openmetadata-ingestion 1.3.2.0rc1__py3-none-any.whl → 1.3.2.0rc2__py3-none-any.whl

metadata/generated/schema/type/entityRelationship.py

@@ -1,6 +1,6 @@
 # generated by datamodel-codegen:
 #   filename:  type/entityRelationship.json
-#   timestamp: 2024-03-25T11:44:26+00:00
+#   timestamp: 2024-03-26T16:14:58+00:00
 
 from __future__ import annotations
 

metadata/generated/schema/type/entityUsage.py

@@ -1,6 +1,6 @@
 # generated by datamodel-codegen:
 #   filename:  type/entityUsage.json
-#   timestamp: 2024-03-25T11:44:26+00:00
+#   timestamp: 2024-03-26T16:14:58+00:00
 
 from __future__ import annotations
 

metadata/generated/schema/type/filterPattern.py

@@ -1,6 +1,6 @@
 # generated by datamodel-codegen:
 #   filename:  type/filterPattern.json
-#   timestamp: 2024-03-25T11:44:26+00:00
+#   timestamp: 2024-03-26T16:14:58+00:00
 
 from __future__ import annotations
 

metadata/generated/schema/type/function.py

@@ -1,6 +1,6 @@
 # generated by datamodel-codegen:
 #   filename:  type/function.json
-#   timestamp: 2024-03-25T11:44:26+00:00
+#   timestamp: 2024-03-26T16:14:58+00:00
 
 from __future__ import annotations
 

metadata/generated/schema/type/include.py

@@ -1,6 +1,6 @@
 # generated by datamodel-codegen:
 #   filename:  type/include.json
-#   timestamp: 2024-03-25T11:44:26+00:00
+#   timestamp: 2024-03-26T16:14:58+00:00
 
 from __future__ import annotations
 

metadata/generated/schema/type/jdbcConnection.py

@@ -1,6 +1,6 @@
 # generated by datamodel-codegen:
 #   filename:  type/jdbcConnection.json
-#   timestamp: 2024-03-25T11:44:26+00:00
+#   timestamp: 2024-03-26T16:14:58+00:00
 
 from __future__ import annotations
 

metadata/generated/schema/type/lifeCycle.py

@@ -1,6 +1,6 @@
 # generated by datamodel-codegen:
 #   filename:  type/lifeCycle.json
-#   timestamp: 2024-03-25T11:44:26+00:00
+#   timestamp: 2024-03-26T16:14:58+00:00
 
 from __future__ import annotations
 

metadata/generated/schema/type/paging.py

@@ -1,6 +1,6 @@
 # generated by datamodel-codegen:
 #   filename:  type/paging.json
-#   timestamp: 2024-03-25T11:44:26+00:00
+#   timestamp: 2024-03-26T16:14:58+00:00
 
 from __future__ import annotations
 

metadata/generated/schema/type/profile.py

@@ -1,6 +1,6 @@
 # generated by datamodel-codegen:
 #   filename:  type/profile.json
-#   timestamp: 2024-03-25T11:44:26+00:00
+#   timestamp: 2024-03-26T16:14:58+00:00
 
 from __future__ import annotations
 

metadata/generated/schema/type/queryParserData.py

@@ -1,6 +1,6 @@
 # generated by datamodel-codegen:
 #   filename:  type/queryParserData.json
-#   timestamp: 2024-03-25T11:44:26+00:00
+#   timestamp: 2024-03-26T16:14:58+00:00
 
 from __future__ import annotations
 

metadata/generated/schema/type/reaction.py

@@ -1,6 +1,6 @@
 # generated by datamodel-codegen:
 #   filename:  type/reaction.json
-#   timestamp: 2024-03-25T11:44:26+00:00
+#   timestamp: 2024-03-26T16:14:58+00:00
 
 from __future__ import annotations
 

metadata/generated/schema/type/schedule.py

@@ -1,6 +1,6 @@
 # generated by datamodel-codegen:
 #   filename:  type/schedule.json
-#   timestamp: 2024-03-25T11:44:26+00:00
+#   timestamp: 2024-03-26T16:14:58+00:00
 
 from __future__ import annotations
 

metadata/generated/schema/type/schema.py

@@ -1,6 +1,6 @@
 # generated by datamodel-codegen:
 #   filename:  type/schema.json
-#   timestamp: 2024-03-25T11:44:26+00:00
+#   timestamp: 2024-03-26T16:14:58+00:00
 
 from __future__ import annotations
 

metadata/generated/schema/type/tableQuery.py

@@ -1,6 +1,6 @@
 # generated by datamodel-codegen:
 #   filename:  type/tableQuery.json
-#   timestamp: 2024-03-25T11:44:26+00:00
+#   timestamp: 2024-03-26T16:14:58+00:00
 
 from __future__ import annotations
 

metadata/generated/schema/type/tableUsageCount.py

@@ -1,6 +1,6 @@
 # generated by datamodel-codegen:
 #   filename:  type/tableUsageCount.json
-#   timestamp: 2024-03-25T11:44:26+00:00
+#   timestamp: 2024-03-26T16:14:58+00:00
 
 from __future__ import annotations
 

metadata/generated/schema/type/tagLabel.py

@@ -1,6 +1,6 @@
 # generated by datamodel-codegen:
 #   filename:  type/tagLabel.json
-#   timestamp: 2024-03-25T11:44:26+00:00
+#   timestamp: 2024-03-26T16:14:58+00:00
 
 from __future__ import annotations
 

metadata/generated/schema/type/usageDetails.py

@@ -1,6 +1,6 @@
 # generated by datamodel-codegen:
 #   filename:  type/usageDetails.json
-#   timestamp: 2024-03-25T11:44:26+00:00
+#   timestamp: 2024-03-26T16:14:58+00:00
 
 from __future__ import annotations
 

metadata/generated/schema/type/usageRequest.py

@@ -1,6 +1,6 @@
 # generated by datamodel-codegen:
 #   filename:  type/usageRequest.json
-#   timestamp: 2024-03-25T11:44:26+00:00
+#   timestamp: 2024-03-26T16:14:58+00:00
 
 from __future__ import annotations
 

metadata/generated/schema/type/votes.py

@@ -1,6 +1,6 @@
 # generated by datamodel-codegen:
 #   filename:  type/votes.json
-#   timestamp: 2024-03-25T11:44:26+00:00
+#   timestamp: 2024-03-26T16:14:58+00:00
 
 from __future__ import annotations
 

metadata/ingestion/api/topology_runner.py

@@ -265,11 +265,12 @@ class TopologyRunnerMixin(Generic[C]):
             if entity:
                 same_fingerprint = True
 
-        create_entity_request_hash = generate_source_hash(
-            create_request=entity_request.right,
-        )
+        create_entity_request_hash = None
 
         if hasattr(entity_request.right, "sourceHash"):
+            create_entity_request_hash = generate_source_hash(
+                create_request=entity_request.right,
+            )
             entity_request.right.sourceHash = create_entity_request_hash
 
         if entity is None and stage.use_cache:

metadata/ingestion/source/database/datalake/metadata.py

@@ -407,7 +407,7 @@ class DatalakeSource(DatabaseServiceSource):
         schema_name = self.context.database_schema
         try:
             table_constraints = None
-            data_frame = fetch_dataframe(
+            data_frame, raw_data = fetch_dataframe(
                 config_source=self.config_source,
                 client=self.client,
                 file_fqn=DatalakeTableSchemaWrapper(
@@ -415,10 +415,11 @@ class DatalakeSource(DatabaseServiceSource):
                     bucket_name=schema_name,
                     file_extension=table_extension,
                 ),
+                fetch_raw_data=True,
             )
             if data_frame:
                 column_parser = DataFrameColumnParser.create(
-                    data_frame[0], table_extension
+                    data_frame[0], table_extension, raw_data=raw_data
                 )
                 columns = column_parser.get_columns()
             else:

metadata/ingestion/source/database/stored_procedures_mixin.py

@@ -138,7 +138,7 @@ class StoredProcedureMixin(ABC):
             return True
 
         if query_type == "INSERT" and re.search(
-            "^.*insert.*into.*select.*$", query_text, re.IGNORECASE
+            "^.*insert.*into.*select.*$", query_text.replace("\n", " "), re.IGNORECASE
         ):
             return True
 

metadata/ingestion/source/storage/storage_service.py

@@ -260,7 +260,7 @@ class StorageServiceSource(TopologyRunnerMixin, Source, ABC):
         metadata_entry: MetadataEntry,
     ) -> List[Column]:
         """Extract Column related metadata from s3"""
-        data_structure_details = fetch_dataframe(
+        data_structure_details, raw_data = fetch_dataframe(
             config_source=config_source,
             client=client,
             file_fqn=DatalakeTableSchemaWrapper(
@@ -269,10 +269,13 @@ class StorageServiceSource(TopologyRunnerMixin, Source, ABC):
                 file_extension=SupportedTypes(metadata_entry.structureFormat),
                 separator=metadata_entry.separator,
             ),
+            fetch_raw_data=True,
         )
         columns = []
         column_parser = DataFrameColumnParser.create(
-            data_structure_details, SupportedTypes(metadata_entry.structureFormat)
+            data_structure_details,
+            SupportedTypes(metadata_entry.structureFormat),
+            raw_data=raw_data,
         )
         columns = column_parser.get_columns()
         return columns

metadata/parsers/json_schema_parser.py

@@ -18,6 +18,8 @@ import traceback
 from enum import Enum
 from typing import List, Optional
 
+from pydantic.main import ModelMetaclass
+
 from metadata.generated.schema.type.schema import FieldModel
 from metadata.utils.logger import ingestion_logger
 
@@ -36,20 +38,25 @@ class JsonSchemaDataTypes(Enum):
     NULL = "null"
     RECORD = "object"
     ARRAY = "array"
+    UNKNOWN = "unknown"
 
 
-def parse_json_schema(schema_text: str) -> Optional[List[FieldModel]]:
+def parse_json_schema(
+    schema_text: str, cls: ModelMetaclass = FieldModel
+) -> Optional[List[FieldModel]]:
     """
     Method to parse the jsonschema
     """
     try:
         json_schema_data = json.loads(schema_text)
         field_models = [
-            FieldModel(
+            cls(
                 name=json_schema_data.get("title", "default"),
                 dataType=JsonSchemaDataTypes(json_schema_data.get("type")).name,
                 description=json_schema_data.get("description"),
-                children=get_json_schema_fields(json_schema_data.get("properties")),
+                children=get_json_schema_fields(
+                    json_schema_data.get("properties", {}), cls=cls
+                ),
             )
         ]
         return field_models
@@ -59,7 +66,9 @@ def parse_json_schema(schema_text: str) -> Optional[List[FieldModel]]:
     return None
 
 
-def get_json_schema_fields(properties) -> Optional[List[FieldModel]]:
+def get_json_schema_fields(
+    properties, cls: ModelMetaclass = FieldModel
+) -> Optional[List[FieldModel]]:
     """
     Recursively convert the parsed schema into required models
     """
@@ -67,9 +76,10 @@ def get_json_schema_fields(properties) -> Optional[List[FieldModel]]:
     for key, value in properties.items():
         try:
             field_models.append(
-                FieldModel(
-                    name=value.get("title", key),
-                    dataType=JsonSchemaDataTypes(value.get("type")).name,
+                cls(
+                    name=key,
+                    displayName=value.get("title"),
+                    dataType=JsonSchemaDataTypes(value.get("type", "unknown")).name,
                     description=value.get("description"),
                     children=get_json_schema_fields(value.get("properties"))
                     if value.get("type") == "object"

metadata/profiler/interface/profiler_interface.py

@@ -33,7 +33,7 @@ from metadata.generated.schema.entity.data.table import (
     TableData,
 )
 from metadata.generated.schema.entity.services.connections.connectionBasicType import (
-    SampleDataStorageConfig,
+    DataStorageConfig,
 )
 from metadata.generated.schema.entity.services.connections.database.datalakeConnection import (
     DatalakeConnection,
@@ -93,7 +93,7 @@ class ProfilerInterface(ABC):
         service_connection_config: Union[DatabaseConnection, DatalakeConnection],
         ometa_client: OpenMetadata,
         entity: Table,
-        storage_config: SampleDataStorageConfig,
+        storage_config: DataStorageConfig,
         profile_sample_config: Optional[ProfileSampleConfig],
         source_config: DatabaseServiceProfilerPipeline,
         sample_query: Optional[str],
@@ -248,7 +248,7 @@ class ProfilerInterface(ABC):
             DatabaseProfilerConfig,
             DatabaseAndSchemaConfig,
         ]
-    ):
+    ) -> Optional[DataStorageConfig]:
         if (
             config
             and config.sampleDataStorageConfig
@@ -264,7 +264,7 @@ class ProfilerInterface(ABC):
         database_profiler_config: Optional[DatabaseProfilerConfig],
         db_service: Optional[DatabaseService],
         profiler_config: ProfilerProcessorConfig,
-    ) -> Optional[SampleDataStorageConfig]:
+    ) -> Optional[DataStorageConfig]:
         """Get config for a specific entity
 
         Args:

metadata/profiler/processor/sample_data_handler.py

@@ -17,8 +17,13 @@ from datetime import datetime
 from functools import singledispatch
 from io import BytesIO
 
+from pydantic.json import ENCODERS_BY_TYPE
+
 from metadata.clients.aws_client import AWSClient
 from metadata.generated.schema.entity.data.table import Table, TableData
+from metadata.generated.schema.entity.services.connections.connectionBasicType import (
+    DataStorageConfig,
+)
 from metadata.generated.schema.security.credentials.awsCredentials import AWSCredentials
 from metadata.profiler.interface.profiler_interface import ProfilerInterface
 from metadata.utils.helpers import clean_uri
@@ -27,15 +32,45 @@ from metadata.utils.logger import profiler_logger
 logger = profiler_logger()
 
 
-def _get_object_key(table: Table, prefix: str, overwrite_data: bool) -> str:
+class PathPatternException(Exception):
+    """
+    Exception class need to validate the file path pattern
+    """
+
+
+def validate_path_pattern(file_path_format: str) -> None:
+    if not (
+        "{service_name}" in file_path_format
+        and "{database_name}" in file_path_format
+        and "{database_schema_name}" in file_path_format
+        and "{table_name}" in file_path_format
+        and file_path_format.endswith(".parquet")
+    ):
+        raise PathPatternException(
+            "Please provide a valid path pattern, "
+            "the pattern should include these components {service_name}, "
+            "{database_name}, {database_schema_name}, {table_name} and "
+            "it should end with extension .parquet"
+        )
+
+
+def _get_object_key(
+    table: Table, prefix: str, overwrite_data: bool, file_path_format: str
+) -> str:
+    validate_path_pattern(file_path_format)
+    file_name = file_path_format.format(
+        service_name=table.service.name,
+        database_name=table.database.name,
+        database_schema_name=table.databaseSchema.name,
+        table_name=table.name.__root__,
+    )
     if not overwrite_data:
-        file_name = f"sample_data_{datetime.now().strftime('%Y_%m_%d')}.parquet"
-    else:
-        file_name = "sample_data.parquet"
-    path = str(table.fullyQualifiedName.__root__).replace(".", "/")
+        file_name = file_name.replace(
+            ".parquet", f"_{datetime.now().strftime('%Y_%m_%d')}.parquet"
+        )
     if prefix:
-        return f"{clean_uri(prefix)}/{path}/{file_name}"
-    return f"{path}/{file_name}"
+        return f"{clean_uri(prefix)}/{file_name}"
+    return file_name
 
 
 def upload_sample_data(data: TableData, profiler_interface: ProfilerInterface) -> None:
@@ -45,9 +80,10 @@ def upload_sample_data(data: TableData, profiler_interface: ProfilerInterface) -
     import pandas as pd  # pylint: disable=import-outside-toplevel
 
     try:
-        sample_storage_config = profiler_interface.storage_config
+        sample_storage_config: DataStorageConfig = profiler_interface.storage_config
         if not sample_storage_config:
             return
+        ENCODERS_BY_TYPE[bytes] = lambda v: v.decode("utf-8", "ignore")
         deserialized_data = json.loads(data.json())
         df = pd.DataFrame(
             data=deserialized_data.get("rows", []),
@@ -59,6 +95,7 @@ def upload_sample_data(data: TableData, profiler_interface: ProfilerInterface) -
             table=profiler_interface.table_entity,
             prefix=sample_storage_config.prefix,
             overwrite_data=sample_storage_config.overwriteData,
+            file_path_format=sample_storage_config.filePathPattern,
         )
         upload_to_storage(
             sample_storage_config.storageConfig,

metadata/readers/dataframe/json.py

@@ -16,7 +16,7 @@ import gzip
 import io
 import json
 import zipfile
-from typing import List, Union
+from typing import Any, Dict, List, Optional, Tuple, Union
 
 from metadata.readers.dataframe.base import DataFrameReader
 from metadata.readers.dataframe.common import dataframe_to_chunks
@@ -47,7 +47,7 @@ class JSONDataFrameReader(DataFrameReader):
     @staticmethod
     def read_from_json(
         key: str, json_text: bytes, decode: bool = False, **__
-    ) -> List["DataFrame"]:
+    ) -> Tuple[List["DataFrame"], Optional[Dict[str, Any]]]:
         """
         Decompress a JSON file (if needed) and read its contents
         as a dataframe.
@@ -60,20 +60,25 @@ class JSONDataFrameReader(DataFrameReader):
         import pandas as pd
 
         json_text = _get_json_text(key=key, text=json_text, decode=decode)
+        raw_data = None
         try:
             data = json.loads(json_text)
+            if isinstance(data, dict) and data.get("$schema"):
+                raw_data = json_text
         except json.decoder.JSONDecodeError:
             logger.debug("Failed to read as JSON object. Trying to read as JSON Lines")
             data = [json.loads(json_obj) for json_obj in json_text.strip().split("\n")]
 
         # if we get a scalar value (e.g. {"a":"b"}) then we need to specify the index
         data = data if not isinstance(data, dict) else [data]
-        return dataframe_to_chunks(pd.DataFrame.from_records(data))
+        return dataframe_to_chunks(pd.DataFrame.from_records(data)), raw_data
 
     def _read(self, *, key: str, bucket_name: str, **kwargs) -> DatalakeColumnWrapper:
         text = self.reader.read(key, bucket_name=bucket_name)
+        dataframes, raw_data = self.read_from_json(
+            key=key, json_text=text, decode=True, **kwargs
+        )
         return DatalakeColumnWrapper(
-            dataframes=self.read_from_json(
-                key=key, json_text=text, decode=True, **kwargs
-            )
+            dataframes=dataframes,
+            raw_data=raw_data,
         )

metadata/readers/dataframe/models.py

@@ -29,6 +29,7 @@ class DatalakeColumnWrapper(BaseModel):
 
     columns: Optional[List[Column]]
     dataframes: Optional[List[Any]]  # pandas.Dataframe does not have any validators
+    raw_data: Any  # in special cases like json schema, we need to store the raw data
 
 
 class DatalakeTableSchemaWrapper(BaseModel):

metadata/utils/datalake/datalake_utils.py

@@ -17,10 +17,11 @@ import ast
 import json
 import random
 import traceback
-from typing import Dict, List, Optional, Union, cast
+from typing import Any, Dict, List, Optional, Union, cast
 
 from metadata.generated.schema.entity.data.table import Column, DataType
 from metadata.ingestion.source.database.column_helpers import truncate_column_name
+from metadata.parsers.json_schema_parser import parse_json_schema
 from metadata.readers.dataframe.models import (
     DatalakeColumnWrapper,
     DatalakeTableSchemaWrapper,
@@ -35,6 +36,7 @@ def fetch_dataframe(
     config_source,
     client,
     file_fqn: DatalakeTableSchemaWrapper,
+    fetch_raw_data: bool = False,
     **kwargs,
 ) -> Optional[List["DataFrame"]]:
     """
@@ -60,6 +62,8 @@ def fetch_dataframe(
                 df_wrapper: DatalakeColumnWrapper = df_reader.read(
                     key=key, bucket_name=bucket_name, **kwargs
                 )
+                if fetch_raw_data:
+                    return df_wrapper.dataframes, df_wrapper.raw_data
                 return df_wrapper.dataframes
             except Exception as err:
                 logger.error(
@@ -73,6 +77,8 @@ def fetch_dataframe(
         # Here we need to blow things up. Without the dataframe we cannot move forward
         raise err
 
+    if fetch_raw_data:
+        return None, None
     return None
 
 
@@ -112,6 +118,7 @@ class DataFrameColumnParser:
         file_type: Optional[SupportedTypes] = None,
         sample: bool = True,
         shuffle: bool = False,
+        raw_data: Any = None,
     ):
         """Instantiate a column parser object with the appropriate parser
 
@@ -126,8 +133,14 @@ class DataFrameColumnParser:
         data_frame = cls._get_data_frame(data_frame, sample, shuffle)
         if file_type == SupportedTypes.PARQUET:
             parser = ParquetDataFrameColumnParser(data_frame)
-            return cls(parser)
-        parser = GenericDataFrameColumnParser(data_frame)
+        elif file_type in {
+            SupportedTypes.JSON,
+            SupportedTypes.JSONGZ,
+            SupportedTypes.JSONZIP,
+        }:
+            parser = JsonDataFrameColumnParser(data_frame, raw_data=raw_data)
+        else:
+            parser = GenericDataFrameColumnParser(data_frame)
         return cls(parser)
 
     @staticmethod
@@ -172,8 +185,9 @@ class GenericDataFrameColumnParser:
         "bytes": DataType.BYTES,
     }
 
-    def __init__(self, data_frame: "DataFrame"):
+    def __init__(self, data_frame: "DataFrame", raw_data: Any = None):
         self.data_frame = data_frame
+        self.raw_data = raw_data
 
     def get_columns(self):
         """
@@ -472,3 +486,19 @@ class ParquetDataFrameColumnParser:
             data_type = self._data_formats.get(str(column.type), DataType.UNKNOWN)
 
         return data_type
+
+
+class JsonDataFrameColumnParser(GenericDataFrameColumnParser):
+    """Given a dataframe object generated from a json file, parse the columns and return a list of Column objects."""
+
+    def get_columns(self):
+        """
+        method to process column details for json files
+        """
+        if self.raw_data:
+            try:
+                return parse_json_schema(schema_text=self.raw_data, cls=Column)
+            except Exception as exc:
+                logger.warning(f"Unable to parse the json schema: {exc}")
+                logger.debug(traceback.format_exc())
+        return self._get_columns(self.data_frame)

metadata/utils/source_hash.py

@@ -14,9 +14,14 @@ Source hash utils module
 """
 
 import hashlib
+import traceback
 from typing import Dict, Optional
 
 from metadata.ingestion.ometa.ometa_api import C
+from metadata.utils.logger import utils_logger
+
+logger = utils_logger()
+
 
 SOURCE_HASH_EXCLUDE_FIELDS = {
     "sourceHash": True,
@@ -25,19 +30,24 @@ SOURCE_HASH_EXCLUDE_FIELDS = {
 
 def generate_source_hash(
     create_request: C, exclude_fields: Optional[Dict] = None
-) -> str:
+) -> Optional[str]:
     """
     Given a create_request model convert it to json string and generate a hash value
     """
-
-    # We always want to exclude the sourceHash when generating the fingerprint
-    exclude_fields = (
-        SOURCE_HASH_EXCLUDE_FIELDS.update(exclude_fields)
-        if exclude_fields
-        else SOURCE_HASH_EXCLUDE_FIELDS
-    )
-
-    create_request_json = create_request.json(exclude=exclude_fields)
-
-    json_bytes = create_request_json.encode("utf-8")
-    return hashlib.md5(json_bytes).hexdigest()
+    try:
+        # We always want to exclude the sourceHash when generating the fingerprint
+        exclude_fields = (
+            SOURCE_HASH_EXCLUDE_FIELDS.update(exclude_fields)
+            if exclude_fields
+            else SOURCE_HASH_EXCLUDE_FIELDS
+        )
+
+        create_request_json = create_request.json(exclude=exclude_fields)
+
+        json_bytes = create_request_json.encode("utf-8")
+        return hashlib.md5(json_bytes).hexdigest()
+
+    except Exception as exc:
+        logger.warning(f"Failed to generate source hash due to - {exc}")
+        logger.debug(traceback.format_exc())
+    return None