Merging upstream version 17.7.0.

Signed-off-by: Daniel Baumann <daniel.baumann@progress-linux.org>

1 files changed, 46 insertions, 18 deletions

diff --git a/sqlglot/parser.py b/sqlglot/parser.py
index 508a273..5adec77 100644
--- a/sqlglot/parser.py
+++ b/sqlglot/parser.py
@@ -327,6 +327,7 @@ class Parser(metaclass=_Parser):
         TokenType.PRIMARY_KEY,
         TokenType.RANGE,
         TokenType.REPLACE,
+        TokenType.RLIKE,
         TokenType.ROW,
         TokenType.UNNEST,
         TokenType.VAR,
@@ -338,6 +339,7 @@ class Parser(metaclass=_Parser):
         TokenType.TIMESTAMP,
         TokenType.TIMESTAMPTZ,
         TokenType.WINDOW,
+        TokenType.XOR,
         *TYPE_TOKENS,
         *SUBQUERY_PREDICATES,
     }
@@ -505,7 +507,6 @@ class Parser(metaclass=_Parser):
         TokenType.DESC: lambda self: self._parse_describe(),
         TokenType.DESCRIBE: lambda self: self._parse_describe(),
         TokenType.DROP: lambda self: self._parse_drop(),
-        TokenType.END: lambda self: self._parse_commit_or_rollback(),
         TokenType.FROM: lambda self: exp.select("*").from_(
             t.cast(exp.From, self._parse_from(skip_from_token=True))
         ),
@@ -716,7 +717,7 @@ class Parser(metaclass=_Parser):
 
     FUNCTIONS_WITH_ALIASED_ARGS = {"STRUCT"}
 
-    FUNCTION_PARSERS: t.Dict[str, t.Callable] = {
+    FUNCTION_PARSERS = {
         "ANY_VALUE": lambda self: self._parse_any_value(),
         "CAST": lambda self: self._parse_cast(self.STRICT_CAST),
         "CONCAT": lambda self: self._parse_concat(),
@@ -1144,6 +1145,7 @@ class Parser(metaclass=_Parser):
 
         return self.expression(
             exp.Drop,
+            comments=start.comments,
             exists=self._parse_exists(),
             this=self._parse_table(schema=True),
             kind=kind,
@@ -1233,11 +1235,14 @@ class Parser(metaclass=_Parser):
             expression = self._parse_ddl_select()
 
             if create_token.token_type == TokenType.TABLE:
+                # exp.Properties.Location.POST_EXPRESSION
+                extend_props(self._parse_properties())
+
                 indexes = []
                 while True:
                     index = self._parse_index()
 
-                    # exp.Properties.Location.POST_EXPRESSION and POST_INDEX
+                    # exp.Properties.Location.POST_INDEX
                     extend_props(self._parse_properties())
 
                     if not index:
@@ -1384,7 +1389,6 @@ class Parser(metaclass=_Parser):
     def _parse_with_property(
         self,
     ) -> t.Optional[exp.Expression] | t.List[t.Optional[exp.Expression]]:
-        self._match(TokenType.WITH)
         if self._match(TokenType.L_PAREN, advance=False):
             return self._parse_wrapped_csv(self._parse_property)
 
@@ -1781,7 +1785,17 @@ class Parser(metaclass=_Parser):
             return None
 
         if self._match_text_seq("SERDE"):
-            return self.expression(exp.RowFormatSerdeProperty, this=self._parse_string())
+            this = self._parse_string()
+
+            serde_properties = None
+            if self._match(TokenType.SERDE_PROPERTIES):
+                serde_properties = self.expression(
+                    exp.SerdeProperties, expressions=self._parse_wrapped_csv(self._parse_property)
+                )
+
+            return self.expression(
+                exp.RowFormatSerdeProperty, this=this, serde_properties=serde_properties
+            )
 
         self._match_text_seq("DELIMITED")
 
@@ -2251,7 +2265,9 @@ class Parser(metaclass=_Parser):
             self._match_set(self.JOIN_KINDS) and self._prev,
         )
 
-    def _parse_join(self, skip_join_token: bool = False) -> t.Optional[exp.Join]:
+    def _parse_join(
+        self, skip_join_token: bool = False, parse_bracket: bool = False
+    ) -> t.Optional[exp.Join]:
         if self._match(TokenType.COMMA):
             return self.expression(exp.Join, this=self._parse_table())
 
@@ -2275,7 +2291,7 @@ class Parser(metaclass=_Parser):
         if outer_apply:
             side = Token(TokenType.LEFT, "LEFT")
 
-        kwargs: t.Dict[str, t.Any] = {"this": self._parse_table()}
+        kwargs: t.Dict[str, t.Any] = {"this": self._parse_table(parse_bracket=parse_bracket)}
 
         if method:
             kwargs["method"] = method.text
@@ -2411,6 +2427,7 @@ class Parser(metaclass=_Parser):
         schema: bool = False,
         joins: bool = False,
         alias_tokens: t.Optional[t.Collection[TokenType]] = None,
+        parse_bracket: bool = False,
     ) -> t.Optional[exp.Expression]:
         lateral = self._parse_lateral()
         if lateral:
@@ -2430,7 +2447,9 @@ class Parser(metaclass=_Parser):
                 subquery.set("pivots", self._parse_pivots())
             return subquery
 
-        this: exp.Expression = self._parse_table_parts(schema=schema)
+        bracket = parse_bracket and self._parse_bracket(None)
+        bracket = self.expression(exp.Table, this=bracket) if bracket else None
+        this: exp.Expression = bracket or self._parse_table_parts(schema=schema)
 
         if schema:
             return self._parse_schema(this=this)
@@ -2758,8 +2777,15 @@ class Parser(metaclass=_Parser):
         self, this: t.Optional[exp.Expression] = None, top: bool = False
     ) -> t.Optional[exp.Expression]:
         if self._match(TokenType.TOP if top else TokenType.LIMIT):
-            limit_paren = self._match(TokenType.L_PAREN)
-            expression = self._parse_number() if top else self._parse_term()
+            comments = self._prev_comments
+            if top:
+                limit_paren = self._match(TokenType.L_PAREN)
+                expression = self._parse_number()
+
+                if limit_paren:
+                    self._match_r_paren()
+            else:
+                expression = self._parse_term()
 
             if self._match(TokenType.COMMA):
                 offset = expression
@@ -2767,10 +2793,9 @@ class Parser(metaclass=_Parser):
             else:
                 offset = None
 
-            limit_exp = self.expression(exp.Limit, this=this, expression=expression, offset=offset)
-
-            if limit_paren:
-                self._match_r_paren()
+            limit_exp = self.expression(
+                exp.Limit, this=this, expression=expression, offset=offset, comments=comments
+            )
 
             return limit_exp
 
@@ -2803,7 +2828,7 @@ class Parser(metaclass=_Parser):
         if not self._match(TokenType.OFFSET):
             return this
 
-        count = self._parse_number()
+        count = self._parse_term()
         self._match_set((TokenType.ROW, TokenType.ROWS))
         return self.expression(exp.Offset, this=this, expression=count)
 
@@ -3320,7 +3345,7 @@ class Parser(metaclass=_Parser):
             else:
                 this = self.expression(exp.Anonymous, this=this, expressions=args)
 
-        self._match_r_paren(this)
+        self._match(TokenType.R_PAREN, expression=this)
         return self._parse_window(this)
 
     def _parse_function_parameter(self) -> t.Optional[exp.Expression]:
@@ -4076,7 +4101,10 @@ class Parser(metaclass=_Parser):
         self, this: t.Optional[exp.Expression], alias: bool = False
     ) -> t.Optional[exp.Expression]:
         if self._match_pair(TokenType.FILTER, TokenType.L_PAREN):
-            this = self.expression(exp.Filter, this=this, expression=self._parse_where())
+            self._match(TokenType.WHERE)
+            this = self.expression(
+                exp.Filter, this=this, expression=self._parse_where(skip_where_token=True)
+            )
             self._match_r_paren()
 
         # T-SQL allows the OVER (...) syntax after WITHIN GROUP.
@@ -4351,7 +4379,7 @@ class Parser(metaclass=_Parser):
             self._parse_set_operations(self._parse_select(nested=True, parse_subquery_alias=False))
         )
 
-    def _parse_transaction(self) -> exp.Transaction:
+    def _parse_transaction(self) -> exp.Transaction | exp.Command:
         this = None
         if self._match_texts(self.TRANSACTION_KIND):
             this = self._prev.text