Merging upstream version 23.16.0.

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

1 files changed, 257 insertions, 57 deletions

diff --git a/sqlglot/parser.py b/sqlglot/parser.py
index b0a2516..67ffd8f 100644
--- a/sqlglot/parser.py
+++ b/sqlglot/parser.py
@@ -61,6 +61,23 @@ def build_logarithm(args: t.List, dialect: Dialect) -> exp.Func:
     return (exp.Ln if dialect.parser_class.LOG_DEFAULTS_TO_LN else exp.Log)(this=this)
 
 
+def build_hex(args: t.List, dialect: Dialect) -> exp.Hex | exp.LowerHex:
+    arg = seq_get(args, 0)
+    return exp.LowerHex(this=arg) if dialect.HEX_LOWERCASE else exp.Hex(this=arg)
+
+
+def build_lower(args: t.List) -> exp.Lower | exp.Hex:
+    # LOWER(HEX(..)) can be simplified to LowerHex to simplify its transpilation
+    arg = seq_get(args, 0)
+    return exp.LowerHex(this=arg.this) if isinstance(arg, exp.Hex) else exp.Lower(this=arg)
+
+
+def build_upper(args: t.List) -> exp.Upper | exp.Hex:
+    # UPPER(HEX(..)) can be simplified to Hex to simplify its transpilation
+    arg = seq_get(args, 0)
+    return exp.Hex(this=arg.this) if isinstance(arg, exp.Hex) else exp.Upper(this=arg)
+
+
 def build_extract_json_with_path(expr_type: t.Type[E]) -> t.Callable[[t.List, Dialect], E]:
     def _builder(args: t.List, dialect: Dialect) -> E:
         expression = expr_type(
@@ -74,6 +91,17 @@ def build_extract_json_with_path(expr_type: t.Type[E]) -> t.Callable[[t.List, Di
     return _builder
 
 
+def build_mod(args: t.List) -> exp.Mod:
+    this = seq_get(args, 0)
+    expression = seq_get(args, 1)
+
+    # Wrap the operands if they are binary nodes, e.g. MOD(a + 1, 7) -> (a + 1) % 7
+    this = exp.Paren(this=this) if isinstance(this, exp.Binary) else this
+    expression = exp.Paren(this=expression) if isinstance(expression, exp.Binary) else expression
+
+    return exp.Mod(this=this, expression=expression)
+
+
 class _Parser(type):
     def __new__(cls, clsname, bases, attrs):
         klass = super().__new__(cls, clsname, bases, attrs)
@@ -123,7 +151,7 @@ class Parser(metaclass=_Parser):
         "LOG": build_logarithm,
         "LOG2": lambda args: exp.Log(this=exp.Literal.number(2), expression=seq_get(args, 0)),
         "LOG10": lambda args: exp.Log(this=exp.Literal.number(10), expression=seq_get(args, 0)),
-        "MOD": lambda args: exp.Mod(this=seq_get(args, 0), expression=seq_get(args, 1)),
+        "MOD": build_mod,
         "TIME_TO_TIME_STR": lambda args: exp.Cast(
             this=seq_get(args, 0),
             to=exp.DataType(this=exp.DataType.Type.TEXT),
@@ -137,6 +165,10 @@ class Parser(metaclass=_Parser):
             length=exp.Literal.number(10),
         ),
         "VAR_MAP": build_var_map,
+        "LOWER": build_lower,
+        "UPPER": build_upper,
+        "HEX": build_hex,
+        "TO_HEX": build_hex,
     }
 
     NO_PAREN_FUNCTIONS = {
@@ -295,22 +327,23 @@ class Parser(metaclass=_Parser):
 
     DB_CREATABLES = {
         TokenType.DATABASE,
-        TokenType.SCHEMA,
-        TokenType.TABLE,
-        TokenType.VIEW,
-        TokenType.MODEL,
         TokenType.DICTIONARY,
+        TokenType.MODEL,
+        TokenType.SCHEMA,
         TokenType.SEQUENCE,
         TokenType.STORAGE_INTEGRATION,
+        TokenType.TABLE,
+        TokenType.TAG,
+        TokenType.VIEW,
     }
 
     CREATABLES = {
         TokenType.COLUMN,
         TokenType.CONSTRAINT,
+        TokenType.FOREIGN_KEY,
         TokenType.FUNCTION,
         TokenType.INDEX,
         TokenType.PROCEDURE,
-        TokenType.FOREIGN_KEY,
         *DB_CREATABLES,
     }
 
@@ -373,6 +406,7 @@ class Parser(metaclass=_Parser):
         TokenType.REFRESH,
         TokenType.REPLACE,
         TokenType.RIGHT,
+        TokenType.ROLLUP,
         TokenType.ROW,
         TokenType.ROWS,
         TokenType.SEMI,
@@ -467,7 +501,6 @@ class Parser(metaclass=_Parser):
     }
 
     EQUALITY = {
-        TokenType.COLON_EQ: exp.PropertyEQ,
         TokenType.EQ: exp.EQ,
         TokenType.NEQ: exp.NEQ,
         TokenType.NULLSAFE_EQ: exp.NullSafeEQ,
@@ -653,6 +686,7 @@ class Parser(metaclass=_Parser):
             kind=self._parse_var_from_options(self.USABLES, raise_unmatched=False),
             this=self._parse_table(schema=False),
         ),
+        TokenType.SEMICOLON: lambda self: self.expression(exp.Semicolon),
     }
 
     UNARY_PARSERS = {
@@ -700,7 +734,12 @@ class Parser(metaclass=_Parser):
         TokenType.FALSE: lambda self, _: self.expression(exp.Boolean, this=False),
         TokenType.SESSION_PARAMETER: lambda self, _: self._parse_session_parameter(),
         TokenType.STAR: lambda self, _: self.expression(
-            exp.Star, **{"except": self._parse_except(), "replace": self._parse_replace()}
+            exp.Star,
+            **{
+                "except": self._parse_star_op("EXCEPT", "EXCLUDE"),
+                "replace": self._parse_star_op("REPLACE"),
+                "rename": self._parse_star_op("RENAME"),
+            },
         ),
     }
 
@@ -729,6 +768,9 @@ class Parser(metaclass=_Parser):
     }
 
     PROPERTY_PARSERS: t.Dict[str, t.Callable] = {
+        "ALLOWED_VALUES": lambda self: self.expression(
+            exp.AllowedValuesProperty, expressions=self._parse_csv(self._parse_primary)
+        ),
         "ALGORITHM": lambda self: self._parse_property_assignment(exp.AlgorithmProperty),
         "AUTO": lambda self: self._parse_auto_property(),
         "AUTO_INCREMENT": lambda self: self._parse_property_assignment(exp.AutoIncrementProperty),
@@ -748,6 +790,7 @@ class Parser(metaclass=_Parser):
         "CONTAINS": lambda self: self._parse_contains_property(),
         "COPY": lambda self: self._parse_copy_property(),
         "DATABLOCKSIZE": lambda self, **kwargs: self._parse_datablocksize(**kwargs),
+        "DATA_DELETION": lambda self: self._parse_data_deletion_property(),
         "DEFINER": lambda self: self._parse_definer(),
         "DETERMINISTIC": lambda self: self.expression(
             exp.StabilityProperty, this=exp.Literal.string("IMMUTABLE")
@@ -796,6 +839,7 @@ class Parser(metaclass=_Parser):
         "READS": lambda self: self._parse_reads_property(),
         "REMOTE": lambda self: self._parse_remote_with_connection(),
         "RETURNS": lambda self: self._parse_returns(),
+        "STRICT": lambda self: self.expression(exp.StrictProperty),
         "ROW": lambda self: self._parse_row(),
         "ROW_FORMAT": lambda self: self._parse_property_assignment(exp.RowFormatProperty),
         "SAMPLE": lambda self: self.expression(
@@ -900,6 +944,14 @@ class Parser(metaclass=_Parser):
         "DELETE": lambda self: self.expression(exp.Delete, where=self._parse_where()),
         "DROP": lambda self: self._parse_alter_table_drop(),
         "RENAME": lambda self: self._parse_alter_table_rename(),
+        "SET": lambda self: self._parse_alter_table_set(),
+    }
+
+    ALTER_ALTER_PARSERS = {
+        "DISTKEY": lambda self: self._parse_alter_diststyle(),
+        "DISTSTYLE": lambda self: self._parse_alter_diststyle(),
+        "SORTKEY": lambda self: self._parse_alter_sortkey(),
+        "COMPOUND": lambda self: self._parse_alter_sortkey(compound=True),
     }
 
     SCHEMA_UNNAMED_CONSTRAINTS = {
@@ -990,6 +1042,8 @@ class Parser(metaclass=_Parser):
         exp.DataType.Type.JSON: lambda self, this, _: self.expression(exp.ParseJSON, this=this),
     }
 
+    TYPE_CONVERTER: t.Dict[exp.DataType.Type, t.Callable[[exp.DataType], exp.DataType]] = {}
+
     DDL_SELECT_TOKENS = {TokenType.SELECT, TokenType.WITH, TokenType.L_PAREN}
 
     PRE_VOLATILE_TOKENS = {TokenType.CREATE, TokenType.REPLACE, TokenType.UNIQUE}
@@ -1084,6 +1138,9 @@ class Parser(metaclass=_Parser):
     # Whether the table sample clause expects CSV syntax
     TABLESAMPLE_CSV = False
 
+    # The default method used for table sampling
+    DEFAULT_SAMPLING_METHOD: t.Optional[str] = None
+
     # Whether the SET command needs a delimiter (e.g. "=") for assignments
     SET_REQUIRES_ASSIGNMENT_DELIMITER = True
 
@@ -1228,6 +1285,9 @@ class Parser(metaclass=_Parser):
 
         for i, token in enumerate(raw_tokens):
             if token.token_type == TokenType.SEMICOLON:
+                if token.comments:
+                    chunks.append([token])
+
                 if i < total - 1:
                     chunks.append([])
             else:
@@ -1471,7 +1531,7 @@ class Parser(metaclass=_Parser):
         if self._match_set(self.STATEMENT_PARSERS):
             return self.STATEMENT_PARSERS[self._prev.token_type](self)
 
-        if self._match_set(Tokenizer.COMMANDS):
+        if self._match_set(self.dialect.tokenizer.COMMANDS):
             return self._parse_command()
 
         expression = self._parse_expression()
@@ -1492,6 +1552,8 @@ class Parser(metaclass=_Parser):
             schema=True, is_db_reference=self._prev.token_type == TokenType.SCHEMA
         )
 
+        cluster = self._parse_on_property() if self._match(TokenType.ON) else None
+
         if self._match(TokenType.L_PAREN, advance=False):
             expressions = self._parse_wrapped_csv(self._parse_types)
         else:
@@ -1503,12 +1565,13 @@ class Parser(metaclass=_Parser):
             exists=if_exists,
             this=table,
             expressions=expressions,
-            kind=kind,
+            kind=kind.upper(),
             temporary=temporary,
             materialized=materialized,
             cascade=self._match_text_seq("CASCADE"),
             constraints=self._match_text_seq("CONSTRAINTS"),
             purge=self._match_text_seq("PURGE"),
+            cluster=cluster,
         )
 
     def _parse_exists(self, not_: bool = False) -> t.Optional[bool]:
@@ -1651,7 +1714,7 @@ class Parser(metaclass=_Parser):
                     exp.Clone, this=self._parse_table(schema=True), shallow=shallow, copy=copy
                 )
 
-        if self._curr:
+        if self._curr and not self._match_set((TokenType.R_PAREN, TokenType.COMMA), advance=False):
             return self._parse_as_command(start)
 
         return self.expression(
@@ -1678,6 +1741,7 @@ class Parser(metaclass=_Parser):
         index = self._index
 
         while self._curr:
+            self._match(TokenType.COMMA)
             if self._match_text_seq("INCREMENT"):
                 self._match_text_seq("BY")
                 self._match_text_seq("=")
@@ -1822,23 +1886,65 @@ class Parser(metaclass=_Parser):
 
         return self.expression(exp.StabilityProperty, this=exp.Literal.string("VOLATILE"))
 
-    def _parse_system_versioning_property(self) -> exp.WithSystemVersioningProperty:
-        self._match_pair(TokenType.EQ, TokenType.ON)
+    def _parse_retention_period(self) -> exp.Var:
+        # Parse TSQL's HISTORY_RETENTION_PERIOD: {INFINITE | <number> DAY | DAYS | MONTH ...}
+        number = self._parse_number()
+        number_str = f"{number} " if number else ""
+        unit = self._parse_var(any_token=True)
+        return exp.var(f"{number_str}{unit}")
+
+    def _parse_system_versioning_property(
+        self, with_: bool = False
+    ) -> exp.WithSystemVersioningProperty:
+        self._match(TokenType.EQ)
+        prop = self.expression(
+            exp.WithSystemVersioningProperty,
+            **{  # type: ignore
+                "on": True,
+                "with": with_,
+            },
+        )
+
+        if self._match_text_seq("OFF"):
+            prop.set("on", False)
+            return prop
 
-        prop = self.expression(exp.WithSystemVersioningProperty)
+        self._match(TokenType.ON)
         if self._match(TokenType.L_PAREN):
-            self._match_text_seq("HISTORY_TABLE", "=")
-            prop.set("this", self._parse_table_parts())
+            while self._curr and not self._match(TokenType.R_PAREN):
+                if self._match_text_seq("HISTORY_TABLE", "="):
+                    prop.set("this", self._parse_table_parts())
+                elif self._match_text_seq("DATA_CONSISTENCY_CHECK", "="):
+                    prop.set("data_consistency", self._advance_any() and self._prev.text.upper())
+                elif self._match_text_seq("HISTORY_RETENTION_PERIOD", "="):
+                    prop.set("retention_period", self._parse_retention_period())
 
-            if self._match(TokenType.COMMA):
-                self._match_text_seq("DATA_CONSISTENCY_CHECK", "=")
-                prop.set("expression", self._advance_any() and self._prev.text.upper())
+                self._match(TokenType.COMMA)
 
-            self._match_r_paren()
+        return prop
+
+    def _parse_data_deletion_property(self) -> exp.DataDeletionProperty:
+        self._match(TokenType.EQ)
+        on = self._match_text_seq("ON") or not self._match_text_seq("OFF")
+        prop = self.expression(exp.DataDeletionProperty, on=on)
+
+        if self._match(TokenType.L_PAREN):
+            while self._curr and not self._match(TokenType.R_PAREN):
+                if self._match_text_seq("FILTER_COLUMN", "="):
+                    prop.set("filter_column", self._parse_column())
+                elif self._match_text_seq("RETENTION_PERIOD", "="):
+                    prop.set("retention_period", self._parse_retention_period())
+
+                self._match(TokenType.COMMA)
 
         return prop
 
     def _parse_with_property(self) -> t.Optional[exp.Expression] | t.List[exp.Expression]:
+        if self._match_text_seq("(", "SYSTEM_VERSIONING"):
+            prop = self._parse_system_versioning_property(with_=True)
+            self._match_r_paren()
+            return prop
+
         if self._match(TokenType.L_PAREN, advance=False):
             return self._parse_wrapped_properties()
 
@@ -1853,6 +1959,9 @@ class Parser(metaclass=_Parser):
         elif self._match_text_seq("NO", "DATA"):
             return self._parse_withdata(no=True)
 
+        if self._match(TokenType.SERDE_PROPERTIES, advance=False):
+            return self._parse_serde_properties(with_=True)
+
         if not self._next:
             return None
 
@@ -2201,6 +2310,7 @@ class Parser(metaclass=_Parser):
 
     def _parse_returns(self) -> exp.ReturnsProperty:
         value: t.Optional[exp.Expression]
+        null = None
         is_table = self._match(TokenType.TABLE)
 
         if is_table:
@@ -2214,10 +2324,13 @@ class Parser(metaclass=_Parser):
                     self.raise_error("Expecting >")
             else:
                 value = self._parse_schema(exp.var("TABLE"))
+        elif self._match_text_seq("NULL", "ON", "NULL", "INPUT"):
+            null = True
+            value = None
         else:
             value = self._parse_types()
 
-        return self.expression(exp.ReturnsProperty, this=value, is_table=is_table)
+        return self.expression(exp.ReturnsProperty, this=value, is_table=is_table, null=null)
 
     def _parse_describe(self) -> exp.Describe:
         kind = self._match_set(self.CREATABLES) and self._prev.text
@@ -2340,6 +2453,21 @@ class Parser(metaclass=_Parser):
             return None
         return self._parse_row_format()
 
+    def _parse_serde_properties(self, with_: bool = False) -> t.Optional[exp.SerdeProperties]:
+        index = self._index
+        with_ = with_ or self._match_text_seq("WITH")
+
+        if not self._match(TokenType.SERDE_PROPERTIES):
+            self._retreat(index)
+            return None
+        return self.expression(
+            exp.SerdeProperties,
+            **{  # type: ignore
+                "expressions": self._parse_wrapped_properties(),
+                "with": with_,
+            },
+        )
+
     def _parse_row_format(
         self, match_row: bool = False
     ) -> t.Optional[exp.RowFormatSerdeProperty | exp.RowFormatDelimitedProperty]:
@@ -2349,11 +2477,7 @@ class Parser(metaclass=_Parser):
         if self._match_text_seq("SERDE"):
             this = self._parse_string()
 
-            serde_properties = None
-            if self._match(TokenType.SERDE_PROPERTIES):
-                serde_properties = self.expression(
-                    exp.SerdeProperties, expressions=self._parse_wrapped_properties()
-                )
+            serde_properties = self._parse_serde_properties()
 
             return self.expression(
                 exp.RowFormatSerdeProperty, this=this, serde_properties=serde_properties
@@ -2672,9 +2796,7 @@ class Parser(metaclass=_Parser):
         )
 
     def _implicit_unnests_to_explicit(self, this: E) -> E:
-        from sqlglot.optimizer.normalize_identifiers import (
-            normalize_identifiers as _norm,
-        )
+        from sqlglot.optimizer.normalize_identifiers import normalize_identifiers as _norm
 
         refs = {_norm(this.args["from"].this.copy(), dialect=self.dialect).alias_or_name}
         for i, join in enumerate(this.args.get("joins") or []):
@@ -3366,6 +3488,9 @@ class Parser(metaclass=_Parser):
         elif self._match_texts(("SEED", "REPEATABLE")):
             seed = self._parse_wrapped(self._parse_number)
 
+        if not method and self.DEFAULT_SAMPLING_METHOD:
+            method = exp.var(self.DEFAULT_SAMPLING_METHOD)
+
         return self.expression(
             exp.TableSample,
             expressions=expressions,
@@ -3519,7 +3644,11 @@ class Parser(metaclass=_Parser):
             elements["all"] = False
 
         while True:
-            expressions = self._parse_csv(self._parse_conjunction)
+            expressions = self._parse_csv(
+                lambda: None
+                if self._match(TokenType.ROLLUP, advance=False)
+                else self._parse_conjunction()
+            )
             if expressions:
                 elements["expressions"].extend(expressions)
 
@@ -3817,7 +3946,24 @@ class Parser(metaclass=_Parser):
         return self._parse_alias(self._parse_conjunction())
 
     def _parse_conjunction(self) -> t.Optional[exp.Expression]:
-        return self._parse_tokens(self._parse_equality, self.CONJUNCTION)
+        this = self._parse_equality()
+
+        if self._match(TokenType.COLON_EQ):
+            this = self.expression(
+                exp.PropertyEQ,
+                this=this,
+                comments=self._prev_comments,
+                expression=self._parse_conjunction(),
+            )
+
+        while self._match_set(self.CONJUNCTION):
+            this = self.expression(
+                self.CONJUNCTION[self._prev.token_type],
+                this=this,
+                comments=self._prev_comments,
+                expression=self._parse_equality(),
+            )
+        return this
 
     def _parse_equality(self) -> t.Optional[exp.Expression]:
         return self._parse_tokens(self._parse_comparison, self.EQUALITY)
@@ -4061,6 +4207,7 @@ class Parser(metaclass=_Parser):
     ) -> t.Optional[exp.Expression]:
         index = self._index
 
+        this: t.Optional[exp.Expression] = None
         prefix = self._match_text_seq("SYSUDTLIB", ".")
 
         if not self._match_set(self.TYPE_TOKENS):
@@ -4081,7 +4228,7 @@ class Parser(metaclass=_Parser):
                     while self._match(TokenType.DOT):
                         type_name = f"{type_name}.{self._advance_any() and self._prev.text}"
 
-                    return exp.DataType.build(type_name, udt=True)
+                    this = exp.DataType.build(type_name, udt=True)
                 else:
                     self._retreat(self._index - 1)
                     return None
@@ -4134,7 +4281,6 @@ class Parser(metaclass=_Parser):
 
             maybe_func = True
 
-        this: t.Optional[exp.Expression] = None
         values: t.Optional[t.List[exp.Expression]] = None
 
         if nested and self._match(TokenType.LT):
@@ -4203,10 +4349,17 @@ class Parser(metaclass=_Parser):
                 values=values,
                 prefix=prefix,
             )
+        elif expressions:
+            this.set("expressions", expressions)
 
         while self._match_pair(TokenType.L_BRACKET, TokenType.R_BRACKET):
             this = exp.DataType(this=exp.DataType.Type.ARRAY, expressions=[this], nested=True)
 
+        if self.TYPE_CONVERTER and isinstance(this.this, exp.DataType.Type):
+            converter = self.TYPE_CONVERTER.get(this.this)
+            if converter:
+                this = converter(t.cast(exp.DataType, this))
+
         return this
 
     def _parse_struct_types(self, type_required: bool = False) -> t.Optional[exp.Expression]:
@@ -4326,9 +4479,7 @@ class Parser(metaclass=_Parser):
             if not this and self._match(TokenType.R_PAREN, advance=False):
                 this = self.expression(exp.Tuple)
             elif isinstance(this, exp.UNWRAPPED_QUERIES):
-                this = self._parse_set_operations(
-                    self._parse_subquery(this=this, parse_alias=False)
-                )
+                this = self._parse_subquery(this=this, parse_alias=False)
             elif isinstance(this, exp.Subquery):
                 this = self._parse_subquery(
                     this=self._parse_set_operations(this), parse_alias=False
@@ -5625,13 +5776,8 @@ class Parser(metaclass=_Parser):
         return self._parse_placeholder()
 
     def _parse_parameter(self) -> exp.Parameter:
-        self._match(TokenType.L_BRACE)
         this = self._parse_identifier() or self._parse_primary_or_var()
-        expression = self._match(TokenType.COLON) and (
-            self._parse_identifier() or self._parse_primary_or_var()
-        )
-        self._match(TokenType.R_BRACE)
-        return self.expression(exp.Parameter, this=this, expression=expression)
+        return self.expression(exp.Parameter, this=this)
 
     def _parse_placeholder(self) -> t.Optional[exp.Expression]:
         if self._match_set(self.PLACEHOLDER_PARSERS):
@@ -5641,23 +5787,14 @@ class Parser(metaclass=_Parser):
             self._advance(-1)
         return None
 
-    def _parse_except(self) -> t.Optional[t.List[exp.Expression]]:
-        if not self._match(TokenType.EXCEPT):
-            return None
-        if self._match(TokenType.L_PAREN, advance=False):
-            return self._parse_wrapped_csv(self._parse_column)
-
-        except_column = self._parse_column()
-        return [except_column] if except_column else None
-
-    def _parse_replace(self) -> t.Optional[t.List[exp.Expression]]:
-        if not self._match(TokenType.REPLACE):
+    def _parse_star_op(self, *keywords: str) -> t.Optional[t.List[exp.Expression]]:
+        if not self._match_texts(keywords):
             return None
         if self._match(TokenType.L_PAREN, advance=False):
             return self._parse_wrapped_csv(self._parse_expression)
 
-        replace_expression = self._parse_expression()
-        return [replace_expression] if replace_expression else None
+        expression = self._parse_expression()
+        return [expression] if expression else None
 
     def _parse_csv(
         self, parse_method: t.Callable, sep: TokenType = TokenType.COMMA
@@ -5812,7 +5949,12 @@ class Parser(metaclass=_Parser):
             return self._parse_wrapped_csv(self._parse_field_def, optional=True)
         return self._parse_wrapped_csv(self._parse_add_column, optional=True)
 
-    def _parse_alter_table_alter(self) -> exp.AlterColumn:
+    def _parse_alter_table_alter(self) -> t.Optional[exp.Expression]:
+        if self._match_texts(self.ALTER_ALTER_PARSERS):
+            return self.ALTER_ALTER_PARSERS[self._prev.text.upper()](self)
+
+        # Many dialects support the ALTER [COLUMN] syntax, so if there is no
+        # keyword after ALTER we default to parsing this statement
         self._match(TokenType.COLUMN)
         column = self._parse_field(any_token=True)
 
@@ -5833,6 +5975,27 @@ class Parser(metaclass=_Parser):
             using=self._match(TokenType.USING) and self._parse_conjunction(),
         )
 
+    def _parse_alter_diststyle(self) -> exp.AlterDistStyle:
+        if self._match_texts(("ALL", "EVEN", "AUTO")):
+            return self.expression(exp.AlterDistStyle, this=exp.var(self._prev.text.upper()))
+
+        self._match_text_seq("KEY", "DISTKEY")
+        return self.expression(exp.AlterDistStyle, this=self._parse_column())
+
+    def _parse_alter_sortkey(self, compound: t.Optional[bool] = None) -> exp.AlterSortKey:
+        if compound:
+            self._match_text_seq("SORTKEY")
+
+        if self._match(TokenType.L_PAREN, advance=False):
+            return self.expression(
+                exp.AlterSortKey, expressions=self._parse_wrapped_id_vars(), compound=compound
+            )
+
+        self._match_texts(("AUTO", "NONE"))
+        return self.expression(
+            exp.AlterSortKey, this=exp.var(self._prev.text.upper()), compound=compound
+        )
+
     def _parse_alter_table_drop(self) -> t.List[exp.Expression]:
         index = self._index - 1
 
@@ -5858,6 +6021,41 @@ class Parser(metaclass=_Parser):
         self._match_text_seq("TO")
         return self.expression(exp.RenameTable, this=self._parse_table(schema=True))
 
+    def _parse_alter_table_set(self) -> exp.AlterSet:
+        alter_set = self.expression(exp.AlterSet)
+
+        if self._match(TokenType.L_PAREN, advance=False) or self._match_text_seq(
+            "TABLE", "PROPERTIES"
+        ):
+            alter_set.set("expressions", self._parse_wrapped_csv(self._parse_conjunction))
+        elif self._match_text_seq("FILESTREAM_ON", advance=False):
+            alter_set.set("expressions", [self._parse_conjunction()])
+        elif self._match_texts(("LOGGED", "UNLOGGED")):
+            alter_set.set("option", exp.var(self._prev.text.upper()))
+        elif self._match_text_seq("WITHOUT") and self._match_texts(("CLUSTER", "OIDS")):
+            alter_set.set("option", exp.var(f"WITHOUT {self._prev.text.upper()}"))
+        elif self._match_text_seq("LOCATION"):
+            alter_set.set("location", self._parse_field())
+        elif self._match_text_seq("ACCESS", "METHOD"):
+            alter_set.set("access_method", self._parse_field())
+        elif self._match_text_seq("TABLESPACE"):
+            alter_set.set("tablespace", self._parse_field())
+        elif self._match_text_seq("FILE", "FORMAT") or self._match_text_seq("FILEFORMAT"):
+            alter_set.set("file_format", [self._parse_field()])
+        elif self._match_text_seq("STAGE_FILE_FORMAT"):
+            alter_set.set("file_format", self._parse_wrapped_options())
+        elif self._match_text_seq("STAGE_COPY_OPTIONS"):
+            alter_set.set("copy_options", self._parse_wrapped_options())
+        elif self._match_text_seq("TAG") or self._match_text_seq("TAGS"):
+            alter_set.set("tag", self._parse_csv(self._parse_conjunction))
+        else:
+            if self._match_text_seq("SERDE"):
+                alter_set.set("serde", self._parse_field())
+
+            alter_set.set("expressions", [self._parse_properties()])
+
+        return alter_set
+
     def _parse_alter(self) -> exp.AlterTable | exp.Command:
         start = self._prev
 
@@ -5867,6 +6065,7 @@ class Parser(metaclass=_Parser):
         exists = self._parse_exists()
         only = self._match_text_seq("ONLY")
         this = self._parse_table(schema=True)
+        cluster = self._parse_on_property() if self._match(TokenType.ON) else None
 
         if self._next:
             self._advance()
@@ -5884,6 +6083,7 @@ class Parser(metaclass=_Parser):
                     actions=actions,
                     only=only,
                     options=options,
+                    cluster=cluster,
                 )
 
         return self._parse_as_command(start)
@@ -5974,7 +6174,7 @@ class Parser(metaclass=_Parser):
         if kind in ("GLOBAL", "SESSION") and self._match_text_seq("TRANSACTION"):
             return self._parse_set_transaction(global_=kind == "GLOBAL")
 
-        left = self._parse_primary() or self._parse_id_var()
+        left = self._parse_primary() or self._parse_column()
         assignment_delimiter = self._match_texts(("=", "TO"))
 
         if not left or (self.SET_REQUIRES_ASSIGNMENT_DELIMITER and not assignment_delimiter):