deltacat 1.1.10__tar.gz → 1.1.12__tar.gz

{deltacat-1.1.10/deltacat.egg-info → deltacat-1.1.12}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.1
 Name: deltacat
-Version: 1.1.10
+Version: 1.1.12
 Summary: A scalable, fast, ACID-compliant Data Catalog powered by Ray.
 Home-page: https://github.com/ray-project/deltacat
 Author: Ray Team

{deltacat-1.1.10 → deltacat-1.1.12}/deltacat/__init__.py

@@ -44,7 +44,7 @@ from deltacat.types.tables import TableWriteMode
 
 deltacat.logs.configure_deltacat_logger(logging.getLogger(__name__))
 
-__version__ = "1.1.10"
+__version__ = "1.1.12"
 
 
 __all__ = [

{deltacat-1.1.10 → deltacat-1.1.12}/deltacat/compute/compactor_v2/utils/primary_key_index.py

@@ -162,7 +162,7 @@ def group_by_pk_hash_bucket(
         len(new_tables) == 1
     ), f"Expected only 1 table in the result but found {len(new_tables)}"
 
-    table = generate_pk_hash_column([table], primary_keys, requires_hash=True)[0]
+    table = new_tables[0]
 
     # group hash bucket record indices
     result = group_record_indices_by_hash_bucket(

{deltacat-1.1.10 → deltacat-1.1.12}/deltacat/exceptions.py

@@ -213,11 +213,14 @@ def categorize_errors(func: Callable):
         except BaseException as e:
             deltacat_storage = None
             deltacat_storage_kwargs = {}
+            all_args = args
             if kwargs:
                 deltacat_storage = kwargs.get(DELTACAT_STORAGE_PARAM)
                 deltacat_storage_kwargs = kwargs.get(DELTACAT_STORAGE_KWARGS_PARAM, {})
-            if not deltacat_storage and args:
-                for arg in args:
+                all_args = all_args + tuple(kwargs.values())
+
+            if not deltacat_storage and all_args:
+                for arg in all_args:
                     if (
                         isinstance(arg, dict)
                         and arg.get(DELTACAT_STORAGE_PARAM) is not None

{deltacat-1.1.10 → deltacat-1.1.12}/deltacat/tests/compute/compactor/utils/test_round_completion_file.py

@@ -179,6 +179,28 @@ class TestReadWriteRoundCompletionFile:
 
         assert rcf == expected_rcf_2
 
+    def test_read_when_none_destination_partition_id(self):
+
+        source_locator = get_test_partition_locator("source")
+        destination_locator = get_test_partition_locator(None)
+
+        expected_rcf = RoundCompletionInfo.of(
+            high_watermark=122,
+            compacted_delta_locator={},
+            compacted_pyarrow_write_result={},
+            sort_keys_bit_width=12,
+        )
+
+        write_round_completion_file(
+            RCF_BUCKET_NAME, source_locator, destination_locator, expected_rcf
+        )
+
+        rcf = read_round_completion_file(
+            RCF_BUCKET_NAME, source_locator, destination_locator
+        )
+
+        assert rcf == expected_rcf
+
     def test_write_when_custom_url_is_passed(self):
         """
         This test case tests the backward compatibility by successfully

{deltacat-1.1.10 → deltacat-1.1.12}/deltacat/tests/utils/test_daft.py

@@ -104,6 +104,44 @@ class TestDaftS3FileToTable(unittest.TestCase):
         self.assertEqual(table.schema.field("MISSING").type, pa.string())
         self.assertEqual(table.num_rows, 100)
 
+    def test_read_from_s3_single_column_with_schema_extra_cols_column_names(self):
+        schema = pa.schema([("a", pa.int8()), ("MISSING", pa.string())])
+        pa_read_func_kwargs_provider = ReadKwargsProviderPyArrowSchemaOverride(
+            schema=schema
+        )
+        table = daft_s3_file_to_table(
+            self.MVP_PATH,
+            content_encoding=ContentEncoding.IDENTITY.value,
+            content_type=ContentType.PARQUET.value,
+            column_names=["a", "MISSING"],
+            pa_read_func_kwargs_provider=pa_read_func_kwargs_provider,
+        )
+        self.assertEqual(
+            table.schema.names, ["a", "MISSING"]
+        )  # NOTE: "MISSING" is padded as a null array
+        self.assertEqual(table.schema.field("a").type, pa.int8())
+        self.assertEqual(table.schema.field("MISSING").type, pa.string())
+        self.assertEqual(table.num_rows, 100)
+
+    def test_read_from_s3_single_column_with_schema_only_missing_col(self):
+        schema = pa.schema([("a", pa.int8()), ("MISSING", pa.string())])
+        pa_read_func_kwargs_provider = ReadKwargsProviderPyArrowSchemaOverride(
+            schema=schema
+        )
+        table = daft_s3_file_to_table(
+            self.MVP_PATH,
+            content_encoding=ContentEncoding.IDENTITY.value,
+            content_type=ContentType.PARQUET.value,
+            include_columns=["MISSING"],
+            column_names=["a", "MISSING"],
+            pa_read_func_kwargs_provider=pa_read_func_kwargs_provider,
+        )
+        self.assertEqual(
+            table.schema.names, ["MISSING"]
+        )  # NOTE: "MISSING" is padded as a null array
+        self.assertEqual(table.schema.field("MISSING").type, pa.string())
+        self.assertEqual(table.num_rows, 0)
+
     def test_read_from_s3_single_column_with_row_groups(self):
 
         metadata = pq.read_metadata(self.MVP_PATH)

{deltacat-1.1.10 → deltacat-1.1.12}/deltacat/tests/utils/test_pyarrow.py

@@ -85,6 +85,43 @@ class TestS3PartialParquetFileToTable(TestCase):
         self.assertEqual(result_schema.field(2).type, "int64")
         self.assertEqual(result_schema.field(2).name, "MISSING")
 
+    def test_s3_partial_parquet_file_to_table_when_schema_missing_columns(self):
+
+        pq_file = ParquetFile(PARQUET_FILE_PATH)
+        partial_parquet_params = PartialParquetParameters.of(
+            pq_metadata=pq_file.metadata
+        )
+        # only first row group to be downloaded
+        partial_parquet_params.row_groups_to_download.pop()
+
+        schema = pa.schema(
+            [
+                pa.field("n_legs", pa.string()),
+                pa.field("animal", pa.string()),
+                # NOTE: This field is not in the parquet file, but will be added on as an all-null column
+                pa.field("MISSING", pa.int64()),
+            ]
+        )
+
+        pa_kwargs_provider = lambda content_type, kwargs: {"schema": schema}
+
+        result = s3_partial_parquet_file_to_table(
+            PARQUET_FILE_PATH,
+            ContentType.PARQUET.value,
+            ContentEncoding.IDENTITY.value,
+            pa_read_func_kwargs_provider=pa_kwargs_provider,
+            partial_file_download_params=partial_parquet_params,
+            column_names=["n_legs", "animal", "MISSING"],
+            include_columns=["MISSING"],
+        )
+
+        self.assertEqual(len(result), 0)
+        self.assertEqual(len(result.column_names), 1)
+
+        result_schema = result.schema
+        self.assertEqual(result_schema.field(0).type, "int64")
+        self.assertEqual(result_schema.field(0).name, "MISSING")
+
     def test_s3_partial_parquet_file_to_table_when_schema_passed_with_include_columns(
         self,
     ):
@@ -234,6 +271,32 @@ class TestReadCSV(TestCase):
             lambda: pyarrow_read_csv(NON_EMPTY_VALID_UTSV_PATH, **kwargs),
         )
 
+    def test_read_csv_when_excess_columns_included(self):
+
+        schema = pa.schema(
+            [
+                ("is_active", pa.string()),
+                ("ship_datetime_utc", pa.timestamp("us")),
+                ("MISSING", pa.string()),
+            ]
+        )
+        kwargs = content_type_to_reader_kwargs(ContentType.UNESCAPED_TSV.value)
+        _add_column_kwargs(
+            ContentType.UNESCAPED_TSV.value,
+            ["is_active", "ship_datetime_utc", "MISSING"],
+            ["is_active", "ship_datetime_utc", "MISSING"],
+            kwargs,
+        )
+
+        read_kwargs_provider = ReadKwargsProviderPyArrowSchemaOverride(schema=schema)
+
+        kwargs = read_kwargs_provider(ContentType.UNESCAPED_TSV.value, kwargs)
+
+        self.assertRaises(
+            pa.lib.ArrowInvalid,
+            lambda: pyarrow_read_csv(NON_EMPTY_VALID_UTSV_PATH, **kwargs),
+        )
+
     def test_read_csv_when_empty_csv_sanity(self):
 
         schema = pa.schema(

{deltacat-1.1.10 → deltacat-1.1.12/deltacat.egg-info}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.1
 Name: deltacat
-Version: 1.1.10
+Version: 1.1.12
 Summary: A scalable, fast, ACID-compliant Data Catalog powered by Ray.
 Home-page: https://github.com/ray-project/deltacat
 Author: Ray Team

{deltacat-1.1.10 → deltacat-1.1.12}/deltacat.egg-info/requires.txt

@@ -4,11 +4,11 @@ numpy==1.21.5
 pandas==1.3.5
 pyarrow==12.0.1
 pydantic==1.10.4
-ray[default]>=2.20.0
+ray>=2.20.0
 s3fs==2024.5.0
 tenacity==8.1.0
 typing-extensions==4.4.0
 pymemcache==4.0.0
 redis==4.6.0
-getdaft==0.2.27
+getdaft==0.2.31
 schedule==1.2.0

{deltacat-1.1.10 → deltacat-1.1.12}/setup.py

@@ -41,13 +41,13 @@ setuptools.setup(
         "pandas == 1.3.5",
         "pyarrow == 12.0.1",
         "pydantic == 1.10.4",
-        "ray[default] >= 2.20.0",
+        "ray >= 2.20.0",
         "s3fs == 2024.5.0",
         "tenacity == 8.1.0",
         "typing-extensions == 4.4.0",
         "pymemcache == 4.0.0",
         "redis == 4.6.0",
-        "getdaft == 0.2.27",
+        "getdaft == 0.2.31",
         "schedule == 1.2.0",
     ],
     setup_requires=["wheel"],