diff --git a/tfx/components/util/tfxio_utils.py b/tfx/components/util/tfxio_utils.py index c2cc04ee2a..9a76c6e54d 100644 --- a/tfx/components/util/tfxio_utils.py +++ b/tfx/components/util/tfxio_utils.py @@ -32,7 +32,10 @@ from tensorflow_metadata.proto.v0 import schema_pb2 -_SUPPORTED_FILE_FORMATS = (example_gen_pb2.FileFormat.FORMAT_PARQUET, example_gen_pb2.FileFormat.FORMAT_TFRECORDS_GZIP) +_SUPPORTED_FILE_FORMATS = ( + example_gen_pb2.FileFormat.FILE_FORMAT_PARQUET, + example_gen_pb2.FileFormat.FORMAT_TFRECORDS_GZIP +) # TODO(b/162532479): switch to support List[str] exclusively, once tfx-bsl # post-0.22 is released. OneOrMorePatterns = Union[str, List[str]] diff --git a/tfx/proto/example_gen.proto b/tfx/proto/example_gen.proto index 93f9179822..4e9f0ee4f2 100644 --- a/tfx/proto/example_gen.proto +++ b/tfx/proto/example_gen.proto @@ -127,7 +127,7 @@ enum FileFormat { // Indicates parquet format files in any of the supported compressions. // https://arrow.apache.org/docs/python/parquet.html#compression-encoding-and-file-compatibility - FORMAT_PARQUET = 16; + FILE_FORMAT_PARQUET = 16; reserved 1 to 4, 7 to 15, 17 to max; }