fix(ingest/powerbi): support comments in m-query grammar (#12177)

Author: sid-acryl

metadata-ingestion/src/datahub/ingestion/source/powerbi/powerbi-lexical-grammar.rule

@@ -21,6 +21,11 @@
 //      |   empty_string
 //      |   empty_string "," argument_list
 // - Added sql_string in any_literal
+// - Added WS_INLINE? in field expression
+//   Added to ignore any comments
+//    %ignore WS             // Ignore whitespace
+//    %ignore CPP_COMMENT    // Ignore single-line comments
+//    %ignore C_COMMENT      // Ignore multi-line comments
 
 lexical_unit:   lexical_elements?
 
@@ -245,6 +250,8 @@ operator_or_punctuator: ","
                     |   "=>" 
                     |   ".." 
                     |   "..."
+                    |   "{{"
+                    |   "}}"
 
 document:   section_document
       |     expression_document
@@ -275,6 +282,7 @@ expression: logical_or_expression
       |     if_expression
       |     error_raising_expression
       |     error_handling_expression
+      |     outer_expression
 
 
 logical_or_expression:  logical_and_expression
@@ -376,6 +384,8 @@ sql_content: /(?:[^\"\\]|\\[\"]|\"\"|\#\(lf\))+/
 
 sql_string: "\"" sql_content "\""
 
+outer_expression: "{{" expression "}}"
+
 argument_list:  WS_INLINE? expression
             |   WS_INLINE? expression WS_INLINE? "," WS_INLINE? argument_list
             |   WS_INLINE? sql_string
@@ -409,7 +419,7 @@ record_expression:  "[" field_list? "]"
 field_list: field
         |   field "," field_list
 
-field:  field_name WS_INLINE? "=" WS_INLINE? expression
+field:  WS_INLINE? field_name WS_INLINE? "=" WS_INLINE? expression
 
 field_name: generalized_identifier
         |   quoted_identifier
@@ -621,4 +631,8 @@ any_literal:      record_literal
 %import common.DIGIT
 %import common.LF
 %import common.CR
-%import common.ESCAPED_STRING
+%import common.ESCAPED_STRING
+
+%ignore WS            // Ignore whitespace
+%ignore CPP_COMMENT    // Ignore single-line comments
+%ignore C_COMMENT      // Ignore multi-line comments

metadata-ingestion/tests/integration/powerbi/test_m_parser.py

@@ -1171,3 +1171,39 @@ def test_m_query_timeout(mock_get_lark_parser):
     assert (
         is_entry_present
     ), 'Warning message "M-Query Parsing Timeout" should be present in reporter'
+
+
+def test_comments_in_m_query():
+    q: str = 'let\n    Source = Snowflake.Databases("xaa48144.snowflakecomputing.com", "COMPUTE_WH", [Role="ACCOUNTADMIN"]),\n    SNOWFLAKE_SAMPLE_DATA_Database = Source{[Name="SNOWFLAKE_SAMPLE_DATA", Kind="Database"]}[Data],\n    TPCDS_SF100TCL_Schema = SNOWFLAKE_SAMPLE_DATA_Database{[Name="TPCDS_SF100TCL", Kind="Schema"]}[Data],\n    ITEM_Table = TPCDS_SF100TCL_Schema{[Name="ITEM", Kind="Table"]}[Data],\n    \n    // Group by I_BRAND and calculate the count\n    BrandCountsTable = Table.Group(ITEM_Table, {"I_BRAND"}, {{"BrandCount", each Table.RowCount(_), Int64.Type}})\nin\n    BrandCountsTable'
+
+    table: powerbi_data_classes.Table = powerbi_data_classes.Table(
+        columns=[],
+        measures=[],
+        expression=q,
+        name="pet_price_index",
+        full_name="datalake.sandbox_pet.pet_price_index",
+    )
+
+    reporter = PowerBiDashboardSourceReport()
+
+    ctx, config, platform_instance_resolver = get_default_instances()
+
+    data_platform_tables: List[DataPlatformTable] = parser.get_upstream_tables(
+        table,
+        reporter,
+        ctx=ctx,
+        config=config,
+        platform_instance_resolver=platform_instance_resolver,
+        parameters={
+            "hostname": "xyz.databricks.com",
+            "http_path": "/sql/1.0/warehouses/abc",
+            "catalog": "cat",
+            "schema": "public",
+        },
+    )[0].upstreams
+
+    assert len(data_platform_tables) == 1
+    assert (
+        data_platform_tables[0].urn
+        == "urn:li:dataset:(urn:li:dataPlatform:snowflake,snowflake_sample_data.tpcds_sf100tcl.item,PROD)"
+    )