fix: parse keywords as names if qualified by period

CHANGELOG.md

@@ -4,6 +4,10 @@ All notable changes to this project will be documented in this file.
 
 ## [Unreleased]
 
+### Formatting Changes and Bug Fixes
+
+- sqlfmt will now parse unquoted reserved keywords as names if they are qualified by a period, e.g., `foo.select` or `foo.case` ([#599](https://github.com/tconbeer/sqlfmt/issues/599) - thank you [@matthieucan](https://github.com/matthieucan)!). 
+
 ## [0.22.0] - 2024-07-25
 
 ### Formatting Changes and Bug Fixes

src/sqlfmt/actions.py

@@ -308,7 +308,41 @@ def handle_number(analyzer: "Analyzer", source_string: str, match: re.Match) ->
     )
 
 
-def handle_nonreserved_keyword(
+def handle_reserved_keyword(
+    analyzer: "Analyzer",
+    source_string: str,
+    match: re.Match,
+    action: Callable[["Analyzer", str, re.Match], None],
+) -> None:
+    """
+    Reserved keywords can be used in most dialects as table or column
+    names without quoting if they are qualified.
+    https://github.com/tconbeer/sqlfmt/issues/599
+
+    This checks if the previous token is a period, and if so, lexes
+    this token as a name; otherwise this action executes the passed
+    action (which likely adds the node as some kind of keyword).
+    """
+    if analyzer.previous_node is None:
+        action(analyzer, source_string, match)
+        return
+
+    previous_token, _ = get_previous_token(analyzer.previous_node)
+    if previous_token is not None and previous_token.type is TokenType.DOT:
+        token = Token.from_match(source_string, match, token_type=TokenType.NAME)
+        if not token.prefix:
+            node = analyzer.node_manager.create_node(
+                token=token, previous_node=analyzer.previous_node
+            )
+            analyzer.node_buffer.append(node)
+            analyzer.pos = token.epos
+            return
+
+    # in all other cases, this is a keyword.
+    action(analyzer, source_string, match)
+
+
+def handle_nonreserved_top_level_keyword(
     analyzer: "Analyzer",
     source_string: str,
     match: re.Match,
@@ -317,6 +351,10 @@ def handle_nonreserved_keyword(
     """
     Checks to see if we're at depth 0 (assuming this is a name); if so, then take the
     passed action, otherwise lex it as a name.
+
+    For example, this allows us to lex these differently:
+    explain select 1;
+    select explain, 1;
     """
     token = Token.from_match(source_string, match, token_type=TokenType.NAME)
     node = analyzer.node_manager.create_node(
@@ -326,7 +364,9 @@ def handle_nonreserved_keyword(
         analyzer.node_buffer.append(node)
         analyzer.pos = token.epos
     else:
-        action(analyzer, source_string, match)
+        handle_reserved_keyword(
+            analyzer=analyzer, source_string=source_string, match=match, action=action
+        )
 
 
 def lex_ruleset(

src/sqlfmt/rules/__init__.py

@@ -25,17 +25,23 @@
         priority=1000,
         pattern=group(r"case") + group(r"\W", r"$"),
         action=partial(
-            actions.add_node_to_buffer, token_type=TokenType.STATEMENT_START
+            actions.handle_reserved_keyword,
+            action=partial(
+                actions.add_node_to_buffer, token_type=TokenType.STATEMENT_START
+            ),
         ),
     ),
     Rule(
         name="statement_end",
         priority=1010,
         pattern=group(r"end") + group(r"\W", r"$"),
         action=partial(
-            actions.safe_add_node_to_buffer,
-            token_type=TokenType.STATEMENT_END,
-            fallback_token_type=TokenType.NAME,
+            actions.handle_reserved_keyword,
+            action=partial(
+                actions.safe_add_node_to_buffer,
+                token_type=TokenType.STATEMENT_END,
+                fallback_token_type=TokenType.NAME,
+            ),
         ),
     ),
     Rule(
@@ -65,7 +71,12 @@
             r"within\s+group",
         )
         + group(r"\W", r"$"),
-        action=partial(actions.add_node_to_buffer, token_type=TokenType.WORD_OPERATOR),
+        action=partial(
+            actions.handle_reserved_keyword,
+            action=partial(
+                actions.add_node_to_buffer, token_type=TokenType.WORD_OPERATOR
+            ),
+        ),
     ),
     Rule(
         name="star_replace_exclude",
@@ -76,8 +87,10 @@
         )
         + group(r"\s+\("),
         action=partial(
-            actions.add_node_to_buffer,
-            token_type=TokenType.WORD_OPERATOR,
+            actions.handle_reserved_keyword,
+            action=partial(
+                actions.add_node_to_buffer, token_type=TokenType.WORD_OPERATOR
+            ),
         ),
     ),
     Rule(
@@ -88,13 +101,21 @@
         name="join_using",
         priority=1110,
         pattern=group(r"using") + group(r"\s*\("),
-        action=partial(actions.add_node_to_buffer, token_type=TokenType.WORD_OPERATOR),
+        action=partial(
+            actions.handle_reserved_keyword,
+            action=partial(
+                actions.add_node_to_buffer, token_type=TokenType.WORD_OPERATOR
+            ),
+        ),
     ),
     Rule(
         name="on",
         priority=1120,
         pattern=group(r"on") + group(r"\W", r"$"),
-        action=partial(actions.add_node_to_buffer, token_type=TokenType.ON),
+        action=partial(
+            actions.handle_reserved_keyword,
+            action=partial(actions.add_node_to_buffer, token_type=TokenType.ON),
+        ),
     ),
     Rule(
         name="boolean_operator",
@@ -106,8 +127,10 @@
         )
         + group(r"\W", r"$"),
         action=partial(
-            actions.add_node_to_buffer,
-            token_type=TokenType.BOOLEAN_OPERATOR,
+            actions.handle_reserved_keyword,
+            action=partial(
+                actions.add_node_to_buffer, token_type=TokenType.BOOLEAN_OPERATOR
+            ),
         ),
     ),
     Rule(
@@ -154,15 +177,25 @@
             r"returning",
         )
         + group(r"\W", r"$"),
-        action=partial(actions.add_node_to_buffer, token_type=TokenType.UNTERM_KEYWORD),
+        action=partial(
+            actions.handle_reserved_keyword,
+            action=partial(
+                actions.add_node_to_buffer, token_type=TokenType.UNTERM_KEYWORD
+            ),
+        ),
     ),
     Rule(
         name="frame_clause",
         priority=1305,
         pattern=group(r"(range|rows|groups)\s+")
         + group(r"(between\s+)?((unbounded|\d+)\s+(preceding|following)|current\s+row)")
         + group(r"\W", r"$"),
-        action=partial(actions.add_node_to_buffer, token_type=TokenType.UNTERM_KEYWORD),
+        action=partial(
+            actions.handle_reserved_keyword,
+            action=partial(
+                actions.add_node_to_buffer, token_type=TokenType.UNTERM_KEYWORD
+            ),
+        ),
     ),
     Rule(
         # BQ arrays use an offset(n) function for
@@ -172,7 +205,12 @@
         name="offset_keyword",
         priority=1310,
         pattern=group(r"offset") + group(r"\s+", r"$"),
-        action=partial(actions.add_node_to_buffer, token_type=TokenType.UNTERM_KEYWORD),
+        action=partial(
+            actions.handle_reserved_keyword,
+            action=partial(
+                actions.add_node_to_buffer, token_type=TokenType.UNTERM_KEYWORD
+            ),
+        ),
     ),
     Rule(
         name="set_operator",
@@ -181,15 +219,18 @@
             r"(union|intersect|except|minus)(\s+(all|distinct))?(\s+by\s+name)?",
         )
         + group(r"\W", r"$"),
-        action=actions.handle_set_operator,
+        action=partial(
+            actions.handle_reserved_keyword,
+            action=actions.handle_set_operator,
+        ),
     ),
     Rule(
         name="explain",
         priority=2000,
         pattern=group(r"explain(\s+(analyze|verbose|using\s+(tabular|json|text)))?")
         + group(r"\W", r"$"),
         action=partial(
-            actions.handle_nonreserved_keyword,
+            actions.handle_nonreserved_top_level_keyword,
             action=partial(
                 actions.add_node_to_buffer, token_type=TokenType.UNTERM_KEYWORD
             ),
@@ -200,7 +241,7 @@
         priority=2010,
         pattern=group(r"grant", r"revoke") + group(r"\W", r"$"),
         action=partial(
-            actions.handle_nonreserved_keyword,
+            actions.handle_nonreserved_top_level_keyword,
             action=partial(actions.lex_ruleset, new_ruleset=GRANT),
         ),
     ),
@@ -209,7 +250,7 @@
         priority=2015,
         pattern=group(CREATE_CLONABLE + r"\s+.+?\s+clone") + group(r"\W", r"$"),
         action=partial(
-            actions.handle_nonreserved_keyword,
+            actions.handle_nonreserved_top_level_keyword,
             action=partial(
                 actions.lex_ruleset,
                 new_ruleset=CLONE,
@@ -221,7 +262,7 @@
         priority=2020,
         pattern=group(CREATE_FUNCTION, ALTER_DROP_FUNCTION) + group(r"\W", r"$"),
         action=partial(
-            actions.handle_nonreserved_keyword,
+            actions.handle_nonreserved_top_level_keyword,
             action=partial(
                 actions.lex_ruleset,
                 new_ruleset=FUNCTION,
@@ -237,7 +278,7 @@
         )
         + group(r"\W", r"$"),
         action=partial(
-            actions.handle_nonreserved_keyword,
+            actions.handle_nonreserved_top_level_keyword,
             action=partial(
                 actions.lex_ruleset,
                 new_ruleset=WAREHOUSE,
@@ -301,7 +342,7 @@
         + r"(?!\()"
         + group(r"\W", r"$"),
         action=partial(
-            actions.handle_nonreserved_keyword,
+            actions.handle_nonreserved_top_level_keyword,
             action=partial(actions.add_node_to_buffer, token_type=TokenType.FMT_OFF),
         ),
     ),

src/sqlfmt/rules/clone.py

@@ -12,7 +12,12 @@
         name="unterm_keyword",
         priority=1300,
         pattern=group(CREATE_CLONABLE, r"clone") + group(r"\W", r"$"),
-        action=partial(actions.add_node_to_buffer, token_type=TokenType.UNTERM_KEYWORD),
+        action=partial(
+            actions.handle_reserved_keyword,
+            action=partial(
+                actions.add_node_to_buffer, token_type=TokenType.UNTERM_KEYWORD
+            ),
+        ),
     ),
     Rule(
         name="word_operator",
@@ -22,6 +27,11 @@
             r"before",
         )
         + group(r"\W", r"$"),
-        action=partial(actions.add_node_to_buffer, token_type=TokenType.WORD_OPERATOR),
+        action=partial(
+            actions.handle_reserved_keyword,
+            action=partial(
+                actions.add_node_to_buffer, token_type=TokenType.WORD_OPERATOR
+            ),
+        ),
     ),
 ]

src/sqlfmt/rules/function.py

@@ -15,7 +15,10 @@
             r"as",
         )
         + group(r"\W", r"$"),
-        action=actions.handle_ddl_as,
+        action=partial(
+            actions.handle_reserved_keyword,
+            action=actions.handle_ddl_as,
+        ),
     ),
     Rule(
         name="word_operator",
@@ -27,7 +30,12 @@
             r"runtime_version",
         )
         + group(r"\W", r"$"),
-        action=partial(actions.add_node_to_buffer, token_type=TokenType.WORD_OPERATOR),
+        action=partial(
+            actions.handle_reserved_keyword,
+            action=partial(
+                actions.add_node_to_buffer, token_type=TokenType.WORD_OPERATOR
+            ),
+        ),
     ),
     Rule(
         name="unterm_keyword",
@@ -85,6 +93,11 @@
             r"(re)?set(\s+all)?",
         )
         + group(r"\W", r"$"),
-        action=partial(actions.add_node_to_buffer, token_type=TokenType.UNTERM_KEYWORD),
+        action=partial(
+            actions.handle_reserved_keyword,
+            action=partial(
+                actions.add_node_to_buffer, token_type=TokenType.UNTERM_KEYWORD
+            ),
+        ),
     ),
 ]

src/sqlfmt/rules/grant.py

@@ -23,6 +23,11 @@
             r"restrict",
         )
         + group(r"\W", r"$"),
-        action=partial(actions.add_node_to_buffer, token_type=TokenType.UNTERM_KEYWORD),
+        action=partial(
+            actions.handle_reserved_keyword,
+            action=partial(
+                actions.add_node_to_buffer, token_type=TokenType.UNTERM_KEYWORD
+            ),
+        ),
     ),
 ]

src/sqlfmt/rules/warehouse.py

@@ -45,6 +45,11 @@
             r"rename\s+to",
         )
         + group(r"\W", r"$"),
-        action=partial(actions.add_node_to_buffer, token_type=TokenType.UNTERM_KEYWORD),
+        action=partial(
+            actions.handle_reserved_keyword,
+            action=partial(
+                actions.add_node_to_buffer, token_type=TokenType.UNTERM_KEYWORD
+            ),
+        ),
     ),
 ]

src/sqlfmt_primer/primer.py

@@ -30,7 +30,7 @@ def get_projects() -> List[SQLProject]:
         SQLProject(
             name="gitlab",
             git_url="https://github.com/tconbeer/gitlab-analytics-sqlfmt.git",
-            git_ref="b1935f4",  # sqlfmt 54b8edd
+            git_ref="5cd49f6",  # sqlfmt aed0f39
             expected_changed=1,
             expected_unchanged=2416,
             expected_errored=0,

tests/data/preformatted/008_reserved_names.sql

@@ -0,0 +1,21 @@
+with
+    "select" as (select * from foo.select),
+    "case" as (select * from foo.case),
+    "end" as (select * from foo.end),
+    "as" as (select * from foo.as),
+    "interval" as (select * from foo.interval),
+    "exclude" as (select * from foo.exclude),
+    "using" as (select * from foo.using),
+    "on" as (select * from foo.on),
+    "and" as (select * from foo.and),
+    "limit" as (select * from foo.limit),
+    "over" as (select * from foo.over),
+    "between" as (select * from foo.between),
+    "union" as (select * from foo.union),
+    "explain" as (select * from foo.explain),
+    "grant" as (select * from foo.grant),
+    "create" as (select * from foo.create),
+    "alter" as (select * from foo.alter),
+    "truncate" as (select * from foo.truncate),
+    "drop" as (select * from foo.drop)
+select 1