/owner/repo - https://github.com/owner/repo.git - https://github.com/owner/repo/tree/branch - github:owner/repo - github:owner/repo@branch - owner/repo (assumes GitHub)

(url: str)

def parse_github_url(url: str) -> Tuple[str, str, Optional[str]]:
    """Parse GitHub URL into (owner, repo, branch).

    Supports:
        - https://github.com/owner/repo
        - https://github.com/owner/repo.git
        - https://github.com/owner/repo/tree/branch
        - github:owner/repo
        - github:owner/repo@branch
        - owner/repo (assumes GitHub)
    """
    url = url.strip()
    branch = None

    # Handle github: shorthand
    if url.startswith("github:"):
        url = url[7:]  # Remove "github:"
        if "@" in url:
            repo_part, branch = url.rsplit("@", 1)
        else:
            repo_part, branch = url, None

        parts = repo_part.split("/")
        if len(parts) != 2:
            raise PluginError(f"Invalid GitHub shorthand: {url}")
        return parts[0], parts[1], branch

    # Handle owner/repo shorthand
    if "/" in url and not url.startswith("http"):
        parts = url.split("/")
        if len(parts) == 2 and not url.startswith("."):
            return parts[0], parts[1], None

    # Handle full URLs with @ref suffix
    if "@" in url and url.startswith(("http://", "https://", "github.com/", "www.github.com/")):
        url, branch = url.rsplit("@", 1)

    # Normalize URLs without scheme (github.com/owner/repo -> https://github.com/owner/repo)
    if url.startswith("github.com/") or url.startswith("www.github.com/"):
        url = "https://" + url

    # Handle full URLs
    parsed = urlparse(url)
    if parsed.netloc not in ("github.com", "www.github.com"):
        raise PluginError(f"Not a GitHub URL: {url}")

    path_parts = parsed.path.strip("/").split("/")
    if len(path_parts) < 2:
        raise PluginError(f"Invalid GitHub URL: {url}")

    owner = path_parts[0]
    repo = path_parts[1].removesuffix(".git")

    # Check for /tree/branch pattern
    if len(path_parts) >= 4 and path_parts[2] == "tree":
        branch = path_parts[3]

    return owner, repo, branch