Function parse_uri

src/websockets/uri.py:60–107

Parse and validate a WebSocket URI. Args: uri: WebSocket URI. Returns: Parsed WebSocket URI. Raises: InvalidURI: If ``uri`` isn't a valid WebSocket URI.

(uri: str)

def parse_uri(uri: str) -> WebSocketURI:
    """
    Parse and validate a WebSocket URI.

    Args:
        uri: WebSocket URI.

    Returns:
        Parsed WebSocket URI.

    Raises:
        InvalidURI: If ``uri`` isn't a valid WebSocket URI.

    """
    parsed = urllib.parse.urlparse(uri)
    if parsed.scheme not in ["ws", "wss"]:
        raise InvalidURI(uri, "scheme isn't ws or wss")
    if parsed.hostname is None:
        raise InvalidURI(uri, "hostname isn't provided")
    if parsed.fragment != "":
        raise InvalidURI(uri, "fragment identifier is meaningless")

    secure = parsed.scheme == "wss"
    host = parsed.hostname
    port = parsed.port or (443 if secure else 80)
    path = parsed.path
    query = parsed.query
    username = parsed.username
    password = parsed.password
    # urllib.parse.urlparse accepts URLs with a username but without a
    # password. This doesn't make sense for HTTP Basic Auth credentials.
    if username is not None and password is None:
        raise InvalidURI(uri, "username provided without password")

    try:
        uri.encode("ascii")
    except UnicodeEncodeError:
        # Input contains non-ASCII characters.
        # It must be an IRI. Convert it to a URI.
        host = host.encode("idna").decode()
        path = urllib.parse.quote(path, safe=DELIMS)
        query = urllib.parse.quote(query, safe=DELIMS)
        if username is not None:
            assert password is not None
            username = urllib.parse.quote(username, safe=DELIMS)
            password = urllib.parse.quote(password, safe=DELIMS)

    return WebSocketURI(secure, host, port, path, query, username, password)