Combined 'array<' and 'array' condition check

open-metadata · Jan 10, 2025 · f7480e1 · f7480e1
1 parent ecb4baa
commit f7480e1
Showing 1 changed file with 11 additions and 8 deletions.
diff --git a/ingestion/src/metadata/ingestion/source/database/column_type_parser.py b/ingestion/src/metadata/ingestion/source/database/column_type_parser.py
@@ -344,7 +344,16 @@ def _parse_datatype_string(  # pylint: disable=too-many-return-statements
         # pylint: disable=too-many-branches
         data_type = data_type.lower().strip()
         data_type = data_type.replace(" ", "")
-        if data_type.startswith("array<"):
+        if data_type.startswith("array"):
+            # Set arrayDataType to UNKNOWN for Snowflake table array columns 
+            # to prevent validation error requiring non-null arrayDataType
+            if data_type == "array":
+                data_type_string = {
+                    "dataType": "ARRAY",
+                    "arrayDataType": DataType.UNKNOWN.value,
+                    "dataTypeDisplay": data_type,
+                }
+                return data_type_string
             if data_type[-1] != ">":
                 raise ValueError(f"expected '>' found: {data_type}")
             arr_data_type = ColumnTypeParser._parse_primitive_datatype_string(
@@ -385,13 +394,7 @@ def _parse_datatype_string(  # pylint: disable=too-many-return-statements
             return ColumnTypeParser._parse_struct_fields_string(data_type[7:-1])
         if ":" in data_type:
             return ColumnTypeParser._parse_struct_fields_string(data_type)
-        if data_type == "array":
-            data_type_string = {
-                "dataType": "ARRAY",
-                "arrayDataType": DataType.UNKNOWN,
-                "dataTypeDisplay": data_type,
-            }
-            return data_type_string
+
 
         return ColumnTypeParser._parse_primitive_datatype_string(data_type)