parser: fix unicode escaped quoted idents

crates/squawk_lexer/src/lib.rs

@@ -222,8 +222,7 @@ impl Cursor<'_> {
             '"' if allows_double => {
                 self.bump();
                 let terminated = self.double_quoted_string();
-                let kind = mk_kind(terminated);
-                TokenKind::Literal { kind }
+                TokenKind::QuotedIdent { terminated }
             }
             _ => self.ident_or_unknown_prefix(),
         }

crates/squawk_lexer/src/snapshots/squawk_lexer__tests__string_unicode_escape.snap

@@ -4,13 +4,13 @@ expression: "lex(r#\"\nU&\"d\\0061t\\+000061\"\n\nU&\"\\0441\\043B\\043E\\043D\"
 ---
 [
     "\n" @ Whitespace,
-    "U&\"d\\0061t\\+000061\"" @ Literal { kind: UnicodeEscStr { terminated: true } },
+    "U&\"d\\0061t\\+000061\"" @ QuotedIdent { terminated: true },
     "\n\n" @ Whitespace,
-    "U&\"\\0441\\043B\\043E\\043D\"" @ Literal { kind: UnicodeEscStr { terminated: true } },
+    "U&\"\\0441\\043B\\043E\\043D\"" @ QuotedIdent { terminated: true },
     "\n\n" @ Whitespace,
     "u&'\\0441\\043B'" @ Literal { kind: UnicodeEscStr { terminated: true } },
     "\n\n" @ Whitespace,
-    "U&\"d!0061t!+000061\"" @ Literal { kind: UnicodeEscStr { terminated: true } },
+    "U&\"d!0061t!+000061\"" @ QuotedIdent { terminated: true },
     " " @ Whitespace,
     "UESCAPE" @ Ident,
     " " @ Whitespace,

crates/squawk_parser/src/grammar.rs

@@ -1845,7 +1845,8 @@ fn name_ref_(p: &mut Parser<'_>) -> Option<CompletedMarker> {
         NAME_REF
     };
     let cm = m.complete(p, if p.at(STRING) { kind } else { NAME_REF });
-    // A path followed by a string is a type cast so we insert a CAST_EXPR
+
+    // A type name followed by a string is a type cast so we insert a CAST_EXPR
     // preceding it to wrap the previously parsed data.
     // e.g., `select numeric '12312'`
     if opt_string_literal(p).is_some() {
@@ -2808,7 +2809,6 @@ fn data_source(p: &mut Parser<'_>) {
             opt_alias(p);
         }
         _ if p.at_ts(FROM_ITEM_KEYWORDS_FIRST) => from_item_name(p),
-
         _ => {
             p.error("expected table reference");
         }
@@ -11308,7 +11308,7 @@ fn vacuum(p: &mut Parser<'_>) -> CompletedMarker {
 //  table_name [ ( column_name [, ...] ) ]
 fn opt_relation_list(p: &mut Parser<'_>) {
     while !p.at(EOF) {
-        if opt_path_name_ref(p).is_none() {
+        if opt_relation_name(p).is_none() {
             break;
         }
         opt_column_list(p);
@@ -12990,6 +12990,12 @@ const NON_RESERVED_WORD: TokenSet = TokenSet::new(&[IDENT])
     .union(TYPE_FUNC_NAME_KEYWORDS);
 
 fn relation_name(p: &mut Parser<'_>) {
+    if opt_relation_name(p).is_none() {
+        p.error("expected relation name");
+    }
+}
+
+fn opt_relation_name(p: &mut Parser<'_>) -> Option<CompletedMarker> {
     let m = p.start();
     if p.eat(ONLY_KW) {
         let trailing_paren = p.eat(L_PAREN);
@@ -12999,10 +13005,13 @@ fn relation_name(p: &mut Parser<'_>) {
             p.expect(R_PAREN);
         }
     } else {
-        path_name_ref(p);
+        if opt_path_name_ref(p).is_none() {
+            m.abandon(p);
+            return None;
+        };
         p.eat(STAR);
     }
-    m.complete(p, RELATION_NAME);
+    Some(m.complete(p, RELATION_NAME))
 }
 
 // ALTER TABLE [ IF EXISTS ] [ ONLY ] name [ * ]

crates/squawk_parser/tests/data/regression_suite/strings.sql

@@ -10,11 +10,11 @@ SELECT 'first line'
 	' - third line'
 	AS "Three lines to one";
 
--- illegal string continuation syntax
-SELECT 'first line'
-' - next line' /* this comment is not allowed here */
-' - third line'
-	AS "Illegal comment within continuation";
+-- -- illegal string continuation syntax
+-- SELECT 'first line'
+-- ' - next line' /* this comment is not allowed here */
+-- ' - third line'
+-- 	AS "Illegal comment within continuation";
 
 -- Unicode escapes
 SET standard_conforming_strings TO on;
@@ -812,16 +812,16 @@ select 'a\bcd' as f1, 'a\b''cd' as f2, 'a\b''''cd' as f3, 'abcd\'   as f4, 'ab\'
 
 set standard_conforming_strings = off;
 
-select 'a\\bcd' as f1, 'a\\b\'cd' as f2, 'a\\b\'''cd' as f3, 'abcd\\'   as f4, 'ab\\\'cd' as f5, '\\\\' as f6;
+-- select 'a\\bcd' as f1, 'a\\b\'cd' as f2, 'a\\b\'''cd' as f3, 'abcd\\'   as f4, 'ab\\\'cd' as f5, '\\\\' as f6;
 
 set escape_string_warning = off;
 set standard_conforming_strings = on;
 
-select 'a\bcd' as f1, 'a\b''cd' as f2, 'a\b''''cd' as f3, 'abcd\'   as f4, 'ab\''cd' as f5, '\\' as f6;
+-- select 'a\bcd' as f1, 'a\b''cd' as f2, 'a\b''''cd' as f3, 'abcd\'   as f4, 'ab\''cd' as f5, '\\' as f6;
 
 set standard_conforming_strings = off;
 
-select 'a\\bcd' as f1, 'a\\b\'cd' as f2, 'a\\b\'''cd' as f3, 'abcd\\'   as f4, 'ab\\\'cd' as f5, '\\\\' as f6;
+-- select 'a\\bcd' as f1, 'a\\b\'cd' as f2, 'a\\b\'''cd' as f3, 'abcd\\'   as f4, 'ab\\\'cd' as f5, '\\\\' as f6;
 
 reset standard_conforming_strings;
 

crates/squawk_parser/tests/data/regression_suite/triggers.sql

@@ -243,24 +243,24 @@ UPDATE some_t SET some_col = TRUE;
 DROP TABLE some_t;
 
 -- bogus cases
-CREATE TRIGGER error_upd_and_col BEFORE UPDATE OR UPDATE OF a ON main_table
-FOR EACH ROW EXECUTE PROCEDURE trigger_func('error_upd_and_col');
-CREATE TRIGGER error_upd_a_a BEFORE UPDATE OF a, a ON main_table
-FOR EACH ROW EXECUTE PROCEDURE trigger_func('error_upd_a_a');
-CREATE TRIGGER error_ins_a BEFORE INSERT OF a ON main_table
-FOR EACH ROW EXECUTE PROCEDURE trigger_func('error_ins_a');
-CREATE TRIGGER error_ins_when BEFORE INSERT OR UPDATE ON main_table
-FOR EACH ROW WHEN (OLD.a <> NEW.a)
-EXECUTE PROCEDURE trigger_func('error_ins_old');
-CREATE TRIGGER error_del_when BEFORE DELETE OR UPDATE ON main_table
-FOR EACH ROW WHEN (OLD.a <> NEW.a)
-EXECUTE PROCEDURE trigger_func('error_del_new');
-CREATE TRIGGER error_del_when BEFORE INSERT OR UPDATE ON main_table
-FOR EACH ROW WHEN (NEW.tableoid <> 0)
-EXECUTE PROCEDURE trigger_func('error_when_sys_column');
-CREATE TRIGGER error_stmt_when BEFORE UPDATE OF a ON main_table
-FOR EACH STATEMENT WHEN (OLD.* IS DISTINCT FROM NEW.*)
-EXECUTE PROCEDURE trigger_func('error_stmt_when');
+-- CREATE TRIGGER error_upd_and_col BEFORE UPDATE OR UPDATE OF a ON main_table
+-- FOR EACH ROW EXECUTE PROCEDURE trigger_func('error_upd_and_col');
+-- CREATE TRIGGER error_upd_a_a BEFORE UPDATE OF a, a ON main_table
+-- FOR EACH ROW EXECUTE PROCEDURE trigger_func('error_upd_a_a');
+-- CREATE TRIGGER error_ins_a BEFORE INSERT OF a ON main_table
+-- FOR EACH ROW EXECUTE PROCEDURE trigger_func('error_ins_a');
+-- CREATE TRIGGER error_ins_when BEFORE INSERT OR UPDATE ON main_table
+-- FOR EACH ROW WHEN (OLD.a <> NEW.a)
+-- EXECUTE PROCEDURE trigger_func('error_ins_old');
+-- CREATE TRIGGER error_del_when BEFORE DELETE OR UPDATE ON main_table
+-- FOR EACH ROW WHEN (OLD.a <> NEW.a)
+-- EXECUTE PROCEDURE trigger_func('error_del_new');
+-- CREATE TRIGGER error_del_when BEFORE INSERT OR UPDATE ON main_table
+-- FOR EACH ROW WHEN (NEW.tableoid <> 0)
+-- EXECUTE PROCEDURE trigger_func('error_when_sys_column');
+-- CREATE TRIGGER error_stmt_when BEFORE UPDATE OF a ON main_table
+-- FOR EACH STATEMENT WHEN (OLD.* IS DISTINCT FROM NEW.*)
+-- EXECUTE PROCEDURE trigger_func('error_stmt_when');
 
 -- check dependency restrictions
 ALTER TABLE main_table DROP COLUMN b;
@@ -1335,12 +1335,12 @@ delete from parted_stmt_trig;
 
 -- insert via copy on the parent
 copy parted_stmt_trig(a) from stdin;
-1
-2
+-- 1
+-- 2
 
 -- insert via copy on the first partition
 copy parted_stmt_trig1(a) from stdin;
-1
+-- 1
 
 -- Disabling a trigger in the parent table should disable children triggers too
 alter table parted_stmt_trig disable trigger trig_ins_after_parent;
@@ -1882,9 +1882,9 @@ delete from child3;
 
 -- copy into parent sees parent-format tuples
 copy parent (a, b) from stdin;
-AAA	42
-BBB	42
-CCC	42
+-- AAA	42
+-- BBB	42
+-- CCC	42
 
 -- DML affecting parent sees tuples collected from children even if
 -- there is no transition table trigger on the children
@@ -1902,9 +1902,9 @@ delete from parent;
 -- copy into parent sees tuples collected from children even if there
 -- is no transition-table trigger on the children
 copy parent (a, b) from stdin;
-AAA	42
-BBB	42
-CCC	42
+-- AAA	42
+-- BBB	42
+-- CCC	42
 
 -- insert into parent with a before trigger on a child tuple before
 -- insertion, and we capture the newly modified row in parent format
@@ -1926,9 +1926,9 @@ insert into parent values ('AAA', 42), ('BBB', 42), ('CCC', 66);
 
 -- copy, parent trigger sees post-modification parent-format tuple
 copy parent (a, b) from stdin;
-AAA	42
-BBB	42
-CCC	234
+-- AAA	42
+-- BBB	42
+-- CCC	234
 
 drop table child1, child2, child3, parent;
 drop function intercept_insert();
@@ -2091,15 +2091,15 @@ delete from child3;
 -- copy into parent sees parent-format tuples (no rerouting, so these
 -- are really inserted into the parent)
 copy parent (a, b) from stdin;
-AAA	42
-BBB	42
-CCC	42
+-- AAA	42
+-- BBB	42
+-- CCC	42
 
 -- same behavior for copy if there is an index (interesting because rows are
 -- captured by a different code path in copyfrom.c if there are indexes)
 create index on parent(b);
 copy parent (a, b) from stdin;
-DDD	42
+-- DDD	42
 
 -- DML affecting parent sees tuples collected from children even if
 -- there is no transition table trigger on the children
@@ -2628,7 +2628,7 @@ select tgrelid::regclass, tgname,
 (select tgname from pg_trigger tr where tr.oid = pg_trigger.tgparentid) parent_tgname
 from pg_trigger where tgrelid in (select relid from pg_partition_tree('grandparent'))
 order by tgname, tgrelid::regclass::text COLLATE "C";
-alter trigger a on only grandparent rename to b;	-- ONLY not supported
+-- alter trigger a on only grandparent rename to b;	-- ONLY not supported
 alter trigger b on middle rename to c;	-- can't rename trigger on partition
 create trigger c after insert on middle
 for each row execute procedure f();

crates/squawk_parser/tests/data/regression_suite/tsearch.sql

@@ -645,12 +645,12 @@ SELECT ts_headline('english',
 --Rewrite sub system
 
 CREATE TABLE test_tsquery (txtkeyword TEXT, txtsample TEXT);
-'New York'	new <-> york | big <-> apple | nyc
-Moscow	moskva | moscow
-'Sanct Peter'	Peterburg | peter | 'Sanct Peterburg'
-foo & bar & qq	foo & (bar | qq) & city
-1 & (2 <-> 3)	2 <-> 4
-5 <-> 6	5 <-> 7
+-- 'New York'	new <-> york | big <-> apple | nyc
+-- Moscow	moskva | moscow
+-- 'Sanct Peter'	Peterburg | peter | 'Sanct Peterburg'
+-- foo & bar & qq	foo & (bar | qq) & city
+-- 1 & (2 <-> 3)	2 <-> 4
+-- 5 <-> 6	5 <-> 7
 
 ALTER TABLE test_tsquery ADD COLUMN keyword tsquery;
 UPDATE test_tsquery SET keyword = to_tsquery('english', txtkeyword);

crates/squawk_parser/tests/snapshots/tests__analyze_ok.snap

@@ -20,26 +20,28 @@ SOURCE_FILE
     WHITESPACE " "
     VERBOSE_KW "verbose"
     WHITESPACE " "
-    PATH
+    RELATION_NAME
       PATH
+        PATH
+          PATH_SEGMENT
+            NAME_REF
+              IDENT "foo"
+        DOT "."
         PATH_SEGMENT
           NAME_REF
-            IDENT "foo"
-      DOT "."
-      PATH_SEGMENT
-        NAME_REF
-          IDENT "bar"
+            IDENT "bar"
     COMMA ","
     WHITESPACE " "
-    PATH
+    RELATION_NAME
       PATH
+        PATH
+          PATH_SEGMENT
+            NAME_REF
+              IDENT "foo"
+        DOT "."
         PATH_SEGMENT
           NAME_REF
-            IDENT "foo"
-      DOT "."
-      PATH_SEGMENT
-        NAME_REF
-          IDENT "bar"
+            IDENT "bar"
     COLUMN_LIST
       L_PAREN "("
       COLUMN
@@ -58,10 +60,11 @@ SOURCE_FILE
       R_PAREN ")"
     COMMA ","
     WHITESPACE " "
-    PATH
-      PATH_SEGMENT
-        NAME_REF
-          IDENT "foo"
+    RELATION_NAME
+      PATH
+        PATH_SEGMENT
+          NAME_REF
+            IDENT "foo"
   SEMICOLON ";"
   WHITESPACE "\n\n"
   COMMENT "-- full_parens"
@@ -85,15 +88,16 @@ SOURCE_FILE
       INT_NUMBER "10"
     R_PAREN ")"
     WHITESPACE " "
-    PATH
+    RELATION_NAME
       PATH
+        PATH
+          PATH_SEGMENT
+            NAME_REF
+              IDENT "foo"
+        DOT "."
         PATH_SEGMENT
           NAME_REF
-            IDENT "foo"
-      DOT "."
-      PATH_SEGMENT
-        NAME_REF
-          IDENT "bar"
+            IDENT "bar"
     COLUMN_LIST
       L_PAREN "("
       COLUMN

crates/squawk_parser/tests/snapshots/tests__misc_ok.snap

@@ -1819,15 +1819,16 @@ SOURCE_FILE
     WHITESPACE " "
     ANALYZE_KW "ANALYZE"
     WHITESPACE " "
-    PATH
+    RELATION_NAME
       PATH
+        PATH
+          PATH_SEGMENT
+            NAME_REF
+              IDENT "partman_test"
+        DOT "."
         PATH_SEGMENT
           NAME_REF
-            IDENT "partman_test"
-      DOT "."
-      PATH_SEGMENT
-        NAME_REF
-          IDENT "time_taptest_table"
+            IDENT "time_taptest_table"
   SEMICOLON ";"
   WHITESPACE "\n\n"
   SELECT

crates/squawk_parser/tests/snapshots/tests__regression_errors.snap

@@ -2,7 +2,7 @@
 source: crates/squawk_parser/tests/tests.rs
 input_file: crates/squawk_parser/tests/data/regression_suite/errors.sql
 ---
-ERROR@948: expected path name
+ERROR@948: expected relation name
 ERROR@1074: expected path name
 ERROR@2188: expected path name
 ERROR@2219: expected path name

crates/squawk_parser/tests/snapshots/tests__regression_merge.snap

@@ -122,7 +122,7 @@ ERROR@41835: expected command, found WHEN_KW
 ERROR@41840: expected command, found MATCHED_KW
 ERROR@41848: expected command, found THEN_KW
 ERROR@41860: expected FROM_KW
-ERROR@41860: expected path name
+ERROR@41860: expected relation name
 ERROR@41953: expected ON_KW
 ERROR@41960: expected WHEN_KW
 ERROR@41960: expected MATCHED, or NOT MATCHED