sqlglot.dialects.clickhouse
1from __future__ import annotations 2 3import typing as t 4 5from sqlglot import exp, generator, parser, tokens, transforms 6from sqlglot.dialects.dialect import ( 7 Dialect, 8 arg_max_or_min_no_count, 9 inline_array_sql, 10 no_pivot_sql, 11 rename_func, 12 var_map_sql, 13) 14from sqlglot.errors import ParseError 15from sqlglot.helper import seq_get 16from sqlglot.parser import parse_var_map 17from sqlglot.tokens import Token, TokenType 18 19 20def _lower_func(sql: str) -> str: 21 index = sql.index("(") 22 return sql[:index].lower() + sql[index:] 23 24 25def _quantile_sql(self, e): 26 quantile = e.args["quantile"] 27 args = f"({self.sql(e, 'this')})" 28 if isinstance(quantile, exp.Array): 29 func = self.func("quantiles", *quantile) 30 else: 31 func = self.func("quantile", quantile) 32 return func + args 33 34 35class ClickHouse(Dialect): 36 NORMALIZE_FUNCTIONS: bool | str = False 37 NULL_ORDERING = "nulls_are_last" 38 STRICT_STRING_CONCAT = True 39 SUPPORTS_USER_DEFINED_TYPES = False 40 SAFE_DIVISION = True 41 42 ESCAPE_SEQUENCES = { 43 "\\0": "\0", 44 } 45 46 class Tokenizer(tokens.Tokenizer): 47 COMMENTS = ["--", "#", "#!", ("/*", "*/")] 48 IDENTIFIERS = ['"', "`"] 49 STRING_ESCAPES = ["'", "\\"] 50 BIT_STRINGS = [("0b", "")] 51 HEX_STRINGS = [("0x", ""), ("0X", "")] 52 HEREDOC_STRINGS = ["$"] 53 54 KEYWORDS = { 55 **tokens.Tokenizer.KEYWORDS, 56 "ATTACH": TokenType.COMMAND, 57 "DATETIME64": TokenType.DATETIME64, 58 "DICTIONARY": TokenType.DICTIONARY, 59 "ENUM": TokenType.ENUM, 60 "ENUM8": TokenType.ENUM8, 61 "ENUM16": TokenType.ENUM16, 62 "FINAL": TokenType.FINAL, 63 "FIXEDSTRING": TokenType.FIXEDSTRING, 64 "FLOAT32": TokenType.FLOAT, 65 "FLOAT64": TokenType.DOUBLE, 66 "GLOBAL": TokenType.GLOBAL, 67 "INT16": TokenType.SMALLINT, 68 "INT256": TokenType.INT256, 69 "INT32": TokenType.INT, 70 "INT64": TokenType.BIGINT, 71 "INT8": TokenType.TINYINT, 72 "LOWCARDINALITY": TokenType.LOWCARDINALITY, 73 "MAP": TokenType.MAP, 74 "NESTED": TokenType.NESTED, 75 "SAMPLE": TokenType.TABLE_SAMPLE, 76 "TUPLE": TokenType.STRUCT, 77 "UINT128": TokenType.UINT128, 78 "UINT16": TokenType.USMALLINT, 79 "UINT256": TokenType.UINT256, 80 "UINT32": TokenType.UINT, 81 "UINT64": TokenType.UBIGINT, 82 "UINT8": TokenType.UTINYINT, 83 } 84 85 SINGLE_TOKENS = { 86 **tokens.Tokenizer.SINGLE_TOKENS, 87 "$": TokenType.HEREDOC_STRING, 88 } 89 90 class Parser(parser.Parser): 91 FUNCTIONS = { 92 **parser.Parser.FUNCTIONS, 93 "ANY": exp.AnyValue.from_arg_list, 94 "DATE_ADD": lambda args: exp.DateAdd( 95 this=seq_get(args, 2), expression=seq_get(args, 1), unit=seq_get(args, 0) 96 ), 97 "DATEADD": lambda args: exp.DateAdd( 98 this=seq_get(args, 2), expression=seq_get(args, 1), unit=seq_get(args, 0) 99 ), 100 "DATE_DIFF": lambda args: exp.DateDiff( 101 this=seq_get(args, 2), expression=seq_get(args, 1), unit=seq_get(args, 0) 102 ), 103 "DATEDIFF": lambda args: exp.DateDiff( 104 this=seq_get(args, 2), expression=seq_get(args, 1), unit=seq_get(args, 0) 105 ), 106 "MAP": parse_var_map, 107 "MATCH": exp.RegexpLike.from_arg_list, 108 "UNIQ": exp.ApproxDistinct.from_arg_list, 109 "XOR": lambda args: exp.Xor(expressions=args), 110 } 111 112 FUNCTIONS_WITH_ALIASED_ARGS = {*parser.Parser.FUNCTIONS_WITH_ALIASED_ARGS, "TUPLE"} 113 114 FUNCTION_PARSERS = { 115 **parser.Parser.FUNCTION_PARSERS, 116 "QUANTILE": lambda self: self._parse_quantile(), 117 } 118 119 FUNCTION_PARSERS.pop("MATCH") 120 121 NO_PAREN_FUNCTION_PARSERS = parser.Parser.NO_PAREN_FUNCTION_PARSERS.copy() 122 NO_PAREN_FUNCTION_PARSERS.pop("ANY") 123 124 RANGE_PARSERS = { 125 **parser.Parser.RANGE_PARSERS, 126 TokenType.GLOBAL: lambda self, this: self._match(TokenType.IN) 127 and self._parse_in(this, is_global=True), 128 } 129 130 # The PLACEHOLDER entry is popped because 1) it doesn't affect Clickhouse (it corresponds to 131 # the postgres-specific JSONBContains parser) and 2) it makes parsing the ternary op simpler. 132 COLUMN_OPERATORS = parser.Parser.COLUMN_OPERATORS.copy() 133 COLUMN_OPERATORS.pop(TokenType.PLACEHOLDER) 134 135 JOIN_KINDS = { 136 *parser.Parser.JOIN_KINDS, 137 TokenType.ANY, 138 TokenType.ASOF, 139 TokenType.ARRAY, 140 } 141 142 TABLE_ALIAS_TOKENS = parser.Parser.TABLE_ALIAS_TOKENS - { 143 TokenType.ANY, 144 TokenType.SETTINGS, 145 TokenType.FORMAT, 146 TokenType.ARRAY, 147 } 148 149 LOG_DEFAULTS_TO_LN = True 150 151 QUERY_MODIFIER_PARSERS = { 152 **parser.Parser.QUERY_MODIFIER_PARSERS, 153 TokenType.SETTINGS: lambda self: ( 154 "settings", 155 self._advance() or self._parse_csv(self._parse_conjunction), 156 ), 157 TokenType.FORMAT: lambda self: ("format", self._advance() or self._parse_id_var()), 158 } 159 160 def _parse_conjunction(self) -> t.Optional[exp.Expression]: 161 this = super()._parse_conjunction() 162 163 if self._match(TokenType.PLACEHOLDER): 164 return self.expression( 165 exp.If, 166 this=this, 167 true=self._parse_conjunction(), 168 false=self._match(TokenType.COLON) and self._parse_conjunction(), 169 ) 170 171 return this 172 173 def _parse_placeholder(self) -> t.Optional[exp.Expression]: 174 """ 175 Parse a placeholder expression like SELECT {abc: UInt32} or FROM {table: Identifier} 176 https://clickhouse.com/docs/en/sql-reference/syntax#defining-and-using-query-parameters 177 """ 178 if not self._match(TokenType.L_BRACE): 179 return None 180 181 this = self._parse_id_var() 182 self._match(TokenType.COLON) 183 kind = self._parse_types(check_func=False, allow_identifiers=False) or ( 184 self._match_text_seq("IDENTIFIER") and "Identifier" 185 ) 186 187 if not kind: 188 self.raise_error("Expecting a placeholder type or 'Identifier' for tables") 189 elif not self._match(TokenType.R_BRACE): 190 self.raise_error("Expecting }") 191 192 return self.expression(exp.Placeholder, this=this, kind=kind) 193 194 def _parse_in(self, this: t.Optional[exp.Expression], is_global: bool = False) -> exp.In: 195 this = super()._parse_in(this) 196 this.set("is_global", is_global) 197 return this 198 199 def _parse_table( 200 self, 201 schema: bool = False, 202 joins: bool = False, 203 alias_tokens: t.Optional[t.Collection[TokenType]] = None, 204 parse_bracket: bool = False, 205 ) -> t.Optional[exp.Expression]: 206 this = super()._parse_table( 207 schema=schema, joins=joins, alias_tokens=alias_tokens, parse_bracket=parse_bracket 208 ) 209 210 if self._match(TokenType.FINAL): 211 this = self.expression(exp.Final, this=this) 212 213 return this 214 215 def _parse_position(self, haystack_first: bool = False) -> exp.StrPosition: 216 return super()._parse_position(haystack_first=True) 217 218 # https://clickhouse.com/docs/en/sql-reference/statements/select/with/ 219 def _parse_cte(self) -> exp.CTE: 220 index = self._index 221 try: 222 # WITH <identifier> AS <subquery expression> 223 return super()._parse_cte() 224 except ParseError: 225 # WITH <expression> AS <identifier> 226 self._retreat(index) 227 statement = self._parse_statement() 228 229 if statement and isinstance(statement.this, exp.Alias): 230 self.raise_error("Expected CTE to have alias") 231 232 return self.expression(exp.CTE, this=statement, alias=statement and statement.this) 233 234 def _parse_join_parts( 235 self, 236 ) -> t.Tuple[t.Optional[Token], t.Optional[Token], t.Optional[Token]]: 237 is_global = self._match(TokenType.GLOBAL) and self._prev 238 kind_pre = self._match_set(self.JOIN_KINDS, advance=False) and self._prev 239 240 if kind_pre: 241 kind = self._match_set(self.JOIN_KINDS) and self._prev 242 side = self._match_set(self.JOIN_SIDES) and self._prev 243 return is_global, side, kind 244 245 return ( 246 is_global, 247 self._match_set(self.JOIN_SIDES) and self._prev, 248 self._match_set(self.JOIN_KINDS) and self._prev, 249 ) 250 251 def _parse_join( 252 self, skip_join_token: bool = False, parse_bracket: bool = False 253 ) -> t.Optional[exp.Join]: 254 join = super()._parse_join(skip_join_token=skip_join_token, parse_bracket=True) 255 256 if join: 257 join.set("global", join.args.pop("method", None)) 258 return join 259 260 def _parse_function( 261 self, 262 functions: t.Optional[t.Dict[str, t.Callable]] = None, 263 anonymous: bool = False, 264 optional_parens: bool = True, 265 ) -> t.Optional[exp.Expression]: 266 func = super()._parse_function( 267 functions=functions, anonymous=anonymous, optional_parens=optional_parens 268 ) 269 270 if isinstance(func, exp.Anonymous): 271 params = self._parse_func_params(func) 272 273 if params: 274 return self.expression( 275 exp.ParameterizedAgg, 276 this=func.this, 277 expressions=func.expressions, 278 params=params, 279 ) 280 281 return func 282 283 def _parse_func_params( 284 self, this: t.Optional[exp.Func] = None 285 ) -> t.Optional[t.List[exp.Expression]]: 286 if self._match_pair(TokenType.R_PAREN, TokenType.L_PAREN): 287 return self._parse_csv(self._parse_lambda) 288 289 if self._match(TokenType.L_PAREN): 290 params = self._parse_csv(self._parse_lambda) 291 self._match_r_paren(this) 292 return params 293 294 return None 295 296 def _parse_quantile(self) -> exp.Quantile: 297 this = self._parse_lambda() 298 params = self._parse_func_params() 299 if params: 300 return self.expression(exp.Quantile, this=params[0], quantile=this) 301 return self.expression(exp.Quantile, this=this, quantile=exp.Literal.number(0.5)) 302 303 def _parse_wrapped_id_vars(self, optional: bool = False) -> t.List[exp.Expression]: 304 return super()._parse_wrapped_id_vars(optional=True) 305 306 def _parse_primary_key( 307 self, wrapped_optional: bool = False, in_props: bool = False 308 ) -> exp.PrimaryKeyColumnConstraint | exp.PrimaryKey: 309 return super()._parse_primary_key( 310 wrapped_optional=wrapped_optional or in_props, in_props=in_props 311 ) 312 313 def _parse_on_property(self) -> t.Optional[exp.Expression]: 314 index = self._index 315 if self._match_text_seq("CLUSTER"): 316 this = self._parse_id_var() 317 if this: 318 return self.expression(exp.OnCluster, this=this) 319 else: 320 self._retreat(index) 321 return None 322 323 class Generator(generator.Generator): 324 QUERY_HINTS = False 325 STRUCT_DELIMITER = ("(", ")") 326 NVL2_SUPPORTED = False 327 TABLESAMPLE_REQUIRES_PARENS = False 328 329 STRING_TYPE_MAPPING = { 330 exp.DataType.Type.CHAR: "String", 331 exp.DataType.Type.LONGBLOB: "String", 332 exp.DataType.Type.LONGTEXT: "String", 333 exp.DataType.Type.MEDIUMBLOB: "String", 334 exp.DataType.Type.MEDIUMTEXT: "String", 335 exp.DataType.Type.TINYBLOB: "String", 336 exp.DataType.Type.TINYTEXT: "String", 337 exp.DataType.Type.TEXT: "String", 338 exp.DataType.Type.VARBINARY: "String", 339 exp.DataType.Type.VARCHAR: "String", 340 } 341 342 TYPE_MAPPING = { 343 **generator.Generator.TYPE_MAPPING, 344 **STRING_TYPE_MAPPING, 345 exp.DataType.Type.ARRAY: "Array", 346 exp.DataType.Type.BIGINT: "Int64", 347 exp.DataType.Type.DATETIME64: "DateTime64", 348 exp.DataType.Type.DOUBLE: "Float64", 349 exp.DataType.Type.ENUM: "Enum", 350 exp.DataType.Type.ENUM8: "Enum8", 351 exp.DataType.Type.ENUM16: "Enum16", 352 exp.DataType.Type.FIXEDSTRING: "FixedString", 353 exp.DataType.Type.FLOAT: "Float32", 354 exp.DataType.Type.INT: "Int32", 355 exp.DataType.Type.MEDIUMINT: "Int32", 356 exp.DataType.Type.INT128: "Int128", 357 exp.DataType.Type.INT256: "Int256", 358 exp.DataType.Type.LOWCARDINALITY: "LowCardinality", 359 exp.DataType.Type.MAP: "Map", 360 exp.DataType.Type.NESTED: "Nested", 361 exp.DataType.Type.NULLABLE: "Nullable", 362 exp.DataType.Type.SMALLINT: "Int16", 363 exp.DataType.Type.STRUCT: "Tuple", 364 exp.DataType.Type.TINYINT: "Int8", 365 exp.DataType.Type.UBIGINT: "UInt64", 366 exp.DataType.Type.UINT: "UInt32", 367 exp.DataType.Type.UINT128: "UInt128", 368 exp.DataType.Type.UINT256: "UInt256", 369 exp.DataType.Type.USMALLINT: "UInt16", 370 exp.DataType.Type.UTINYINT: "UInt8", 371 } 372 373 TRANSFORMS = { 374 **generator.Generator.TRANSFORMS, 375 exp.Select: transforms.preprocess([transforms.eliminate_qualify]), 376 exp.AnyValue: rename_func("any"), 377 exp.ApproxDistinct: rename_func("uniq"), 378 exp.ArgMax: arg_max_or_min_no_count("argMax"), 379 exp.ArgMin: arg_max_or_min_no_count("argMin"), 380 exp.Array: inline_array_sql, 381 exp.CastToStrType: rename_func("CAST"), 382 exp.CurrentDate: lambda self, e: self.func("CURRENT_DATE"), 383 exp.DateAdd: lambda self, e: self.func( 384 "DATE_ADD", exp.Literal.string(e.text("unit") or "day"), e.expression, e.this 385 ), 386 exp.DateDiff: lambda self, e: self.func( 387 "DATE_DIFF", exp.Literal.string(e.text("unit") or "day"), e.expression, e.this 388 ), 389 exp.Final: lambda self, e: f"{self.sql(e, 'this')} FINAL", 390 exp.IsNan: rename_func("isNaN"), 391 exp.Map: lambda self, e: _lower_func(var_map_sql(self, e)), 392 exp.Nullif: rename_func("nullIf"), 393 exp.PartitionedByProperty: lambda self, e: f"PARTITION BY {self.sql(e, 'this')}", 394 exp.Pivot: no_pivot_sql, 395 exp.Quantile: _quantile_sql, 396 exp.RegexpLike: lambda self, e: f"match({self.format_args(e.this, e.expression)})", 397 exp.StartsWith: rename_func("startsWith"), 398 exp.StrPosition: lambda self, e: f"position({self.format_args(e.this, e.args.get('substr'), e.args.get('position'))})", 399 exp.VarMap: lambda self, e: _lower_func(var_map_sql(self, e)), 400 exp.Xor: lambda self, e: self.func("xor", e.this, e.expression, *e.expressions), 401 } 402 403 PROPERTIES_LOCATION = { 404 **generator.Generator.PROPERTIES_LOCATION, 405 exp.VolatileProperty: exp.Properties.Location.UNSUPPORTED, 406 exp.PartitionedByProperty: exp.Properties.Location.POST_SCHEMA, 407 exp.OnCluster: exp.Properties.Location.POST_NAME, 408 } 409 410 JOIN_HINTS = False 411 TABLE_HINTS = False 412 EXPLICIT_UNION = True 413 GROUPINGS_SEP = "" 414 415 # there's no list in docs, but it can be found in Clickhouse code 416 # see `ClickHouse/src/Parsers/ParserCreate*.cpp` 417 ON_CLUSTER_TARGETS = { 418 "DATABASE", 419 "TABLE", 420 "VIEW", 421 "DICTIONARY", 422 "INDEX", 423 "FUNCTION", 424 "NAMED COLLECTION", 425 } 426 427 def _any_to_has( 428 self, 429 expression: exp.EQ | exp.NEQ, 430 default: t.Callable[[t.Any], str], 431 prefix: str = "", 432 ) -> str: 433 if isinstance(expression.left, exp.Any): 434 arr = expression.left 435 this = expression.right 436 elif isinstance(expression.right, exp.Any): 437 arr = expression.right 438 this = expression.left 439 else: 440 return default(expression) 441 return prefix + self.func("has", arr.this.unnest(), this) 442 443 def eq_sql(self, expression: exp.EQ) -> str: 444 return self._any_to_has(expression, super().eq_sql) 445 446 def neq_sql(self, expression: exp.NEQ) -> str: 447 return self._any_to_has(expression, super().neq_sql, "NOT ") 448 449 def regexpilike_sql(self, expression: exp.RegexpILike) -> str: 450 # Manually add a flag to make the search case-insensitive 451 regex = self.func("CONCAT", "'(?i)'", expression.expression) 452 return f"match({self.format_args(expression.this, regex)})" 453 454 def datatype_sql(self, expression: exp.DataType) -> str: 455 # String is the standard ClickHouse type, every other variant is just an alias. 456 # Additionally, any supplied length parameter will be ignored. 457 # 458 # https://clickhouse.com/docs/en/sql-reference/data-types/string 459 if expression.this in self.STRING_TYPE_MAPPING: 460 return "String" 461 462 return super().datatype_sql(expression) 463 464 def safeconcat_sql(self, expression: exp.SafeConcat) -> str: 465 # Clickhouse errors out if we try to cast a NULL value to TEXT 466 return self.func( 467 "CONCAT", 468 *[ 469 exp.func("if", e.is_(exp.null()), e, exp.cast(e, "text")) 470 for e in t.cast(t.List[exp.Condition], expression.expressions) 471 ], 472 ) 473 474 def cte_sql(self, expression: exp.CTE) -> str: 475 if isinstance(expression.this, exp.Alias): 476 return self.sql(expression, "this") 477 478 return super().cte_sql(expression) 479 480 def after_limit_modifiers(self, expression: exp.Expression) -> t.List[str]: 481 return super().after_limit_modifiers(expression) + [ 482 self.seg("SETTINGS ") + self.expressions(expression, key="settings", flat=True) 483 if expression.args.get("settings") 484 else "", 485 self.seg("FORMAT ") + self.sql(expression, "format") 486 if expression.args.get("format") 487 else "", 488 ] 489 490 def parameterizedagg_sql(self, expression: exp.Anonymous) -> str: 491 params = self.expressions(expression, key="params", flat=True) 492 return self.func(expression.name, *expression.expressions) + f"({params})" 493 494 def placeholder_sql(self, expression: exp.Placeholder) -> str: 495 return f"{{{expression.name}: {self.sql(expression, 'kind')}}}" 496 497 def oncluster_sql(self, expression: exp.OnCluster) -> str: 498 return f"ON CLUSTER {self.sql(expression, 'this')}" 499 500 def createable_sql(self, expression: exp.Create, locations: t.DefaultDict) -> str: 501 kind = self.sql(expression, "kind").upper() 502 if kind in self.ON_CLUSTER_TARGETS and locations.get(exp.Properties.Location.POST_NAME): 503 this_name = self.sql(expression.this, "this") 504 this_properties = " ".join( 505 [self.sql(prop) for prop in locations[exp.Properties.Location.POST_NAME]] 506 ) 507 this_schema = self.schema_columns_sql(expression.this) 508 return f"{this_name}{self.sep()}{this_properties}{self.sep()}{this_schema}" 509 510 return super().createable_sql(expression, locations)
36class ClickHouse(Dialect): 37 NORMALIZE_FUNCTIONS: bool | str = False 38 NULL_ORDERING = "nulls_are_last" 39 STRICT_STRING_CONCAT = True 40 SUPPORTS_USER_DEFINED_TYPES = False 41 SAFE_DIVISION = True 42 43 ESCAPE_SEQUENCES = { 44 "\\0": "\0", 45 } 46 47 class Tokenizer(tokens.Tokenizer): 48 COMMENTS = ["--", "#", "#!", ("/*", "*/")] 49 IDENTIFIERS = ['"', "`"] 50 STRING_ESCAPES = ["'", "\\"] 51 BIT_STRINGS = [("0b", "")] 52 HEX_STRINGS = [("0x", ""), ("0X", "")] 53 HEREDOC_STRINGS = ["$"] 54 55 KEYWORDS = { 56 **tokens.Tokenizer.KEYWORDS, 57 "ATTACH": TokenType.COMMAND, 58 "DATETIME64": TokenType.DATETIME64, 59 "DICTIONARY": TokenType.DICTIONARY, 60 "ENUM": TokenType.ENUM, 61 "ENUM8": TokenType.ENUM8, 62 "ENUM16": TokenType.ENUM16, 63 "FINAL": TokenType.FINAL, 64 "FIXEDSTRING": TokenType.FIXEDSTRING, 65 "FLOAT32": TokenType.FLOAT, 66 "FLOAT64": TokenType.DOUBLE, 67 "GLOBAL": TokenType.GLOBAL, 68 "INT16": TokenType.SMALLINT, 69 "INT256": TokenType.INT256, 70 "INT32": TokenType.INT, 71 "INT64": TokenType.BIGINT, 72 "INT8": TokenType.TINYINT, 73 "LOWCARDINALITY": TokenType.LOWCARDINALITY, 74 "MAP": TokenType.MAP, 75 "NESTED": TokenType.NESTED, 76 "SAMPLE": TokenType.TABLE_SAMPLE, 77 "TUPLE": TokenType.STRUCT, 78 "UINT128": TokenType.UINT128, 79 "UINT16": TokenType.USMALLINT, 80 "UINT256": TokenType.UINT256, 81 "UINT32": TokenType.UINT, 82 "UINT64": TokenType.UBIGINT, 83 "UINT8": TokenType.UTINYINT, 84 } 85 86 SINGLE_TOKENS = { 87 **tokens.Tokenizer.SINGLE_TOKENS, 88 "$": TokenType.HEREDOC_STRING, 89 } 90 91 class Parser(parser.Parser): 92 FUNCTIONS = { 93 **parser.Parser.FUNCTIONS, 94 "ANY": exp.AnyValue.from_arg_list, 95 "DATE_ADD": lambda args: exp.DateAdd( 96 this=seq_get(args, 2), expression=seq_get(args, 1), unit=seq_get(args, 0) 97 ), 98 "DATEADD": lambda args: exp.DateAdd( 99 this=seq_get(args, 2), expression=seq_get(args, 1), unit=seq_get(args, 0) 100 ), 101 "DATE_DIFF": lambda args: exp.DateDiff( 102 this=seq_get(args, 2), expression=seq_get(args, 1), unit=seq_get(args, 0) 103 ), 104 "DATEDIFF": lambda args: exp.DateDiff( 105 this=seq_get(args, 2), expression=seq_get(args, 1), unit=seq_get(args, 0) 106 ), 107 "MAP": parse_var_map, 108 "MATCH": exp.RegexpLike.from_arg_list, 109 "UNIQ": exp.ApproxDistinct.from_arg_list, 110 "XOR": lambda args: exp.Xor(expressions=args), 111 } 112 113 FUNCTIONS_WITH_ALIASED_ARGS = {*parser.Parser.FUNCTIONS_WITH_ALIASED_ARGS, "TUPLE"} 114 115 FUNCTION_PARSERS = { 116 **parser.Parser.FUNCTION_PARSERS, 117 "QUANTILE": lambda self: self._parse_quantile(), 118 } 119 120 FUNCTION_PARSERS.pop("MATCH") 121 122 NO_PAREN_FUNCTION_PARSERS = parser.Parser.NO_PAREN_FUNCTION_PARSERS.copy() 123 NO_PAREN_FUNCTION_PARSERS.pop("ANY") 124 125 RANGE_PARSERS = { 126 **parser.Parser.RANGE_PARSERS, 127 TokenType.GLOBAL: lambda self, this: self._match(TokenType.IN) 128 and self._parse_in(this, is_global=True), 129 } 130 131 # The PLACEHOLDER entry is popped because 1) it doesn't affect Clickhouse (it corresponds to 132 # the postgres-specific JSONBContains parser) and 2) it makes parsing the ternary op simpler. 133 COLUMN_OPERATORS = parser.Parser.COLUMN_OPERATORS.copy() 134 COLUMN_OPERATORS.pop(TokenType.PLACEHOLDER) 135 136 JOIN_KINDS = { 137 *parser.Parser.JOIN_KINDS, 138 TokenType.ANY, 139 TokenType.ASOF, 140 TokenType.ARRAY, 141 } 142 143 TABLE_ALIAS_TOKENS = parser.Parser.TABLE_ALIAS_TOKENS - { 144 TokenType.ANY, 145 TokenType.SETTINGS, 146 TokenType.FORMAT, 147 TokenType.ARRAY, 148 } 149 150 LOG_DEFAULTS_TO_LN = True 151 152 QUERY_MODIFIER_PARSERS = { 153 **parser.Parser.QUERY_MODIFIER_PARSERS, 154 TokenType.SETTINGS: lambda self: ( 155 "settings", 156 self._advance() or self._parse_csv(self._parse_conjunction), 157 ), 158 TokenType.FORMAT: lambda self: ("format", self._advance() or self._parse_id_var()), 159 } 160 161 def _parse_conjunction(self) -> t.Optional[exp.Expression]: 162 this = super()._parse_conjunction() 163 164 if self._match(TokenType.PLACEHOLDER): 165 return self.expression( 166 exp.If, 167 this=this, 168 true=self._parse_conjunction(), 169 false=self._match(TokenType.COLON) and self._parse_conjunction(), 170 ) 171 172 return this 173 174 def _parse_placeholder(self) -> t.Optional[exp.Expression]: 175 """ 176 Parse a placeholder expression like SELECT {abc: UInt32} or FROM {table: Identifier} 177 https://clickhouse.com/docs/en/sql-reference/syntax#defining-and-using-query-parameters 178 """ 179 if not self._match(TokenType.L_BRACE): 180 return None 181 182 this = self._parse_id_var() 183 self._match(TokenType.COLON) 184 kind = self._parse_types(check_func=False, allow_identifiers=False) or ( 185 self._match_text_seq("IDENTIFIER") and "Identifier" 186 ) 187 188 if not kind: 189 self.raise_error("Expecting a placeholder type or 'Identifier' for tables") 190 elif not self._match(TokenType.R_BRACE): 191 self.raise_error("Expecting }") 192 193 return self.expression(exp.Placeholder, this=this, kind=kind) 194 195 def _parse_in(self, this: t.Optional[exp.Expression], is_global: bool = False) -> exp.In: 196 this = super()._parse_in(this) 197 this.set("is_global", is_global) 198 return this 199 200 def _parse_table( 201 self, 202 schema: bool = False, 203 joins: bool = False, 204 alias_tokens: t.Optional[t.Collection[TokenType]] = None, 205 parse_bracket: bool = False, 206 ) -> t.Optional[exp.Expression]: 207 this = super()._parse_table( 208 schema=schema, joins=joins, alias_tokens=alias_tokens, parse_bracket=parse_bracket 209 ) 210 211 if self._match(TokenType.FINAL): 212 this = self.expression(exp.Final, this=this) 213 214 return this 215 216 def _parse_position(self, haystack_first: bool = False) -> exp.StrPosition: 217 return super()._parse_position(haystack_first=True) 218 219 # https://clickhouse.com/docs/en/sql-reference/statements/select/with/ 220 def _parse_cte(self) -> exp.CTE: 221 index = self._index 222 try: 223 # WITH <identifier> AS <subquery expression> 224 return super()._parse_cte() 225 except ParseError: 226 # WITH <expression> AS <identifier> 227 self._retreat(index) 228 statement = self._parse_statement() 229 230 if statement and isinstance(statement.this, exp.Alias): 231 self.raise_error("Expected CTE to have alias") 232 233 return self.expression(exp.CTE, this=statement, alias=statement and statement.this) 234 235 def _parse_join_parts( 236 self, 237 ) -> t.Tuple[t.Optional[Token], t.Optional[Token], t.Optional[Token]]: 238 is_global = self._match(TokenType.GLOBAL) and self._prev 239 kind_pre = self._match_set(self.JOIN_KINDS, advance=False) and self._prev 240 241 if kind_pre: 242 kind = self._match_set(self.JOIN_KINDS) and self._prev 243 side = self._match_set(self.JOIN_SIDES) and self._prev 244 return is_global, side, kind 245 246 return ( 247 is_global, 248 self._match_set(self.JOIN_SIDES) and self._prev, 249 self._match_set(self.JOIN_KINDS) and self._prev, 250 ) 251 252 def _parse_join( 253 self, skip_join_token: bool = False, parse_bracket: bool = False 254 ) -> t.Optional[exp.Join]: 255 join = super()._parse_join(skip_join_token=skip_join_token, parse_bracket=True) 256 257 if join: 258 join.set("global", join.args.pop("method", None)) 259 return join 260 261 def _parse_function( 262 self, 263 functions: t.Optional[t.Dict[str, t.Callable]] = None, 264 anonymous: bool = False, 265 optional_parens: bool = True, 266 ) -> t.Optional[exp.Expression]: 267 func = super()._parse_function( 268 functions=functions, anonymous=anonymous, optional_parens=optional_parens 269 ) 270 271 if isinstance(func, exp.Anonymous): 272 params = self._parse_func_params(func) 273 274 if params: 275 return self.expression( 276 exp.ParameterizedAgg, 277 this=func.this, 278 expressions=func.expressions, 279 params=params, 280 ) 281 282 return func 283 284 def _parse_func_params( 285 self, this: t.Optional[exp.Func] = None 286 ) -> t.Optional[t.List[exp.Expression]]: 287 if self._match_pair(TokenType.R_PAREN, TokenType.L_PAREN): 288 return self._parse_csv(self._parse_lambda) 289 290 if self._match(TokenType.L_PAREN): 291 params = self._parse_csv(self._parse_lambda) 292 self._match_r_paren(this) 293 return params 294 295 return None 296 297 def _parse_quantile(self) -> exp.Quantile: 298 this = self._parse_lambda() 299 params = self._parse_func_params() 300 if params: 301 return self.expression(exp.Quantile, this=params[0], quantile=this) 302 return self.expression(exp.Quantile, this=this, quantile=exp.Literal.number(0.5)) 303 304 def _parse_wrapped_id_vars(self, optional: bool = False) -> t.List[exp.Expression]: 305 return super()._parse_wrapped_id_vars(optional=True) 306 307 def _parse_primary_key( 308 self, wrapped_optional: bool = False, in_props: bool = False 309 ) -> exp.PrimaryKeyColumnConstraint | exp.PrimaryKey: 310 return super()._parse_primary_key( 311 wrapped_optional=wrapped_optional or in_props, in_props=in_props 312 ) 313 314 def _parse_on_property(self) -> t.Optional[exp.Expression]: 315 index = self._index 316 if self._match_text_seq("CLUSTER"): 317 this = self._parse_id_var() 318 if this: 319 return self.expression(exp.OnCluster, this=this) 320 else: 321 self._retreat(index) 322 return None 323 324 class Generator(generator.Generator): 325 QUERY_HINTS = False 326 STRUCT_DELIMITER = ("(", ")") 327 NVL2_SUPPORTED = False 328 TABLESAMPLE_REQUIRES_PARENS = False 329 330 STRING_TYPE_MAPPING = { 331 exp.DataType.Type.CHAR: "String", 332 exp.DataType.Type.LONGBLOB: "String", 333 exp.DataType.Type.LONGTEXT: "String", 334 exp.DataType.Type.MEDIUMBLOB: "String", 335 exp.DataType.Type.MEDIUMTEXT: "String", 336 exp.DataType.Type.TINYBLOB: "String", 337 exp.DataType.Type.TINYTEXT: "String", 338 exp.DataType.Type.TEXT: "String", 339 exp.DataType.Type.VARBINARY: "String", 340 exp.DataType.Type.VARCHAR: "String", 341 } 342 343 TYPE_MAPPING = { 344 **generator.Generator.TYPE_MAPPING, 345 **STRING_TYPE_MAPPING, 346 exp.DataType.Type.ARRAY: "Array", 347 exp.DataType.Type.BIGINT: "Int64", 348 exp.DataType.Type.DATETIME64: "DateTime64", 349 exp.DataType.Type.DOUBLE: "Float64", 350 exp.DataType.Type.ENUM: "Enum", 351 exp.DataType.Type.ENUM8: "Enum8", 352 exp.DataType.Type.ENUM16: "Enum16", 353 exp.DataType.Type.FIXEDSTRING: "FixedString", 354 exp.DataType.Type.FLOAT: "Float32", 355 exp.DataType.Type.INT: "Int32", 356 exp.DataType.Type.MEDIUMINT: "Int32", 357 exp.DataType.Type.INT128: "Int128", 358 exp.DataType.Type.INT256: "Int256", 359 exp.DataType.Type.LOWCARDINALITY: "LowCardinality", 360 exp.DataType.Type.MAP: "Map", 361 exp.DataType.Type.NESTED: "Nested", 362 exp.DataType.Type.NULLABLE: "Nullable", 363 exp.DataType.Type.SMALLINT: "Int16", 364 exp.DataType.Type.STRUCT: "Tuple", 365 exp.DataType.Type.TINYINT: "Int8", 366 exp.DataType.Type.UBIGINT: "UInt64", 367 exp.DataType.Type.UINT: "UInt32", 368 exp.DataType.Type.UINT128: "UInt128", 369 exp.DataType.Type.UINT256: "UInt256", 370 exp.DataType.Type.USMALLINT: "UInt16", 371 exp.DataType.Type.UTINYINT: "UInt8", 372 } 373 374 TRANSFORMS = { 375 **generator.Generator.TRANSFORMS, 376 exp.Select: transforms.preprocess([transforms.eliminate_qualify]), 377 exp.AnyValue: rename_func("any"), 378 exp.ApproxDistinct: rename_func("uniq"), 379 exp.ArgMax: arg_max_or_min_no_count("argMax"), 380 exp.ArgMin: arg_max_or_min_no_count("argMin"), 381 exp.Array: inline_array_sql, 382 exp.CastToStrType: rename_func("CAST"), 383 exp.CurrentDate: lambda self, e: self.func("CURRENT_DATE"), 384 exp.DateAdd: lambda self, e: self.func( 385 "DATE_ADD", exp.Literal.string(e.text("unit") or "day"), e.expression, e.this 386 ), 387 exp.DateDiff: lambda self, e: self.func( 388 "DATE_DIFF", exp.Literal.string(e.text("unit") or "day"), e.expression, e.this 389 ), 390 exp.Final: lambda self, e: f"{self.sql(e, 'this')} FINAL", 391 exp.IsNan: rename_func("isNaN"), 392 exp.Map: lambda self, e: _lower_func(var_map_sql(self, e)), 393 exp.Nullif: rename_func("nullIf"), 394 exp.PartitionedByProperty: lambda self, e: f"PARTITION BY {self.sql(e, 'this')}", 395 exp.Pivot: no_pivot_sql, 396 exp.Quantile: _quantile_sql, 397 exp.RegexpLike: lambda self, e: f"match({self.format_args(e.this, e.expression)})", 398 exp.StartsWith: rename_func("startsWith"), 399 exp.StrPosition: lambda self, e: f"position({self.format_args(e.this, e.args.get('substr'), e.args.get('position'))})", 400 exp.VarMap: lambda self, e: _lower_func(var_map_sql(self, e)), 401 exp.Xor: lambda self, e: self.func("xor", e.this, e.expression, *e.expressions), 402 } 403 404 PROPERTIES_LOCATION = { 405 **generator.Generator.PROPERTIES_LOCATION, 406 exp.VolatileProperty: exp.Properties.Location.UNSUPPORTED, 407 exp.PartitionedByProperty: exp.Properties.Location.POST_SCHEMA, 408 exp.OnCluster: exp.Properties.Location.POST_NAME, 409 } 410 411 JOIN_HINTS = False 412 TABLE_HINTS = False 413 EXPLICIT_UNION = True 414 GROUPINGS_SEP = "" 415 416 # there's no list in docs, but it can be found in Clickhouse code 417 # see `ClickHouse/src/Parsers/ParserCreate*.cpp` 418 ON_CLUSTER_TARGETS = { 419 "DATABASE", 420 "TABLE", 421 "VIEW", 422 "DICTIONARY", 423 "INDEX", 424 "FUNCTION", 425 "NAMED COLLECTION", 426 } 427 428 def _any_to_has( 429 self, 430 expression: exp.EQ | exp.NEQ, 431 default: t.Callable[[t.Any], str], 432 prefix: str = "", 433 ) -> str: 434 if isinstance(expression.left, exp.Any): 435 arr = expression.left 436 this = expression.right 437 elif isinstance(expression.right, exp.Any): 438 arr = expression.right 439 this = expression.left 440 else: 441 return default(expression) 442 return prefix + self.func("has", arr.this.unnest(), this) 443 444 def eq_sql(self, expression: exp.EQ) -> str: 445 return self._any_to_has(expression, super().eq_sql) 446 447 def neq_sql(self, expression: exp.NEQ) -> str: 448 return self._any_to_has(expression, super().neq_sql, "NOT ") 449 450 def regexpilike_sql(self, expression: exp.RegexpILike) -> str: 451 # Manually add a flag to make the search case-insensitive 452 regex = self.func("CONCAT", "'(?i)'", expression.expression) 453 return f"match({self.format_args(expression.this, regex)})" 454 455 def datatype_sql(self, expression: exp.DataType) -> str: 456 # String is the standard ClickHouse type, every other variant is just an alias. 457 # Additionally, any supplied length parameter will be ignored. 458 # 459 # https://clickhouse.com/docs/en/sql-reference/data-types/string 460 if expression.this in self.STRING_TYPE_MAPPING: 461 return "String" 462 463 return super().datatype_sql(expression) 464 465 def safeconcat_sql(self, expression: exp.SafeConcat) -> str: 466 # Clickhouse errors out if we try to cast a NULL value to TEXT 467 return self.func( 468 "CONCAT", 469 *[ 470 exp.func("if", e.is_(exp.null()), e, exp.cast(e, "text")) 471 for e in t.cast(t.List[exp.Condition], expression.expressions) 472 ], 473 ) 474 475 def cte_sql(self, expression: exp.CTE) -> str: 476 if isinstance(expression.this, exp.Alias): 477 return self.sql(expression, "this") 478 479 return super().cte_sql(expression) 480 481 def after_limit_modifiers(self, expression: exp.Expression) -> t.List[str]: 482 return super().after_limit_modifiers(expression) + [ 483 self.seg("SETTINGS ") + self.expressions(expression, key="settings", flat=True) 484 if expression.args.get("settings") 485 else "", 486 self.seg("FORMAT ") + self.sql(expression, "format") 487 if expression.args.get("format") 488 else "", 489 ] 490 491 def parameterizedagg_sql(self, expression: exp.Anonymous) -> str: 492 params = self.expressions(expression, key="params", flat=True) 493 return self.func(expression.name, *expression.expressions) + f"({params})" 494 495 def placeholder_sql(self, expression: exp.Placeholder) -> str: 496 return f"{{{expression.name}: {self.sql(expression, 'kind')}}}" 497 498 def oncluster_sql(self, expression: exp.OnCluster) -> str: 499 return f"ON CLUSTER {self.sql(expression, 'this')}" 500 501 def createable_sql(self, expression: exp.Create, locations: t.DefaultDict) -> str: 502 kind = self.sql(expression, "kind").upper() 503 if kind in self.ON_CLUSTER_TARGETS and locations.get(exp.Properties.Location.POST_NAME): 504 this_name = self.sql(expression.this, "this") 505 this_properties = " ".join( 506 [self.sql(prop) for prop in locations[exp.Properties.Location.POST_NAME]] 507 ) 508 this_schema = self.schema_columns_sql(expression.this) 509 return f"{this_name}{self.sep()}{this_properties}{self.sep()}{this_schema}" 510 511 return super().createable_sql(expression, locations)
tokenizer_class =
<class 'ClickHouse.Tokenizer'>
parser_class =
<class 'ClickHouse.Parser'>
generator_class =
<class 'ClickHouse.Generator'>
Inherited Members
- sqlglot.dialects.dialect.Dialect
- INDEX_OFFSET
- UNNEST_COLUMN_ONLY
- ALIAS_POST_TABLESAMPLE
- RESOLVES_IDENTIFIERS_AS_UPPERCASE
- IDENTIFIERS_CAN_START_WITH_DIGIT
- DPIPE_IS_STRING_CONCAT
- SUPPORTS_SEMI_ANTI_JOIN
- LOG_BASE_FIRST
- TYPED_DIVISION
- DATE_FORMAT
- DATEINT_FORMAT
- TIME_FORMAT
- TIME_MAPPING
- FORMAT_MAPPING
- PSEUDOCOLUMNS
- get_or_raise
- format_time
- normalize_identifier
- case_sensitive
- can_identify
- quote_identifier
- parse
- parse_into
- generate
- transpile
- tokenize
- tokenizer
- parser
- generator
47 class Tokenizer(tokens.Tokenizer): 48 COMMENTS = ["--", "#", "#!", ("/*", "*/")] 49 IDENTIFIERS = ['"', "`"] 50 STRING_ESCAPES = ["'", "\\"] 51 BIT_STRINGS = [("0b", "")] 52 HEX_STRINGS = [("0x", ""), ("0X", "")] 53 HEREDOC_STRINGS = ["$"] 54 55 KEYWORDS = { 56 **tokens.Tokenizer.KEYWORDS, 57 "ATTACH": TokenType.COMMAND, 58 "DATETIME64": TokenType.DATETIME64, 59 "DICTIONARY": TokenType.DICTIONARY, 60 "ENUM": TokenType.ENUM, 61 "ENUM8": TokenType.ENUM8, 62 "ENUM16": TokenType.ENUM16, 63 "FINAL": TokenType.FINAL, 64 "FIXEDSTRING": TokenType.FIXEDSTRING, 65 "FLOAT32": TokenType.FLOAT, 66 "FLOAT64": TokenType.DOUBLE, 67 "GLOBAL": TokenType.GLOBAL, 68 "INT16": TokenType.SMALLINT, 69 "INT256": TokenType.INT256, 70 "INT32": TokenType.INT, 71 "INT64": TokenType.BIGINT, 72 "INT8": TokenType.TINYINT, 73 "LOWCARDINALITY": TokenType.LOWCARDINALITY, 74 "MAP": TokenType.MAP, 75 "NESTED": TokenType.NESTED, 76 "SAMPLE": TokenType.TABLE_SAMPLE, 77 "TUPLE": TokenType.STRUCT, 78 "UINT128": TokenType.UINT128, 79 "UINT16": TokenType.USMALLINT, 80 "UINT256": TokenType.UINT256, 81 "UINT32": TokenType.UINT, 82 "UINT64": TokenType.UBIGINT, 83 "UINT8": TokenType.UTINYINT, 84 } 85 86 SINGLE_TOKENS = { 87 **tokens.Tokenizer.SINGLE_TOKENS, 88 "$": TokenType.HEREDOC_STRING, 89 }
KEYWORDS =
{'{%': <TokenType.BLOCK_START: 'BLOCK_START'>, '{%+': <TokenType.BLOCK_START: 'BLOCK_START'>, '{%-': <TokenType.BLOCK_START: 'BLOCK_START'>, '%}': <TokenType.BLOCK_END: 'BLOCK_END'>, '+%}': <TokenType.BLOCK_END: 'BLOCK_END'>, '-%}': <TokenType.BLOCK_END: 'BLOCK_END'>, '{{+': <TokenType.BLOCK_START: 'BLOCK_START'>, '{{-': <TokenType.BLOCK_START: 'BLOCK_START'>, '+}}': <TokenType.BLOCK_END: 'BLOCK_END'>, '-}}': <TokenType.BLOCK_END: 'BLOCK_END'>, '/*+': <TokenType.HINT: 'HINT'>, '==': <TokenType.EQ: 'EQ'>, '::': <TokenType.DCOLON: 'DCOLON'>, '||': <TokenType.DPIPE: 'DPIPE'>, '>=': <TokenType.GTE: 'GTE'>, '<=': <TokenType.LTE: 'LTE'>, '<>': <TokenType.NEQ: 'NEQ'>, '!=': <TokenType.NEQ: 'NEQ'>, '<=>': <TokenType.NULLSAFE_EQ: 'NULLSAFE_EQ'>, '->': <TokenType.ARROW: 'ARROW'>, '->>': <TokenType.DARROW: 'DARROW'>, '=>': <TokenType.FARROW: 'FARROW'>, '#>': <TokenType.HASH_ARROW: 'HASH_ARROW'>, '#>>': <TokenType.DHASH_ARROW: 'DHASH_ARROW'>, '<->': <TokenType.LR_ARROW: 'LR_ARROW'>, '&&': <TokenType.DAMP: 'DAMP'>, '??': <TokenType.DQMARK: 'DQMARK'>, 'ALL': <TokenType.ALL: 'ALL'>, 'ALWAYS': <TokenType.ALWAYS: 'ALWAYS'>, 'AND': <TokenType.AND: 'AND'>, 'ANTI': <TokenType.ANTI: 'ANTI'>, 'ANY': <TokenType.ANY: 'ANY'>, 'ASC': <TokenType.ASC: 'ASC'>, 'AS': <TokenType.ALIAS: 'ALIAS'>, 'ASOF': <TokenType.ASOF: 'ASOF'>, 'AUTOINCREMENT': <TokenType.AUTO_INCREMENT: 'AUTO_INCREMENT'>, 'AUTO_INCREMENT': <TokenType.AUTO_INCREMENT: 'AUTO_INCREMENT'>, 'BEGIN': <TokenType.BEGIN: 'BEGIN'>, 'BETWEEN': <TokenType.BETWEEN: 'BETWEEN'>, 'CACHE': <TokenType.CACHE: 'CACHE'>, 'UNCACHE': <TokenType.UNCACHE: 'UNCACHE'>, 'CASE': <TokenType.CASE: 'CASE'>, 'CHARACTER SET': <TokenType.CHARACTER_SET: 'CHARACTER_SET'>, 'CLUSTER BY': <TokenType.CLUSTER_BY: 'CLUSTER_BY'>, 'COLLATE': <TokenType.COLLATE: 'COLLATE'>, 'COLUMN': <TokenType.COLUMN: 'COLUMN'>, 'COMMIT': <TokenType.COMMIT: 'COMMIT'>, 'CONNECT BY': <TokenType.CONNECT_BY: 'CONNECT_BY'>, 'CONSTRAINT': <TokenType.CONSTRAINT: 'CONSTRAINT'>, 'CREATE': <TokenType.CREATE: 'CREATE'>, 'CROSS': <TokenType.CROSS: 'CROSS'>, 'CUBE': <TokenType.CUBE: 'CUBE'>, 'CURRENT_DATE': <TokenType.CURRENT_DATE: 'CURRENT_DATE'>, 'CURRENT_TIME': <TokenType.CURRENT_TIME: 'CURRENT_TIME'>, 'CURRENT_TIMESTAMP': <TokenType.CURRENT_TIMESTAMP: 'CURRENT_TIMESTAMP'>, 'CURRENT_USER': <TokenType.CURRENT_USER: 'CURRENT_USER'>, 'DATABASE': <TokenType.DATABASE: 'DATABASE'>, 'DEFAULT': <TokenType.DEFAULT: 'DEFAULT'>, 'DELETE': <TokenType.DELETE: 'DELETE'>, 'DESC': <TokenType.DESC: 'DESC'>, 'DESCRIBE': <TokenType.DESCRIBE: 'DESCRIBE'>, 'DISTINCT': <TokenType.DISTINCT: 'DISTINCT'>, 'DISTRIBUTE BY': <TokenType.DISTRIBUTE_BY: 'DISTRIBUTE_BY'>, 'DIV': <TokenType.DIV: 'DIV'>, 'DROP': <TokenType.DROP: 'DROP'>, 'ELSE': <TokenType.ELSE: 'ELSE'>, 'END': <TokenType.END: 'END'>, 'ESCAPE': <TokenType.ESCAPE: 'ESCAPE'>, 'EXCEPT': <TokenType.EXCEPT: 'EXCEPT'>, 'EXECUTE': <TokenType.EXECUTE: 'EXECUTE'>, 'EXISTS': <TokenType.EXISTS: 'EXISTS'>, 'FALSE': <TokenType.FALSE: 'FALSE'>, 'FETCH': <TokenType.FETCH: 'FETCH'>, 'FILTER': <TokenType.FILTER: 'FILTER'>, 'FIRST': <TokenType.FIRST: 'FIRST'>, 'FULL': <TokenType.FULL: 'FULL'>, 'FUNCTION': <TokenType.FUNCTION: 'FUNCTION'>, 'FOR': <TokenType.FOR: 'FOR'>, 'FOREIGN KEY': <TokenType.FOREIGN_KEY: 'FOREIGN_KEY'>, 'FORMAT': <TokenType.FORMAT: 'FORMAT'>, 'FROM': <TokenType.FROM: 'FROM'>, 'GEOGRAPHY': <TokenType.GEOGRAPHY: 'GEOGRAPHY'>, 'GEOMETRY': <TokenType.GEOMETRY: 'GEOMETRY'>, 'GLOB': <TokenType.GLOB: 'GLOB'>, 'GROUP BY': <TokenType.GROUP_BY: 'GROUP_BY'>, 'GROUPING SETS': <TokenType.GROUPING_SETS: 'GROUPING_SETS'>, 'HAVING': <TokenType.HAVING: 'HAVING'>, 'ILIKE': <TokenType.ILIKE: 'ILIKE'>, 'IN': <TokenType.IN: 'IN'>, 'INDEX': <TokenType.INDEX: 'INDEX'>, 'INET': <TokenType.INET: 'INET'>, 'INNER': <TokenType.INNER: 'INNER'>, 'INSERT': <TokenType.INSERT: 'INSERT'>, 'INTERVAL': <TokenType.INTERVAL: 'INTERVAL'>, 'INTERSECT': <TokenType.INTERSECT: 'INTERSECT'>, 'INTO': <TokenType.INTO: 'INTO'>, 'IS': <TokenType.IS: 'IS'>, 'ISNULL': <TokenType.ISNULL: 'ISNULL'>, 'JOIN': <TokenType.JOIN: 'JOIN'>, 'KEEP': <TokenType.KEEP: 'KEEP'>, 'KILL': <TokenType.KILL: 'KILL'>, 'LATERAL': <TokenType.LATERAL: 'LATERAL'>, 'LEFT': <TokenType.LEFT: 'LEFT'>, 'LIKE': <TokenType.LIKE: 'LIKE'>, 'LIMIT': <TokenType.LIMIT: 'LIMIT'>, 'LOAD': <TokenType.LOAD: 'LOAD'>, 'LOCK': <TokenType.LOCK: 'LOCK'>, 'MERGE': <TokenType.MERGE: 'MERGE'>, 'NATURAL': <TokenType.NATURAL: 'NATURAL'>, 'NEXT': <TokenType.NEXT: 'NEXT'>, 'NOT': <TokenType.NOT: 'NOT'>, 'NOTNULL': <TokenType.NOTNULL: 'NOTNULL'>, 'NULL': <TokenType.NULL: 'NULL'>, 'OBJECT': <TokenType.OBJECT: 'OBJECT'>, 'OFFSET': <TokenType.OFFSET: 'OFFSET'>, 'ON': <TokenType.ON: 'ON'>, 'OR': <TokenType.OR: 'OR'>, 'XOR': <TokenType.XOR: 'XOR'>, 'ORDER BY': <TokenType.ORDER_BY: 'ORDER_BY'>, 'ORDINALITY': <TokenType.ORDINALITY: 'ORDINALITY'>, 'OUTER': <TokenType.OUTER: 'OUTER'>, 'OVER': <TokenType.OVER: 'OVER'>, 'OVERLAPS': <TokenType.OVERLAPS: 'OVERLAPS'>, 'OVERWRITE': <TokenType.OVERWRITE: 'OVERWRITE'>, 'PARTITION': <TokenType.PARTITION: 'PARTITION'>, 'PARTITION BY': <TokenType.PARTITION_BY: 'PARTITION_BY'>, 'PARTITIONED BY': <TokenType.PARTITION_BY: 'PARTITION_BY'>, 'PARTITIONED_BY': <TokenType.PARTITION_BY: 'PARTITION_BY'>, 'PERCENT': <TokenType.PERCENT: 'PERCENT'>, 'PIVOT': <TokenType.PIVOT: 'PIVOT'>, 'PRAGMA': <TokenType.PRAGMA: 'PRAGMA'>, 'PRIMARY KEY': <TokenType.PRIMARY_KEY: 'PRIMARY_KEY'>, 'PROCEDURE': <TokenType.PROCEDURE: 'PROCEDURE'>, 'QUALIFY': <TokenType.QUALIFY: 'QUALIFY'>, 'RANGE': <TokenType.RANGE: 'RANGE'>, 'RECURSIVE': <TokenType.RECURSIVE: 'RECURSIVE'>, 'REGEXP': <TokenType.RLIKE: 'RLIKE'>, 'REPLACE': <TokenType.REPLACE: 'REPLACE'>, 'RETURNING': <TokenType.RETURNING: 'RETURNING'>, 'REFERENCES': <TokenType.REFERENCES: 'REFERENCES'>, 'RIGHT': <TokenType.RIGHT: 'RIGHT'>, 'RLIKE': <TokenType.RLIKE: 'RLIKE'>, 'ROLLBACK': <TokenType.ROLLBACK: 'ROLLBACK'>, 'ROLLUP': <TokenType.ROLLUP: 'ROLLUP'>, 'ROW': <TokenType.ROW: 'ROW'>, 'ROWS': <TokenType.ROWS: 'ROWS'>, 'SCHEMA': <TokenType.SCHEMA: 'SCHEMA'>, 'SELECT': <TokenType.SELECT: 'SELECT'>, 'SEMI': <TokenType.SEMI: 'SEMI'>, 'SET': <TokenType.SET: 'SET'>, 'SETTINGS': <TokenType.SETTINGS: 'SETTINGS'>, 'SHOW': <TokenType.SHOW: 'SHOW'>, 'SIMILAR TO': <TokenType.SIMILAR_TO: 'SIMILAR_TO'>, 'SOME': <TokenType.SOME: 'SOME'>, 'SORT BY': <TokenType.SORT_BY: 'SORT_BY'>, 'START WITH': <TokenType.START_WITH: 'START_WITH'>, 'TABLE': <TokenType.TABLE: 'TABLE'>, 'TABLESAMPLE': <TokenType.TABLE_SAMPLE: 'TABLE_SAMPLE'>, 'TEMP': <TokenType.TEMPORARY: 'TEMPORARY'>, 'TEMPORARY': <TokenType.TEMPORARY: 'TEMPORARY'>, 'THEN': <TokenType.THEN: 'THEN'>, 'TRUE': <TokenType.TRUE: 'TRUE'>, 'UNION': <TokenType.UNION: 'UNION'>, 'UNKNOWN': <TokenType.UNKNOWN: 'UNKNOWN'>, 'UNNEST': <TokenType.UNNEST: 'UNNEST'>, 'UNPIVOT': <TokenType.UNPIVOT: 'UNPIVOT'>, 'UPDATE': <TokenType.UPDATE: 'UPDATE'>, 'USE': <TokenType.USE: 'USE'>, 'USING': <TokenType.USING: 'USING'>, 'UUID': <TokenType.UUID: 'UUID'>, 'VALUES': <TokenType.VALUES: 'VALUES'>, 'VIEW': <TokenType.VIEW: 'VIEW'>, 'VOLATILE': <TokenType.VOLATILE: 'VOLATILE'>, 'WHEN': <TokenType.WHEN: 'WHEN'>, 'WHERE': <TokenType.WHERE: 'WHERE'>, 'WINDOW': <TokenType.WINDOW: 'WINDOW'>, 'WITH': <TokenType.WITH: 'WITH'>, 'APPLY': <TokenType.APPLY: 'APPLY'>, 'ARRAY': <TokenType.ARRAY: 'ARRAY'>, 'BIT': <TokenType.BIT: 'BIT'>, 'BOOL': <TokenType.BOOLEAN: 'BOOLEAN'>, 'BOOLEAN': <TokenType.BOOLEAN: 'BOOLEAN'>, 'BYTE': <TokenType.TINYINT: 'TINYINT'>, 'MEDIUMINT': <TokenType.MEDIUMINT: 'MEDIUMINT'>, 'TINYINT': <TokenType.TINYINT: 'TINYINT'>, 'SHORT': <TokenType.SMALLINT: 'SMALLINT'>, 'SMALLINT': <TokenType.SMALLINT: 'SMALLINT'>, 'INT128': <TokenType.INT128: 'INT128'>, 'INT2': <TokenType.SMALLINT: 'SMALLINT'>, 'INTEGER': <TokenType.INT: 'INT'>, 'INT': <TokenType.INT: 'INT'>, 'INT4': <TokenType.INT: 'INT'>, 'LONG': <TokenType.BIGINT: 'BIGINT'>, 'BIGINT': <TokenType.BIGINT: 'BIGINT'>, 'INT8': <TokenType.TINYINT: 'TINYINT'>, 'DEC': <TokenType.DECIMAL: 'DECIMAL'>, 'DECIMAL': <TokenType.DECIMAL: 'DECIMAL'>, 'BIGDECIMAL': <TokenType.BIGDECIMAL: 'BIGDECIMAL'>, 'BIGNUMERIC': <TokenType.BIGDECIMAL: 'BIGDECIMAL'>, 'MAP': <TokenType.MAP: 'MAP'>, 'NULLABLE': <TokenType.NULLABLE: 'NULLABLE'>, 'NUMBER': <TokenType.DECIMAL: 'DECIMAL'>, 'NUMERIC': <TokenType.DECIMAL: 'DECIMAL'>, 'FIXED': <TokenType.DECIMAL: 'DECIMAL'>, 'REAL': <TokenType.FLOAT: 'FLOAT'>, 'FLOAT': <TokenType.FLOAT: 'FLOAT'>, 'FLOAT4': <TokenType.FLOAT: 'FLOAT'>, 'FLOAT8': <TokenType.DOUBLE: 'DOUBLE'>, 'DOUBLE': <TokenType.DOUBLE: 'DOUBLE'>, 'DOUBLE PRECISION': <TokenType.DOUBLE: 'DOUBLE'>, 'JSON': <TokenType.JSON: 'JSON'>, 'CHAR': <TokenType.CHAR: 'CHAR'>, 'CHARACTER': <TokenType.CHAR: 'CHAR'>, 'NCHAR': <TokenType.NCHAR: 'NCHAR'>, 'VARCHAR': <TokenType.VARCHAR: 'VARCHAR'>, 'VARCHAR2': <TokenType.VARCHAR: 'VARCHAR'>, 'NVARCHAR': <TokenType.NVARCHAR: 'NVARCHAR'>, 'NVARCHAR2': <TokenType.NVARCHAR: 'NVARCHAR'>, 'STR': <TokenType.TEXT: 'TEXT'>, 'STRING': <TokenType.TEXT: 'TEXT'>, 'TEXT': <TokenType.TEXT: 'TEXT'>, 'LONGTEXT': <TokenType.LONGTEXT: 'LONGTEXT'>, 'MEDIUMTEXT': <TokenType.MEDIUMTEXT: 'MEDIUMTEXT'>, 'TINYTEXT': <TokenType.TINYTEXT: 'TINYTEXT'>, 'CLOB': <TokenType.TEXT: 'TEXT'>, 'LONGVARCHAR': <TokenType.TEXT: 'TEXT'>, 'BINARY': <TokenType.BINARY: 'BINARY'>, 'BLOB': <TokenType.VARBINARY: 'VARBINARY'>, 'LONGBLOB': <TokenType.LONGBLOB: 'LONGBLOB'>, 'MEDIUMBLOB': <TokenType.MEDIUMBLOB: 'MEDIUMBLOB'>, 'TINYBLOB': <TokenType.TINYBLOB: 'TINYBLOB'>, 'BYTEA': <TokenType.VARBINARY: 'VARBINARY'>, 'VARBINARY': <TokenType.VARBINARY: 'VARBINARY'>, 'TIME': <TokenType.TIME: 'TIME'>, 'TIMETZ': <TokenType.TIMETZ: 'TIMETZ'>, 'TIMESTAMP': <TokenType.TIMESTAMP: 'TIMESTAMP'>, 'TIMESTAMPTZ': <TokenType.TIMESTAMPTZ: 'TIMESTAMPTZ'>, 'TIMESTAMPLTZ': <TokenType.TIMESTAMPLTZ: 'TIMESTAMPLTZ'>, 'DATE': <TokenType.DATE: 'DATE'>, 'DATETIME': <TokenType.DATETIME: 'DATETIME'>, 'INT4RANGE': <TokenType.INT4RANGE: 'INT4RANGE'>, 'INT4MULTIRANGE': <TokenType.INT4MULTIRANGE: 'INT4MULTIRANGE'>, 'INT8RANGE': <TokenType.INT8RANGE: 'INT8RANGE'>, 'INT8MULTIRANGE': <TokenType.INT8MULTIRANGE: 'INT8MULTIRANGE'>, 'NUMRANGE': <TokenType.NUMRANGE: 'NUMRANGE'>, 'NUMMULTIRANGE': <TokenType.NUMMULTIRANGE: 'NUMMULTIRANGE'>, 'TSRANGE': <TokenType.TSRANGE: 'TSRANGE'>, 'TSMULTIRANGE': <TokenType.TSMULTIRANGE: 'TSMULTIRANGE'>, 'TSTZRANGE': <TokenType.TSTZRANGE: 'TSTZRANGE'>, 'TSTZMULTIRANGE': <TokenType.TSTZMULTIRANGE: 'TSTZMULTIRANGE'>, 'DATERANGE': <TokenType.DATERANGE: 'DATERANGE'>, 'DATEMULTIRANGE': <TokenType.DATEMULTIRANGE: 'DATEMULTIRANGE'>, 'UNIQUE': <TokenType.UNIQUE: 'UNIQUE'>, 'STRUCT': <TokenType.STRUCT: 'STRUCT'>, 'VARIANT': <TokenType.VARIANT: 'VARIANT'>, 'ALTER': <TokenType.ALTER: 'ALTER'>, 'ANALYZE': <TokenType.COMMAND: 'COMMAND'>, 'CALL': <TokenType.COMMAND: 'COMMAND'>, 'COMMENT': <TokenType.COMMENT: 'COMMENT'>, 'COPY': <TokenType.COMMAND: 'COMMAND'>, 'EXPLAIN': <TokenType.COMMAND: 'COMMAND'>, 'GRANT': <TokenType.COMMAND: 'COMMAND'>, 'OPTIMIZE': <TokenType.COMMAND: 'COMMAND'>, 'PREPARE': <TokenType.COMMAND: 'COMMAND'>, 'TRUNCATE': <TokenType.COMMAND: 'COMMAND'>, 'VACUUM': <TokenType.COMMAND: 'COMMAND'>, 'USER-DEFINED': <TokenType.USERDEFINED: 'USERDEFINED'>, 'FOR VERSION': <TokenType.VERSION_SNAPSHOT: 'VERSION_SNAPSHOT'>, 'FOR TIMESTAMP': <TokenType.TIMESTAMP_SNAPSHOT: 'TIMESTAMP_SNAPSHOT'>, 'ATTACH': <TokenType.COMMAND: 'COMMAND'>, 'DATETIME64': <TokenType.DATETIME64: 'DATETIME64'>, 'DICTIONARY': <TokenType.DICTIONARY: 'DICTIONARY'>, 'ENUM': <TokenType.ENUM: 'ENUM'>, 'ENUM8': <TokenType.ENUM8: 'ENUM8'>, 'ENUM16': <TokenType.ENUM16: 'ENUM16'>, 'FINAL': <TokenType.FINAL: 'FINAL'>, 'FIXEDSTRING': <TokenType.FIXEDSTRING: 'FIXEDSTRING'>, 'FLOAT32': <TokenType.FLOAT: 'FLOAT'>, 'FLOAT64': <TokenType.DOUBLE: 'DOUBLE'>, 'GLOBAL': <TokenType.GLOBAL: 'GLOBAL'>, 'INT16': <TokenType.SMALLINT: 'SMALLINT'>, 'INT256': <TokenType.INT256: 'INT256'>, 'INT32': <TokenType.INT: 'INT'>, 'INT64': <TokenType.BIGINT: 'BIGINT'>, 'LOWCARDINALITY': <TokenType.LOWCARDINALITY: 'LOWCARDINALITY'>, 'NESTED': <TokenType.NESTED: 'NESTED'>, 'SAMPLE': <TokenType.TABLE_SAMPLE: 'TABLE_SAMPLE'>, 'TUPLE': <TokenType.STRUCT: 'STRUCT'>, 'UINT128': <TokenType.UINT128: 'UINT128'>, 'UINT16': <TokenType.USMALLINT: 'USMALLINT'>, 'UINT256': <TokenType.UINT256: 'UINT256'>, 'UINT32': <TokenType.UINT: 'UINT'>, 'UINT64': <TokenType.UBIGINT: 'UBIGINT'>, 'UINT8': <TokenType.UTINYINT: 'UTINYINT'>}
SINGLE_TOKENS =
{'(': <TokenType.L_PAREN: 'L_PAREN'>, ')': <TokenType.R_PAREN: 'R_PAREN'>, '[': <TokenType.L_BRACKET: 'L_BRACKET'>, ']': <TokenType.R_BRACKET: 'R_BRACKET'>, '{': <TokenType.L_BRACE: 'L_BRACE'>, '}': <TokenType.R_BRACE: 'R_BRACE'>, '&': <TokenType.AMP: 'AMP'>, '^': <TokenType.CARET: 'CARET'>, ':': <TokenType.COLON: 'COLON'>, ',': <TokenType.COMMA: 'COMMA'>, '.': <TokenType.DOT: 'DOT'>, '-': <TokenType.DASH: 'DASH'>, '=': <TokenType.EQ: 'EQ'>, '>': <TokenType.GT: 'GT'>, '<': <TokenType.LT: 'LT'>, '%': <TokenType.MOD: 'MOD'>, '!': <TokenType.NOT: 'NOT'>, '|': <TokenType.PIPE: 'PIPE'>, '+': <TokenType.PLUS: 'PLUS'>, ';': <TokenType.SEMICOLON: 'SEMICOLON'>, '/': <TokenType.SLASH: 'SLASH'>, '\\': <TokenType.BACKSLASH: 'BACKSLASH'>, '*': <TokenType.STAR: 'STAR'>, '~': <TokenType.TILDA: 'TILDA'>, '?': <TokenType.PLACEHOLDER: 'PLACEHOLDER'>, '@': <TokenType.PARAMETER: 'PARAMETER'>, "'": <TokenType.QUOTE: 'QUOTE'>, '`': <TokenType.IDENTIFIER: 'IDENTIFIER'>, '"': <TokenType.IDENTIFIER: 'IDENTIFIER'>, '#': <TokenType.HASH: 'HASH'>, '$': <TokenType.HEREDOC_STRING: 'HEREDOC_STRING'>}
91 class Parser(parser.Parser): 92 FUNCTIONS = { 93 **parser.Parser.FUNCTIONS, 94 "ANY": exp.AnyValue.from_arg_list, 95 "DATE_ADD": lambda args: exp.DateAdd( 96 this=seq_get(args, 2), expression=seq_get(args, 1), unit=seq_get(args, 0) 97 ), 98 "DATEADD": lambda args: exp.DateAdd( 99 this=seq_get(args, 2), expression=seq_get(args, 1), unit=seq_get(args, 0) 100 ), 101 "DATE_DIFF": lambda args: exp.DateDiff( 102 this=seq_get(args, 2), expression=seq_get(args, 1), unit=seq_get(args, 0) 103 ), 104 "DATEDIFF": lambda args: exp.DateDiff( 105 this=seq_get(args, 2), expression=seq_get(args, 1), unit=seq_get(args, 0) 106 ), 107 "MAP": parse_var_map, 108 "MATCH": exp.RegexpLike.from_arg_list, 109 "UNIQ": exp.ApproxDistinct.from_arg_list, 110 "XOR": lambda args: exp.Xor(expressions=args), 111 } 112 113 FUNCTIONS_WITH_ALIASED_ARGS = {*parser.Parser.FUNCTIONS_WITH_ALIASED_ARGS, "TUPLE"} 114 115 FUNCTION_PARSERS = { 116 **parser.Parser.FUNCTION_PARSERS, 117 "QUANTILE": lambda self: self._parse_quantile(), 118 } 119 120 FUNCTION_PARSERS.pop("MATCH") 121 122 NO_PAREN_FUNCTION_PARSERS = parser.Parser.NO_PAREN_FUNCTION_PARSERS.copy() 123 NO_PAREN_FUNCTION_PARSERS.pop("ANY") 124 125 RANGE_PARSERS = { 126 **parser.Parser.RANGE_PARSERS, 127 TokenType.GLOBAL: lambda self, this: self._match(TokenType.IN) 128 and self._parse_in(this, is_global=True), 129 } 130 131 # The PLACEHOLDER entry is popped because 1) it doesn't affect Clickhouse (it corresponds to 132 # the postgres-specific JSONBContains parser) and 2) it makes parsing the ternary op simpler. 133 COLUMN_OPERATORS = parser.Parser.COLUMN_OPERATORS.copy() 134 COLUMN_OPERATORS.pop(TokenType.PLACEHOLDER) 135 136 JOIN_KINDS = { 137 *parser.Parser.JOIN_KINDS, 138 TokenType.ANY, 139 TokenType.ASOF, 140 TokenType.ARRAY, 141 } 142 143 TABLE_ALIAS_TOKENS = parser.Parser.TABLE_ALIAS_TOKENS - { 144 TokenType.ANY, 145 TokenType.SETTINGS, 146 TokenType.FORMAT, 147 TokenType.ARRAY, 148 } 149 150 LOG_DEFAULTS_TO_LN = True 151 152 QUERY_MODIFIER_PARSERS = { 153 **parser.Parser.QUERY_MODIFIER_PARSERS, 154 TokenType.SETTINGS: lambda self: ( 155 "settings", 156 self._advance() or self._parse_csv(self._parse_conjunction), 157 ), 158 TokenType.FORMAT: lambda self: ("format", self._advance() or self._parse_id_var()), 159 } 160 161 def _parse_conjunction(self) -> t.Optional[exp.Expression]: 162 this = super()._parse_conjunction() 163 164 if self._match(TokenType.PLACEHOLDER): 165 return self.expression( 166 exp.If, 167 this=this, 168 true=self._parse_conjunction(), 169 false=self._match(TokenType.COLON) and self._parse_conjunction(), 170 ) 171 172 return this 173 174 def _parse_placeholder(self) -> t.Optional[exp.Expression]: 175 """ 176 Parse a placeholder expression like SELECT {abc: UInt32} or FROM {table: Identifier} 177 https://clickhouse.com/docs/en/sql-reference/syntax#defining-and-using-query-parameters 178 """ 179 if not self._match(TokenType.L_BRACE): 180 return None 181 182 this = self._parse_id_var() 183 self._match(TokenType.COLON) 184 kind = self._parse_types(check_func=False, allow_identifiers=False) or ( 185 self._match_text_seq("IDENTIFIER") and "Identifier" 186 ) 187 188 if not kind: 189 self.raise_error("Expecting a placeholder type or 'Identifier' for tables") 190 elif not self._match(TokenType.R_BRACE): 191 self.raise_error("Expecting }") 192 193 return self.expression(exp.Placeholder, this=this, kind=kind) 194 195 def _parse_in(self, this: t.Optional[exp.Expression], is_global: bool = False) -> exp.In: 196 this = super()._parse_in(this) 197 this.set("is_global", is_global) 198 return this 199 200 def _parse_table( 201 self, 202 schema: bool = False, 203 joins: bool = False, 204 alias_tokens: t.Optional[t.Collection[TokenType]] = None, 205 parse_bracket: bool = False, 206 ) -> t.Optional[exp.Expression]: 207 this = super()._parse_table( 208 schema=schema, joins=joins, alias_tokens=alias_tokens, parse_bracket=parse_bracket 209 ) 210 211 if self._match(TokenType.FINAL): 212 this = self.expression(exp.Final, this=this) 213 214 return this 215 216 def _parse_position(self, haystack_first: bool = False) -> exp.StrPosition: 217 return super()._parse_position(haystack_first=True) 218 219 # https://clickhouse.com/docs/en/sql-reference/statements/select/with/ 220 def _parse_cte(self) -> exp.CTE: 221 index = self._index 222 try: 223 # WITH <identifier> AS <subquery expression> 224 return super()._parse_cte() 225 except ParseError: 226 # WITH <expression> AS <identifier> 227 self._retreat(index) 228 statement = self._parse_statement() 229 230 if statement and isinstance(statement.this, exp.Alias): 231 self.raise_error("Expected CTE to have alias") 232 233 return self.expression(exp.CTE, this=statement, alias=statement and statement.this) 234 235 def _parse_join_parts( 236 self, 237 ) -> t.Tuple[t.Optional[Token], t.Optional[Token], t.Optional[Token]]: 238 is_global = self._match(TokenType.GLOBAL) and self._prev 239 kind_pre = self._match_set(self.JOIN_KINDS, advance=False) and self._prev 240 241 if kind_pre: 242 kind = self._match_set(self.JOIN_KINDS) and self._prev 243 side = self._match_set(self.JOIN_SIDES) and self._prev 244 return is_global, side, kind 245 246 return ( 247 is_global, 248 self._match_set(self.JOIN_SIDES) and self._prev, 249 self._match_set(self.JOIN_KINDS) and self._prev, 250 ) 251 252 def _parse_join( 253 self, skip_join_token: bool = False, parse_bracket: bool = False 254 ) -> t.Optional[exp.Join]: 255 join = super()._parse_join(skip_join_token=skip_join_token, parse_bracket=True) 256 257 if join: 258 join.set("global", join.args.pop("method", None)) 259 return join 260 261 def _parse_function( 262 self, 263 functions: t.Optional[t.Dict[str, t.Callable]] = None, 264 anonymous: bool = False, 265 optional_parens: bool = True, 266 ) -> t.Optional[exp.Expression]: 267 func = super()._parse_function( 268 functions=functions, anonymous=anonymous, optional_parens=optional_parens 269 ) 270 271 if isinstance(func, exp.Anonymous): 272 params = self._parse_func_params(func) 273 274 if params: 275 return self.expression( 276 exp.ParameterizedAgg, 277 this=func.this, 278 expressions=func.expressions, 279 params=params, 280 ) 281 282 return func 283 284 def _parse_func_params( 285 self, this: t.Optional[exp.Func] = None 286 ) -> t.Optional[t.List[exp.Expression]]: 287 if self._match_pair(TokenType.R_PAREN, TokenType.L_PAREN): 288 return self._parse_csv(self._parse_lambda) 289 290 if self._match(TokenType.L_PAREN): 291 params = self._parse_csv(self._parse_lambda) 292 self._match_r_paren(this) 293 return params 294 295 return None 296 297 def _parse_quantile(self) -> exp.Quantile: 298 this = self._parse_lambda() 299 params = self._parse_func_params() 300 if params: 301 return self.expression(exp.Quantile, this=params[0], quantile=this) 302 return self.expression(exp.Quantile, this=this, quantile=exp.Literal.number(0.5)) 303 304 def _parse_wrapped_id_vars(self, optional: bool = False) -> t.List[exp.Expression]: 305 return super()._parse_wrapped_id_vars(optional=True) 306 307 def _parse_primary_key( 308 self, wrapped_optional: bool = False, in_props: bool = False 309 ) -> exp.PrimaryKeyColumnConstraint | exp.PrimaryKey: 310 return super()._parse_primary_key( 311 wrapped_optional=wrapped_optional or in_props, in_props=in_props 312 ) 313 314 def _parse_on_property(self) -> t.Optional[exp.Expression]: 315 index = self._index 316 if self._match_text_seq("CLUSTER"): 317 this = self._parse_id_var() 318 if this: 319 return self.expression(exp.OnCluster, this=this) 320 else: 321 self._retreat(index) 322 return None
Parser consumes a list of tokens produced by the Tokenizer and produces a parsed syntax tree.
Arguments:
- error_level: The desired error level. Default: ErrorLevel.IMMEDIATE
- error_message_context: Determines the amount of context to capture from a query string when displaying the error message (in number of characters). Default: 100
- max_errors: Maximum number of error messages to include in a raised ParseError. This is only relevant if error_level is ErrorLevel.RAISE. Default: 3
FUNCTIONS =
{'ABS': <bound method Func.from_arg_list of <class 'sqlglot.expressions.Abs'>>, 'ANY_VALUE': <bound method Func.from_arg_list of <class 'sqlglot.expressions.AnyValue'>>, 'APPROX_DISTINCT': <bound method Func.from_arg_list of <class 'sqlglot.expressions.ApproxDistinct'>>, 'APPROX_COUNT_DISTINCT': <bound method Func.from_arg_list of <class 'sqlglot.expressions.ApproxDistinct'>>, 'APPROX_QUANTILE': <bound method Func.from_arg_list of <class 'sqlglot.expressions.ApproxQuantile'>>, 'APPROX_TOP_K': <bound method Func.from_arg_list of <class 'sqlglot.expressions.ApproxTopK'>>, 'ARG_MAX': <bound method Func.from_arg_list of <class 'sqlglot.expressions.ArgMax'>>, 'ARGMAX': <bound method Func.from_arg_list of <class 'sqlglot.expressions.ArgMax'>>, 'MAX_BY': <bound method Func.from_arg_list of <class 'sqlglot.expressions.ArgMax'>>, 'ARG_MIN': <bound method Func.from_arg_list of <class 'sqlglot.expressions.ArgMin'>>, 'ARGMIN': <bound method Func.from_arg_list of <class 'sqlglot.expressions.ArgMin'>>, 'MIN_BY': <bound method Func.from_arg_list of <class 'sqlglot.expressions.ArgMin'>>, 'ARRAY': <bound method Func.from_arg_list of <class 'sqlglot.expressions.Array'>>, 'ARRAY_AGG': <bound method Func.from_arg_list of <class 'sqlglot.expressions.ArrayAgg'>>, 'ARRAY_ALL': <bound method Func.from_arg_list of <class 'sqlglot.expressions.ArrayAll'>>, 'ARRAY_ANY': <bound method Func.from_arg_list of <class 'sqlglot.expressions.ArrayAny'>>, 'ARRAY_CONCAT': <bound method Func.from_arg_list of <class 'sqlglot.expressions.ArrayConcat'>>, 'ARRAY_CAT': <bound method Func.from_arg_list of <class 'sqlglot.expressions.ArrayConcat'>>, 'ARRAY_CONTAINS': <bound method Func.from_arg_list of <class 'sqlglot.expressions.ArrayContains'>>, 'FILTER': <bound method Func.from_arg_list of <class 'sqlglot.expressions.ArrayFilter'>>, 'ARRAY_FILTER': <bound method Func.from_arg_list of <class 'sqlglot.expressions.ArrayFilter'>>, 'ARRAY_JOIN': <bound method Func.from_arg_list of <class 'sqlglot.expressions.ArrayJoin'>>, 'ARRAY_SIZE': <bound method Func.from_arg_list of <class 'sqlglot.expressions.ArraySize'>>, 'ARRAY_SORT': <bound method Func.from_arg_list of <class 'sqlglot.expressions.ArraySort'>>, 'ARRAY_SUM': <bound method Func.from_arg_list of <class 'sqlglot.expressions.ArraySum'>>, 'ARRAY_UNION_AGG': <bound method Func.from_arg_list of <class 'sqlglot.expressions.ArrayUnionAgg'>>, 'AVG': <bound method Func.from_arg_list of <class 'sqlglot.expressions.Avg'>>, 'CASE': <bound method Func.from_arg_list of <class 'sqlglot.expressions.Case'>>, 'CAST': <bound method Func.from_arg_list of <class 'sqlglot.expressions.Cast'>>, 'CAST_TO_STR_TYPE': <bound method Func.from_arg_list of <class 'sqlglot.expressions.CastToStrType'>>, 'CEIL': <bound method Func.from_arg_list of <class 'sqlglot.expressions.Ceil'>>, 'CEILING': <bound method Func.from_arg_list of <class 'sqlglot.expressions.Ceil'>>, 'CHR': <bound method Func.from_arg_list of <class 'sqlglot.expressions.Chr'>>, 'CHAR': <bound method Func.from_arg_list of <class 'sqlglot.expressions.Chr'>>, 'COALESCE': <bound method Func.from_arg_list of <class 'sqlglot.expressions.Coalesce'>>, 'IFNULL': <bound method Func.from_arg_list of <class 'sqlglot.expressions.Coalesce'>>, 'NVL': <bound method Func.from_arg_list of <class 'sqlglot.expressions.Coalesce'>>, 'COLLATE': <bound method Func.from_arg_list of <class 'sqlglot.expressions.Collate'>>, 'CONCAT': <bound method Func.from_arg_list of <class 'sqlglot.expressions.Concat'>>, 'CONCAT_WS': <bound method Func.from_arg_list of <class 'sqlglot.expressions.ConcatWs'>>, 'COUNT': <bound method Func.from_arg_list of <class 'sqlglot.expressions.Count'>>, 'COUNT_IF': <bound method Func.from_arg_list of <class 'sqlglot.expressions.CountIf'>>, 'CURRENT_DATE': <bound method Func.from_arg_list of <class 'sqlglot.expressions.CurrentDate'>>, 'CURRENT_DATETIME': <bound method Func.from_arg_list of <class 'sqlglot.expressions.CurrentDatetime'>>, 'CURRENT_TIME': <bound method Func.from_arg_list of <class 'sqlglot.expressions.CurrentTime'>>, 'CURRENT_TIMESTAMP': <bound method Func.from_arg_list of <class 'sqlglot.expressions.CurrentTimestamp'>>, 'CURRENT_USER': <bound method Func.from_arg_list of <class 'sqlglot.expressions.CurrentUser'>>, 'DATE': <bound method Func.from_arg_list of <class 'sqlglot.expressions.Date'>>, 'DATE_ADD': <function ClickHouse.Parser.<lambda>>, 'DATEDIFF': <function ClickHouse.Parser.<lambda>>, 'DATE_DIFF': <function ClickHouse.Parser.<lambda>>, 'DATEFROMPARTS': <bound method Func.from_arg_list of <class 'sqlglot.expressions.DateFromParts'>>, 'DATE_STR_TO_DATE': <bound method Func.from_arg_list of <class 'sqlglot.expressions.DateStrToDate'>>, 'DATE_SUB': <bound method Func.from_arg_list of <class 'sqlglot.expressions.DateSub'>>, 'DATE_TO_DATE_STR': <function Parser.<lambda>>, 'DATE_TO_DI': <bound method Func.from_arg_list of <class 'sqlglot.expressions.DateToDi'>>, 'DATE_TRUNC': <bound method Func.from_arg_list of <class 'sqlglot.expressions.DateTrunc'>>, 'DATETIME_ADD': <bound method Func.from_arg_list of <class 'sqlglot.expressions.DatetimeAdd'>>, 'DATETIME_DIFF': <bound method Func.from_arg_list of <class 'sqlglot.expressions.DatetimeDiff'>>, 'DATETIME_SUB': <bound method Func.from_arg_list of <class 'sqlglot.expressions.DatetimeSub'>>, 'DATETIME_TRUNC': <bound method Func.from_arg_list of <class 'sqlglot.expressions.DatetimeTrunc'>>, 'DAY': <bound method Func.from_arg_list of <class 'sqlglot.expressions.Day'>>, 'DAY_OF_MONTH': <bound method Func.from_arg_list of <class 'sqlglot.expressions.DayOfMonth'>>, 'DAYOFMONTH': <bound method Func.from_arg_list of <class 'sqlglot.expressions.DayOfMonth'>>, 'DAY_OF_WEEK': <bound method Func.from_arg_list of <class 'sqlglot.expressions.DayOfWeek'>>, 'DAYOFWEEK': <bound method Func.from_arg_list of <class 'sqlglot.expressions.DayOfWeek'>>, 'DAY_OF_YEAR': <bound method Func.from_arg_list of <class 'sqlglot.expressions.DayOfYear'>>, 'DAYOFYEAR': <bound method Func.from_arg_list of <class 'sqlglot.expressions.DayOfYear'>>, 'DECODE': <bound method Func.from_arg_list of <class 'sqlglot.expressions.Decode'>>, 'DI_TO_DATE': <bound method Func.from_arg_list of <class 'sqlglot.expressions.DiToDate'>>, 'ENCODE': <bound method Func.from_arg_list of <class 'sqlglot.expressions.Encode'>>, 'EXP': <bound method Func.from_arg_list of <class 'sqlglot.expressions.Exp'>>, 'EXPLODE': <bound method Func.from_arg_list of <class 'sqlglot.expressions.Explode'>>, 'EXPLODE_OUTER': <bound method Func.from_arg_list of <class 'sqlglot.expressions.ExplodeOuter'>>, 'EXTRACT': <bound method Func.from_arg_list of <class 'sqlglot.expressions.Extract'>>, 'FIRST': <bound method Func.from_arg_list of <class 'sqlglot.expressions.First'>>, 'FLATTEN': <bound method Func.from_arg_list of <class 'sqlglot.expressions.Flatten'>>, 'FLOOR': <bound method Func.from_arg_list of <class 'sqlglot.expressions.Floor'>>, 'FROM_BASE': <bound method Func.from_arg_list of <class 'sqlglot.expressions.FromBase'>>, 'FROM_BASE64': <bound method Func.from_arg_list of <class 'sqlglot.expressions.FromBase64'>>, 'GENERATE_SERIES': <bound method Func.from_arg_list of <class 'sqlglot.expressions.GenerateSeries'>>, 'GREATEST': <bound method Func.from_arg_list of <class 'sqlglot.expressions.Greatest'>>, 'GROUP_CONCAT': <bound method Func.from_arg_list of <class 'sqlglot.expressions.GroupConcat'>>, 'HEX': <bound method Func.from_arg_list of <class 'sqlglot.expressions.Hex'>>, 'HLL': <bound method Func.from_arg_list of <class 'sqlglot.expressions.Hll'>>, 'IF': <bound method Func.from_arg_list of <class 'sqlglot.expressions.If'>>, 'INITCAP': <bound method Func.from_arg_list of <class 'sqlglot.expressions.Initcap'>>, 'IS_NAN': <bound method Func.from_arg_list of <class 'sqlglot.expressions.IsNan'>>, 'ISNAN': <bound method Func.from_arg_list of <class 'sqlglot.expressions.IsNan'>>, 'J_S_O_N_ARRAY': <bound method Func.from_arg_list of <class 'sqlglot.expressions.JSONArray'>>, 'J_S_O_N_ARRAY_AGG': <bound method Func.from_arg_list of <class 'sqlglot.expressions.JSONArrayAgg'>>, 'JSON_ARRAY_CONTAINS': <bound method Func.from_arg_list of <class 'sqlglot.expressions.JSONArrayContains'>>, 'JSONB_EXTRACT': <bound method Func.from_arg_list of <class 'sqlglot.expressions.JSONBExtract'>>, 'JSONB_EXTRACT_SCALAR': <bound method Func.from_arg_list of <class 'sqlglot.expressions.JSONBExtractScalar'>>, 'JSON_EXTRACT': <bound method Func.from_arg_list of <class 'sqlglot.expressions.JSONExtract'>>, 'JSON_EXTRACT_SCALAR': <bound method Func.from_arg_list of <class 'sqlglot.expressions.JSONExtractScalar'>>, 'JSON_FORMAT': <bound method Func.from_arg_list of <class 'sqlglot.expressions.JSONFormat'>>, 'J_S_O_N_OBJECT': <bound method Func.from_arg_list of <class 'sqlglot.expressions.JSONObject'>>, 'J_S_O_N_TABLE': <bound method Func.from_arg_list of <class 'sqlglot.expressions.JSONTable'>>, 'LAST': <bound method Func.from_arg_list of <class 'sqlglot.expressions.Last'>>, 'LAST_DATE_OF_MONTH': <bound method Func.from_arg_list of <class 'sqlglot.expressions.LastDateOfMonth'>>, 'LEAST': <bound method Func.from_arg_list of <class 'sqlglot.expressions.Least'>>, 'LEFT': <bound method Func.from_arg_list of <class 'sqlglot.expressions.Left'>>, 'LENGTH': <bound method Func.from_arg_list of <class 'sqlglot.expressions.Length'>>, 'LEN': <bound method Func.from_arg_list of <class 'sqlglot.expressions.Length'>>, 'LEVENSHTEIN': <bound method Func.from_arg_list of <class 'sqlglot.expressions.Levenshtein'>>, 'LN': <bound method Func.from_arg_list of <class 'sqlglot.expressions.Ln'>>, 'LOG': <bound method Func.from_arg_list of <class 'sqlglot.expressions.Log'>>, 'LOG10': <bound method Func.from_arg_list of <class 'sqlglot.expressions.Log10'>>, 'LOG2': <bound method Func.from_arg_list of <class 'sqlglot.expressions.Log2'>>, 'LOGICAL_AND': <bound method Func.from_arg_list of <class 'sqlglot.expressions.LogicalAnd'>>, 'BOOL_AND': <bound method Func.from_arg_list of <class 'sqlglot.expressions.LogicalAnd'>>, 'BOOLAND_AGG': <bound method Func.from_arg_list of <class 'sqlglot.expressions.LogicalAnd'>>, 'LOGICAL_OR': <bound method Func.from_arg_list of <class 'sqlglot.expressions.LogicalOr'>>, 'BOOL_OR': <bound method Func.from_arg_list of <class 'sqlglot.expressions.LogicalOr'>>, 'BOOLOR_AGG': <bound method Func.from_arg_list of <class 'sqlglot.expressions.LogicalOr'>>, 'LOWER': <bound method Func.from_arg_list of <class 'sqlglot.expressions.Lower'>>, 'LCASE': <bound method Func.from_arg_list of <class 'sqlglot.expressions.Lower'>>, 'MD5': <bound method Func.from_arg_list of <class 'sqlglot.expressions.MD5'>>, 'MD5_DIGEST': <bound method Func.from_arg_list of <class 'sqlglot.expressions.MD5Digest'>>, 'MAP': <function parse_var_map>, 'MAP_FROM_ENTRIES': <bound method Func.from_arg_list of <class 'sqlglot.expressions.MapFromEntries'>>, 'MATCH_AGAINST': <bound method Func.from_arg_list of <class 'sqlglot.expressions.MatchAgainst'>>, 'MAX': <bound method Func.from_arg_list of <class 'sqlglot.expressions.Max'>>, 'MIN': <bound method Func.from_arg_list of <class 'sqlglot.expressions.Min'>>, 'MONTH': <bound method Func.from_arg_list of <class 'sqlglot.expressions.Month'>>, 'MONTHS_BETWEEN': <bound method Func.from_arg_list of <class 'sqlglot.expressions.MonthsBetween'>>, 'NEXT_VALUE_FOR': <bound method Func.from_arg_list of <class 'sqlglot.expressions.NextValueFor'>>, 'NULLIF': <bound method Func.from_arg_list of <class 'sqlglot.expressions.Nullif'>>, 'NUMBER_TO_STR': <bound method Func.from_arg_list of <class 'sqlglot.expressions.NumberToStr'>>, 'NVL2': <bound method Func.from_arg_list of <class 'sqlglot.expressions.Nvl2'>>, 'OPEN_J_S_O_N': <bound method Func.from_arg_list of <class 'sqlglot.expressions.OpenJSON'>>, 'PARAMETERIZED_AGG': <bound method Func.from_arg_list of <class 'sqlglot.expressions.ParameterizedAgg'>>, 'PARSE_JSON': <bound method Func.from_arg_list of <class 'sqlglot.expressions.ParseJSON'>>, 'JSON_PARSE': <bound method Func.from_arg_list of <class 'sqlglot.expressions.ParseJSON'>>, 'PERCENTILE_CONT': <bound method Func.from_arg_list of <class 'sqlglot.expressions.PercentileCont'>>, 'PERCENTILE_DISC': <bound method Func.from_arg_list of <class 'sqlglot.expressions.PercentileDisc'>>, 'POSEXPLODE': <bound method Func.from_arg_list of <class 'sqlglot.expressions.Posexplode'>>, 'POSEXPLODE_OUTER': <bound method Func.from_arg_list of <class 'sqlglot.expressions.PosexplodeOuter'>>, 'POWER': <bound method Func.from_arg_list of <class 'sqlglot.expressions.Pow'>>, 'POW': <bound method Func.from_arg_list of <class 'sqlglot.expressions.Pow'>>, 'PREDICT': <bound method Func.from_arg_list of <class 'sqlglot.expressions.Predict'>>, 'QUANTILE': <bound method Func.from_arg_list of <class 'sqlglot.expressions.Quantile'>>, 'RANGE_N': <bound method Func.from_arg_list of <class 'sqlglot.expressions.RangeN'>>, 'READ_CSV': <bound method Func.from_arg_list of <class 'sqlglot.expressions.ReadCSV'>>, 'REDUCE': <bound method Func.from_arg_list of <class 'sqlglot.expressions.Reduce'>>, 'REGEXP_EXTRACT': <bound method Func.from_arg_list of <class 'sqlglot.expressions.RegexpExtract'>>, 'REGEXP_I_LIKE': <bound method Func.from_arg_list of <class 'sqlglot.expressions.RegexpILike'>>, 'REGEXP_LIKE': <bound method Func.from_arg_list of <class 'sqlglot.expressions.RegexpLike'>>, 'REGEXP_REPLACE': <bound method Func.from_arg_list of <class 'sqlglot.expressions.RegexpReplace'>>, 'REGEXP_SPLIT': <bound method Func.from_arg_list of <class 'sqlglot.expressions.RegexpSplit'>>, 'REPEAT': <bound method Func.from_arg_list of <class 'sqlglot.expressions.Repeat'>>, 'RIGHT': <bound method Func.from_arg_list of <class 'sqlglot.expressions.Right'>>, 'ROUND': <bound method Func.from_arg_list of <class 'sqlglot.expressions.Round'>>, 'ROW_NUMBER': <bound method Func.from_arg_list of <class 'sqlglot.expressions.RowNumber'>>, 'SHA': <bound method Func.from_arg_list of <class 'sqlglot.expressions.SHA'>>, 'SHA1': <bound method Func.from_arg_list of <class 'sqlglot.expressions.SHA'>>, 'SHA2': <bound method Func.from_arg_list of <class 'sqlglot.expressions.SHA2'>>, 'SAFE_CONCAT': <bound method Func.from_arg_list of <class 'sqlglot.expressions.SafeConcat'>>, 'SAFE_DIVIDE': <bound method Func.from_arg_list of <class 'sqlglot.expressions.SafeDivide'>>, 'SET_AGG': <bound method Func.from_arg_list of <class 'sqlglot.expressions.SetAgg'>>, 'SORT_ARRAY': <bound method Func.from_arg_list of <class 'sqlglot.expressions.SortArray'>>, 'SPLIT': <bound method Func.from_arg_list of <class 'sqlglot.expressions.Split'>>, 'SQRT': <bound method Func.from_arg_list of <class 'sqlglot.expressions.Sqrt'>>, 'STANDARD_HASH': <bound method Func.from_arg_list of <class 'sqlglot.expressions.StandardHash'>>, 'STAR_MAP': <bound method Func.from_arg_list of <class 'sqlglot.expressions.StarMap'>>, 'STARTS_WITH': <bound method Func.from_arg_list of <class 'sqlglot.expressions.StartsWith'>>, 'STARTSWITH': <bound method Func.from_arg_list of <class 'sqlglot.expressions.StartsWith'>>, 'STDDEV': <bound method Func.from_arg_list of <class 'sqlglot.expressions.Stddev'>>, 'STDDEV_POP': <bound method Func.from_arg_list of <class 'sqlglot.expressions.StddevPop'>>, 'STDDEV_SAMP': <bound method Func.from_arg_list of <class 'sqlglot.expressions.StddevSamp'>>, 'STR_POSITION': <bound method Func.from_arg_list of <class 'sqlglot.expressions.StrPosition'>>, 'STR_TO_DATE': <bound method Func.from_arg_list of <class 'sqlglot.expressions.StrToDate'>>, 'STR_TO_MAP': <bound method Func.from_arg_list of <class 'sqlglot.expressions.StrToMap'>>, 'STR_TO_TIME': <bound method Func.from_arg_list of <class 'sqlglot.expressions.StrToTime'>>, 'STR_TO_UNIX': <bound method Func.from_arg_list of <class 'sqlglot.expressions.StrToUnix'>>, 'STRUCT': <bound method Func.from_arg_list of <class 'sqlglot.expressions.Struct'>>, 'STRUCT_EXTRACT': <bound method Func.from_arg_list of <class 'sqlglot.expressions.StructExtract'>>, 'STUFF': <bound method Func.from_arg_list of <class 'sqlglot.expressions.Stuff'>>, 'INSERT': <bound method Func.from_arg_list of <class 'sqlglot.expressions.Stuff'>>, 'SUBSTRING': <bound method Func.from_arg_list of <class 'sqlglot.expressions.Substring'>>, 'SUM': <bound method Func.from_arg_list of <class 'sqlglot.expressions.Sum'>>, 'TIME_ADD': <bound method Func.from_arg_list of <class 'sqlglot.expressions.TimeAdd'>>, 'TIME_DIFF': <bound method Func.from_arg_list of <class 'sqlglot.expressions.TimeDiff'>>, 'TIME_STR_TO_DATE': <bound method Func.from_arg_list of <class 'sqlglot.expressions.TimeStrToDate'>>, 'TIME_STR_TO_TIME': <bound method Func.from_arg_list of <class 'sqlglot.expressions.TimeStrToTime'>>, 'TIME_STR_TO_UNIX': <bound method Func.from_arg_list of <class 'sqlglot.expressions.TimeStrToUnix'>>, 'TIME_SUB': <bound method Func.from_arg_list of <class 'sqlglot.expressions.TimeSub'>>, 'TIME_TO_STR': <bound method Func.from_arg_list of <class 'sqlglot.expressions.TimeToStr'>>, 'TIME_TO_TIME_STR': <function Parser.<lambda>>, 'TIME_TO_UNIX': <bound method Func.from_arg_list of <class 'sqlglot.expressions.TimeToUnix'>>, 'TIME_TRUNC': <bound method Func.from_arg_list of <class 'sqlglot.expressions.TimeTrunc'>>, 'TIMESTAMP': <bound method Func.from_arg_list of <class 'sqlglot.expressions.Timestamp'>>, 'TIMESTAMP_ADD': <bound method Func.from_arg_list of <class 'sqlglot.expressions.TimestampAdd'>>, 'TIMESTAMP_DIFF': <bound method Func.from_arg_list of <class 'sqlglot.expressions.TimestampDiff'>>, 'TIMESTAMP_SUB': <bound method Func.from_arg_list of <class 'sqlglot.expressions.TimestampSub'>>, 'TIMESTAMP_TRUNC': <bound method Func.from_arg_list of <class 'sqlglot.expressions.TimestampTrunc'>>, 'TO_BASE64': <bound method Func.from_arg_list of <class 'sqlglot.expressions.ToBase64'>>, 'TO_CHAR': <bound method Func.from_arg_list of <class 'sqlglot.expressions.ToChar'>>, 'TO_DAYS': <bound method Func.from_arg_list of <class 'sqlglot.expressions.ToDays'>>, 'TRANSFORM': <bound method Func.from_arg_list of <class 'sqlglot.expressions.Transform'>>, 'TRIM': <bound method Func.from_arg_list of <class 'sqlglot.expressions.Trim'>>, 'TRY_CAST': <bound method Func.from_arg_list of <class 'sqlglot.expressions.TryCast'>>, 'TS_OR_DI_TO_DI': <bound method Func.from_arg_list of <class 'sqlglot.expressions.TsOrDiToDi'>>, 'TS_OR_DS_ADD': <bound method Func.from_arg_list of <class 'sqlglot.expressions.TsOrDsAdd'>>, 'TS_OR_DS_TO_DATE': <bound method Func.from_arg_list of <class 'sqlglot.expressions.TsOrDsToDate'>>, 'TS_OR_DS_TO_DATE_STR': <function Parser.<lambda>>, 'UNHEX': <bound method Func.from_arg_list of <class 'sqlglot.expressions.Unhex'>>, 'UNIX_TO_STR': <bound method Func.from_arg_list of <class 'sqlglot.expressions.UnixToStr'>>, 'UNIX_TO_TIME': <bound method Func.from_arg_list of <class 'sqlglot.expressions.UnixToTime'>>, 'UNIX_TO_TIME_STR': <bound method Func.from_arg_list of <class 'sqlglot.expressions.UnixToTimeStr'>>, 'UPPER': <bound method Func.from_arg_list of <class 'sqlglot.expressions.Upper'>>, 'UCASE': <bound method Func.from_arg_list of <class 'sqlglot.expressions.Upper'>>, 'VAR_MAP': <function parse_var_map>, 'VARIANCE': <bound method Func.from_arg_list of <class 'sqlglot.expressions.Variance'>>, 'VARIANCE_SAMP': <bound method Func.from_arg_list of <class 'sqlglot.expressions.Variance'>>, 'VAR_SAMP': <bound method Func.from_arg_list of <class 'sqlglot.expressions.Variance'>>, 'VARIANCE_POP': <bound method Func.from_arg_list of <class 'sqlglot.expressions.VariancePop'>>, 'VAR_POP': <bound method Func.from_arg_list of <class 'sqlglot.expressions.VariancePop'>>, 'WEEK': <bound method Func.from_arg_list of <class 'sqlglot.expressions.Week'>>, 'WEEK_OF_YEAR': <bound method Func.from_arg_list of <class 'sqlglot.expressions.WeekOfYear'>>, 'WEEKOFYEAR': <bound method Func.from_arg_list of <class 'sqlglot.expressions.WeekOfYear'>>, 'WHEN': <bound method Func.from_arg_list of <class 'sqlglot.expressions.When'>>, 'X_M_L_TABLE': <bound method Func.from_arg_list of <class 'sqlglot.expressions.XMLTable'>>, 'XOR': <function ClickHouse.Parser.<lambda>>, 'YEAR': <bound method Func.from_arg_list of <class 'sqlglot.expressions.Year'>>, 'GLOB': <function Parser.<lambda>>, 'LIKE': <function parse_like>, 'ANY': <bound method Func.from_arg_list of <class 'sqlglot.expressions.AnyValue'>>, 'DATEADD': <function ClickHouse.Parser.<lambda>>, 'MATCH': <bound method Func.from_arg_list of <class 'sqlglot.expressions.RegexpLike'>>, 'UNIQ': <bound method Func.from_arg_list of <class 'sqlglot.expressions.ApproxDistinct'>>}
FUNCTION_PARSERS =
{'ANY_VALUE': <function Parser.<lambda>>, 'CAST': <function Parser.<lambda>>, 'CONCAT': <function Parser.<lambda>>, 'CONCAT_WS': <function Parser.<lambda>>, 'CONVERT': <function Parser.<lambda>>, 'DECODE': <function Parser.<lambda>>, 'EXTRACT': <function Parser.<lambda>>, 'JSON_OBJECT': <function Parser.<lambda>>, 'JSON_TABLE': <function Parser.<lambda>>, 'LOG': <function Parser.<lambda>>, 'OPENJSON': <function Parser.<lambda>>, 'POSITION': <function Parser.<lambda>>, 'PREDICT': <function Parser.<lambda>>, 'SAFE_CAST': <function Parser.<lambda>>, 'STRING_AGG': <function Parser.<lambda>>, 'SUBSTRING': <function Parser.<lambda>>, 'TRIM': <function Parser.<lambda>>, 'TRY_CAST': <function Parser.<lambda>>, 'TRY_CONVERT': <function Parser.<lambda>>, 'QUANTILE': <function ClickHouse.Parser.<lambda>>}
NO_PAREN_FUNCTION_PARSERS =
{'CASE': <function Parser.<lambda>>, 'IF': <function Parser.<lambda>>, 'NEXT': <function Parser.<lambda>>}
RANGE_PARSERS =
{<TokenType.BETWEEN: 'BETWEEN'>: <function Parser.<lambda>>, <TokenType.GLOB: 'GLOB'>: <function binary_range_parser.<locals>.<lambda>>, <TokenType.ILIKE: 'ILIKE'>: <function binary_range_parser.<locals>.<lambda>>, <TokenType.IN: 'IN'>: <function Parser.<lambda>>, <TokenType.IRLIKE: 'IRLIKE'>: <function binary_range_parser.<locals>.<lambda>>, <TokenType.IS: 'IS'>: <function Parser.<lambda>>, <TokenType.LIKE: 'LIKE'>: <function binary_range_parser.<locals>.<lambda>>, <TokenType.OVERLAPS: 'OVERLAPS'>: <function binary_range_parser.<locals>.<lambda>>, <TokenType.RLIKE: 'RLIKE'>: <function binary_range_parser.<locals>.<lambda>>, <TokenType.SIMILAR_TO: 'SIMILAR_TO'>: <function binary_range_parser.<locals>.<lambda>>, <TokenType.FOR: 'FOR'>: <function Parser.<lambda>>, <TokenType.GLOBAL: 'GLOBAL'>: <function ClickHouse.Parser.<lambda>>}
COLUMN_OPERATORS =
{<TokenType.DOT: 'DOT'>: None, <TokenType.DCOLON: 'DCOLON'>: <function Parser.<lambda>>, <TokenType.ARROW: 'ARROW'>: <function Parser.<lambda>>, <TokenType.DARROW: 'DARROW'>: <function Parser.<lambda>>, <TokenType.HASH_ARROW: 'HASH_ARROW'>: <function Parser.<lambda>>, <TokenType.DHASH_ARROW: 'DHASH_ARROW'>: <function Parser.<lambda>>}
JOIN_KINDS =
{<TokenType.ANY: 'ANY'>, <TokenType.ASOF: 'ASOF'>, <TokenType.ARRAY: 'ARRAY'>, <TokenType.OUTER: 'OUTER'>, <TokenType.SEMI: 'SEMI'>, <TokenType.INNER: 'INNER'>, <TokenType.ANTI: 'ANTI'>, <TokenType.CROSS: 'CROSS'>}
TABLE_ALIAS_TOKENS =
{<TokenType.BIT: 'BIT'>, <TokenType.TINYINT: 'TINYINT'>, <TokenType.LONGBLOB: 'LONGBLOB'>, <TokenType.GEOGRAPHY: 'GEOGRAPHY'>, <TokenType.JSONB: 'JSONB'>, <TokenType.AUTO_INCREMENT: 'AUTO_INCREMENT'>, <TokenType.OVERWRITE: 'OVERWRITE'>, <TokenType.MEDIUMBLOB: 'MEDIUMBLOB'>, <TokenType.UNKNOWN: 'UNKNOWN'>, <TokenType.DESCRIBE: 'DESCRIBE'>, <TokenType.DATABASE: 'DATABASE'>, <TokenType.DATETIME: 'DATETIME'>, <TokenType.OBJECT: 'OBJECT'>, <TokenType.TEXT: 'TEXT'>, <TokenType.FALSE: 'FALSE'>, <TokenType.VARIANT: 'VARIANT'>, <TokenType.DIV: 'DIV'>, <TokenType.RECURSIVE: 'RECURSIVE'>, <TokenType.YEAR: 'YEAR'>, <TokenType.TSRANGE: 'TSRANGE'>, <TokenType.PRAGMA: 'PRAGMA'>, <TokenType.GEOMETRY: 'GEOMETRY'>, <TokenType.LOAD: 'LOAD'>, <TokenType.INDEX: 'INDEX'>, <TokenType.TIMESTAMP: 'TIMESTAMP'>, <TokenType.NULLABLE: 'NULLABLE'>, <TokenType.BINARY: 'BINARY'>, <TokenType.TIMETZ: 'TIMETZ'>, <TokenType.CACHE: 'CACHE'>, <TokenType.TSTZMULTIRANGE: 'TSTZMULTIRANGE'>, <TokenType.TIMESTAMP_NS: 'TIMESTAMP_NS'>, <TokenType.REFRESH: 'REFRESH'>, <TokenType.COMMAND: 'COMMAND'>, <TokenType.MERGE: 'MERGE'>, <TokenType.CONSTRAINT: 'CONSTRAINT'>, <TokenType.BIGINT: 'BIGINT'>, <TokenType.INT8RANGE: 'INT8RANGE'>, <TokenType.INTERVAL: 'INTERVAL'>, <TokenType.BIGDECIMAL: 'BIGDECIMAL'>, <TokenType.ROWVERSION: 'ROWVERSION'>, <TokenType.DATETIME64: 'DATETIME64'>, <TokenType.USMALLINT: 'USMALLINT'>, <TokenType.ROW: 'ROW'>, <TokenType.DICTIONARY: 'DICTIONARY'>, <TokenType.ESCAPE: 'ESCAPE'>, <TokenType.KILL: 'KILL'>, <TokenType.ALL: 'ALL'>, <TokenType.DECIMAL: 'DECIMAL'>, <TokenType.DESC: 'DESC'>, <TokenType.UTINYINT: 'UTINYINT'>, <TokenType.ENUM16: 'ENUM16'>, <TokenType.DELETE: 'DELETE'>, <TokenType.ENUM: 'ENUM'>, <TokenType.OVERLAPS: 'OVERLAPS'>, <TokenType.TINYTEXT: 'TINYTEXT'>, <TokenType.IMAGE: 'IMAGE'>, <TokenType.ISNULL: 'ISNULL'>, <TokenType.INT4RANGE: 'INT4RANGE'>, <TokenType.SET: 'SET'>, <TokenType.EXECUTE: 'EXECUTE'>, <TokenType.UINT256: 'UINT256'>, <TokenType.VARBINARY: 'VARBINARY'>, <TokenType.UNPIVOT: 'UNPIVOT'>, <TokenType.SOME: 'SOME'>, <TokenType.FILTER: 'FILTER'>, <TokenType.BEGIN: 'BEGIN'>, <TokenType.TSTZRANGE: 'TSTZRANGE'>, <TokenType.UMEDIUMINT: 'UMEDIUMINT'>, <TokenType.NCHAR: 'NCHAR'>, <TokenType.NUMMULTIRANGE: 'NUMMULTIRANGE'>, <TokenType.SHOW: 'SHOW'>, <TokenType.UDECIMAL: 'UDECIMAL'>, <TokenType.ENUM8: 'ENUM8'>, <TokenType.UUID: 'UUID'>, <TokenType.CURRENT_DATETIME: 'CURRENT_DATETIME'>, <TokenType.COLLATE: 'COLLATE'>, <TokenType.SUPER: 'SUPER'>, <TokenType.CURRENT_TIMESTAMP: 'CURRENT_TIMESTAMP'>, <TokenType.TABLE: 'TABLE'>, <TokenType.PROCEDURE: 'PROCEDURE'>, <TokenType.FIRST: 'FIRST'>, <TokenType.BIGSERIAL: 'BIGSERIAL'>, <TokenType.DEFAULT: 'DEFAULT'>, <TokenType.EXISTS: 'EXISTS'>, <TokenType.COMMIT: 'COMMIT'>, <TokenType.JSON: 'JSON'>, <TokenType.INT256: 'INT256'>, <TokenType.MEDIUMINT: 'MEDIUMINT'>, <TokenType.ASC: 'ASC'>, <TokenType.PERCENT: 'PERCENT'>, <TokenType.NUMRANGE: 'NUMRANGE'>, <TokenType.TRUE: 'TRUE'>, <TokenType.COMMENT: 'COMMENT'>, <TokenType.PARTITION: 'PARTITION'>, <TokenType.TIME: 'TIME'>, <TokenType.SERIAL: 'SERIAL'>, <TokenType.MEDIUMTEXT: 'MEDIUMTEXT'>, <TokenType.DATERANGE: 'DATERANGE'>, <TokenType.FLOAT: 'FLOAT'>, <TokenType.HLLSKETCH: 'HLLSKETCH'>, <TokenType.IPPREFIX: 'IPPREFIX'>, <TokenType.MAP: 'MAP'>, <TokenType.VIEW: 'VIEW'>, <TokenType.SCHEMA: 'SCHEMA'>, <TokenType.TSMULTIRANGE: 'TSMULTIRANGE'>, <TokenType.LONGTEXT: 'LONGTEXT'>, <TokenType.STRUCT: 'STRUCT'>, <TokenType.CASE: 'CASE'>, <TokenType.RANGE: 'RANGE'>, <TokenType.INT4MULTIRANGE: 'INT4MULTIRANGE'>, <TokenType.INT: 'INT'>, <TokenType.REFERENCES: 'REFERENCES'>, <TokenType.UINT128: 'UINT128'>, <TokenType.UNIQUEIDENTIFIER: 'UNIQUEIDENTIFIER'>, <TokenType.CURRENT_DATE: 'CURRENT_DATE'>, <TokenType.INT128: 'INT128'>, <TokenType.END: 'END'>, <TokenType.IPADDRESS: 'IPADDRESS'>, <TokenType.CURRENT_USER: 'CURRENT_USER'>, <TokenType.UNIQUE: 'UNIQUE'>, <TokenType.TEMPORARY: 'TEMPORARY'>, <TokenType.VARCHAR: 'VARCHAR'>, <TokenType.ORDINALITY: 'ORDINALITY'>, <TokenType.DOUBLE: 'DOUBLE'>, <TokenType.INT8MULTIRANGE: 'INT8MULTIRANGE'>, <TokenType.XML: 'XML'>, <TokenType.TINYBLOB: 'TINYBLOB'>, <TokenType.TIMESTAMPLTZ: 'TIMESTAMPLTZ'>, <TokenType.UBIGINT: 'UBIGINT'>, <TokenType.PSEUDO_TYPE: 'PSEUDO_TYPE'>, <TokenType.NULL: 'NULL'>, <TokenType.BOOLEAN: 'BOOLEAN'>, <TokenType.SMALLINT: 'SMALLINT'>, <TokenType.FUNCTION: 'FUNCTION'>, <TokenType.NEXT: 'NEXT'>, <TokenType.PIVOT: 'PIVOT'>, <TokenType.MODEL: 'MODEL'>, <TokenType.TIMESTAMP_MS: 'TIMESTAMP_MS'>, <TokenType.CHAR: 'CHAR'>, <TokenType.DATEMULTIRANGE: 'DATEMULTIRANGE'>, <TokenType.UINT: 'UINT'>, <TokenType.FIXEDSTRING: 'FIXEDSTRING'>, <TokenType.UPDATE: 'UPDATE'>, <TokenType.DATE: 'DATE'>, <TokenType.NESTED: 'NESTED'>, <TokenType.HSTORE: 'HSTORE'>, <TokenType.KEEP: 'KEEP'>, <TokenType.COLUMN: 'COLUMN'>, <TokenType.SMALLSERIAL: 'SMALLSERIAL'>, <TokenType.VOLATILE: 'VOLATILE'>, <TokenType.LOWCARDINALITY: 'LOWCARDINALITY'>, <TokenType.IS: 'IS'>, <TokenType.USERDEFINED: 'USERDEFINED'>, <TokenType.ROWS: 'ROWS'>, <TokenType.TIMESTAMP_S: 'TIMESTAMP_S'>, <TokenType.SMALLMONEY: 'SMALLMONEY'>, <TokenType.NVARCHAR: 'NVARCHAR'>, <TokenType.MONEY: 'MONEY'>, <TokenType.CURRENT_TIME: 'CURRENT_TIME'>, <TokenType.INET: 'INET'>, <TokenType.TOP: 'TOP'>, <TokenType.OBJECT_IDENTIFIER: 'OBJECT_IDENTIFIER'>, <TokenType.VAR: 'VAR'>, <TokenType.USE: 'USE'>, <TokenType.TIMESTAMPTZ: 'TIMESTAMPTZ'>}
QUERY_MODIFIER_PARSERS =
{<TokenType.MATCH_RECOGNIZE: 'MATCH_RECOGNIZE'>: <function Parser.<lambda>>, <TokenType.WHERE: 'WHERE'>: <function Parser.<lambda>>, <TokenType.GROUP_BY: 'GROUP_BY'>: <function Parser.<lambda>>, <TokenType.HAVING: 'HAVING'>: <function Parser.<lambda>>, <TokenType.QUALIFY: 'QUALIFY'>: <function Parser.<lambda>>, <TokenType.WINDOW: 'WINDOW'>: <function Parser.<lambda>>, <TokenType.ORDER_BY: 'ORDER_BY'>: <function Parser.<lambda>>, <TokenType.LIMIT: 'LIMIT'>: <function Parser.<lambda>>, <TokenType.FETCH: 'FETCH'>: <function Parser.<lambda>>, <TokenType.OFFSET: 'OFFSET'>: <function Parser.<lambda>>, <TokenType.FOR: 'FOR'>: <function Parser.<lambda>>, <TokenType.LOCK: 'LOCK'>: <function Parser.<lambda>>, <TokenType.TABLE_SAMPLE: 'TABLE_SAMPLE'>: <function Parser.<lambda>>, <TokenType.USING: 'USING'>: <function Parser.<lambda>>, <TokenType.CLUSTER_BY: 'CLUSTER_BY'>: <function Parser.<lambda>>, <TokenType.DISTRIBUTE_BY: 'DISTRIBUTE_BY'>: <function Parser.<lambda>>, <TokenType.SORT_BY: 'SORT_BY'>: <function Parser.<lambda>>, <TokenType.CONNECT_BY: 'CONNECT_BY'>: <function Parser.<lambda>>, <TokenType.START_WITH: 'START_WITH'>: <function Parser.<lambda>>, <TokenType.SETTINGS: 'SETTINGS'>: <function ClickHouse.Parser.<lambda>>, <TokenType.FORMAT: 'FORMAT'>: <function ClickHouse.Parser.<lambda>>}
SET_TRIE: Dict =
{'GLOBAL': {0: True}, 'LOCAL': {0: True}, 'SESSION': {0: True}, 'TRANSACTION': {0: True}}
Inherited Members
- sqlglot.parser.Parser
- Parser
- NO_PAREN_FUNCTIONS
- STRUCT_TYPE_TOKENS
- NESTED_TYPE_TOKENS
- ENUM_TYPE_TOKENS
- TYPE_TOKENS
- SIGNED_TO_UNSIGNED_TYPE_TOKEN
- SUBQUERY_PREDICATES
- RESERVED_KEYWORDS
- DB_CREATABLES
- CREATABLES
- ID_VAR_TOKENS
- INTERVAL_VARS
- COMMENT_TABLE_ALIAS_TOKENS
- UPDATE_ALIAS_TOKENS
- TRIM_TYPES
- FUNC_TOKENS
- CONJUNCTION
- EQUALITY
- COMPARISON
- BITWISE
- TERM
- FACTOR
- EXPONENT
- TIMES
- TIMESTAMPS
- SET_OPERATIONS
- JOIN_METHODS
- JOIN_SIDES
- JOIN_HINTS
- LAMBDAS
- EXPRESSION_PARSERS
- STATEMENT_PARSERS
- UNARY_PARSERS
- PRIMARY_PARSERS
- PLACEHOLDER_PARSERS
- PROPERTY_PARSERS
- CONSTRAINT_PARSERS
- ALTER_PARSERS
- SCHEMA_UNNAMED_CONSTRAINTS
- INVALID_FUNC_NAME_TOKENS
- SET_PARSERS
- SHOW_PARSERS
- TYPE_LITERAL_PARSERS
- MODIFIABLES
- DDL_SELECT_TOKENS
- PRE_VOLATILE_TOKENS
- TRANSACTION_KIND
- TRANSACTION_CHARACTERISTICS
- INSERT_ALTERNATIVES
- CLONE_KEYWORDS
- CLONE_KINDS
- OPCLASS_FOLLOW_KEYWORDS
- OPTYPE_FOLLOW_TOKENS
- TABLE_INDEX_HINT_TOKENS
- WINDOW_ALIAS_TOKENS
- WINDOW_BEFORE_PAREN_TOKENS
- WINDOW_SIDES
- FETCH_TOKENS
- ADD_CONSTRAINT_TOKENS
- DISTINCT_TOKENS
- NULL_TOKENS
- UNNEST_OFFSET_ALIAS_TOKENS
- STRICT_CAST
- CONCAT_NULL_OUTPUTS_STRING
- PREFIXED_PIVOT_COLUMNS
- IDENTIFY_PIVOT_STRINGS
- LOG_BASE_FIRST
- ALTER_TABLE_ADD_COLUMN_KEYWORD
- TABLESAMPLE_CSV
- SET_REQUIRES_ASSIGNMENT_DELIMITER
- TRIM_PATTERN_FIRST
- TYPED_DIVISION
- INDEX_OFFSET
- UNNEST_COLUMN_ONLY
- ALIAS_POST_TABLESAMPLE
- FORMAT_MAPPING
- TIME_MAPPING
- error_level
- error_message_context
- max_errors
- reset
- parse
- parse_into
- check_errors
- raise_error
- expression
- validate_expression
- errors
- sql
324 class Generator(generator.Generator): 325 QUERY_HINTS = False 326 STRUCT_DELIMITER = ("(", ")") 327 NVL2_SUPPORTED = False 328 TABLESAMPLE_REQUIRES_PARENS = False 329 330 STRING_TYPE_MAPPING = { 331 exp.DataType.Type.CHAR: "String", 332 exp.DataType.Type.LONGBLOB: "String", 333 exp.DataType.Type.LONGTEXT: "String", 334 exp.DataType.Type.MEDIUMBLOB: "String", 335 exp.DataType.Type.MEDIUMTEXT: "String", 336 exp.DataType.Type.TINYBLOB: "String", 337 exp.DataType.Type.TINYTEXT: "String", 338 exp.DataType.Type.TEXT: "String", 339 exp.DataType.Type.VARBINARY: "String", 340 exp.DataType.Type.VARCHAR: "String", 341 } 342 343 TYPE_MAPPING = { 344 **generator.Generator.TYPE_MAPPING, 345 **STRING_TYPE_MAPPING, 346 exp.DataType.Type.ARRAY: "Array", 347 exp.DataType.Type.BIGINT: "Int64", 348 exp.DataType.Type.DATETIME64: "DateTime64", 349 exp.DataType.Type.DOUBLE: "Float64", 350 exp.DataType.Type.ENUM: "Enum", 351 exp.DataType.Type.ENUM8: "Enum8", 352 exp.DataType.Type.ENUM16: "Enum16", 353 exp.DataType.Type.FIXEDSTRING: "FixedString", 354 exp.DataType.Type.FLOAT: "Float32", 355 exp.DataType.Type.INT: "Int32", 356 exp.DataType.Type.MEDIUMINT: "Int32", 357 exp.DataType.Type.INT128: "Int128", 358 exp.DataType.Type.INT256: "Int256", 359 exp.DataType.Type.LOWCARDINALITY: "LowCardinality", 360 exp.DataType.Type.MAP: "Map", 361 exp.DataType.Type.NESTED: "Nested", 362 exp.DataType.Type.NULLABLE: "Nullable", 363 exp.DataType.Type.SMALLINT: "Int16", 364 exp.DataType.Type.STRUCT: "Tuple", 365 exp.DataType.Type.TINYINT: "Int8", 366 exp.DataType.Type.UBIGINT: "UInt64", 367 exp.DataType.Type.UINT: "UInt32", 368 exp.DataType.Type.UINT128: "UInt128", 369 exp.DataType.Type.UINT256: "UInt256", 370 exp.DataType.Type.USMALLINT: "UInt16", 371 exp.DataType.Type.UTINYINT: "UInt8", 372 } 373 374 TRANSFORMS = { 375 **generator.Generator.TRANSFORMS, 376 exp.Select: transforms.preprocess([transforms.eliminate_qualify]), 377 exp.AnyValue: rename_func("any"), 378 exp.ApproxDistinct: rename_func("uniq"), 379 exp.ArgMax: arg_max_or_min_no_count("argMax"), 380 exp.ArgMin: arg_max_or_min_no_count("argMin"), 381 exp.Array: inline_array_sql, 382 exp.CastToStrType: rename_func("CAST"), 383 exp.CurrentDate: lambda self, e: self.func("CURRENT_DATE"), 384 exp.DateAdd: lambda self, e: self.func( 385 "DATE_ADD", exp.Literal.string(e.text("unit") or "day"), e.expression, e.this 386 ), 387 exp.DateDiff: lambda self, e: self.func( 388 "DATE_DIFF", exp.Literal.string(e.text("unit") or "day"), e.expression, e.this 389 ), 390 exp.Final: lambda self, e: f"{self.sql(e, 'this')} FINAL", 391 exp.IsNan: rename_func("isNaN"), 392 exp.Map: lambda self, e: _lower_func(var_map_sql(self, e)), 393 exp.Nullif: rename_func("nullIf"), 394 exp.PartitionedByProperty: lambda self, e: f"PARTITION BY {self.sql(e, 'this')}", 395 exp.Pivot: no_pivot_sql, 396 exp.Quantile: _quantile_sql, 397 exp.RegexpLike: lambda self, e: f"match({self.format_args(e.this, e.expression)})", 398 exp.StartsWith: rename_func("startsWith"), 399 exp.StrPosition: lambda self, e: f"position({self.format_args(e.this, e.args.get('substr'), e.args.get('position'))})", 400 exp.VarMap: lambda self, e: _lower_func(var_map_sql(self, e)), 401 exp.Xor: lambda self, e: self.func("xor", e.this, e.expression, *e.expressions), 402 } 403 404 PROPERTIES_LOCATION = { 405 **generator.Generator.PROPERTIES_LOCATION, 406 exp.VolatileProperty: exp.Properties.Location.UNSUPPORTED, 407 exp.PartitionedByProperty: exp.Properties.Location.POST_SCHEMA, 408 exp.OnCluster: exp.Properties.Location.POST_NAME, 409 } 410 411 JOIN_HINTS = False 412 TABLE_HINTS = False 413 EXPLICIT_UNION = True 414 GROUPINGS_SEP = "" 415 416 # there's no list in docs, but it can be found in Clickhouse code 417 # see `ClickHouse/src/Parsers/ParserCreate*.cpp` 418 ON_CLUSTER_TARGETS = { 419 "DATABASE", 420 "TABLE", 421 "VIEW", 422 "DICTIONARY", 423 "INDEX", 424 "FUNCTION", 425 "NAMED COLLECTION", 426 } 427 428 def _any_to_has( 429 self, 430 expression: exp.EQ | exp.NEQ, 431 default: t.Callable[[t.Any], str], 432 prefix: str = "", 433 ) -> str: 434 if isinstance(expression.left, exp.Any): 435 arr = expression.left 436 this = expression.right 437 elif isinstance(expression.right, exp.Any): 438 arr = expression.right 439 this = expression.left 440 else: 441 return default(expression) 442 return prefix + self.func("has", arr.this.unnest(), this) 443 444 def eq_sql(self, expression: exp.EQ) -> str: 445 return self._any_to_has(expression, super().eq_sql) 446 447 def neq_sql(self, expression: exp.NEQ) -> str: 448 return self._any_to_has(expression, super().neq_sql, "NOT ") 449 450 def regexpilike_sql(self, expression: exp.RegexpILike) -> str: 451 # Manually add a flag to make the search case-insensitive 452 regex = self.func("CONCAT", "'(?i)'", expression.expression) 453 return f"match({self.format_args(expression.this, regex)})" 454 455 def datatype_sql(self, expression: exp.DataType) -> str: 456 # String is the standard ClickHouse type, every other variant is just an alias. 457 # Additionally, any supplied length parameter will be ignored. 458 # 459 # https://clickhouse.com/docs/en/sql-reference/data-types/string 460 if expression.this in self.STRING_TYPE_MAPPING: 461 return "String" 462 463 return super().datatype_sql(expression) 464 465 def safeconcat_sql(self, expression: exp.SafeConcat) -> str: 466 # Clickhouse errors out if we try to cast a NULL value to TEXT 467 return self.func( 468 "CONCAT", 469 *[ 470 exp.func("if", e.is_(exp.null()), e, exp.cast(e, "text")) 471 for e in t.cast(t.List[exp.Condition], expression.expressions) 472 ], 473 ) 474 475 def cte_sql(self, expression: exp.CTE) -> str: 476 if isinstance(expression.this, exp.Alias): 477 return self.sql(expression, "this") 478 479 return super().cte_sql(expression) 480 481 def after_limit_modifiers(self, expression: exp.Expression) -> t.List[str]: 482 return super().after_limit_modifiers(expression) + [ 483 self.seg("SETTINGS ") + self.expressions(expression, key="settings", flat=True) 484 if expression.args.get("settings") 485 else "", 486 self.seg("FORMAT ") + self.sql(expression, "format") 487 if expression.args.get("format") 488 else "", 489 ] 490 491 def parameterizedagg_sql(self, expression: exp.Anonymous) -> str: 492 params = self.expressions(expression, key="params", flat=True) 493 return self.func(expression.name, *expression.expressions) + f"({params})" 494 495 def placeholder_sql(self, expression: exp.Placeholder) -> str: 496 return f"{{{expression.name}: {self.sql(expression, 'kind')}}}" 497 498 def oncluster_sql(self, expression: exp.OnCluster) -> str: 499 return f"ON CLUSTER {self.sql(expression, 'this')}" 500 501 def createable_sql(self, expression: exp.Create, locations: t.DefaultDict) -> str: 502 kind = self.sql(expression, "kind").upper() 503 if kind in self.ON_CLUSTER_TARGETS and locations.get(exp.Properties.Location.POST_NAME): 504 this_name = self.sql(expression.this, "this") 505 this_properties = " ".join( 506 [self.sql(prop) for prop in locations[exp.Properties.Location.POST_NAME]] 507 ) 508 this_schema = self.schema_columns_sql(expression.this) 509 return f"{this_name}{self.sep()}{this_properties}{self.sep()}{this_schema}" 510 511 return super().createable_sql(expression, locations)
Generator converts a given syntax tree to the corresponding SQL string.
Arguments:
- pretty: Whether or not to format the produced SQL string. Default: False.
- identify: Determines when an identifier should be quoted. Possible values are: False (default): Never quote, except in cases where it's mandatory by the dialect. True or 'always': Always quote. 'safe': Only quote identifiers that are case insensitive.
- normalize: Whether or not to normalize identifiers to lowercase. Default: False.
- pad: Determines the pad size in a formatted string. Default: 2.
- indent: Determines the indentation size in a formatted string. Default: 2.
- normalize_functions: Whether or not to normalize all function names. Possible values are: "upper" or True (default): Convert names to uppercase. "lower": Convert names to lowercase. False: Disables function name normalization.
- unsupported_level: Determines the generator's behavior when it encounters unsupported expressions. Default ErrorLevel.WARN.
- max_unsupported: Maximum number of unsupported messages to include in a raised UnsupportedError. This is only relevant if unsupported_level is ErrorLevel.RAISE. Default: 3
- leading_comma: Determines whether or not the comma is leading or trailing in select expressions. This is only relevant when generating in pretty mode. Default: False
- max_text_width: The max number of characters in a segment before creating new lines in pretty mode. The default is on the smaller end because the length only represents a segment and not the true line length. Default: 80
- comments: Whether or not to preserve comments in the output SQL code. Default: True
STRING_TYPE_MAPPING =
{<Type.CHAR: 'CHAR'>: 'String', <Type.LONGBLOB: 'LONGBLOB'>: 'String', <Type.LONGTEXT: 'LONGTEXT'>: 'String', <Type.MEDIUMBLOB: 'MEDIUMBLOB'>: 'String', <Type.MEDIUMTEXT: 'MEDIUMTEXT'>: 'String', <Type.TINYBLOB: 'TINYBLOB'>: 'String', <Type.TINYTEXT: 'TINYTEXT'>: 'String', <Type.TEXT: 'TEXT'>: 'String', <Type.VARBINARY: 'VARBINARY'>: 'String', <Type.VARCHAR: 'VARCHAR'>: 'String'}
TYPE_MAPPING =
{<Type.NCHAR: 'NCHAR'>: 'CHAR', <Type.NVARCHAR: 'NVARCHAR'>: 'VARCHAR', <Type.MEDIUMTEXT: 'MEDIUMTEXT'>: 'String', <Type.LONGTEXT: 'LONGTEXT'>: 'String', <Type.TINYTEXT: 'TINYTEXT'>: 'String', <Type.MEDIUMBLOB: 'MEDIUMBLOB'>: 'String', <Type.LONGBLOB: 'LONGBLOB'>: 'String', <Type.TINYBLOB: 'TINYBLOB'>: 'String', <Type.INET: 'INET'>: 'INET', <Type.CHAR: 'CHAR'>: 'String', <Type.TEXT: 'TEXT'>: 'String', <Type.VARBINARY: 'VARBINARY'>: 'String', <Type.VARCHAR: 'VARCHAR'>: 'String', <Type.ARRAY: 'ARRAY'>: 'Array', <Type.BIGINT: 'BIGINT'>: 'Int64', <Type.DATETIME64: 'DATETIME64'>: 'DateTime64', <Type.DOUBLE: 'DOUBLE'>: 'Float64', <Type.ENUM: 'ENUM'>: 'Enum', <Type.ENUM8: 'ENUM8'>: 'Enum8', <Type.ENUM16: 'ENUM16'>: 'Enum16', <Type.FIXEDSTRING: 'FIXEDSTRING'>: 'FixedString', <Type.FLOAT: 'FLOAT'>: 'Float32', <Type.INT: 'INT'>: 'Int32', <Type.MEDIUMINT: 'MEDIUMINT'>: 'Int32', <Type.INT128: 'INT128'>: 'Int128', <Type.INT256: 'INT256'>: 'Int256', <Type.LOWCARDINALITY: 'LOWCARDINALITY'>: 'LowCardinality', <Type.MAP: 'MAP'>: 'Map', <Type.NESTED: 'NESTED'>: 'Nested', <Type.NULLABLE: 'NULLABLE'>: 'Nullable', <Type.SMALLINT: 'SMALLINT'>: 'Int16', <Type.STRUCT: 'STRUCT'>: 'Tuple', <Type.TINYINT: 'TINYINT'>: 'Int8', <Type.UBIGINT: 'UBIGINT'>: 'UInt64', <Type.UINT: 'UINT'>: 'UInt32', <Type.UINT128: 'UINT128'>: 'UInt128', <Type.UINT256: 'UINT256'>: 'UInt256', <Type.USMALLINT: 'USMALLINT'>: 'UInt16', <Type.UTINYINT: 'UTINYINT'>: 'UInt8'}
TRANSFORMS =
{<class 'sqlglot.expressions.DateAdd'>: <function ClickHouse.Generator.<lambda>>, <class 'sqlglot.expressions.TsOrDsAdd'>: <function Generator.<lambda>>, <class 'sqlglot.expressions.CaseSpecificColumnConstraint'>: <function Generator.<lambda>>, <class 'sqlglot.expressions.CharacterSetColumnConstraint'>: <function Generator.<lambda>>, <class 'sqlglot.expressions.CharacterSetProperty'>: <function Generator.<lambda>>, <class 'sqlglot.expressions.CheckColumnConstraint'>: <function Generator.<lambda>>, <class 'sqlglot.expressions.ClusteredColumnConstraint'>: <function Generator.<lambda>>, <class 'sqlglot.expressions.CollateColumnConstraint'>: <function Generator.<lambda>>, <class 'sqlglot.expressions.CopyGrantsProperty'>: <function Generator.<lambda>>, <class 'sqlglot.expressions.CommentColumnConstraint'>: <function Generator.<lambda>>, <class 'sqlglot.expressions.DateFormatColumnConstraint'>: <function Generator.<lambda>>, <class 'sqlglot.expressions.DefaultColumnConstraint'>: <function Generator.<lambda>>, <class 'sqlglot.expressions.EncodeColumnConstraint'>: <function Generator.<lambda>>, <class 'sqlglot.expressions.ExecuteAsProperty'>: <function Generator.<lambda>>, <class 'sqlglot.expressions.ExternalProperty'>: <function Generator.<lambda>>, <class 'sqlglot.expressions.HeapProperty'>: <function Generator.<lambda>>, <class 'sqlglot.expressions.InlineLengthColumnConstraint'>: <function Generator.<lambda>>, <class 'sqlglot.expressions.InputModelProperty'>: <function Generator.<lambda>>, <class 'sqlglot.expressions.IntervalSpan'>: <function Generator.<lambda>>, <class 'sqlglot.expressions.LanguageProperty'>: <function Generator.<lambda>>, <class 'sqlglot.expressions.LocationProperty'>: <function Generator.<lambda>>, <class 'sqlglot.expressions.LogProperty'>: <function Generator.<lambda>>, <class 'sqlglot.expressions.MaterializedProperty'>: <function Generator.<lambda>>, <class 'sqlglot.expressions.NoPrimaryIndexProperty'>: <function Generator.<lambda>>, <class 'sqlglot.expressions.NonClusteredColumnConstraint'>: <function Generator.<lambda>>, <class 'sqlglot.expressions.NotForReplicationColumnConstraint'>: <function Generator.<lambda>>, <class 'sqlglot.expressions.OnCommitProperty'>: <function Generator.<lambda>>, <class 'sqlglot.expressions.OnProperty'>: <function Generator.<lambda>>, <class 'sqlglot.expressions.OnUpdateColumnConstraint'>: <function Generator.<lambda>>, <class 'sqlglot.expressions.OutputModelProperty'>: <function Generator.<lambda>>, <class 'sqlglot.expressions.PathColumnConstraint'>: <function Generator.<lambda>>, <class 'sqlglot.expressions.RemoteWithConnectionModelProperty'>: <function Generator.<lambda>>, <class 'sqlglot.expressions.ReturnsProperty'>: <function Generator.<lambda>>, <class 'sqlglot.expressions.SampleProperty'>: <function Generator.<lambda>>, <class 'sqlglot.expressions.SetProperty'>: <function Generator.<lambda>>, <class 'sqlglot.expressions.SettingsProperty'>: <function Generator.<lambda>>, <class 'sqlglot.expressions.SqlSecurityProperty'>: <function Generator.<lambda>>, <class 'sqlglot.expressions.StabilityProperty'>: <function Generator.<lambda>>, <class 'sqlglot.expressions.TemporaryProperty'>: <function Generator.<lambda>>, <class 'sqlglot.expressions.ToTableProperty'>: <function Generator.<lambda>>, <class 'sqlglot.expressions.TransientProperty'>: <function Generator.<lambda>>, <class 'sqlglot.expressions.TransformModelProperty'>: <function Generator.<lambda>>, <class 'sqlglot.expressions.TitleColumnConstraint'>: <function Generator.<lambda>>, <class 'sqlglot.expressions.UppercaseColumnConstraint'>: <function Generator.<lambda>>, <class 'sqlglot.expressions.VarMap'>: <function ClickHouse.Generator.<lambda>>, <class 'sqlglot.expressions.VolatileProperty'>: <function Generator.<lambda>>, <class 'sqlglot.expressions.WithJournalTableProperty'>: <function Generator.<lambda>>, <class 'sqlglot.expressions.Select'>: <function preprocess.<locals>._to_sql>, <class 'sqlglot.expressions.AnyValue'>: <function rename_func.<locals>.<lambda>>, <class 'sqlglot.expressions.ApproxDistinct'>: <function rename_func.<locals>.<lambda>>, <class 'sqlglot.expressions.ArgMax'>: <function arg_max_or_min_no_count.<locals>._arg_max_or_min_sql>, <class 'sqlglot.expressions.ArgMin'>: <function arg_max_or_min_no_count.<locals>._arg_max_or_min_sql>, <class 'sqlglot.expressions.Array'>: <function inline_array_sql>, <class 'sqlglot.expressions.CastToStrType'>: <function rename_func.<locals>.<lambda>>, <class 'sqlglot.expressions.CurrentDate'>: <function ClickHouse.Generator.<lambda>>, <class 'sqlglot.expressions.DateDiff'>: <function ClickHouse.Generator.<lambda>>, <class 'sqlglot.expressions.Final'>: <function ClickHouse.Generator.<lambda>>, <class 'sqlglot.expressions.IsNan'>: <function rename_func.<locals>.<lambda>>, <class 'sqlglot.expressions.Map'>: <function ClickHouse.Generator.<lambda>>, <class 'sqlglot.expressions.Nullif'>: <function rename_func.<locals>.<lambda>>, <class 'sqlglot.expressions.PartitionedByProperty'>: <function ClickHouse.Generator.<lambda>>, <class 'sqlglot.expressions.Pivot'>: <function no_pivot_sql>, <class 'sqlglot.expressions.Quantile'>: <function _quantile_sql>, <class 'sqlglot.expressions.RegexpLike'>: <function ClickHouse.Generator.<lambda>>, <class 'sqlglot.expressions.StartsWith'>: <function rename_func.<locals>.<lambda>>, <class 'sqlglot.expressions.StrPosition'>: <function ClickHouse.Generator.<lambda>>, <class 'sqlglot.expressions.Xor'>: <function ClickHouse.Generator.<lambda>>}
PROPERTIES_LOCATION =
{<class 'sqlglot.expressions.AlgorithmProperty'>: <Location.POST_CREATE: 'POST_CREATE'>, <class 'sqlglot.expressions.AutoIncrementProperty'>: <Location.POST_SCHEMA: 'POST_SCHEMA'>, <class 'sqlglot.expressions.BlockCompressionProperty'>: <Location.POST_NAME: 'POST_NAME'>, <class 'sqlglot.expressions.CharacterSetProperty'>: <Location.POST_SCHEMA: 'POST_SCHEMA'>, <class 'sqlglot.expressions.ChecksumProperty'>: <Location.POST_NAME: 'POST_NAME'>, <class 'sqlglot.expressions.CollateProperty'>: <Location.POST_SCHEMA: 'POST_SCHEMA'>, <class 'sqlglot.expressions.CopyGrantsProperty'>: <Location.POST_SCHEMA: 'POST_SCHEMA'>, <class 'sqlglot.expressions.Cluster'>: <Location.POST_SCHEMA: 'POST_SCHEMA'>, <class 'sqlglot.expressions.ClusteredByProperty'>: <Location.POST_SCHEMA: 'POST_SCHEMA'>, <class 'sqlglot.expressions.DataBlocksizeProperty'>: <Location.POST_NAME: 'POST_NAME'>, <class 'sqlglot.expressions.DefinerProperty'>: <Location.POST_CREATE: 'POST_CREATE'>, <class 'sqlglot.expressions.DictRange'>: <Location.POST_SCHEMA: 'POST_SCHEMA'>, <class 'sqlglot.expressions.DictProperty'>: <Location.POST_SCHEMA: 'POST_SCHEMA'>, <class 'sqlglot.expressions.DistKeyProperty'>: <Location.POST_SCHEMA: 'POST_SCHEMA'>, <class 'sqlglot.expressions.DistStyleProperty'>: <Location.POST_SCHEMA: 'POST_SCHEMA'>, <class 'sqlglot.expressions.EngineProperty'>: <Location.POST_SCHEMA: 'POST_SCHEMA'>, <class 'sqlglot.expressions.ExecuteAsProperty'>: <Location.POST_SCHEMA: 'POST_SCHEMA'>, <class 'sqlglot.expressions.ExternalProperty'>: <Location.POST_CREATE: 'POST_CREATE'>, <class 'sqlglot.expressions.FallbackProperty'>: <Location.POST_NAME: 'POST_NAME'>, <class 'sqlglot.expressions.FileFormatProperty'>: <Location.POST_WITH: 'POST_WITH'>, <class 'sqlglot.expressions.FreespaceProperty'>: <Location.POST_NAME: 'POST_NAME'>, <class 'sqlglot.expressions.HeapProperty'>: <Location.POST_WITH: 'POST_WITH'>, <class 'sqlglot.expressions.InputModelProperty'>: <Location.POST_SCHEMA: 'POST_SCHEMA'>, <class 'sqlglot.expressions.IsolatedLoadingProperty'>: <Location.POST_NAME: 'POST_NAME'>, <class 'sqlglot.expressions.JournalProperty'>: <Location.POST_NAME: 'POST_NAME'>, <class 'sqlglot.expressions.LanguageProperty'>: <Location.POST_SCHEMA: 'POST_SCHEMA'>, <class 'sqlglot.expressions.LikeProperty'>: <Location.POST_SCHEMA: 'POST_SCHEMA'>, <class 'sqlglot.expressions.LocationProperty'>: <Location.POST_SCHEMA: 'POST_SCHEMA'>, <class 'sqlglot.expressions.LockingProperty'>: <Location.POST_ALIAS: 'POST_ALIAS'>, <class 'sqlglot.expressions.LogProperty'>: <Location.POST_NAME: 'POST_NAME'>, <class 'sqlglot.expressions.MaterializedProperty'>: <Location.POST_CREATE: 'POST_CREATE'>, <class 'sqlglot.expressions.MergeBlockRatioProperty'>: <Location.POST_NAME: 'POST_NAME'>, <class 'sqlglot.expressions.NoPrimaryIndexProperty'>: <Location.POST_EXPRESSION: 'POST_EXPRESSION'>, <class 'sqlglot.expressions.OnProperty'>: <Location.POST_SCHEMA: 'POST_SCHEMA'>, <class 'sqlglot.expressions.OnCommitProperty'>: <Location.POST_EXPRESSION: 'POST_EXPRESSION'>, <class 'sqlglot.expressions.Order'>: <Location.POST_SCHEMA: 'POST_SCHEMA'>, <class 'sqlglot.expressions.OutputModelProperty'>: <Location.POST_SCHEMA: 'POST_SCHEMA'>, <class 'sqlglot.expressions.PartitionedByProperty'>: <Location.POST_SCHEMA: 'POST_SCHEMA'>, <class 'sqlglot.expressions.PartitionedOfProperty'>: <Location.POST_SCHEMA: 'POST_SCHEMA'>, <class 'sqlglot.expressions.PrimaryKey'>: <Location.POST_SCHEMA: 'POST_SCHEMA'>, <class 'sqlglot.expressions.Property'>: <Location.POST_WITH: 'POST_WITH'>, <class 'sqlglot.expressions.RemoteWithConnectionModelProperty'>: <Location.POST_SCHEMA: 'POST_SCHEMA'>, <class 'sqlglot.expressions.ReturnsProperty'>: <Location.POST_SCHEMA: 'POST_SCHEMA'>, <class 'sqlglot.expressions.RowFormatProperty'>: <Location.POST_SCHEMA: 'POST_SCHEMA'>, <class 'sqlglot.expressions.RowFormatDelimitedProperty'>: <Location.POST_SCHEMA: 'POST_SCHEMA'>, <class 'sqlglot.expressions.RowFormatSerdeProperty'>: <Location.POST_SCHEMA: 'POST_SCHEMA'>, <class 'sqlglot.expressions.SampleProperty'>: <Location.POST_SCHEMA: 'POST_SCHEMA'>, <class 'sqlglot.expressions.SchemaCommentProperty'>: <Location.POST_SCHEMA: 'POST_SCHEMA'>, <class 'sqlglot.expressions.SerdeProperties'>: <Location.POST_SCHEMA: 'POST_SCHEMA'>, <class 'sqlglot.expressions.Set'>: <Location.POST_SCHEMA: 'POST_SCHEMA'>, <class 'sqlglot.expressions.SettingsProperty'>: <Location.POST_SCHEMA: 'POST_SCHEMA'>, <class 'sqlglot.expressions.SetProperty'>: <Location.POST_CREATE: 'POST_CREATE'>, <class 'sqlglot.expressions.SortKeyProperty'>: <Location.POST_SCHEMA: 'POST_SCHEMA'>, <class 'sqlglot.expressions.SqlSecurityProperty'>: <Location.POST_CREATE: 'POST_CREATE'>, <class 'sqlglot.expressions.StabilityProperty'>: <Location.POST_SCHEMA: 'POST_SCHEMA'>, <class 'sqlglot.expressions.TemporaryProperty'>: <Location.POST_CREATE: 'POST_CREATE'>, <class 'sqlglot.expressions.ToTableProperty'>: <Location.POST_SCHEMA: 'POST_SCHEMA'>, <class 'sqlglot.expressions.TransientProperty'>: <Location.POST_CREATE: 'POST_CREATE'>, <class 'sqlglot.expressions.TransformModelProperty'>: <Location.POST_SCHEMA: 'POST_SCHEMA'>, <class 'sqlglot.expressions.MergeTreeTTL'>: <Location.POST_SCHEMA: 'POST_SCHEMA'>, <class 'sqlglot.expressions.VolatileProperty'>: <Location.UNSUPPORTED: 'UNSUPPORTED'>, <class 'sqlglot.expressions.WithDataProperty'>: <Location.POST_EXPRESSION: 'POST_EXPRESSION'>, <class 'sqlglot.expressions.WithJournalTableProperty'>: <Location.POST_NAME: 'POST_NAME'>, <class 'sqlglot.expressions.WithSystemVersioningProperty'>: <Location.POST_SCHEMA: 'POST_SCHEMA'>, <class 'sqlglot.expressions.OnCluster'>: <Location.POST_NAME: 'POST_NAME'>}
ON_CLUSTER_TARGETS =
{'VIEW', 'INDEX', 'DICTIONARY', 'FUNCTION', 'TABLE', 'NAMED COLLECTION', 'DATABASE'}
455 def datatype_sql(self, expression: exp.DataType) -> str: 456 # String is the standard ClickHouse type, every other variant is just an alias. 457 # Additionally, any supplied length parameter will be ignored. 458 # 459 # https://clickhouse.com/docs/en/sql-reference/data-types/string 460 if expression.this in self.STRING_TYPE_MAPPING: 461 return "String" 462 463 return super().datatype_sql(expression)
465 def safeconcat_sql(self, expression: exp.SafeConcat) -> str: 466 # Clickhouse errors out if we try to cast a NULL value to TEXT 467 return self.func( 468 "CONCAT", 469 *[ 470 exp.func("if", e.is_(exp.null()), e, exp.cast(e, "text")) 471 for e in t.cast(t.List[exp.Condition], expression.expressions) 472 ], 473 )
481 def after_limit_modifiers(self, expression: exp.Expression) -> t.List[str]: 482 return super().after_limit_modifiers(expression) + [ 483 self.seg("SETTINGS ") + self.expressions(expression, key="settings", flat=True) 484 if expression.args.get("settings") 485 else "", 486 self.seg("FORMAT ") + self.sql(expression, "format") 487 if expression.args.get("format") 488 else "", 489 ]
501 def createable_sql(self, expression: exp.Create, locations: t.DefaultDict) -> str: 502 kind = self.sql(expression, "kind").upper() 503 if kind in self.ON_CLUSTER_TARGETS and locations.get(exp.Properties.Location.POST_NAME): 504 this_name = self.sql(expression.this, "this") 505 this_properties = " ".join( 506 [self.sql(prop) for prop in locations[exp.Properties.Location.POST_NAME]] 507 ) 508 this_schema = self.schema_columns_sql(expression.this) 509 return f"{this_name}{self.sep()}{this_properties}{self.sep()}{this_schema}" 510 511 return super().createable_sql(expression, locations)
@classmethod
def
can_identify(text: str, identify: str | bool = 'safe') -> bool:
288 @classmethod 289 def can_identify(cls, text: str, identify: str | bool = "safe") -> bool: 290 """Checks if text can be identified given an identify option. 291 292 Args: 293 text: The text to check. 294 identify: 295 "always" or `True`: Always returns true. 296 "safe": True if the identifier is case-insensitive. 297 298 Returns: 299 Whether or not the given text can be identified. 300 """ 301 if identify is True or identify == "always": 302 return True 303 304 if identify == "safe": 305 return not cls.case_sensitive(text) 306 307 return False
Checks if text can be identified given an identify option.
Arguments:
- text: The text to check.
- identify: "always" or
True
: Always returns true. "safe": True if the identifier is case-insensitive.
Returns:
Whether or not the given text can be identified.
TOKENIZER_CLASS =
<class 'ClickHouse.Tokenizer'>
Inherited Members
- sqlglot.generator.Generator
- Generator
- LOG_BASE_FIRST
- NULL_ORDERING_SUPPORTED
- LOCKING_READS_SUPPORTED
- WRAP_DERIVED_VALUES
- CREATE_FUNCTION_RETURN_AS
- MATCHED_BY_SOURCE
- SINGLE_STRING_INTERVAL
- INTERVAL_ALLOWS_PLURAL_FORM
- TABLESAMPLE_WITH_METHOD
- TABLESAMPLE_SIZE_IS_PERCENT
- LIMIT_FETCH
- LIMIT_ONLY_LITERALS
- RENAME_TABLE_WITH_DB
- INDEX_ON
- QUERY_HINT_SEP
- IS_BOOL_ALLOWED
- DUPLICATE_KEY_UPDATE_WITH_SET
- LIMIT_IS_TOP
- RETURNING_END
- COLUMN_JOIN_MARKS_SUPPORTED
- EXTRACT_ALLOWS_QUOTES
- TZ_TO_WITH_TIME_ZONE
- VALUES_AS_TABLE
- ALTER_TABLE_ADD_COLUMN_KEYWORD
- UNNEST_WITH_ORDINALITY
- AGGREGATE_FILTER_SUPPORTED
- SEMI_ANTI_JOIN_WITH_SIDE
- SUPPORTS_PARAMETERS
- COMPUTED_COLUMN_WITH_TYPE
- SUPPORTS_TABLE_COPY
- COLLATE_IS_FUNC
- DATA_TYPE_SPECIFIERS_ALLOWED
- SUPPORTS_NESTED_CTES
- ENSURE_BOOLS
- CTE_RECURSIVE_KEYWORD_REQUIRED
- TYPED_DIVISION
- STAR_MAPPING
- TIME_PART_SINGULARS
- TOKEN_MAPPING
- PARAMETER_TOKEN
- RESERVED_KEYWORDS
- WITH_SEPARATED_COMMENTS
- EXCLUDE_COMMENTS
- UNWRAPPED_INTERVAL_VALUES
- KEY_VALUE_DEFINITONS
- SENTINEL_LINE_BREAK
- INDEX_OFFSET
- UNNEST_COLUMN_ONLY
- ALIAS_POST_TABLESAMPLE
- IDENTIFIERS_CAN_START_WITH_DIGIT
- pretty
- identify
- normalize
- pad
- unsupported_level
- max_unsupported
- leading_comma
- max_text_width
- comments
- normalize_functions
- unsupported_messages
- generate
- unsupported
- sep
- seg
- pad_comment
- maybe_comment
- wrap
- no_identify
- normalize_func
- indent
- sql
- uncache_sql
- cache_sql
- characterset_sql
- column_sql
- columnposition_sql
- columndef_sql
- columnconstraint_sql
- computedcolumnconstraint_sql
- autoincrementcolumnconstraint_sql
- compresscolumnconstraint_sql
- generatedasidentitycolumnconstraint_sql
- generatedasrowcolumnconstraint_sql
- periodforsystemtimeconstraint_sql
- notnullcolumnconstraint_sql
- primarykeycolumnconstraint_sql
- uniquecolumnconstraint_sql
- create_sql
- clone_sql
- describe_sql
- prepend_ctes
- with_sql
- tablealias_sql
- bitstring_sql
- hexstring_sql
- bytestring_sql
- rawstring_sql
- datatypeparam_sql
- directory_sql
- delete_sql
- drop_sql
- except_sql
- except_op
- fetch_sql
- filter_sql
- hint_sql
- index_sql
- identifier_sql
- inputoutputformat_sql
- national_sql
- partition_sql
- properties_sql
- root_properties
- properties
- with_properties
- locate_properties
- property_name
- property_sql
- likeproperty_sql
- fallbackproperty_sql
- journalproperty_sql
- freespaceproperty_sql
- checksumproperty_sql
- mergeblockratioproperty_sql
- datablocksizeproperty_sql
- blockcompressionproperty_sql
- isolatedloadingproperty_sql
- partitionboundspec_sql
- partitionedofproperty_sql
- lockingproperty_sql
- withdataproperty_sql
- withsystemversioningproperty_sql
- insert_sql
- intersect_sql
- intersect_op
- introducer_sql
- kill_sql
- pseudotype_sql
- objectidentifier_sql
- onconflict_sql
- returning_sql
- rowformatdelimitedproperty_sql
- withtablehint_sql
- indextablehint_sql
- table_sql
- tablesample_sql
- pivot_sql
- version_sql
- tuple_sql
- update_sql
- values_sql
- var_sql
- into_sql
- from_sql
- group_sql
- having_sql
- connect_sql
- prior_sql
- join_sql
- lambda_sql
- lateral_sql
- limit_sql
- offset_sql
- setitem_sql
- set_sql
- pragma_sql
- lock_sql
- literal_sql
- escape_str
- loaddata_sql
- null_sql
- boolean_sql
- order_sql
- cluster_sql
- distribute_sql
- sort_sql
- ordered_sql
- matchrecognize_sql
- query_modifiers
- offset_limit_modifiers
- after_having_modifiers
- select_sql
- schema_sql
- schema_columns_sql
- star_sql
- parameter_sql
- sessionparameter_sql
- subquery_sql
- qualify_sql
- union_sql
- union_op
- unnest_sql
- where_sql
- window_sql
- partition_by_sql
- windowspec_sql
- withingroup_sql
- between_sql
- bracket_sql
- safebracket_sql
- all_sql
- any_sql
- exists_sql
- case_sql
- constraint_sql
- nextvaluefor_sql
- extract_sql
- trim_sql
- check_sql
- foreignkey_sql
- primarykey_sql
- if_sql
- matchagainst_sql
- jsonkeyvalue_sql
- formatjson_sql
- jsonobject_sql
- jsonarray_sql
- jsonarrayagg_sql
- jsoncolumndef_sql
- jsonschema_sql
- jsontable_sql
- openjsoncolumndef_sql
- openjson_sql
- in_sql
- in_unnest_op
- interval_sql
- return_sql
- reference_sql
- anonymous_sql
- paren_sql
- neg_sql
- not_sql
- alias_sql
- aliases_sql
- attimezone_sql
- add_sql
- and_sql
- xor_sql
- connector_sql
- bitwiseand_sql
- bitwiseleftshift_sql
- bitwisenot_sql
- bitwiseor_sql
- bitwiserightshift_sql
- bitwisexor_sql
- cast_sql
- currentdate_sql
- collate_sql
- command_sql
- comment_sql
- mergetreettlaction_sql
- mergetreettl_sql
- transaction_sql
- commit_sql
- rollback_sql
- altercolumn_sql
- renametable_sql
- altertable_sql
- add_column_sql
- droppartition_sql
- addconstraint_sql
- distinct_sql
- ignorenulls_sql
- respectnulls_sql
- intdiv_sql
- dpipe_sql
- safedpipe_sql
- div_sql
- overlaps_sql
- distance_sql
- dot_sql
- propertyeq_sql
- escape_sql
- glob_sql
- gt_sql
- gte_sql
- ilike_sql
- ilikeany_sql
- is_sql
- like_sql
- likeany_sql
- similarto_sql
- lt_sql
- lte_sql
- mod_sql
- mul_sql
- nullsafeeq_sql
- nullsafeneq_sql
- or_sql
- slice_sql
- sub_sql
- trycast_sql
- log_sql
- use_sql
- binary
- function_fallback_sql
- func
- format_args
- text_width
- format_time
- expressions
- op_expressions
- naked_property
- set_operation
- tag_sql
- token_sql
- userdefinedfunction_sql
- joinhint_sql
- kwarg_sql
- when_sql
- merge_sql
- tochar_sql
- dictproperty_sql
- dictrange_sql
- dictsubproperty_sql
- clusteredbyproperty_sql
- anyvalue_sql
- querytransform_sql
- indexconstraintoption_sql
- indexcolumnconstraint_sql
- nvl2_sql
- comprehension_sql
- columnprefix_sql
- opclass_sql
- predict_sql
- forin_sql
- refresh_sql