acryl-datahub-cloud 0.3.14.1rc2__py3-none-any.whl → 0.3.14.1rc4__py3-none-any.whl
This diff represents the content of publicly available package versions that have been released to one of the supported registries. The information contained in this diff is provided for informational purposes only and reflects changes between package versions as they appear in their respective public registries.
Potentially problematic release.
This version of acryl-datahub-cloud might be problematic. Click here for more details.
- acryl_datahub_cloud/_codegen_config.json +1 -1
- acryl_datahub_cloud/datahub_reporting/datahub_dataset.py +3 -3
- acryl_datahub_cloud/metadata/_urns/urn_defs.py +1981 -1981
- acryl_datahub_cloud/metadata/com/linkedin/pegasus2avro/settings/global/__init__.py +8 -0
- acryl_datahub_cloud/metadata/schema.avsc +25682 -26059
- acryl_datahub_cloud/metadata/schema_classes.py +876 -681
- acryl_datahub_cloud/metadata/schemas/CorpUserSettings.avsc +127 -0
- acryl_datahub_cloud/metadata/schemas/GlobalSettingsInfo.avsc +136 -0
- acryl_datahub_cloud/metadata/schemas/__init__.py +3 -3
- {acryl_datahub_cloud-0.3.14.1rc2.dist-info → acryl_datahub_cloud-0.3.14.1rc4.dist-info}/METADATA +47 -47
- {acryl_datahub_cloud-0.3.14.1rc2.dist-info → acryl_datahub_cloud-0.3.14.1rc4.dist-info}/RECORD +14 -14
- {acryl_datahub_cloud-0.3.14.1rc2.dist-info → acryl_datahub_cloud-0.3.14.1rc4.dist-info}/WHEEL +0 -0
- {acryl_datahub_cloud-0.3.14.1rc2.dist-info → acryl_datahub_cloud-0.3.14.1rc4.dist-info}/entry_points.txt +0 -0
- {acryl_datahub_cloud-0.3.14.1rc2.dist-info → acryl_datahub_cloud-0.3.14.1rc4.dist-info}/top_level.txt +0 -0
|
@@ -271,7 +271,7 @@ class DataHubBasedS3Dataset:
|
|
|
271
271
|
)
|
|
272
272
|
|
|
273
273
|
def _generate_schema_metadata(
|
|
274
|
-
self, duckdb_columns: List[Tuple[str,
|
|
274
|
+
self, duckdb_columns: List[Tuple[str, Any]]
|
|
275
275
|
) -> SchemaMetadataClass:
|
|
276
276
|
def get_type_from_dtype(dtype: str) -> SchemaFieldDataTypeClass:
|
|
277
277
|
if "int" in dtype or "float" in dtype or "number" in dtype:
|
|
@@ -302,7 +302,7 @@ class DataHubBasedS3Dataset:
|
|
|
302
302
|
)
|
|
303
303
|
for column in duckdb_columns:
|
|
304
304
|
# generate data type
|
|
305
|
-
data_type = column[1].lower()
|
|
305
|
+
data_type = str(column[1]).lower()
|
|
306
306
|
schema_metadata.fields.append(
|
|
307
307
|
SchemaFieldClass(
|
|
308
308
|
fieldPath=column[0],
|
|
@@ -341,7 +341,7 @@ class DataHubBasedS3Dataset:
|
|
|
341
341
|
# generate min, max, avg, distinct count, null count
|
|
342
342
|
column_name = column[0]
|
|
343
343
|
logger.info(f"Generating field profile for {column_name}")
|
|
344
|
-
data_type = column[1].lower()
|
|
344
|
+
data_type = str(column[1]).lower()
|
|
345
345
|
if "int" in data_type or "float" in data_type:
|
|
346
346
|
query = (
|
|
347
347
|
f"SELECT COUNT(DISTINCT {column_name}), COUNT(*) - COUNT({column_name}), MIN({column_name}), MAX({column_name}), AVG({column_name})"
|