fix issue parsing structs

CHANGELOG.md

@@ -2,12 +2,20 @@
 
 ### Fixes
 - Enhanced get_columns_in_relation method to handle a bug in open source deltalake which doesnt return schema details in `show table extended in databasename like '*'` query output. This impacts dbt snapshots if file format is open source deltalake ([#207](https://github.com/dbt-labs/dbt-spark/pull/207))
+- Parse properly columns when there are struct fields to avoid considering inner fields: Issue ([#202](https://github.com/dbt-labs/dbt-spark/issues/202))
+
+### Contributors
+- [@harryharanb](https://github.com/harryharanb) ([#207](https://github.com/dbt-labs/dbt-spark/pull/207))
+- [@SCouto](https://github.com/Scouto) ([#204](https://github.com/dbt-labs/dbt-spark/pull/204))
+
+## dbt-spark 0.21.0b2 (August 20, 2021)
+
+### Fixes
 - Add pyodbc import error message to dbt.exceptions.RuntimeException to get more detailed information when running `dbt debug` ([#192](https://github.com/dbt-labs/dbt-spark/pull/192))
 - Add support for ODBC Server Side Parameters, allowing options that need to be set with the `SET` statement to be used ([#201](https://github.com/dbt-labs/dbt-spark/pull/201))
 - Add `retry_all` configuration setting to retry all connection issues, not just when the `_is_retryable_error` function determines ([#194](https://github.com/dbt-labs/dbt-spark/pull/194))
 
 ### Contributors
-- [@harryharanb](https://github.com/harryharanb) ([#207](https://github.com/dbt-labs/dbt-spark/pull/207))
 - [@JCZuurmond](https://github.com/JCZuurmond) ([#192](https://github.com/fishtown-analytics/dbt-spark/pull/192))
 - [@jethron](https://github.com/jethron) ([#201](https://github.com/fishtown-analytics/dbt-spark/pull/201))
 - [@gregingenii](https://github.com/gregingenii) ([#194](https://github.com/dbt-labs/dbt-spark/pull/194))
@@ -24,6 +32,7 @@
 ### Contributors
 - [@ali-tny](https://github.com/ali-tny) ([#197](https://github.com/fishtown-analytics/dbt-spark/pull/197))
 
+
 ## dbt-spark 0.20.0 (July 12, 2021)
 
 ## dbt-spark 0.20.0rc2 (July 7, 2021)

dbt/adapters/spark/impl.py

@@ -64,7 +64,7 @@ class SparkAdapter(SQLAdapter):
         'stats:rows:include',
     )
     INFORMATION_COLUMNS_REGEX = re.compile(
-        r"\|-- (.*): (.*) \(nullable = (.*)\b", re.MULTILINE)
+        r"^ \|-- (.*): (.*) \(nullable = (.*)\b", re.MULTILINE)
     INFORMATION_OWNER_REGEX = re.compile(r"^Owner: (.*)$", re.MULTILINE)
     INFORMATION_STATISTICS_REGEX = re.compile(
         r"^Statistics: (.*)$", re.MULTILINE)

test/unit/test_adapter.py

@@ -275,6 +275,7 @@ def test_parse_relation(self):
             ('col1', 'decimal(22,0)'),
             ('col2', 'string',),
             ('dt', 'date'),
+            ('struct_col', 'struct<struct_inner_col:string>'),
             ('# Partition Information', 'data_type'),
             ('# col_name', 'data_type'),
             ('dt', 'date'),
@@ -299,7 +300,7 @@ def test_parse_relation(self):
         config = self._get_target_http(self.project_cfg)
         rows = SparkAdapter(config).parse_describe_extended(
             relation, input_cols)
-        self.assertEqual(len(rows), 3)
+        self.assertEqual(len(rows), 4)
         self.assertEqual(rows[0].to_column_dict(omit_none=False), {
             'table_database': None,
             'table_schema': relation.schema,
@@ -342,6 +343,20 @@ def test_parse_relation(self):
             'char_size': None
         })
 
+        self.assertEqual(rows[3].to_column_dict(omit_none=False), {
+            'table_database': None,
+            'table_schema': relation.schema,
+            'table_name': relation.name,
+            'table_type': rel_type,
+            'table_owner': 'root',
+            'column': 'struct_col',
+            'column_index': 3,
+            'dtype': 'struct<struct_inner_col:string>',
+            'numeric_scale': None,
+            'numeric_precision': None,
+            'char_size': None
+        })
+
     def test_parse_relation_with_integer_owner(self):
         self.maxDiff = None
         rel_type = SparkRelation.get_relation_type.Table
@@ -507,6 +522,8 @@ def test_parse_columns_from_information_with_table_type_and_delta_provider(self)
             " |-- col1: decimal(22,0) (nullable = true)\n"
             " |-- col2: string (nullable = true)\n"
             " |-- dt: date (nullable = true)\n"
+            " |-- struct_col: struct (nullable = true)\n"
+            " |    |-- struct_inner_col: string (nullable = true)\n"
         )
         relation = SparkRelation.create(
             schema='default_schema',
@@ -518,7 +535,7 @@ def test_parse_columns_from_information_with_table_type_and_delta_provider(self)
         config = self._get_target_http(self.project_cfg)
         columns = SparkAdapter(config).parse_columns_from_information(
             relation)
-        self.assertEqual(len(columns), 3)
+        self.assertEqual(len(columns), 4)
         self.assertEqual(columns[0].to_column_dict(omit_none=False), {
             'table_database': None,
             'table_schema': relation.schema,
@@ -538,6 +555,25 @@ def test_parse_columns_from_information_with_table_type_and_delta_provider(self)
             'stats:bytes:value': 123456789,
         })
 
+        self.assertEqual(columns[3].to_column_dict(omit_none=False), {
+            'table_database': None,
+            'table_schema': relation.schema,
+            'table_name': relation.name,
+            'table_type': rel_type,
+            'table_owner': 'root',
+            'column': 'struct_col',
+            'column_index': 3,
+            'dtype': 'struct',
+            'numeric_scale': None,
+            'numeric_precision': None,
+            'char_size': None,
+
+            'stats:bytes:description': '',
+            'stats:bytes:include': True,
+            'stats:bytes:label': 'bytes',
+            'stats:bytes:value': 123456789,
+        })
+
     def test_parse_columns_from_information_with_view_type(self):
         self.maxDiff = None
         rel_type = SparkRelation.get_relation_type.View
@@ -571,6 +607,8 @@ def test_parse_columns_from_information_with_view_type(self):
             " |-- col1: decimal(22,0) (nullable = true)\n"
             " |-- col2: string (nullable = true)\n"
             " |-- dt: date (nullable = true)\n"
+            " |-- struct_col: struct (nullable = true)\n"
+            " |    |-- struct_inner_col: string (nullable = true)\n"
         )
         relation = SparkRelation.create(
             schema='default_schema',
@@ -582,7 +620,7 @@ def test_parse_columns_from_information_with_view_type(self):
         config = self._get_target_http(self.project_cfg)
         columns = SparkAdapter(config).parse_columns_from_information(
             relation)
-        self.assertEqual(len(columns), 3)
+        self.assertEqual(len(columns), 4)
         self.assertEqual(columns[1].to_column_dict(omit_none=False), {
             'table_database': None,
             'table_schema': relation.schema,
@@ -597,6 +635,20 @@ def test_parse_columns_from_information_with_view_type(self):
             'char_size': None
         })
 
+        self.assertEqual(columns[3].to_column_dict(omit_none=False), {
+            'table_database': None,
+            'table_schema': relation.schema,
+            'table_name': relation.name,
+            'table_type': rel_type,
+            'table_owner': 'root',
+            'column': 'struct_col',
+            'column_index': 3,
+            'dtype': 'struct',
+            'numeric_scale': None,
+            'numeric_precision': None,
+            'char_size': None
+        })
+
     def test_parse_columns_from_information_with_table_type_and_parquet_provider(self):
         self.maxDiff = None
         rel_type = SparkRelation.get_relation_type.Table
@@ -619,6 +671,8 @@ def test_parse_columns_from_information_with_table_type_and_parquet_provider(sel
             " |-- col1: decimal(22,0) (nullable = true)\n"
             " |-- col2: string (nullable = true)\n"
             " |-- dt: date (nullable = true)\n"
+            " |-- struct_col: struct (nullable = true)\n"
+            " |    |-- struct_inner_col: string (nullable = true)\n"
         )
         relation = SparkRelation.create(
             schema='default_schema',
@@ -630,7 +684,7 @@ def test_parse_columns_from_information_with_table_type_and_parquet_provider(sel
         config = self._get_target_http(self.project_cfg)
         columns = SparkAdapter(config).parse_columns_from_information(
             relation)
-        self.assertEqual(len(columns), 3)
+        self.assertEqual(len(columns), 4)
         self.assertEqual(columns[2].to_column_dict(omit_none=False), {
             'table_database': None,
             'table_schema': relation.schema,
@@ -655,3 +709,27 @@ def test_parse_columns_from_information_with_table_type_and_parquet_provider(sel
             'stats:rows:value': 12345678
         })
 
+        self.assertEqual(columns[3].to_column_dict(omit_none=False), {
+            'table_database': None,
+            'table_schema': relation.schema,
+            'table_name': relation.name,
+            'table_type': rel_type,
+            'table_owner': 'root',
+            'column': 'struct_col',
+            'column_index': 3,
+            'dtype': 'struct',
+            'numeric_scale': None,
+            'numeric_precision': None,
+            'char_size': None,
+
+            'stats:bytes:description': '',
+            'stats:bytes:include': True,
+            'stats:bytes:label': 'bytes',
+            'stats:bytes:value': 1234567890,
+
+            'stats:rows:description': '',
+            'stats:rows:include': True,
+            'stats:rows:label': 'rows',
+            'stats:rows:value': 12345678
+        })
+