Method ToFileInfo

cpp/src/arrow/filesystem/gcsfs.cc:681–699

The normalize_directories parameter is needed because how a directory is listed. If a specific path is asked for with a trailing slash it is expected to have a trailing slash [1] but for recursive listings it is expected that directories have their path normalized [2]. [1] https://github.com/apache/arrow/blob/3eaa7dd0e8b3dabc5438203331f05e3e6c011e37/python/pyarrow/tests/test_fs.py#L688 [2] https:

  // [2]
  // https://github.com/apache/arrow/blob/3eaa7dd0e8b3dabc5438203331f05e3e6c011e37/cpp/src/arrow/filesystem/test_util.cc#L767
  static FileInfo ToFileInfo(const std::string& full_path,
                             const gcs::ObjectMetadata& meta,
                             bool normalize_directories = false) {
    if (IsDirectory(meta) || (!full_path.empty() && full_path.back() == '/')) {
      if (normalize_directories) {
        auto normalized = std::string(internal::RemoveTrailingSlash(full_path));
        return FileInfo(std::move(normalized), FileType::Directory);
      } else {
        return FileInfo(full_path, FileType::Directory);
      }
    }
    auto info = FileInfo(full_path, FileType::File);
    info.set_size(static_cast<int64_t>(meta.size()));
    // An object has multiple "time" attributes, including the time when its data was
    // created, and the time when its metadata was last updated. We use the object
    // creation time because the data for an object cannot be changed once created.
    info.set_mtime(meta.time_created());
    return info;
  }

  GcsOptions options_;
  gcs::Client client_;