Snowflake: Support TOP N select clause modifier

src/sqlfluff/dialects/dialect_snowflake.py

@@ -877,6 +877,21 @@ class SelectStatementSegment(ansi_dialect.get_segment("SelectStatementSegment"))
     )
 
 
+@snowflake_dialect.segment(replace=True)
+class SelectClauseModifierSegment(BaseSegment):
+    """Things that come after SELECT but before the columns, specifically for Snowflake.
+
+    https://docs.snowflake.com/en/sql-reference/constructs.html
+    """
+
+    type = "select_clause_modifier"
+    match_grammar = Sequence(
+        OneOf("DISTINCT", "ALL", optional=True),
+        # TOP N is unique to Snowflake, and we can optionally add DISTINCT/ALL in front of it.
+        Sequence("TOP", Ref("NumericLiteralSegment"), optional=True),
+    )
+
+
 @snowflake_dialect.segment()
 class AlterTableColumnStatementSegment(BaseSegment):
     """An `ALTER TABLE .. ALTER COLUMN` statement.

src/sqlfluff/dialects/dialect_snowflake_keywords.py

@@ -378,6 +378,7 @@
 TEXT
 TIME
 TIMESTAMP
+TOP
 TRANSACTION
 TRANSACTIONS
 TRANSIENT

test/fixtures/dialects/snowflake/snowflake_select_clause_modifiers.sql

@@ -0,0 +1,21 @@
+SELECT *, col1, col2, my_table.col1, my_table.* FROM my_table;
+
+SELECT DISTINCT * FROM my_table;
+
+SELECT DISTINCT col1 FROM my_table;
+
+SELECT ALL my_table.* FROM my_table;
+
+SELECT TOP 1 * FROM my_table;
+
+SELECT TOP 2 col1 FROM my_table;
+
+SELECT TOP 3 col1, my_table.* FROM my_table;
+
+SELECT ALL TOP 10 col1 FROM my_table;
+
+SELECT DISTINCT TOP 20 my_table.col1 FROM my_table;
+
+SELECT DISTINCT TOP 30 * FROM my_table;
+
+SELECT DISTINCT TOP 40 col1, my_table.* FROM my_table;

test/fixtures/dialects/snowflake/snowflake_select_clause_modifiers.yml

@@ -0,0 +1,247 @@
+# YML test files are auto-generated from SQL files and should not be edited by
+# hand. To help enforce this, the "hash" field in the file must match a hash
+# computed by SQLFluff when running the tests. Please run
+# `python test/generate_parse_fixture_yml.py`  to generate them after adding or
+# altering SQL files.
+_hash: c2f7708096f526a51c7f5d0e0afe420cdbff71fc0cc1ef0d9605edc31577b578
+file:
+- statement:
+    select_statement:
+      select_clause:
+      - keyword: SELECT
+      - select_clause_element:
+          wildcard_expression:
+            wildcard_identifier:
+              star: '*'
+      - comma: ','
+      - select_clause_element:
+          column_reference:
+            identifier: col1
+      - comma: ','
+      - select_clause_element:
+          column_reference:
+            identifier: col2
+      - comma: ','
+      - select_clause_element:
+          column_reference:
+          - identifier: my_table
+          - dot: .
+          - identifier: col1
+      - comma: ','
+      - select_clause_element:
+          wildcard_expression:
+            wildcard_identifier:
+              identifier: my_table
+              dot: .
+              star: '*'
+      from_clause:
+        keyword: FROM
+        from_expression:
+          from_expression_element:
+            table_expression:
+              table_reference:
+                identifier: my_table
+- statement_terminator: ;
+- statement:
+    select_statement:
+      select_clause:
+        keyword: SELECT
+        select_clause_modifier:
+          keyword: DISTINCT
+        select_clause_element:
+          wildcard_expression:
+            wildcard_identifier:
+              star: '*'
+      from_clause:
+        keyword: FROM
+        from_expression:
+          from_expression_element:
+            table_expression:
+              table_reference:
+                identifier: my_table
+- statement_terminator: ;
+- statement:
+    select_statement:
+      select_clause:
+        keyword: SELECT
+        select_clause_modifier:
+          keyword: DISTINCT
+        select_clause_element:
+          column_reference:
+            identifier: col1
+      from_clause:
+        keyword: FROM
+        from_expression:
+          from_expression_element:
+            table_expression:
+              table_reference:
+                identifier: my_table
+- statement_terminator: ;
+- statement:
+    select_statement:
+      select_clause:
+        keyword: SELECT
+        select_clause_modifier:
+          keyword: ALL
+        select_clause_element:
+          wildcard_expression:
+            wildcard_identifier:
+              identifier: my_table
+              dot: .
+              star: '*'
+      from_clause:
+        keyword: FROM
+        from_expression:
+          from_expression_element:
+            table_expression:
+              table_reference:
+                identifier: my_table
+- statement_terminator: ;
+- statement:
+    select_statement:
+      select_clause:
+        keyword: SELECT
+        select_clause_modifier:
+          keyword: TOP
+          literal: '1'
+        select_clause_element:
+          wildcard_expression:
+            wildcard_identifier:
+              star: '*'
+      from_clause:
+        keyword: FROM
+        from_expression:
+          from_expression_element:
+            table_expression:
+              table_reference:
+                identifier: my_table
+- statement_terminator: ;
+- statement:
+    select_statement:
+      select_clause:
+        keyword: SELECT
+        select_clause_modifier:
+          keyword: TOP
+          literal: '2'
+        select_clause_element:
+          column_reference:
+            identifier: col1
+      from_clause:
+        keyword: FROM
+        from_expression:
+          from_expression_element:
+            table_expression:
+              table_reference:
+                identifier: my_table
+- statement_terminator: ;
+- statement:
+    select_statement:
+      select_clause:
+      - keyword: SELECT
+      - select_clause_modifier:
+          keyword: TOP
+          literal: '3'
+      - select_clause_element:
+          column_reference:
+            identifier: col1
+      - comma: ','
+      - select_clause_element:
+          wildcard_expression:
+            wildcard_identifier:
+              identifier: my_table
+              dot: .
+              star: '*'
+      from_clause:
+        keyword: FROM
+        from_expression:
+          from_expression_element:
+            table_expression:
+              table_reference:
+                identifier: my_table
+- statement_terminator: ;
+- statement:
+    select_statement:
+      select_clause:
+        keyword: SELECT
+        select_clause_modifier:
+        - keyword: ALL
+        - keyword: TOP
+        - literal: '10'
+        select_clause_element:
+          column_reference:
+            identifier: col1
+      from_clause:
+        keyword: FROM
+        from_expression:
+          from_expression_element:
+            table_expression:
+              table_reference:
+                identifier: my_table
+- statement_terminator: ;
+- statement:
+    select_statement:
+      select_clause:
+        keyword: SELECT
+        select_clause_modifier:
+        - keyword: DISTINCT
+        - keyword: TOP
+        - literal: '20'
+        select_clause_element:
+          column_reference:
+          - identifier: my_table
+          - dot: .
+          - identifier: col1
+      from_clause:
+        keyword: FROM
+        from_expression:
+          from_expression_element:
+            table_expression:
+              table_reference:
+                identifier: my_table
+- statement_terminator: ;
+- statement:
+    select_statement:
+      select_clause:
+        keyword: SELECT
+        select_clause_modifier:
+        - keyword: DISTINCT
+        - keyword: TOP
+        - literal: '30'
+        select_clause_element:
+          wildcard_expression:
+            wildcard_identifier:
+              star: '*'
+      from_clause:
+        keyword: FROM
+        from_expression:
+          from_expression_element:
+            table_expression:
+              table_reference:
+                identifier: my_table
+- statement_terminator: ;
+- statement:
+    select_statement:
+      select_clause:
+      - keyword: SELECT
+      - select_clause_modifier:
+        - keyword: DISTINCT
+        - keyword: TOP
+        - literal: '40'
+      - select_clause_element:
+          column_reference:
+            identifier: col1
+      - comma: ','
+      - select_clause_element:
+          wildcard_expression:
+            wildcard_identifier:
+              identifier: my_table
+              dot: .
+              star: '*'
+      from_clause:
+        keyword: FROM
+        from_expression:
+          from_expression_element:
+            table_expression:
+              table_reference:
+                identifier: my_table
+- statement_terminator: ;