allensdk.brain_observatory.data_release_utils.metadata_utils.utils module¶
- allensdk.brain_observatory.data_release_utils.metadata_utils.utils.add_file_paths_to_metadata_table(metadata_table: DataFrame, id_generator: FileIDGenerator, file_dir: Path, file_prefix: str | None, index_col: str, data_dir_col: str | None, on_missing_file: str, file_suffix: str = 'nwb', file_stem: str | None = None) DataFrame [source]¶
Add file_id and file_path columns to session dataframe.
- Parameters:
- metadata_table: pd.DataFrame
The dataframe to which we are adding file_id and file_path
- id_generator: FileIDGenerator
For maintaining a unique mapping between file_path and file_id
- file_dir: pathlib.Path
directory where files will be found
- file_prefix: str
Prefix of file names
- index_col: str
Column in metadata_table used to index files
- data_dir_col
Column in metadata_table denoting directory structure of data For example if data is stored under each session_id
<session_id> / file_a <session_id> / file_b …
then give the name of the session_id col here
If None, data is assumed to be stored flat
- on_missing_file: str
- Specifies how to handle missing files
‘error’ -> raise an exception ‘warning’ -> assign dummy file_id and warn ‘skip’ -> drop that row from the table and warn
- file_suffix
- file_stem
Explicit file stem. Overrides dynamic naming of files
- Returns:
- metadata_table:
The same as the input dataframe but with file_id and file_path columns added
Notes
Files are assumed to be named like {file_dir}/{file_prefix}_{metadata_table.index_col}.{file_suffix}