ungoogled-software/ungoogled-chromium · error · FileExistsError

FileExistsError

Error message

FileExistsError

What it means

apply_substitution() raises FileExistsError(domainsub_cache) when a cache archive path is supplied and a file already exists at that location, to avoid overwriting an existing domain-substitution cache tar.

Source

Thrown at utils/domain_substitution.py:205

    regex_path is a pathlib.Path to domain_regex.list
    files_path is a pathlib.Path to domain_substitution.list
    source_tree is a pathlib.Path to the source tree.
    domainsub_cache is a pathlib.Path to the domain substitution cache.

    Raises NotADirectoryError if the patches directory is not a directory or does not exist
    Raises FileNotFoundError if the source tree or required directory does not exist.
    Raises FileExistsError if the domain substitution cache already exists.
    Raises ValueError if an entry in the domain substitution list contains the file index
        hash delimiter.
    """
    if not source_tree.exists():
        raise FileNotFoundError(source_tree)
    if not regex_path.exists():
        raise FileNotFoundError(regex_path)
    if not files_path.exists():
        raise FileNotFoundError(files_path)
    if domainsub_cache and domainsub_cache.exists():
        raise FileExistsError(domainsub_cache)
    resolved_tree = source_tree.resolve()
    regex_pairs = DomainRegexList(regex_path).regex_pairs
    fileindex_content = io.BytesIO()
    with tarfile.open(str(domainsub_cache), f'w:{domainsub_cache.suffix[1:]}',
                      compresslevel=1) if domainsub_cache else open(
                          os.devnull, 'w', encoding=ENCODING) as cache_tar:
        for relative_path in filter(len, files_path.read_text().splitlines()):
            if _INDEX_HASH_DELIMITER in relative_path:
                if domainsub_cache:
                    # Cache tar will be incomplete; remove it for convenience
                    cache_tar.close()
                    domainsub_cache.unlink()
                raise ValueError(f'Path "{relative_path}" contains '
                                 f'the file index hash delimiter "{_INDEX_HASH_DELIMITER}"')
            path = resolved_tree / relative_path
            if not path.exists():
                get_logger().warning('Skipping non-existent path: %s', path)
                continue

View on GitHub (pinned to f85e84a480)

Solutions

1. Delete the existing cache file before re-running apply_substitution()
2. Pass a new/unique cache path (or domainsub_cache=None to skip caching)
3. Only pass domainsub_cache when starting a fresh substitution

Example fix

// before
cache = Path('domsub_cache.tar.gz')
apply_substitution(tree, regexes, files, domainsub_cache=cache)
// after
cache = Path('domsub_cache.tar.gz')
if cache.exists():
    cache.unlink()
apply_substitution(tree, regexes, files, domainsub_cache=cache)

cache = Path('domsub_cache.tar.gz').resolve()
if cache.exists():
    cache.unlink()  # or choose a new unique name

try:
    apply_substitution(tree, regexes, files, domainsub_cache=cache)
except FileExistsError:
    cache.unlink()
    apply_substitution(tree, regexes, files, domainsub_cache=cache)

When it happens

Trigger: Calling apply_substitution() with a non-None domainsub_cache whose path .exists() — e.g. re-running apply after a previous run created the cache, or a stale cache file left behind.

Common situations: Re-running substitution after an aborted/previous run; resuming a pipeline where the cache wasn't cleaned; reusing the same cache filename for multiple source trees.

Related errors

• FileNotFoundError
• Could not find relative_to directory in extracted files: %s
• Temporary unpacking directory already exists: %s
• Unable to decode with any encoding: {path}
• Path "{relative_path}" contains the file index hash delimite

AI-assisted analysis of ungoogled-software/ungoogled-chromium@f85e84a480 (2026-08-29). Data as JSON: /api/errors/ebc67b9e62b6ac20. Report an issue: GitHub.