Class FileSource

sdk/python/feast/infra/offline_stores/file_source.py:32–273

A FileSource object defines a data source that a DaskOfflineStore or DuckDBOfflineStore class can use.

@typechecked
class FileSource(DataSource):
    """A FileSource object defines a data source that a DaskOfflineStore or DuckDBOfflineStore class can use."""

    def source_type(self) -> DataSourceProto.SourceType.ValueType:
        return DataSourceProto.BATCH_FILE

    def __init__(
        self,
        *,
        path: str,
        name: Optional[str] = "",
        event_timestamp_column: Optional[str] = "",
        file_format: Optional[FileFormat] = None,
        created_timestamp_column: Optional[str] = "",
        field_mapping: Optional[Dict[str, str]] = None,
        s3_endpoint_override: Optional[str] = None,
        description: Optional[str] = "",
        tags: Optional[Dict[str, str]] = None,
        owner: Optional[str] = "",
        timestamp_field: Optional[str] = "",
    ):
        """
        Creates a FileSource object.

        Args:
            path: File path to file containing feature data. Must contain an event_timestamp column, entity columns and
                feature columns.
            name (optional): Name for the file source. Defaults to the path.
            event_timestamp_column (optional): (Deprecated in favor of timestamp_field) Event
                timestamp column used for point in time joins of feature values.
            created_timestamp_column (optional): Timestamp column when row was created, used for deduplicating rows.
            file_format (optional): Explicitly set the file format. Allows Feast to bypass inferring the file format.
            field_mapping: A dictionary mapping of column names in this data source to feature names in a feature table
                or view. Only used for feature columns, not entities or timestamp columns.
            s3_endpoint_override (optional): Overrides AWS S3 enpoint with custom S3 storage
            description (optional): A human-readable description.
            tags (optional): A dictionary of key-value pairs to store arbitrary metadata.
            owner (optional): The owner of the file source, typically the email of the primary
                maintainer.
            timestamp_field (optional): Event timestamp field used for point in time
                joins of feature values.

        Examples:
            >>> from feast import FileSource
            >>> file_source = FileSource(path="my_features.parquet", timestamp_field="event_timestamp")
        """
        self.file_options = FileOptions(
            file_format=file_format,
            uri=path,
            s3_endpoint_override=s3_endpoint_override,
        )

        super().__init__(
            name=name if name else path,
            timestamp_field=timestamp_field,
            created_timestamp_column=created_timestamp_column,
            field_mapping=field_mapping,
            description=description,