def sanitize_sheets_url()

in google/generativeai/notebook/sheets_sanitize_url.py [0:0]

• 30 lines of code
• 8 McCabe index (conditional complexity)

def sanitize_sheets_url(url: str) -> str:
    """Sanitize a Sheets URL.

    Run some saftey checks to check whether `url` is a Sheets URL. This is not a
    general-purpose URL sanitizer. Rather, it makes use of the fact that we know
    the URL has to be for Sheets so we can make a few assumptions about (e.g. the
    domain).

    Args:
      url: The url to sanitize.

    Returns:
      The sanitized url.

    Raises:
      ValueError: If `url` does not match the expected restrictions for a Sheets
      URL.
    """
    parse_result = parse.urlparse(url)
    if parse_result.scheme != "https":
        raise ValueError(
            'Scheme for Sheets url must be "https", got "{}"'.format(parse_result.scheme)
        )
    if parse_result.netloc not in ("docs.google.com", "sheets.googleapis.com"):
        raise ValueError(
            'Domain for Sheets url must be "docs.google.com", got "{}"'.format(parse_result.netloc)
        )

    # Path component.
    try:
        for fragment in parse_result.path.split("/"):
            _validate_url_part(fragment)
    except ValueError as exc:
        raise ValueError('Invalid path for Sheets url, got "{}"'.format(parse_result.path)) from exc

    # Params component.
    if parse_result.params:
        raise ValueError('Params component must be empty, got "{}"'.format(parse_result.params))

    # Query component.
    try:
        _validate_url_query_or_fragment(parse_result.query)
    except ValueError as exc:
        raise ValueError(
            'Invalid query for Sheets url, got "{}"'.format(parse_result.query)
        ) from exc

    # Fragment component.
    try:
        _validate_url_query_or_fragment(parse_result.fragment)
    except ValueError as exc:
        raise ValueError(
            'Invalid fragment for Sheets url, got "{}"'.format(parse_result.fragment)
        ) from exc

    return url