sqlglot.dialects.duckdb
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 JSON_EXTRACT_TYPE, 9 NormalizationStrategy, 10 approx_count_distinct_sql, 11 arg_max_or_min_no_count, 12 arrow_json_extract_sql, 13 binary_from_function, 14 bool_xor_sql, 15 build_default_decimal_type, 16 date_trunc_to_time, 17 datestrtodate_sql, 18 encode_decode_sql, 19 build_formatted_time, 20 inline_array_unless_query, 21 no_comment_column_constraint_sql, 22 no_safe_divide_sql, 23 no_timestamp_sql, 24 pivot_column_names, 25 regexp_extract_sql, 26 rename_func, 27 str_position_sql, 28 str_to_time_sql, 29 timestamptrunc_sql, 30 timestrtotime_sql, 31 unit_to_var, 32) 33from sqlglot.helper import seq_get 34from sqlglot.tokens import TokenType 35 36 37def _ts_or_ds_add_sql(self: DuckDB.Generator, expression: exp.TsOrDsAdd) -> str: 38 this = self.sql(expression, "this") 39 interval = self.sql(exp.Interval(this=expression.expression, unit=unit_to_var(expression))) 40 return f"CAST({this} AS {self.sql(expression.return_type)}) + {interval}" 41 42 43def _date_delta_sql( 44 self: DuckDB.Generator, expression: exp.DateAdd | exp.DateSub | exp.TimeAdd 45) -> str: 46 this = self.sql(expression, "this") 47 unit = unit_to_var(expression) 48 op = "+" if isinstance(expression, (exp.DateAdd, exp.TimeAdd)) else "-" 49 return f"{this} {op} {self.sql(exp.Interval(this=expression.expression, unit=unit))}" 50 51 52# BigQuery -> DuckDB conversion for the DATE function 53def _date_sql(self: DuckDB.Generator, expression: exp.Date) -> str: 54 result = f"CAST({self.sql(expression, 'this')} AS DATE)" 55 zone = self.sql(expression, "zone") 56 57 if zone: 58 date_str = self.func("STRFTIME", result, "'%d/%m/%Y'") 59 date_str = f"{date_str} || ' ' || {zone}" 60 61 # This will create a TIMESTAMP with time zone information 62 result = self.func("STRPTIME", date_str, "'%d/%m/%Y %Z'") 63 64 return result 65 66 67def _array_sort_sql(self: DuckDB.Generator, expression: exp.ArraySort) -> str: 68 if expression.expression: 69 self.unsupported("DuckDB ARRAY_SORT does not support a comparator") 70 return self.func("ARRAY_SORT", expression.this) 71 72 73def _sort_array_sql(self: DuckDB.Generator, expression: exp.SortArray) -> str: 74 name = "ARRAY_REVERSE_SORT" if expression.args.get("asc") == exp.false() else "ARRAY_SORT" 75 return self.func(name, expression.this) 76 77 78def _build_sort_array_desc(args: t.List) -> exp.Expression: 79 return exp.SortArray(this=seq_get(args, 0), asc=exp.false()) 80 81 82def _build_date_diff(args: t.List) -> exp.Expression: 83 return exp.DateDiff(this=seq_get(args, 2), expression=seq_get(args, 1), unit=seq_get(args, 0)) 84 85 86def _build_generate_series(end_exclusive: bool = False) -> t.Callable[[t.List], exp.GenerateSeries]: 87 def _builder(args: t.List) -> exp.GenerateSeries: 88 # Check https://duckdb.org/docs/sql/functions/nested.html#range-functions 89 if len(args) == 1: 90 # DuckDB uses 0 as a default for the series' start when it's omitted 91 args.insert(0, exp.Literal.number("0")) 92 93 gen_series = exp.GenerateSeries.from_arg_list(args) 94 gen_series.set("is_end_exclusive", end_exclusive) 95 96 return gen_series 97 98 return _builder 99 100 101def _build_make_timestamp(args: t.List) -> exp.Expression: 102 if len(args) == 1: 103 return exp.UnixToTime(this=seq_get(args, 0), scale=exp.UnixToTime.MICROS) 104 105 return exp.TimestampFromParts( 106 year=seq_get(args, 0), 107 month=seq_get(args, 1), 108 day=seq_get(args, 2), 109 hour=seq_get(args, 3), 110 min=seq_get(args, 4), 111 sec=seq_get(args, 5), 112 ) 113 114 115def _struct_sql(self: DuckDB.Generator, expression: exp.Struct) -> str: 116 args: t.List[str] = [] 117 for i, expr in enumerate(expression.expressions): 118 if isinstance(expr, exp.PropertyEQ): 119 key = expr.name 120 value = expr.expression 121 else: 122 key = f"_{i}" 123 value = expr 124 125 args.append(f"{self.sql(exp.Literal.string(key))}: {self.sql(value)}") 126 127 return f"{{{', '.join(args)}}}" 128 129 130def _datatype_sql(self: DuckDB.Generator, expression: exp.DataType) -> str: 131 if expression.is_type("array"): 132 return f"{self.expressions(expression, flat=True)}[]" 133 134 # Type TIMESTAMP / TIME WITH TIME ZONE does not support any modifiers 135 if expression.is_type("timestamptz", "timetz"): 136 return expression.this.value 137 138 return self.datatype_sql(expression) 139 140 141def _json_format_sql(self: DuckDB.Generator, expression: exp.JSONFormat) -> str: 142 sql = self.func("TO_JSON", expression.this, expression.args.get("options")) 143 return f"CAST({sql} AS TEXT)" 144 145 146def _unix_to_time_sql(self: DuckDB.Generator, expression: exp.UnixToTime) -> str: 147 scale = expression.args.get("scale") 148 timestamp = expression.this 149 150 if scale in (None, exp.UnixToTime.SECONDS): 151 return self.func("TO_TIMESTAMP", timestamp) 152 if scale == exp.UnixToTime.MILLIS: 153 return self.func("EPOCH_MS", timestamp) 154 if scale == exp.UnixToTime.MICROS: 155 return self.func("MAKE_TIMESTAMP", timestamp) 156 157 return self.func("TO_TIMESTAMP", exp.Div(this=timestamp, expression=exp.func("POW", 10, scale))) 158 159 160def _arrow_json_extract_sql(self: DuckDB.Generator, expression: JSON_EXTRACT_TYPE) -> str: 161 arrow_sql = arrow_json_extract_sql(self, expression) 162 if not expression.same_parent and isinstance(expression.parent, exp.Binary): 163 arrow_sql = self.wrap(arrow_sql) 164 return arrow_sql 165 166 167class DuckDB(Dialect): 168 NULL_ORDERING = "nulls_are_last" 169 SUPPORTS_USER_DEFINED_TYPES = False 170 SAFE_DIVISION = True 171 INDEX_OFFSET = 1 172 CONCAT_COALESCE = True 173 174 # https://duckdb.org/docs/sql/introduction.html#creating-a-new-table 175 NORMALIZATION_STRATEGY = NormalizationStrategy.CASE_INSENSITIVE 176 177 def to_json_path(self, path: t.Optional[exp.Expression]) -> t.Optional[exp.Expression]: 178 if isinstance(path, exp.Literal): 179 # DuckDB also supports the JSON pointer syntax, where every path starts with a `/`. 180 # Additionally, it allows accessing the back of lists using the `[#-i]` syntax. 181 # This check ensures we'll avoid trying to parse these as JSON paths, which can 182 # either result in a noisy warning or in an invalid representation of the path. 183 path_text = path.name 184 if path_text.startswith("/") or "[#" in path_text: 185 return path 186 187 return super().to_json_path(path) 188 189 class Tokenizer(tokens.Tokenizer): 190 HEREDOC_STRINGS = ["$"] 191 192 HEREDOC_TAG_IS_IDENTIFIER = True 193 HEREDOC_STRING_ALTERNATIVE = TokenType.PARAMETER 194 195 KEYWORDS = { 196 **tokens.Tokenizer.KEYWORDS, 197 "//": TokenType.DIV, 198 "ATTACH": TokenType.COMMAND, 199 "BINARY": TokenType.VARBINARY, 200 "BITSTRING": TokenType.BIT, 201 "BPCHAR": TokenType.TEXT, 202 "CHAR": TokenType.TEXT, 203 "CHARACTER VARYING": TokenType.TEXT, 204 "EXCLUDE": TokenType.EXCEPT, 205 "LOGICAL": TokenType.BOOLEAN, 206 "ONLY": TokenType.ONLY, 207 "PIVOT_WIDER": TokenType.PIVOT, 208 "POSITIONAL": TokenType.POSITIONAL, 209 "SIGNED": TokenType.INT, 210 "STRING": TokenType.VARCHAR, 211 "UBIGINT": TokenType.UBIGINT, 212 "UINTEGER": TokenType.UINT, 213 "USMALLINT": TokenType.USMALLINT, 214 "UTINYINT": TokenType.UTINYINT, 215 "TIMESTAMP_S": TokenType.TIMESTAMP_S, 216 "TIMESTAMP_MS": TokenType.TIMESTAMP_MS, 217 "TIMESTAMP_NS": TokenType.TIMESTAMP_NS, 218 "TIMESTAMP_US": TokenType.TIMESTAMP, 219 } 220 221 SINGLE_TOKENS = { 222 **tokens.Tokenizer.SINGLE_TOKENS, 223 "$": TokenType.PARAMETER, 224 } 225 226 class Parser(parser.Parser): 227 BITWISE = { 228 **parser.Parser.BITWISE, 229 TokenType.TILDA: exp.RegexpLike, 230 } 231 232 FUNCTIONS_WITH_ALIASED_ARGS = {*parser.Parser.FUNCTIONS_WITH_ALIASED_ARGS, "STRUCT_PACK"} 233 234 FUNCTIONS = { 235 **parser.Parser.FUNCTIONS, 236 "ARRAY_REVERSE_SORT": _build_sort_array_desc, 237 "ARRAY_SORT": exp.SortArray.from_arg_list, 238 "DATEDIFF": _build_date_diff, 239 "DATE_DIFF": _build_date_diff, 240 "DATE_TRUNC": date_trunc_to_time, 241 "DATETRUNC": date_trunc_to_time, 242 "DECODE": lambda args: exp.Decode( 243 this=seq_get(args, 0), charset=exp.Literal.string("utf-8") 244 ), 245 "ENCODE": lambda args: exp.Encode( 246 this=seq_get(args, 0), charset=exp.Literal.string("utf-8") 247 ), 248 "EPOCH": exp.TimeToUnix.from_arg_list, 249 "EPOCH_MS": lambda args: exp.UnixToTime( 250 this=seq_get(args, 0), scale=exp.UnixToTime.MILLIS 251 ), 252 "JSON": exp.ParseJSON.from_arg_list, 253 "JSON_EXTRACT_PATH": parser.build_extract_json_with_path(exp.JSONExtract), 254 "JSON_EXTRACT_STRING": parser.build_extract_json_with_path(exp.JSONExtractScalar), 255 "LIST_HAS": exp.ArrayContains.from_arg_list, 256 "LIST_REVERSE_SORT": _build_sort_array_desc, 257 "LIST_SORT": exp.SortArray.from_arg_list, 258 "LIST_VALUE": exp.Array.from_arg_list, 259 "MAKE_TIME": exp.TimeFromParts.from_arg_list, 260 "MAKE_TIMESTAMP": _build_make_timestamp, 261 "MEDIAN": lambda args: exp.PercentileCont( 262 this=seq_get(args, 0), expression=exp.Literal.number(0.5) 263 ), 264 "QUANTILE_CONT": exp.PercentileCont.from_arg_list, 265 "QUANTILE_DISC": exp.PercentileDisc.from_arg_list, 266 "REGEXP_EXTRACT": lambda args: exp.RegexpExtract( 267 this=seq_get(args, 0), expression=seq_get(args, 1), group=seq_get(args, 2) 268 ), 269 "REGEXP_MATCHES": exp.RegexpLike.from_arg_list, 270 "REGEXP_REPLACE": lambda args: exp.RegexpReplace( 271 this=seq_get(args, 0), 272 expression=seq_get(args, 1), 273 replacement=seq_get(args, 2), 274 modifiers=seq_get(args, 3), 275 ), 276 "STRFTIME": build_formatted_time(exp.TimeToStr, "duckdb"), 277 "STRING_SPLIT": exp.Split.from_arg_list, 278 "STRING_SPLIT_REGEX": exp.RegexpSplit.from_arg_list, 279 "STRING_TO_ARRAY": exp.Split.from_arg_list, 280 "STRPTIME": build_formatted_time(exp.StrToTime, "duckdb"), 281 "STRUCT_PACK": exp.Struct.from_arg_list, 282 "STR_SPLIT": exp.Split.from_arg_list, 283 "STR_SPLIT_REGEX": exp.RegexpSplit.from_arg_list, 284 "TO_TIMESTAMP": exp.UnixToTime.from_arg_list, 285 "UNNEST": exp.Explode.from_arg_list, 286 "XOR": binary_from_function(exp.BitwiseXor), 287 "GENERATE_SERIES": _build_generate_series(), 288 "RANGE": _build_generate_series(end_exclusive=True), 289 } 290 291 FUNCTIONS.pop("DATE_SUB") 292 293 FUNCTION_PARSERS = parser.Parser.FUNCTION_PARSERS.copy() 294 FUNCTION_PARSERS.pop("DECODE") 295 296 NO_PAREN_FUNCTION_PARSERS = { 297 **parser.Parser.NO_PAREN_FUNCTION_PARSERS, 298 "MAP": lambda self: self._parse_map(), 299 } 300 301 TABLE_ALIAS_TOKENS = parser.Parser.TABLE_ALIAS_TOKENS - { 302 TokenType.SEMI, 303 TokenType.ANTI, 304 } 305 306 PLACEHOLDER_PARSERS = { 307 **parser.Parser.PLACEHOLDER_PARSERS, 308 TokenType.PARAMETER: lambda self: ( 309 self.expression(exp.Placeholder, this=self._prev.text) 310 if self._match(TokenType.NUMBER) or self._match_set(self.ID_VAR_TOKENS) 311 else None 312 ), 313 } 314 315 TYPE_CONVERTER = { 316 # https://duckdb.org/docs/sql/data_types/numeric 317 exp.DataType.Type.DECIMAL: build_default_decimal_type(precision=18, scale=3), 318 } 319 320 def _parse_table_sample(self, as_modifier: bool = False) -> t.Optional[exp.TableSample]: 321 # https://duckdb.org/docs/sql/samples.html 322 sample = super()._parse_table_sample(as_modifier=as_modifier) 323 if sample and not sample.args.get("method"): 324 if sample.args.get("size"): 325 sample.set("method", exp.var("RESERVOIR")) 326 else: 327 sample.set("method", exp.var("SYSTEM")) 328 329 return sample 330 331 def _parse_bracket( 332 self, this: t.Optional[exp.Expression] = None 333 ) -> t.Optional[exp.Expression]: 334 bracket = super()._parse_bracket(this) 335 if isinstance(bracket, exp.Bracket): 336 bracket.set("returns_list_for_maps", True) 337 338 return bracket 339 340 def _parse_map(self) -> exp.ToMap | exp.Map: 341 if self._match(TokenType.L_BRACE, advance=False): 342 return self.expression(exp.ToMap, this=self._parse_bracket()) 343 344 args = self._parse_wrapped_csv(self._parse_conjunction) 345 return self.expression(exp.Map, keys=seq_get(args, 0), values=seq_get(args, 1)) 346 347 def _parse_struct_types(self, type_required: bool = False) -> t.Optional[exp.Expression]: 348 return self._parse_field_def() 349 350 def _pivot_column_names(self, aggregations: t.List[exp.Expression]) -> t.List[str]: 351 if len(aggregations) == 1: 352 return super()._pivot_column_names(aggregations) 353 return pivot_column_names(aggregations, dialect="duckdb") 354 355 class Generator(generator.Generator): 356 PARAMETER_TOKEN = "$" 357 NAMED_PLACEHOLDER_TOKEN = "$" 358 JOIN_HINTS = False 359 TABLE_HINTS = False 360 QUERY_HINTS = False 361 LIMIT_FETCH = "LIMIT" 362 STRUCT_DELIMITER = ("(", ")") 363 RENAME_TABLE_WITH_DB = False 364 NVL2_SUPPORTED = False 365 SEMI_ANTI_JOIN_WITH_SIDE = False 366 TABLESAMPLE_KEYWORDS = "USING SAMPLE" 367 TABLESAMPLE_SEED_KEYWORD = "REPEATABLE" 368 LAST_DAY_SUPPORTS_DATE_PART = False 369 JSON_KEY_VALUE_PAIR_SEP = "," 370 IGNORE_NULLS_IN_FUNC = True 371 JSON_PATH_BRACKETED_KEY_SUPPORTED = False 372 SUPPORTS_CREATE_TABLE_LIKE = False 373 MULTI_ARG_DISTINCT = False 374 CAN_IMPLEMENT_ARRAY_ANY = True 375 SUPPORTS_TO_NUMBER = False 376 COPY_HAS_INTO_KEYWORD = False 377 STAR_EXCEPT = "EXCLUDE" 378 379 TRANSFORMS = { 380 **generator.Generator.TRANSFORMS, 381 exp.ApproxDistinct: approx_count_distinct_sql, 382 exp.Array: inline_array_unless_query, 383 exp.ArrayContainsAll: rename_func("ARRAY_HAS_ALL"), 384 exp.ArrayFilter: rename_func("LIST_FILTER"), 385 exp.ArraySize: rename_func("ARRAY_LENGTH"), 386 exp.ArgMax: arg_max_or_min_no_count("ARG_MAX"), 387 exp.ArgMin: arg_max_or_min_no_count("ARG_MIN"), 388 exp.ArraySort: _array_sort_sql, 389 exp.ArraySum: rename_func("LIST_SUM"), 390 exp.BitwiseXor: rename_func("XOR"), 391 exp.CommentColumnConstraint: no_comment_column_constraint_sql, 392 exp.CurrentDate: lambda *_: "CURRENT_DATE", 393 exp.CurrentTime: lambda *_: "CURRENT_TIME", 394 exp.CurrentTimestamp: lambda *_: "CURRENT_TIMESTAMP", 395 exp.DayOfMonth: rename_func("DAYOFMONTH"), 396 exp.DayOfWeek: rename_func("DAYOFWEEK"), 397 exp.DayOfYear: rename_func("DAYOFYEAR"), 398 exp.DataType: _datatype_sql, 399 exp.Date: _date_sql, 400 exp.DateAdd: _date_delta_sql, 401 exp.DateFromParts: rename_func("MAKE_DATE"), 402 exp.DateSub: _date_delta_sql, 403 exp.DateDiff: lambda self, e: self.func( 404 "DATE_DIFF", f"'{e.args.get('unit') or 'DAY'}'", e.expression, e.this 405 ), 406 exp.DateStrToDate: datestrtodate_sql, 407 exp.DateToDi: lambda self, 408 e: f"CAST(STRFTIME({self.sql(e, 'this')}, {DuckDB.DATEINT_FORMAT}) AS INT)", 409 exp.Decode: lambda self, e: encode_decode_sql(self, e, "DECODE", replace=False), 410 exp.DiToDate: lambda self, 411 e: f"CAST(STRPTIME(CAST({self.sql(e, 'this')} AS TEXT), {DuckDB.DATEINT_FORMAT}) AS DATE)", 412 exp.Encode: lambda self, e: encode_decode_sql(self, e, "ENCODE", replace=False), 413 exp.Explode: rename_func("UNNEST"), 414 exp.IntDiv: lambda self, e: self.binary(e, "//"), 415 exp.IsInf: rename_func("ISINF"), 416 exp.IsNan: rename_func("ISNAN"), 417 exp.JSONExtract: _arrow_json_extract_sql, 418 exp.JSONExtractScalar: _arrow_json_extract_sql, 419 exp.JSONFormat: _json_format_sql, 420 exp.LogicalOr: rename_func("BOOL_OR"), 421 exp.LogicalAnd: rename_func("BOOL_AND"), 422 exp.MD5Digest: lambda self, e: self.func("UNHEX", self.func("MD5", e.this)), 423 exp.MonthsBetween: lambda self, e: self.func( 424 "DATEDIFF", 425 "'month'", 426 exp.cast(e.expression, exp.DataType.Type.TIMESTAMP, copy=True), 427 exp.cast(e.this, exp.DataType.Type.TIMESTAMP, copy=True), 428 ), 429 exp.ParseJSON: rename_func("JSON"), 430 exp.PercentileCont: rename_func("QUANTILE_CONT"), 431 exp.PercentileDisc: rename_func("QUANTILE_DISC"), 432 # DuckDB doesn't allow qualified columns inside of PIVOT expressions. 433 # See: https://github.com/duckdb/duckdb/blob/671faf92411182f81dce42ac43de8bfb05d9909e/src/planner/binder/tableref/bind_pivot.cpp#L61-L62 434 exp.Pivot: transforms.preprocess([transforms.unqualify_columns]), 435 exp.RegexpExtract: regexp_extract_sql, 436 exp.RegexpReplace: lambda self, e: self.func( 437 "REGEXP_REPLACE", 438 e.this, 439 e.expression, 440 e.args.get("replacement"), 441 e.args.get("modifiers"), 442 ), 443 exp.RegexpLike: rename_func("REGEXP_MATCHES"), 444 exp.RegexpSplit: rename_func("STR_SPLIT_REGEX"), 445 exp.Rand: rename_func("RANDOM"), 446 exp.SafeDivide: no_safe_divide_sql, 447 exp.Split: rename_func("STR_SPLIT"), 448 exp.SortArray: _sort_array_sql, 449 exp.StrPosition: str_position_sql, 450 exp.StrToDate: lambda self, e: f"CAST({str_to_time_sql(self, e)} AS DATE)", 451 exp.StrToTime: str_to_time_sql, 452 exp.StrToUnix: lambda self, e: self.func( 453 "EPOCH", self.func("STRPTIME", e.this, self.format_time(e)) 454 ), 455 exp.Struct: _struct_sql, 456 exp.TimeAdd: _date_delta_sql, 457 exp.Timestamp: no_timestamp_sql, 458 exp.TimestampDiff: lambda self, e: self.func( 459 "DATE_DIFF", exp.Literal.string(e.unit), e.expression, e.this 460 ), 461 exp.TimestampTrunc: timestamptrunc_sql(), 462 exp.TimeStrToDate: lambda self, e: self.sql(exp.cast(e.this, exp.DataType.Type.DATE)), 463 exp.TimeStrToTime: timestrtotime_sql, 464 exp.TimeStrToUnix: lambda self, e: self.func( 465 "EPOCH", exp.cast(e.this, exp.DataType.Type.TIMESTAMP) 466 ), 467 exp.TimeToStr: lambda self, e: self.func("STRFTIME", e.this, self.format_time(e)), 468 exp.TimeToUnix: rename_func("EPOCH"), 469 exp.TsOrDiToDi: lambda self, 470 e: f"CAST(SUBSTR(REPLACE(CAST({self.sql(e, 'this')} AS TEXT), '-', ''), 1, 8) AS INT)", 471 exp.TsOrDsAdd: _ts_or_ds_add_sql, 472 exp.TsOrDsDiff: lambda self, e: self.func( 473 "DATE_DIFF", 474 f"'{e.args.get('unit') or 'DAY'}'", 475 exp.cast(e.expression, exp.DataType.Type.TIMESTAMP), 476 exp.cast(e.this, exp.DataType.Type.TIMESTAMP), 477 ), 478 exp.UnixToStr: lambda self, e: self.func( 479 "STRFTIME", self.func("TO_TIMESTAMP", e.this), self.format_time(e) 480 ), 481 exp.UnixToTime: _unix_to_time_sql, 482 exp.UnixToTimeStr: lambda self, e: f"CAST(TO_TIMESTAMP({self.sql(e, 'this')}) AS TEXT)", 483 exp.VariancePop: rename_func("VAR_POP"), 484 exp.WeekOfYear: rename_func("WEEKOFYEAR"), 485 exp.Xor: bool_xor_sql, 486 } 487 488 SUPPORTED_JSON_PATH_PARTS = { 489 exp.JSONPathKey, 490 exp.JSONPathRoot, 491 exp.JSONPathSubscript, 492 exp.JSONPathWildcard, 493 } 494 495 TYPE_MAPPING = { 496 **generator.Generator.TYPE_MAPPING, 497 exp.DataType.Type.BINARY: "BLOB", 498 exp.DataType.Type.CHAR: "TEXT", 499 exp.DataType.Type.FLOAT: "REAL", 500 exp.DataType.Type.NCHAR: "TEXT", 501 exp.DataType.Type.NVARCHAR: "TEXT", 502 exp.DataType.Type.UINT: "UINTEGER", 503 exp.DataType.Type.VARBINARY: "BLOB", 504 exp.DataType.Type.ROWVERSION: "BLOB", 505 exp.DataType.Type.VARCHAR: "TEXT", 506 exp.DataType.Type.TIMESTAMP_S: "TIMESTAMP_S", 507 exp.DataType.Type.TIMESTAMP_MS: "TIMESTAMP_MS", 508 exp.DataType.Type.TIMESTAMP_NS: "TIMESTAMP_NS", 509 } 510 511 UNWRAPPED_INTERVAL_VALUES = (exp.Literal, exp.Paren) 512 513 # DuckDB doesn't generally support CREATE TABLE .. properties 514 # https://duckdb.org/docs/sql/statements/create_table.html 515 PROPERTIES_LOCATION = { 516 prop: exp.Properties.Location.UNSUPPORTED 517 for prop in generator.Generator.PROPERTIES_LOCATION 518 } 519 520 # There are a few exceptions (e.g. temporary tables) which are supported or 521 # can be transpiled to DuckDB, so we explicitly override them accordingly 522 PROPERTIES_LOCATION[exp.LikeProperty] = exp.Properties.Location.POST_SCHEMA 523 PROPERTIES_LOCATION[exp.TemporaryProperty] = exp.Properties.Location.POST_CREATE 524 525 def timefromparts_sql(self, expression: exp.TimeFromParts) -> str: 526 nano = expression.args.get("nano") 527 if nano is not None: 528 expression.set( 529 "sec", expression.args["sec"] + nano.pop() / exp.Literal.number(1000000000.0) 530 ) 531 532 return rename_func("MAKE_TIME")(self, expression) 533 534 def timestampfromparts_sql(self, expression: exp.TimestampFromParts) -> str: 535 sec = expression.args["sec"] 536 537 milli = expression.args.get("milli") 538 if milli is not None: 539 sec += milli.pop() / exp.Literal.number(1000.0) 540 541 nano = expression.args.get("nano") 542 if nano is not None: 543 sec += nano.pop() / exp.Literal.number(1000000000.0) 544 545 if milli or nano: 546 expression.set("sec", sec) 547 548 return rename_func("MAKE_TIMESTAMP")(self, expression) 549 550 def tablesample_sql( 551 self, 552 expression: exp.TableSample, 553 sep: str = " AS ", 554 tablesample_keyword: t.Optional[str] = None, 555 ) -> str: 556 if not isinstance(expression.parent, exp.Select): 557 # This sample clause only applies to a single source, not the entire resulting relation 558 tablesample_keyword = "TABLESAMPLE" 559 560 if expression.args.get("size"): 561 method = expression.args.get("method") 562 if method and method.name.upper() != "RESERVOIR": 563 self.unsupported( 564 f"Sampling method {method} is not supported with a discrete sample count, " 565 "defaulting to reservoir sampling" 566 ) 567 expression.set("method", exp.var("RESERVOIR")) 568 569 return super().tablesample_sql( 570 expression, sep=sep, tablesample_keyword=tablesample_keyword 571 ) 572 573 def interval_sql(self, expression: exp.Interval) -> str: 574 multiplier: t.Optional[int] = None 575 unit = expression.text("unit").lower() 576 577 if unit.startswith("week"): 578 multiplier = 7 579 if unit.startswith("quarter"): 580 multiplier = 90 581 582 if multiplier: 583 return f"({multiplier} * {super().interval_sql(exp.Interval(this=expression.this, unit=exp.var('DAY')))})" 584 585 return super().interval_sql(expression) 586 587 def columndef_sql(self, expression: exp.ColumnDef, sep: str = " ") -> str: 588 if isinstance(expression.parent, exp.UserDefinedFunction): 589 return self.sql(expression, "this") 590 return super().columndef_sql(expression, sep) 591 592 def join_sql(self, expression: exp.Join) -> str: 593 if ( 594 expression.side == "LEFT" 595 and not expression.args.get("on") 596 and isinstance(expression.this, exp.Unnest) 597 ): 598 # Some dialects support `LEFT JOIN UNNEST(...)` without an explicit ON clause 599 # DuckDB doesn't, but we can just add a dummy ON clause that is always true 600 return super().join_sql(expression.on(exp.true())) 601 602 return super().join_sql(expression) 603 604 def generateseries_sql(self, expression: exp.GenerateSeries) -> str: 605 # GENERATE_SERIES(a, b) -> [a, b], RANGE(a, b) -> [a, b) 606 if expression.args.get("is_end_exclusive"): 607 expression.set("is_end_exclusive", None) 608 return rename_func("RANGE")(self, expression) 609 610 return super().generateseries_sql(expression) 611 612 def bracket_sql(self, expression: exp.Bracket) -> str: 613 this = expression.this 614 if isinstance(this, exp.Array): 615 this.replace(exp.paren(this)) 616 617 bracket = super().bracket_sql(expression) 618 619 if not expression.args.get("returns_list_for_maps"): 620 if not this.type: 621 from sqlglot.optimizer.annotate_types import annotate_types 622 623 this = annotate_types(this) 624 625 if this.is_type(exp.DataType.Type.MAP): 626 bracket = f"({bracket})[1]" 627 628 return bracket 629 630 def withingroup_sql(self, expression: exp.WithinGroup) -> str: 631 expression_sql = self.sql(expression, "expression") 632 633 func = expression.this 634 if isinstance(func, exp.PERCENTILES): 635 # Make the order key the first arg and slide the fraction to the right 636 # https://duckdb.org/docs/sql/aggregates#ordered-set-aggregate-functions 637 order_col = expression.find(exp.Ordered) 638 if order_col: 639 func.set("expression", func.this) 640 func.set("this", order_col.this) 641 642 this = self.sql(expression, "this").rstrip(")") 643 644 return f"{this}{expression_sql})"
168class DuckDB(Dialect): 169 NULL_ORDERING = "nulls_are_last" 170 SUPPORTS_USER_DEFINED_TYPES = False 171 SAFE_DIVISION = True 172 INDEX_OFFSET = 1 173 CONCAT_COALESCE = True 174 175 # https://duckdb.org/docs/sql/introduction.html#creating-a-new-table 176 NORMALIZATION_STRATEGY = NormalizationStrategy.CASE_INSENSITIVE 177 178 def to_json_path(self, path: t.Optional[exp.Expression]) -> t.Optional[exp.Expression]: 179 if isinstance(path, exp.Literal): 180 # DuckDB also supports the JSON pointer syntax, where every path starts with a `/`. 181 # Additionally, it allows accessing the back of lists using the `[#-i]` syntax. 182 # This check ensures we'll avoid trying to parse these as JSON paths, which can 183 # either result in a noisy warning or in an invalid representation of the path. 184 path_text = path.name 185 if path_text.startswith("/") or "[#" in path_text: 186 return path 187 188 return super().to_json_path(path) 189 190 class Tokenizer(tokens.Tokenizer): 191 HEREDOC_STRINGS = ["$"] 192 193 HEREDOC_TAG_IS_IDENTIFIER = True 194 HEREDOC_STRING_ALTERNATIVE = TokenType.PARAMETER 195 196 KEYWORDS = { 197 **tokens.Tokenizer.KEYWORDS, 198 "//": TokenType.DIV, 199 "ATTACH": TokenType.COMMAND, 200 "BINARY": TokenType.VARBINARY, 201 "BITSTRING": TokenType.BIT, 202 "BPCHAR": TokenType.TEXT, 203 "CHAR": TokenType.TEXT, 204 "CHARACTER VARYING": TokenType.TEXT, 205 "EXCLUDE": TokenType.EXCEPT, 206 "LOGICAL": TokenType.BOOLEAN, 207 "ONLY": TokenType.ONLY, 208 "PIVOT_WIDER": TokenType.PIVOT, 209 "POSITIONAL": TokenType.POSITIONAL, 210 "SIGNED": TokenType.INT, 211 "STRING": TokenType.VARCHAR, 212 "UBIGINT": TokenType.UBIGINT, 213 "UINTEGER": TokenType.UINT, 214 "USMALLINT": TokenType.USMALLINT, 215 "UTINYINT": TokenType.UTINYINT, 216 "TIMESTAMP_S": TokenType.TIMESTAMP_S, 217 "TIMESTAMP_MS": TokenType.TIMESTAMP_MS, 218 "TIMESTAMP_NS": TokenType.TIMESTAMP_NS, 219 "TIMESTAMP_US": TokenType.TIMESTAMP, 220 } 221 222 SINGLE_TOKENS = { 223 **tokens.Tokenizer.SINGLE_TOKENS, 224 "$": TokenType.PARAMETER, 225 } 226 227 class Parser(parser.Parser): 228 BITWISE = { 229 **parser.Parser.BITWISE, 230 TokenType.TILDA: exp.RegexpLike, 231 } 232 233 FUNCTIONS_WITH_ALIASED_ARGS = {*parser.Parser.FUNCTIONS_WITH_ALIASED_ARGS, "STRUCT_PACK"} 234 235 FUNCTIONS = { 236 **parser.Parser.FUNCTIONS, 237 "ARRAY_REVERSE_SORT": _build_sort_array_desc, 238 "ARRAY_SORT": exp.SortArray.from_arg_list, 239 "DATEDIFF": _build_date_diff, 240 "DATE_DIFF": _build_date_diff, 241 "DATE_TRUNC": date_trunc_to_time, 242 "DATETRUNC": date_trunc_to_time, 243 "DECODE": lambda args: exp.Decode( 244 this=seq_get(args, 0), charset=exp.Literal.string("utf-8") 245 ), 246 "ENCODE": lambda args: exp.Encode( 247 this=seq_get(args, 0), charset=exp.Literal.string("utf-8") 248 ), 249 "EPOCH": exp.TimeToUnix.from_arg_list, 250 "EPOCH_MS": lambda args: exp.UnixToTime( 251 this=seq_get(args, 0), scale=exp.UnixToTime.MILLIS 252 ), 253 "JSON": exp.ParseJSON.from_arg_list, 254 "JSON_EXTRACT_PATH": parser.build_extract_json_with_path(exp.JSONExtract), 255 "JSON_EXTRACT_STRING": parser.build_extract_json_with_path(exp.JSONExtractScalar), 256 "LIST_HAS": exp.ArrayContains.from_arg_list, 257 "LIST_REVERSE_SORT": _build_sort_array_desc, 258 "LIST_SORT": exp.SortArray.from_arg_list, 259 "LIST_VALUE": exp.Array.from_arg_list, 260 "MAKE_TIME": exp.TimeFromParts.from_arg_list, 261 "MAKE_TIMESTAMP": _build_make_timestamp, 262 "MEDIAN": lambda args: exp.PercentileCont( 263 this=seq_get(args, 0), expression=exp.Literal.number(0.5) 264 ), 265 "QUANTILE_CONT": exp.PercentileCont.from_arg_list, 266 "QUANTILE_DISC": exp.PercentileDisc.from_arg_list, 267 "REGEXP_EXTRACT": lambda args: exp.RegexpExtract( 268 this=seq_get(args, 0), expression=seq_get(args, 1), group=seq_get(args, 2) 269 ), 270 "REGEXP_MATCHES": exp.RegexpLike.from_arg_list, 271 "REGEXP_REPLACE": lambda args: exp.RegexpReplace( 272 this=seq_get(args, 0), 273 expression=seq_get(args, 1), 274 replacement=seq_get(args, 2), 275 modifiers=seq_get(args, 3), 276 ), 277 "STRFTIME": build_formatted_time(exp.TimeToStr, "duckdb"), 278 "STRING_SPLIT": exp.Split.from_arg_list, 279 "STRING_SPLIT_REGEX": exp.RegexpSplit.from_arg_list, 280 "STRING_TO_ARRAY": exp.Split.from_arg_list, 281 "STRPTIME": build_formatted_time(exp.StrToTime, "duckdb"), 282 "STRUCT_PACK": exp.Struct.from_arg_list, 283 "STR_SPLIT": exp.Split.from_arg_list, 284 "STR_SPLIT_REGEX": exp.RegexpSplit.from_arg_list, 285 "TO_TIMESTAMP": exp.UnixToTime.from_arg_list, 286 "UNNEST": exp.Explode.from_arg_list, 287 "XOR": binary_from_function(exp.BitwiseXor), 288 "GENERATE_SERIES": _build_generate_series(), 289 "RANGE": _build_generate_series(end_exclusive=True), 290 } 291 292 FUNCTIONS.pop("DATE_SUB") 293 294 FUNCTION_PARSERS = parser.Parser.FUNCTION_PARSERS.copy() 295 FUNCTION_PARSERS.pop("DECODE") 296 297 NO_PAREN_FUNCTION_PARSERS = { 298 **parser.Parser.NO_PAREN_FUNCTION_PARSERS, 299 "MAP": lambda self: self._parse_map(), 300 } 301 302 TABLE_ALIAS_TOKENS = parser.Parser.TABLE_ALIAS_TOKENS - { 303 TokenType.SEMI, 304 TokenType.ANTI, 305 } 306 307 PLACEHOLDER_PARSERS = { 308 **parser.Parser.PLACEHOLDER_PARSERS, 309 TokenType.PARAMETER: lambda self: ( 310 self.expression(exp.Placeholder, this=self._prev.text) 311 if self._match(TokenType.NUMBER) or self._match_set(self.ID_VAR_TOKENS) 312 else None 313 ), 314 } 315 316 TYPE_CONVERTER = { 317 # https://duckdb.org/docs/sql/data_types/numeric 318 exp.DataType.Type.DECIMAL: build_default_decimal_type(precision=18, scale=3), 319 } 320 321 def _parse_table_sample(self, as_modifier: bool = False) -> t.Optional[exp.TableSample]: 322 # https://duckdb.org/docs/sql/samples.html 323 sample = super()._parse_table_sample(as_modifier=as_modifier) 324 if sample and not sample.args.get("method"): 325 if sample.args.get("size"): 326 sample.set("method", exp.var("RESERVOIR")) 327 else: 328 sample.set("method", exp.var("SYSTEM")) 329 330 return sample 331 332 def _parse_bracket( 333 self, this: t.Optional[exp.Expression] = None 334 ) -> t.Optional[exp.Expression]: 335 bracket = super()._parse_bracket(this) 336 if isinstance(bracket, exp.Bracket): 337 bracket.set("returns_list_for_maps", True) 338 339 return bracket 340 341 def _parse_map(self) -> exp.ToMap | exp.Map: 342 if self._match(TokenType.L_BRACE, advance=False): 343 return self.expression(exp.ToMap, this=self._parse_bracket()) 344 345 args = self._parse_wrapped_csv(self._parse_conjunction) 346 return self.expression(exp.Map, keys=seq_get(args, 0), values=seq_get(args, 1)) 347 348 def _parse_struct_types(self, type_required: bool = False) -> t.Optional[exp.Expression]: 349 return self._parse_field_def() 350 351 def _pivot_column_names(self, aggregations: t.List[exp.Expression]) -> t.List[str]: 352 if len(aggregations) == 1: 353 return super()._pivot_column_names(aggregations) 354 return pivot_column_names(aggregations, dialect="duckdb") 355 356 class Generator(generator.Generator): 357 PARAMETER_TOKEN = "$" 358 NAMED_PLACEHOLDER_TOKEN = "$" 359 JOIN_HINTS = False 360 TABLE_HINTS = False 361 QUERY_HINTS = False 362 LIMIT_FETCH = "LIMIT" 363 STRUCT_DELIMITER = ("(", ")") 364 RENAME_TABLE_WITH_DB = False 365 NVL2_SUPPORTED = False 366 SEMI_ANTI_JOIN_WITH_SIDE = False 367 TABLESAMPLE_KEYWORDS = "USING SAMPLE" 368 TABLESAMPLE_SEED_KEYWORD = "REPEATABLE" 369 LAST_DAY_SUPPORTS_DATE_PART = False 370 JSON_KEY_VALUE_PAIR_SEP = "," 371 IGNORE_NULLS_IN_FUNC = True 372 JSON_PATH_BRACKETED_KEY_SUPPORTED = False 373 SUPPORTS_CREATE_TABLE_LIKE = False 374 MULTI_ARG_DISTINCT = False 375 CAN_IMPLEMENT_ARRAY_ANY = True 376 SUPPORTS_TO_NUMBER = False 377 COPY_HAS_INTO_KEYWORD = False 378 STAR_EXCEPT = "EXCLUDE" 379 380 TRANSFORMS = { 381 **generator.Generator.TRANSFORMS, 382 exp.ApproxDistinct: approx_count_distinct_sql, 383 exp.Array: inline_array_unless_query, 384 exp.ArrayContainsAll: rename_func("ARRAY_HAS_ALL"), 385 exp.ArrayFilter: rename_func("LIST_FILTER"), 386 exp.ArraySize: rename_func("ARRAY_LENGTH"), 387 exp.ArgMax: arg_max_or_min_no_count("ARG_MAX"), 388 exp.ArgMin: arg_max_or_min_no_count("ARG_MIN"), 389 exp.ArraySort: _array_sort_sql, 390 exp.ArraySum: rename_func("LIST_SUM"), 391 exp.BitwiseXor: rename_func("XOR"), 392 exp.CommentColumnConstraint: no_comment_column_constraint_sql, 393 exp.CurrentDate: lambda *_: "CURRENT_DATE", 394 exp.CurrentTime: lambda *_: "CURRENT_TIME", 395 exp.CurrentTimestamp: lambda *_: "CURRENT_TIMESTAMP", 396 exp.DayOfMonth: rename_func("DAYOFMONTH"), 397 exp.DayOfWeek: rename_func("DAYOFWEEK"), 398 exp.DayOfYear: rename_func("DAYOFYEAR"), 399 exp.DataType: _datatype_sql, 400 exp.Date: _date_sql, 401 exp.DateAdd: _date_delta_sql, 402 exp.DateFromParts: rename_func("MAKE_DATE"), 403 exp.DateSub: _date_delta_sql, 404 exp.DateDiff: lambda self, e: self.func( 405 "DATE_DIFF", f"'{e.args.get('unit') or 'DAY'}'", e.expression, e.this 406 ), 407 exp.DateStrToDate: datestrtodate_sql, 408 exp.DateToDi: lambda self, 409 e: f"CAST(STRFTIME({self.sql(e, 'this')}, {DuckDB.DATEINT_FORMAT}) AS INT)", 410 exp.Decode: lambda self, e: encode_decode_sql(self, e, "DECODE", replace=False), 411 exp.DiToDate: lambda self, 412 e: f"CAST(STRPTIME(CAST({self.sql(e, 'this')} AS TEXT), {DuckDB.DATEINT_FORMAT}) AS DATE)", 413 exp.Encode: lambda self, e: encode_decode_sql(self, e, "ENCODE", replace=False), 414 exp.Explode: rename_func("UNNEST"), 415 exp.IntDiv: lambda self, e: self.binary(e, "//"), 416 exp.IsInf: rename_func("ISINF"), 417 exp.IsNan: rename_func("ISNAN"), 418 exp.JSONExtract: _arrow_json_extract_sql, 419 exp.JSONExtractScalar: _arrow_json_extract_sql, 420 exp.JSONFormat: _json_format_sql, 421 exp.LogicalOr: rename_func("BOOL_OR"), 422 exp.LogicalAnd: rename_func("BOOL_AND"), 423 exp.MD5Digest: lambda self, e: self.func("UNHEX", self.func("MD5", e.this)), 424 exp.MonthsBetween: lambda self, e: self.func( 425 "DATEDIFF", 426 "'month'", 427 exp.cast(e.expression, exp.DataType.Type.TIMESTAMP, copy=True), 428 exp.cast(e.this, exp.DataType.Type.TIMESTAMP, copy=True), 429 ), 430 exp.ParseJSON: rename_func("JSON"), 431 exp.PercentileCont: rename_func("QUANTILE_CONT"), 432 exp.PercentileDisc: rename_func("QUANTILE_DISC"), 433 # DuckDB doesn't allow qualified columns inside of PIVOT expressions. 434 # See: https://github.com/duckdb/duckdb/blob/671faf92411182f81dce42ac43de8bfb05d9909e/src/planner/binder/tableref/bind_pivot.cpp#L61-L62 435 exp.Pivot: transforms.preprocess([transforms.unqualify_columns]), 436 exp.RegexpExtract: regexp_extract_sql, 437 exp.RegexpReplace: lambda self, e: self.func( 438 "REGEXP_REPLACE", 439 e.this, 440 e.expression, 441 e.args.get("replacement"), 442 e.args.get("modifiers"), 443 ), 444 exp.RegexpLike: rename_func("REGEXP_MATCHES"), 445 exp.RegexpSplit: rename_func("STR_SPLIT_REGEX"), 446 exp.Rand: rename_func("RANDOM"), 447 exp.SafeDivide: no_safe_divide_sql, 448 exp.Split: rename_func("STR_SPLIT"), 449 exp.SortArray: _sort_array_sql, 450 exp.StrPosition: str_position_sql, 451 exp.StrToDate: lambda self, e: f"CAST({str_to_time_sql(self, e)} AS DATE)", 452 exp.StrToTime: str_to_time_sql, 453 exp.StrToUnix: lambda self, e: self.func( 454 "EPOCH", self.func("STRPTIME", e.this, self.format_time(e)) 455 ), 456 exp.Struct: _struct_sql, 457 exp.TimeAdd: _date_delta_sql, 458 exp.Timestamp: no_timestamp_sql, 459 exp.TimestampDiff: lambda self, e: self.func( 460 "DATE_DIFF", exp.Literal.string(e.unit), e.expression, e.this 461 ), 462 exp.TimestampTrunc: timestamptrunc_sql(), 463 exp.TimeStrToDate: lambda self, e: self.sql(exp.cast(e.this, exp.DataType.Type.DATE)), 464 exp.TimeStrToTime: timestrtotime_sql, 465 exp.TimeStrToUnix: lambda self, e: self.func( 466 "EPOCH", exp.cast(e.this, exp.DataType.Type.TIMESTAMP) 467 ), 468 exp.TimeToStr: lambda self, e: self.func("STRFTIME", e.this, self.format_time(e)), 469 exp.TimeToUnix: rename_func("EPOCH"), 470 exp.TsOrDiToDi: lambda self, 471 e: f"CAST(SUBSTR(REPLACE(CAST({self.sql(e, 'this')} AS TEXT), '-', ''), 1, 8) AS INT)", 472 exp.TsOrDsAdd: _ts_or_ds_add_sql, 473 exp.TsOrDsDiff: lambda self, e: self.func( 474 "DATE_DIFF", 475 f"'{e.args.get('unit') or 'DAY'}'", 476 exp.cast(e.expression, exp.DataType.Type.TIMESTAMP), 477 exp.cast(e.this, exp.DataType.Type.TIMESTAMP), 478 ), 479 exp.UnixToStr: lambda self, e: self.func( 480 "STRFTIME", self.func("TO_TIMESTAMP", e.this), self.format_time(e) 481 ), 482 exp.UnixToTime: _unix_to_time_sql, 483 exp.UnixToTimeStr: lambda self, e: f"CAST(TO_TIMESTAMP({self.sql(e, 'this')}) AS TEXT)", 484 exp.VariancePop: rename_func("VAR_POP"), 485 exp.WeekOfYear: rename_func("WEEKOFYEAR"), 486 exp.Xor: bool_xor_sql, 487 } 488 489 SUPPORTED_JSON_PATH_PARTS = { 490 exp.JSONPathKey, 491 exp.JSONPathRoot, 492 exp.JSONPathSubscript, 493 exp.JSONPathWildcard, 494 } 495 496 TYPE_MAPPING = { 497 **generator.Generator.TYPE_MAPPING, 498 exp.DataType.Type.BINARY: "BLOB", 499 exp.DataType.Type.CHAR: "TEXT", 500 exp.DataType.Type.FLOAT: "REAL", 501 exp.DataType.Type.NCHAR: "TEXT", 502 exp.DataType.Type.NVARCHAR: "TEXT", 503 exp.DataType.Type.UINT: "UINTEGER", 504 exp.DataType.Type.VARBINARY: "BLOB", 505 exp.DataType.Type.ROWVERSION: "BLOB", 506 exp.DataType.Type.VARCHAR: "TEXT", 507 exp.DataType.Type.TIMESTAMP_S: "TIMESTAMP_S", 508 exp.DataType.Type.TIMESTAMP_MS: "TIMESTAMP_MS", 509 exp.DataType.Type.TIMESTAMP_NS: "TIMESTAMP_NS", 510 } 511 512 UNWRAPPED_INTERVAL_VALUES = (exp.Literal, exp.Paren) 513 514 # DuckDB doesn't generally support CREATE TABLE .. properties 515 # https://duckdb.org/docs/sql/statements/create_table.html 516 PROPERTIES_LOCATION = { 517 prop: exp.Properties.Location.UNSUPPORTED 518 for prop in generator.Generator.PROPERTIES_LOCATION 519 } 520 521 # There are a few exceptions (e.g. temporary tables) which are supported or 522 # can be transpiled to DuckDB, so we explicitly override them accordingly 523 PROPERTIES_LOCATION[exp.LikeProperty] = exp.Properties.Location.POST_SCHEMA 524 PROPERTIES_LOCATION[exp.TemporaryProperty] = exp.Properties.Location.POST_CREATE 525 526 def timefromparts_sql(self, expression: exp.TimeFromParts) -> str: 527 nano = expression.args.get("nano") 528 if nano is not None: 529 expression.set( 530 "sec", expression.args["sec"] + nano.pop() / exp.Literal.number(1000000000.0) 531 ) 532 533 return rename_func("MAKE_TIME")(self, expression) 534 535 def timestampfromparts_sql(self, expression: exp.TimestampFromParts) -> str: 536 sec = expression.args["sec"] 537 538 milli = expression.args.get("milli") 539 if milli is not None: 540 sec += milli.pop() / exp.Literal.number(1000.0) 541 542 nano = expression.args.get("nano") 543 if nano is not None: 544 sec += nano.pop() / exp.Literal.number(1000000000.0) 545 546 if milli or nano: 547 expression.set("sec", sec) 548 549 return rename_func("MAKE_TIMESTAMP")(self, expression) 550 551 def tablesample_sql( 552 self, 553 expression: exp.TableSample, 554 sep: str = " AS ", 555 tablesample_keyword: t.Optional[str] = None, 556 ) -> str: 557 if not isinstance(expression.parent, exp.Select): 558 # This sample clause only applies to a single source, not the entire resulting relation 559 tablesample_keyword = "TABLESAMPLE" 560 561 if expression.args.get("size"): 562 method = expression.args.get("method") 563 if method and method.name.upper() != "RESERVOIR": 564 self.unsupported( 565 f"Sampling method {method} is not supported with a discrete sample count, " 566 "defaulting to reservoir sampling" 567 ) 568 expression.set("method", exp.var("RESERVOIR")) 569 570 return super().tablesample_sql( 571 expression, sep=sep, tablesample_keyword=tablesample_keyword 572 ) 573 574 def interval_sql(self, expression: exp.Interval) -> str: 575 multiplier: t.Optional[int] = None 576 unit = expression.text("unit").lower() 577 578 if unit.startswith("week"): 579 multiplier = 7 580 if unit.startswith("quarter"): 581 multiplier = 90 582 583 if multiplier: 584 return f"({multiplier} * {super().interval_sql(exp.Interval(this=expression.this, unit=exp.var('DAY')))})" 585 586 return super().interval_sql(expression) 587 588 def columndef_sql(self, expression: exp.ColumnDef, sep: str = " ") -> str: 589 if isinstance(expression.parent, exp.UserDefinedFunction): 590 return self.sql(expression, "this") 591 return super().columndef_sql(expression, sep) 592 593 def join_sql(self, expression: exp.Join) -> str: 594 if ( 595 expression.side == "LEFT" 596 and not expression.args.get("on") 597 and isinstance(expression.this, exp.Unnest) 598 ): 599 # Some dialects support `LEFT JOIN UNNEST(...)` without an explicit ON clause 600 # DuckDB doesn't, but we can just add a dummy ON clause that is always true 601 return super().join_sql(expression.on(exp.true())) 602 603 return super().join_sql(expression) 604 605 def generateseries_sql(self, expression: exp.GenerateSeries) -> str: 606 # GENERATE_SERIES(a, b) -> [a, b], RANGE(a, b) -> [a, b) 607 if expression.args.get("is_end_exclusive"): 608 expression.set("is_end_exclusive", None) 609 return rename_func("RANGE")(self, expression) 610 611 return super().generateseries_sql(expression) 612 613 def bracket_sql(self, expression: exp.Bracket) -> str: 614 this = expression.this 615 if isinstance(this, exp.Array): 616 this.replace(exp.paren(this)) 617 618 bracket = super().bracket_sql(expression) 619 620 if not expression.args.get("returns_list_for_maps"): 621 if not this.type: 622 from sqlglot.optimizer.annotate_types import annotate_types 623 624 this = annotate_types(this) 625 626 if this.is_type(exp.DataType.Type.MAP): 627 bracket = f"({bracket})[1]" 628 629 return bracket 630 631 def withingroup_sql(self, expression: exp.WithinGroup) -> str: 632 expression_sql = self.sql(expression, "expression") 633 634 func = expression.this 635 if isinstance(func, exp.PERCENTILES): 636 # Make the order key the first arg and slide the fraction to the right 637 # https://duckdb.org/docs/sql/aggregates#ordered-set-aggregate-functions 638 order_col = expression.find(exp.Ordered) 639 if order_col: 640 func.set("expression", func.this) 641 func.set("this", order_col.this) 642 643 this = self.sql(expression, "this").rstrip(")") 644 645 return f"{this}{expression_sql})"
NULL_ORDERING =
'nulls_are_last'
Default NULL
ordering method to use if not explicitly set.
Possible values: "nulls_are_small"
, "nulls_are_large"
, "nulls_are_last"
CONCAT_COALESCE =
True
A NULL
arg in CONCAT
yields NULL
by default, but in some dialects it yields an empty string.
NORMALIZATION_STRATEGY =
<NormalizationStrategy.CASE_INSENSITIVE: 'CASE_INSENSITIVE'>
Specifies the strategy according to which identifiers should be normalized.
def
to_json_path( self, path: Optional[sqlglot.expressions.Expression]) -> Optional[sqlglot.expressions.Expression]:
178 def to_json_path(self, path: t.Optional[exp.Expression]) -> t.Optional[exp.Expression]: 179 if isinstance(path, exp.Literal): 180 # DuckDB also supports the JSON pointer syntax, where every path starts with a `/`. 181 # Additionally, it allows accessing the back of lists using the `[#-i]` syntax. 182 # This check ensures we'll avoid trying to parse these as JSON paths, which can 183 # either result in a noisy warning or in an invalid representation of the path. 184 path_text = path.name 185 if path_text.startswith("/") or "[#" in path_text: 186 return path 187 188 return super().to_json_path(path)
tokenizer_class =
<class 'DuckDB.Tokenizer'>
parser_class =
<class 'DuckDB.Parser'>
generator_class =
<class 'DuckDB.Generator'>
Inherited Members
- sqlglot.dialects.dialect.Dialect
- Dialect
- WEEK_OFFSET
- UNNEST_COLUMN_ONLY
- ALIAS_POST_TABLESAMPLE
- TABLESAMPLE_SIZE_IS_PERCENT
- IDENTIFIERS_CAN_START_WITH_DIGIT
- DPIPE_IS_STRING_CONCAT
- STRICT_STRING_CONCAT
- SUPPORTS_SEMI_ANTI_JOIN
- NORMALIZE_FUNCTIONS
- LOG_BASE_FIRST
- TYPED_DIVISION
- HEX_LOWERCASE
- DATE_FORMAT
- DATEINT_FORMAT
- TIME_FORMAT
- TIME_MAPPING
- FORMAT_MAPPING
- UNESCAPED_SEQUENCES
- PSEUDOCOLUMNS
- PREFER_CTE_ALIAS_COLUMN
- COPY_PARAMS_ARE_CSV
- get_or_raise
- format_time
- normalize_identifier
- case_sensitive
- can_identify
- quote_identifier
- parse
- parse_into
- generate
- transpile
- tokenize
- tokenizer
- parser
- generator
190 class Tokenizer(tokens.Tokenizer): 191 HEREDOC_STRINGS = ["$"] 192 193 HEREDOC_TAG_IS_IDENTIFIER = True 194 HEREDOC_STRING_ALTERNATIVE = TokenType.PARAMETER 195 196 KEYWORDS = { 197 **tokens.Tokenizer.KEYWORDS, 198 "//": TokenType.DIV, 199 "ATTACH": TokenType.COMMAND, 200 "BINARY": TokenType.VARBINARY, 201 "BITSTRING": TokenType.BIT, 202 "BPCHAR": TokenType.TEXT, 203 "CHAR": TokenType.TEXT, 204 "CHARACTER VARYING": TokenType.TEXT, 205 "EXCLUDE": TokenType.EXCEPT, 206 "LOGICAL": TokenType.BOOLEAN, 207 "ONLY": TokenType.ONLY, 208 "PIVOT_WIDER": TokenType.PIVOT, 209 "POSITIONAL": TokenType.POSITIONAL, 210 "SIGNED": TokenType.INT, 211 "STRING": TokenType.VARCHAR, 212 "UBIGINT": TokenType.UBIGINT, 213 "UINTEGER": TokenType.UINT, 214 "USMALLINT": TokenType.USMALLINT, 215 "UTINYINT": TokenType.UTINYINT, 216 "TIMESTAMP_S": TokenType.TIMESTAMP_S, 217 "TIMESTAMP_MS": TokenType.TIMESTAMP_MS, 218 "TIMESTAMP_NS": TokenType.TIMESTAMP_NS, 219 "TIMESTAMP_US": TokenType.TIMESTAMP, 220 } 221 222 SINGLE_TOKENS = { 223 **tokens.Tokenizer.SINGLE_TOKENS, 224 "$": TokenType.PARAMETER, 225 }
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.COLON_EQ: 'COLON_EQ'>, '<=>': <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'>, 'COPY': <TokenType.COPY: 'COPY'>, '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'>, 'ENUM': <TokenType.ENUM: 'ENUM'>, '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'>, 'TRUNCATE': <TokenType.TRUNCATE: 'TRUNCATE'>, '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'>, 'INT1': <TokenType.TINYINT: 'TINYINT'>, 'TINYINT': <TokenType.TINYINT: 'TINYINT'>, 'INT16': <TokenType.SMALLINT: 'SMALLINT'>, 'SHORT': <TokenType.SMALLINT: 'SMALLINT'>, 'SMALLINT': <TokenType.SMALLINT: 'SMALLINT'>, 'INT128': <TokenType.INT128: 'INT128'>, 'HUGEINT': <TokenType.INT128: 'INT128'>, 'INT2': <TokenType.SMALLINT: 'SMALLINT'>, 'INTEGER': <TokenType.INT: 'INT'>, 'INT': <TokenType.INT: 'INT'>, 'INT4': <TokenType.INT: 'INT'>, 'INT32': <TokenType.INT: 'INT'>, 'INT64': <TokenType.BIGINT: 'BIGINT'>, 'LONG': <TokenType.BIGINT: 'BIGINT'>, 'BIGINT': <TokenType.BIGINT: 'BIGINT'>, 'INT8': <TokenType.TINYINT: 'TINYINT'>, 'UINT': <TokenType.UINT: 'UINT'>, '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'>, 'JSONB': <TokenType.JSONB: 'JSONB'>, 'CHAR': <TokenType.TEXT: 'TEXT'>, 'CHARACTER': <TokenType.CHAR: 'CHAR'>, 'NCHAR': <TokenType.NCHAR: 'NCHAR'>, 'VARCHAR': <TokenType.VARCHAR: 'VARCHAR'>, 'VARCHAR2': <TokenType.VARCHAR: 'VARCHAR'>, 'NVARCHAR': <TokenType.NVARCHAR: 'NVARCHAR'>, 'NVARCHAR2': <TokenType.NVARCHAR: 'NVARCHAR'>, 'BPCHAR': <TokenType.TEXT: 'TEXT'>, 'STR': <TokenType.TEXT: 'TEXT'>, 'STRING': <TokenType.VARCHAR: 'VARCHAR'>, '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.VARBINARY: 'VARBINARY'>, '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'>, 'TIMESTAMP_LTZ': <TokenType.TIMESTAMPLTZ: 'TIMESTAMPLTZ'>, 'TIMESTAMPNTZ': <TokenType.TIMESTAMPNTZ: 'TIMESTAMPNTZ'>, 'TIMESTAMP_NTZ': <TokenType.TIMESTAMPNTZ: 'TIMESTAMPNTZ'>, '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'>, 'SEQUENCE': <TokenType.SEQUENCE: 'SEQUENCE'>, 'VARIANT': <TokenType.VARIANT: 'VARIANT'>, 'ALTER': <TokenType.ALTER: 'ALTER'>, 'ANALYZE': <TokenType.COMMAND: 'COMMAND'>, 'CALL': <TokenType.COMMAND: 'COMMAND'>, 'COMMENT': <TokenType.COMMENT: 'COMMENT'>, 'EXPLAIN': <TokenType.COMMAND: 'COMMAND'>, 'GRANT': <TokenType.COMMAND: 'COMMAND'>, 'OPTIMIZE': <TokenType.COMMAND: 'COMMAND'>, 'PREPARE': <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'>, '//': <TokenType.DIV: 'DIV'>, 'ATTACH': <TokenType.COMMAND: 'COMMAND'>, 'BITSTRING': <TokenType.BIT: 'BIT'>, 'CHARACTER VARYING': <TokenType.TEXT: 'TEXT'>, 'EXCLUDE': <TokenType.EXCEPT: 'EXCEPT'>, 'LOGICAL': <TokenType.BOOLEAN: 'BOOLEAN'>, 'ONLY': <TokenType.ONLY: 'ONLY'>, 'PIVOT_WIDER': <TokenType.PIVOT: 'PIVOT'>, 'POSITIONAL': <TokenType.POSITIONAL: 'POSITIONAL'>, 'SIGNED': <TokenType.INT: 'INT'>, 'UBIGINT': <TokenType.UBIGINT: 'UBIGINT'>, 'UINTEGER': <TokenType.UINT: 'UINT'>, 'USMALLINT': <TokenType.USMALLINT: 'USMALLINT'>, 'UTINYINT': <TokenType.UTINYINT: 'UTINYINT'>, 'TIMESTAMP_S': <TokenType.TIMESTAMP_S: 'TIMESTAMP_S'>, 'TIMESTAMP_MS': <TokenType.TIMESTAMP_MS: 'TIMESTAMP_MS'>, 'TIMESTAMP_NS': <TokenType.TIMESTAMP_NS: 'TIMESTAMP_NS'>, 'TIMESTAMP_US': <TokenType.TIMESTAMP: 'TIMESTAMP'>}
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.HASH: 'HASH'>, "'": <TokenType.UNKNOWN: 'UNKNOWN'>, '`': <TokenType.UNKNOWN: 'UNKNOWN'>, '"': <TokenType.UNKNOWN: 'UNKNOWN'>, '$': <TokenType.PARAMETER: 'PARAMETER'>}
Inherited Members
227 class Parser(parser.Parser): 228 BITWISE = { 229 **parser.Parser.BITWISE, 230 TokenType.TILDA: exp.RegexpLike, 231 } 232 233 FUNCTIONS_WITH_ALIASED_ARGS = {*parser.Parser.FUNCTIONS_WITH_ALIASED_ARGS, "STRUCT_PACK"} 234 235 FUNCTIONS = { 236 **parser.Parser.FUNCTIONS, 237 "ARRAY_REVERSE_SORT": _build_sort_array_desc, 238 "ARRAY_SORT": exp.SortArray.from_arg_list, 239 "DATEDIFF": _build_date_diff, 240 "DATE_DIFF": _build_date_diff, 241 "DATE_TRUNC": date_trunc_to_time, 242 "DATETRUNC": date_trunc_to_time, 243 "DECODE": lambda args: exp.Decode( 244 this=seq_get(args, 0), charset=exp.Literal.string("utf-8") 245 ), 246 "ENCODE": lambda args: exp.Encode( 247 this=seq_get(args, 0), charset=exp.Literal.string("utf-8") 248 ), 249 "EPOCH": exp.TimeToUnix.from_arg_list, 250 "EPOCH_MS": lambda args: exp.UnixToTime( 251 this=seq_get(args, 0), scale=exp.UnixToTime.MILLIS 252 ), 253 "JSON": exp.ParseJSON.from_arg_list, 254 "JSON_EXTRACT_PATH": parser.build_extract_json_with_path(exp.JSONExtract), 255 "JSON_EXTRACT_STRING": parser.build_extract_json_with_path(exp.JSONExtractScalar), 256 "LIST_HAS": exp.ArrayContains.from_arg_list, 257 "LIST_REVERSE_SORT": _build_sort_array_desc, 258 "LIST_SORT": exp.SortArray.from_arg_list, 259 "LIST_VALUE": exp.Array.from_arg_list, 260 "MAKE_TIME": exp.TimeFromParts.from_arg_list, 261 "MAKE_TIMESTAMP": _build_make_timestamp, 262 "MEDIAN": lambda args: exp.PercentileCont( 263 this=seq_get(args, 0), expression=exp.Literal.number(0.5) 264 ), 265 "QUANTILE_CONT": exp.PercentileCont.from_arg_list, 266 "QUANTILE_DISC": exp.PercentileDisc.from_arg_list, 267 "REGEXP_EXTRACT": lambda args: exp.RegexpExtract( 268 this=seq_get(args, 0), expression=seq_get(args, 1), group=seq_get(args, 2) 269 ), 270 "REGEXP_MATCHES": exp.RegexpLike.from_arg_list, 271 "REGEXP_REPLACE": lambda args: exp.RegexpReplace( 272 this=seq_get(args, 0), 273 expression=seq_get(args, 1), 274 replacement=seq_get(args, 2), 275 modifiers=seq_get(args, 3), 276 ), 277 "STRFTIME": build_formatted_time(exp.TimeToStr, "duckdb"), 278 "STRING_SPLIT": exp.Split.from_arg_list, 279 "STRING_SPLIT_REGEX": exp.RegexpSplit.from_arg_list, 280 "STRING_TO_ARRAY": exp.Split.from_arg_list, 281 "STRPTIME": build_formatted_time(exp.StrToTime, "duckdb"), 282 "STRUCT_PACK": exp.Struct.from_arg_list, 283 "STR_SPLIT": exp.Split.from_arg_list, 284 "STR_SPLIT_REGEX": exp.RegexpSplit.from_arg_list, 285 "TO_TIMESTAMP": exp.UnixToTime.from_arg_list, 286 "UNNEST": exp.Explode.from_arg_list, 287 "XOR": binary_from_function(exp.BitwiseXor), 288 "GENERATE_SERIES": _build_generate_series(), 289 "RANGE": _build_generate_series(end_exclusive=True), 290 } 291 292 FUNCTIONS.pop("DATE_SUB") 293 294 FUNCTION_PARSERS = parser.Parser.FUNCTION_PARSERS.copy() 295 FUNCTION_PARSERS.pop("DECODE") 296 297 NO_PAREN_FUNCTION_PARSERS = { 298 **parser.Parser.NO_PAREN_FUNCTION_PARSERS, 299 "MAP": lambda self: self._parse_map(), 300 } 301 302 TABLE_ALIAS_TOKENS = parser.Parser.TABLE_ALIAS_TOKENS - { 303 TokenType.SEMI, 304 TokenType.ANTI, 305 } 306 307 PLACEHOLDER_PARSERS = { 308 **parser.Parser.PLACEHOLDER_PARSERS, 309 TokenType.PARAMETER: lambda self: ( 310 self.expression(exp.Placeholder, this=self._prev.text) 311 if self._match(TokenType.NUMBER) or self._match_set(self.ID_VAR_TOKENS) 312 else None 313 ), 314 } 315 316 TYPE_CONVERTER = { 317 # https://duckdb.org/docs/sql/data_types/numeric 318 exp.DataType.Type.DECIMAL: build_default_decimal_type(precision=18, scale=3), 319 } 320 321 def _parse_table_sample(self, as_modifier: bool = False) -> t.Optional[exp.TableSample]: 322 # https://duckdb.org/docs/sql/samples.html 323 sample = super()._parse_table_sample(as_modifier=as_modifier) 324 if sample and not sample.args.get("method"): 325 if sample.args.get("size"): 326 sample.set("method", exp.var("RESERVOIR")) 327 else: 328 sample.set("method", exp.var("SYSTEM")) 329 330 return sample 331 332 def _parse_bracket( 333 self, this: t.Optional[exp.Expression] = None 334 ) -> t.Optional[exp.Expression]: 335 bracket = super()._parse_bracket(this) 336 if isinstance(bracket, exp.Bracket): 337 bracket.set("returns_list_for_maps", True) 338 339 return bracket 340 341 def _parse_map(self) -> exp.ToMap | exp.Map: 342 if self._match(TokenType.L_BRACE, advance=False): 343 return self.expression(exp.ToMap, this=self._parse_bracket()) 344 345 args = self._parse_wrapped_csv(self._parse_conjunction) 346 return self.expression(exp.Map, keys=seq_get(args, 0), values=seq_get(args, 1)) 347 348 def _parse_struct_types(self, type_required: bool = False) -> t.Optional[exp.Expression]: 349 return self._parse_field_def() 350 351 def _pivot_column_names(self, aggregations: t.List[exp.Expression]) -> t.List[str]: 352 if len(aggregations) == 1: 353 return super()._pivot_column_names(aggregations) 354 return pivot_column_names(aggregations, dialect="duckdb")
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: 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
BITWISE =
{<TokenType.AMP: 'AMP'>: <class 'sqlglot.expressions.BitwiseAnd'>, <TokenType.CARET: 'CARET'>: <class 'sqlglot.expressions.BitwiseXor'>, <TokenType.PIPE: 'PIPE'>: <class 'sqlglot.expressions.BitwiseOr'>, <TokenType.TILDA: 'TILDA'>: <class 'sqlglot.expressions.RegexpLike'>}
FUNCTIONS =
{'ABS': <bound method Func.from_arg_list of <class 'sqlglot.expressions.Abs'>>, 'ADD_MONTHS': <bound method Func.from_arg_list of <class 'sqlglot.expressions.AddMonths'>>, 'ANONYMOUS_AGG_FUNC': <bound method Func.from_arg_list of <class 'sqlglot.expressions.AnonymousAggFunc'>>, '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_CONSTRUCT_COMPACT': <bound method Func.from_arg_list of <class 'sqlglot.expressions.ArrayConstructCompact'>>, 'ARRAY_CONTAINS': <bound method Func.from_arg_list of <class 'sqlglot.expressions.ArrayContains'>>, 'ARRAY_HAS': <bound method Func.from_arg_list of <class 'sqlglot.expressions.ArrayContains'>>, 'ARRAY_CONTAINS_ALL': <bound method Func.from_arg_list of <class 'sqlglot.expressions.ArrayContainsAll'>>, 'ARRAY_HAS_ALL': <bound method Func.from_arg_list of <class 'sqlglot.expressions.ArrayContainsAll'>>, '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_OVERLAPS': <bound method Func.from_arg_list of <class 'sqlglot.expressions.ArrayOverlaps'>>, 'ARRAY_SIZE': <bound method Func.from_arg_list of <class 'sqlglot.expressions.ArraySize'>>, 'ARRAY_LENGTH': <bound method Func.from_arg_list of <class 'sqlglot.expressions.ArraySize'>>, 'ARRAY_SORT': <bound method Func.from_arg_list of <class 'sqlglot.expressions.SortArray'>>, 'ARRAY_SUM': <bound method Func.from_arg_list of <class 'sqlglot.expressions.ArraySum'>>, 'ARRAY_TO_STRING': <bound method Func.from_arg_list of <class 'sqlglot.expressions.ArrayToString'>>, 'ARRAY_JOIN': <bound method Func.from_arg_list of <class 'sqlglot.expressions.ArrayToString'>>, 'ARRAY_UNION_AGG': <bound method Func.from_arg_list of <class 'sqlglot.expressions.ArrayUnionAgg'>>, 'ARRAY_UNIQUE_AGG': <bound method Func.from_arg_list of <class 'sqlglot.expressions.ArrayUniqueAgg'>>, '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'>>, 'CBRT': <bound method Func.from_arg_list of <class 'sqlglot.expressions.Cbrt'>>, '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'>>, 'COMBINED_AGG_FUNC': <bound method Func.from_arg_list of <class 'sqlglot.expressions.CombinedAggFunc'>>, 'COMBINED_PARAMETERIZED_AGG': <bound method Func.from_arg_list of <class 'sqlglot.expressions.CombinedParameterizedAgg'>>, 'CONCAT': <function Parser.<lambda>>, 'CONCAT_WS': <function Parser.<lambda>>, 'CONNECT_BY_ROOT': <bound method Func.from_arg_list of <class 'sqlglot.expressions.ConnectByRoot'>>, 'CONVERT': <bound method Func.from_arg_list of <class 'sqlglot.expressions.Convert'>>, 'CORR': <bound method Func.from_arg_list of <class 'sqlglot.expressions.Corr'>>, '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'>>, 'COUNTIF': <bound method Func.from_arg_list of <class 'sqlglot.expressions.CountIf'>>, 'COVAR_POP': <bound method Func.from_arg_list of <class 'sqlglot.expressions.CovarPop'>>, 'COVAR_SAMP': <bound method Func.from_arg_list of <class 'sqlglot.expressions.CovarSamp'>>, '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': <bound method Func.from_arg_list of <class 'sqlglot.expressions.DateAdd'>>, 'DATEDIFF': <function _build_date_diff>, 'DATE_DIFF': <function _build_date_diff>, 'DATE_FROM_PARTS': <bound method Func.from_arg_list of <class 'sqlglot.expressions.DateFromParts'>>, '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_TO_DATE_STR': <function Parser.<lambda>>, 'DATE_TO_DI': <bound method Func.from_arg_list of <class 'sqlglot.expressions.DateToDi'>>, 'DATE_TRUNC': <function date_trunc_to_time>, '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': <function DuckDB.Parser.<lambda>>, 'DI_TO_DATE': <bound method Func.from_arg_list of <class 'sqlglot.expressions.DiToDate'>>, 'ENCODE': <function DuckDB.Parser.<lambda>>, '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'>>, 'FIRST_VALUE': <bound method Func.from_arg_list of <class 'sqlglot.expressions.FirstValue'>>, '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_DATE_ARRAY': <bound method Func.from_arg_list of <class 'sqlglot.expressions.GenerateDateArray'>>, 'GENERATE_SERIES': <function _build_generate_series.<locals>._builder>, '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': <function build_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'>>, 'IIF': <bound method Func.from_arg_list of <class 'sqlglot.expressions.If'>>, 'INITCAP': <bound method Func.from_arg_list of <class 'sqlglot.expressions.Initcap'>>, 'IS_INF': <bound method Func.from_arg_list of <class 'sqlglot.expressions.IsInf'>>, 'ISINF': <bound method Func.from_arg_list of <class 'sqlglot.expressions.IsInf'>>, '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': <function build_extract_json_with_path.<locals>._builder>, 'JSON_EXTRACT_SCALAR': <function build_extract_json_with_path.<locals>._builder>, '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_OBJECT_AGG': <bound method Func.from_arg_list of <class 'sqlglot.expressions.JSONObjectAgg'>>, 'J_S_O_N_TABLE': <bound method Func.from_arg_list of <class 'sqlglot.expressions.JSONTable'>>, 'LAG': <bound method Func.from_arg_list of <class 'sqlglot.expressions.Lag'>>, 'LAST': <bound method Func.from_arg_list of <class 'sqlglot.expressions.Last'>>, 'LAST_DAY': <bound method Func.from_arg_list of <class 'sqlglot.expressions.LastDay'>>, 'LAST_DAY_OF_MONTH': <bound method Func.from_arg_list of <class 'sqlglot.expressions.LastDay'>>, 'LAST_VALUE': <bound method Func.from_arg_list of <class 'sqlglot.expressions.LastValue'>>, 'LEAD': <bound method Func.from_arg_list of <class 'sqlglot.expressions.Lead'>>, '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': <function build_logarithm>, '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': <function build_lower>, 'LCASE': <bound method Func.from_arg_list of <class 'sqlglot.expressions.Lower'>>, 'LOWER_HEX': <bound method Func.from_arg_list of <class 'sqlglot.expressions.LowerHex'>>, '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': <bound method Func.from_arg_list of <class 'sqlglot.expressions.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'>>, 'NTH_VALUE': <bound method Func.from_arg_list of <class 'sqlglot.expressions.NthValue'>>, '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'>>, 'QUARTER': <bound method Func.from_arg_list of <class 'sqlglot.expressions.Quarter'>>, 'RAND': <bound method Func.from_arg_list of <class 'sqlglot.expressions.Rand'>>, 'RANDOM': <bound method Func.from_arg_list of <class 'sqlglot.expressions.Rand'>>, 'RANDN': <bound method Func.from_arg_list of <class 'sqlglot.expressions.Randn'>>, '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': <function DuckDB.Parser.<lambda>>, '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': <function DuckDB.Parser.<lambda>>, '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_DIVIDE': <bound method Func.from_arg_list of <class 'sqlglot.expressions.SafeDivide'>>, 'SIGN': <bound method Func.from_arg_list of <class 'sqlglot.expressions.Sign'>>, 'SIGNUM': <bound method Func.from_arg_list of <class 'sqlglot.expressions.Sign'>>, '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'>>, 'STRING_TO_ARRAY': <bound method Func.from_arg_list of <class 'sqlglot.expressions.Split'>>, 'SPLIT_BY_STRING': <bound method Func.from_arg_list of <class 'sqlglot.expressions.StringToArray'>>, '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_FROM_PARTS': <bound method Func.from_arg_list of <class 'sqlglot.expressions.TimeFromParts'>>, 'TIMEFROMPARTS': <bound method Func.from_arg_list of <class 'sqlglot.expressions.TimeFromParts'>>, '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'>>, 'TIMESTAMPDIFF': <bound method Func.from_arg_list of <class 'sqlglot.expressions.TimestampDiff'>>, 'TIMESTAMP_DIFF': <bound method Func.from_arg_list of <class 'sqlglot.expressions.TimestampDiff'>>, 'TIMESTAMP_FROM_PARTS': <bound method Func.from_arg_list of <class 'sqlglot.expressions.TimestampFromParts'>>, 'TIMESTAMPFROMPARTS': <bound method Func.from_arg_list of <class 'sqlglot.expressions.TimestampFromParts'>>, '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_ARRAY': <bound method Func.from_arg_list of <class 'sqlglot.expressions.ToArray'>>, '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'>>, 'TO_MAP': <bound method Func.from_arg_list of <class 'sqlglot.expressions.ToMap'>>, 'TO_NUMBER': <bound method Func.from_arg_list of <class 'sqlglot.expressions.ToNumber'>>, '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': <bound method Func.from_arg_list of <class 'sqlglot.expressions.Try'>>, '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_DIFF': <bound method Func.from_arg_list of <class 'sqlglot.expressions.TsOrDsDiff'>>, '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>>, 'TS_OR_DS_TO_TIME': <bound method Func.from_arg_list of <class 'sqlglot.expressions.TsOrDsToTime'>>, 'TS_OR_DS_TO_TIMESTAMP': <bound method Func.from_arg_list of <class 'sqlglot.expressions.TsOrDsToTimestamp'>>, 'UNHEX': <bound method Func.from_arg_list of <class 'sqlglot.expressions.Unhex'>>, 'UNIX_DATE': <bound method Func.from_arg_list of <class 'sqlglot.expressions.UnixDate'>>, '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': <function build_upper>, 'UCASE': <bound method Func.from_arg_list of <class 'sqlglot.expressions.Upper'>>, 'VAR_MAP': <function build_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 binary_from_function.<locals>.<lambda>>, 'YEAR': <bound method Func.from_arg_list of <class 'sqlglot.expressions.Year'>>, 'GLOB': <function Parser.<lambda>>, 'JSON_EXTRACT_PATH_TEXT': <function build_extract_json_with_path.<locals>._builder>, 'LIKE': <function build_like>, 'LOG2': <function Parser.<lambda>>, 'LOG10': <function Parser.<lambda>>, 'MOD': <function build_mod>, 'TO_HEX': <function build_hex>, 'ARRAY_REVERSE_SORT': <function _build_sort_array_desc>, 'DATETRUNC': <function date_trunc_to_time>, 'EPOCH': <bound method Func.from_arg_list of <class 'sqlglot.expressions.TimeToUnix'>>, 'EPOCH_MS': <function DuckDB.Parser.<lambda>>, 'JSON': <bound method Func.from_arg_list of <class 'sqlglot.expressions.ParseJSON'>>, 'JSON_EXTRACT_PATH': <function build_extract_json_with_path.<locals>._builder>, 'JSON_EXTRACT_STRING': <function build_extract_json_with_path.<locals>._builder>, 'LIST_HAS': <bound method Func.from_arg_list of <class 'sqlglot.expressions.ArrayContains'>>, 'LIST_REVERSE_SORT': <function _build_sort_array_desc>, 'LIST_SORT': <bound method Func.from_arg_list of <class 'sqlglot.expressions.SortArray'>>, 'LIST_VALUE': <bound method Func.from_arg_list of <class 'sqlglot.expressions.Array'>>, 'MAKE_TIME': <bound method Func.from_arg_list of <class 'sqlglot.expressions.TimeFromParts'>>, 'MAKE_TIMESTAMP': <function _build_make_timestamp>, 'MEDIAN': <function DuckDB.Parser.<lambda>>, 'QUANTILE_CONT': <bound method Func.from_arg_list of <class 'sqlglot.expressions.PercentileCont'>>, 'QUANTILE_DISC': <bound method Func.from_arg_list of <class 'sqlglot.expressions.PercentileDisc'>>, 'REGEXP_MATCHES': <bound method Func.from_arg_list of <class 'sqlglot.expressions.RegexpLike'>>, 'STRFTIME': <function build_formatted_time.<locals>._builder>, 'STRING_SPLIT': <bound method Func.from_arg_list of <class 'sqlglot.expressions.Split'>>, 'STRING_SPLIT_REGEX': <bound method Func.from_arg_list of <class 'sqlglot.expressions.RegexpSplit'>>, 'STRPTIME': <function build_formatted_time.<locals>._builder>, 'STRUCT_PACK': <bound method Func.from_arg_list of <class 'sqlglot.expressions.Struct'>>, 'STR_SPLIT': <bound method Func.from_arg_list of <class 'sqlglot.expressions.Split'>>, 'STR_SPLIT_REGEX': <bound method Func.from_arg_list of <class 'sqlglot.expressions.RegexpSplit'>>, 'TO_TIMESTAMP': <bound method Func.from_arg_list of <class 'sqlglot.expressions.UnixToTime'>>, 'UNNEST': <bound method Func.from_arg_list of <class 'sqlglot.expressions.Explode'>>, 'RANGE': <function _build_generate_series.<locals>._builder>}
FUNCTION_PARSERS =
{'CAST': <function Parser.<lambda>>, 'CONVERT': <function Parser.<lambda>>, 'EXTRACT': <function Parser.<lambda>>, 'JSON_OBJECT': <function Parser.<lambda>>, 'JSON_OBJECTAGG': <function Parser.<lambda>>, 'JSON_TABLE': <function Parser.<lambda>>, 'MATCH': <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>>}
NO_PAREN_FUNCTION_PARSERS =
{'ANY': <function Parser.<lambda>>, 'CASE': <function Parser.<lambda>>, 'IF': <function Parser.<lambda>>, 'NEXT': <function Parser.<lambda>>, 'MAP': <function DuckDB.Parser.<lambda>>}
TABLE_ALIAS_TOKENS =
{<TokenType.DATETIME64: 'DATETIME64'>, <TokenType.CASE: 'CASE'>, <TokenType.FUNCTION: 'FUNCTION'>, <TokenType.FILTER: 'FILTER'>, <TokenType.DATABASE: 'DATABASE'>, <TokenType.ANY: 'ANY'>, <TokenType.CURRENT_DATETIME: 'CURRENT_DATETIME'>, <TokenType.TRUNCATE: 'TRUNCATE'>, <TokenType.FOREIGN_KEY: 'FOREIGN_KEY'>, <TokenType.BOOLEAN: 'BOOLEAN'>, <TokenType.IS: 'IS'>, <TokenType.SUPER: 'SUPER'>, <TokenType.NUMRANGE: 'NUMRANGE'>, <TokenType.FIRST: 'FIRST'>, <TokenType.DATEMULTIRANGE: 'DATEMULTIRANGE'>, <TokenType.NESTED: 'NESTED'>, <TokenType.DESC: 'DESC'>, <TokenType.UINT: 'UINT'>, <TokenType.PSEUDO_TYPE: 'PSEUDO_TYPE'>, <TokenType.TINYBLOB: 'TINYBLOB'>, <TokenType.IPV4: 'IPV4'>, <TokenType.REPLACE: 'REPLACE'>, <TokenType.MONEY: 'MONEY'>, <TokenType.PARTITION: 'PARTITION'>, <TokenType.SHOW: 'SHOW'>, <TokenType.COLUMN: 'COLUMN'>, <TokenType.HLLSKETCH: 'HLLSKETCH'>, <TokenType.BPCHAR: 'BPCHAR'>, <TokenType.USMALLINT: 'USMALLINT'>, <TokenType.NVARCHAR: 'NVARCHAR'>, <TokenType.INTERVAL: 'INTERVAL'>, <TokenType.IDENTIFIER: 'IDENTIFIER'>, <TokenType.TABLE: 'TABLE'>, <TokenType.SEQUENCE: 'SEQUENCE'>, <TokenType.USE: 'USE'>, <TokenType.FORMAT: 'FORMAT'>, <TokenType.ESCAPE: 'ESCAPE'>, <TokenType.XML: 'XML'>, <TokenType.MEDIUMBLOB: 'MEDIUMBLOB'>, <TokenType.NULLABLE: 'NULLABLE'>, <TokenType.DECIMAL: 'DECIMAL'>, <TokenType.MAP: 'MAP'>, <TokenType.CONSTRAINT: 'CONSTRAINT'>, <TokenType.INT256: 'INT256'>, <TokenType.VIEW: 'VIEW'>, <TokenType.COMMENT: 'COMMENT'>, <TokenType.SCHEMA: 'SCHEMA'>, <TokenType.SMALLINT: 'SMALLINT'>, <TokenType.ENUM: 'ENUM'>, <TokenType.INT8RANGE: 'INT8RANGE'>, <TokenType.JSON: 'JSON'>, <TokenType.ROWVERSION: 'ROWVERSION'>, <TokenType.SIMPLEAGGREGATEFUNCTION: 'SIMPLEAGGREGATEFUNCTION'>, <TokenType.COPY: 'COPY'>, <TokenType.IPADDRESS: 'IPADDRESS'>, <TokenType.SERIAL: 'SERIAL'>, <TokenType.DESCRIBE: 'DESCRIBE'>, <TokenType.DATERANGE: 'DATERANGE'>, <TokenType.NEXT: 'NEXT'>, <TokenType.PROCEDURE: 'PROCEDURE'>, <TokenType.TOP: 'TOP'>, <TokenType.LOAD: 'LOAD'>, <TokenType.USERDEFINED: 'USERDEFINED'>, <TokenType.EXECUTE: 'EXECUTE'>, <TokenType.TDIGEST: 'TDIGEST'>, <TokenType.UNKNOWN: 'UNKNOWN'>, <TokenType.STRUCT: 'STRUCT'>, <TokenType.BIT: 'BIT'>, <TokenType.MEDIUMTEXT: 'MEDIUMTEXT'>, <TokenType.BINARY: 'BINARY'>, <TokenType.DATE32: 'DATE32'>, <TokenType.TSMULTIRANGE: 'TSMULTIRANGE'>, <TokenType.TSTZMULTIRANGE: 'TSTZMULTIRANGE'>, <TokenType.COMMIT: 'COMMIT'>, <TokenType.PIVOT: 'PIVOT'>, <TokenType.OPERATOR: 'OPERATOR'>, <TokenType.NUMMULTIRANGE: 'NUMMULTIRANGE'>, <TokenType.OBJECT: 'OBJECT'>, <TokenType.EXISTS: 'EXISTS'>, <TokenType.ENUM16: 'ENUM16'>, <TokenType.IMAGE: 'IMAGE'>, <TokenType.IPPREFIX: 'IPPREFIX'>, <TokenType.STORAGE_INTEGRATION: 'STORAGE_INTEGRATION'>, <TokenType.LONGBLOB: 'LONGBLOB'>, <TokenType.ALL: 'ALL'>, <TokenType.RANGE: 'RANGE'>, <TokenType.INT128: 'INT128'>, <TokenType.REFRESH: 'REFRESH'>, <TokenType.YEAR: 'YEAR'>, <TokenType.OVERLAPS: 'OVERLAPS'>, <TokenType.OBJECT_IDENTIFIER: 'OBJECT_IDENTIFIER'>, <TokenType.TIMESTAMP: 'TIMESTAMP'>, <TokenType.TIMESTAMP_MS: 'TIMESTAMP_MS'>, <TokenType.INT8MULTIRANGE: 'INT8MULTIRANGE'>, <TokenType.MEDIUMINT: 'MEDIUMINT'>, <TokenType.TINYINT: 'TINYINT'>, <TokenType.ROW: 'ROW'>, <TokenType.UINT256: 'UINT256'>, <TokenType.HSTORE: 'HSTORE'>, <TokenType.UTINYINT: 'UTINYINT'>, <TokenType.JSONB: 'JSONB'>, <TokenType.TSTZRANGE: 'TSTZRANGE'>, <TokenType.TIMESTAMP_S: 'TIMESTAMP_S'>, <TokenType.MODEL: 'MODEL'>, <TokenType.DEFAULT: 'DEFAULT'>, <TokenType.INDEX: 'INDEX'>, <TokenType.ARRAY: 'ARRAY'>, <TokenType.TIMESTAMPNTZ: 'TIMESTAMPNTZ'>, <TokenType.UMEDIUMINT: 'UMEDIUMINT'>, <TokenType.TRUE: 'TRUE'>, <TokenType.OVERWRITE: 'OVERWRITE'>, <TokenType.CHAR: 'CHAR'>, <TokenType.COLLATE: 'COLLATE'>, <TokenType.UPDATE: 'UPDATE'>, <TokenType.TAG: 'TAG'>, <TokenType.WAREHOUSE: 'WAREHOUSE'>, <TokenType.UNIQUE: 'UNIQUE'>, <TokenType.REFERENCES: 'REFERENCES'>, <TokenType.LOWCARDINALITY: 'LOWCARDINALITY'>, <TokenType.PERCENT: 'PERCENT'>, <TokenType.SET: 'SET'>, <TokenType.ROWS: 'ROWS'>, <TokenType.TEXT: 'TEXT'>, <TokenType.COMMAND: 'COMMAND'>, <TokenType.SMALLSERIAL: 'SMALLSERIAL'>, <TokenType.FINAL: 'FINAL'>, <TokenType.STREAMLIT: 'STREAMLIT'>, <TokenType.END: 'END'>, <TokenType.CURRENT_TIMESTAMP: 'CURRENT_TIMESTAMP'>, <TokenType.BIGINT: 'BIGINT'>, <TokenType.LONGTEXT: 'LONGTEXT'>, <TokenType.ORDINALITY: 'ORDINALITY'>, <TokenType.AGGREGATEFUNCTION: 'AGGREGATEFUNCTION'>, <TokenType.CURRENT_TIME: 'CURRENT_TIME'>, <TokenType.MERGE: 'MERGE'>, <TokenType.ISNULL: 'ISNULL'>, <TokenType.FALSE: 'FALSE'>, <TokenType.TINYTEXT: 'TINYTEXT'>, <TokenType.TIMETZ: 'TIMETZ'>, <TokenType.SETTINGS: 'SETTINGS'>, <TokenType.VOLATILE: 'VOLATILE'>, <TokenType.DOUBLE: 'DOUBLE'>, <TokenType.VARCHAR: 'VARCHAR'>, <TokenType.NULL: 'NULL'>, <TokenType.UBIGINT: 'UBIGINT'>, <TokenType.BEGIN: 'BEGIN'>, <TokenType.DATETIME: 'DATETIME'>, <TokenType.ROLLUP: 'ROLLUP'>, <TokenType.TSRANGE: 'TSRANGE'>, <TokenType.UUID: 'UUID'>, <TokenType.GEOGRAPHY: 'GEOGRAPHY'>, <TokenType.BIGSERIAL: 'BIGSERIAL'>, <TokenType.UINT128: 'UINT128'>, <TokenType.NAME: 'NAME'>, <TokenType.ASC: 'ASC'>, <TokenType.INT: 'INT'>, <TokenType.KEEP: 'KEEP'>, <TokenType.UNIQUEIDENTIFIER: 'UNIQUEIDENTIFIER'>, <TokenType.KILL: 'KILL'>, <TokenType.FLOAT: 'FLOAT'>, <TokenType.VARBINARY: 'VARBINARY'>, <TokenType.TIMESTAMPLTZ: 'TIMESTAMPLTZ'>, <TokenType.FIXEDSTRING: 'FIXEDSTRING'>, <TokenType.BIGDECIMAL: 'BIGDECIMAL'>, <TokenType.DATE: 'DATE'>, <TokenType.SOME: 'SOME'>, <TokenType.DELETE: 'DELETE'>, <TokenType.INT4MULTIRANGE: 'INT4MULTIRANGE'>, <TokenType.DIV: 'DIV'>, <TokenType.UDECIMAL: 'UDECIMAL'>, <TokenType.TIME: 'TIME'>, <TokenType.TIMESTAMPTZ: 'TIMESTAMPTZ'>, <TokenType.SMALLMONEY: 'SMALLMONEY'>, <TokenType.DICTIONARY: 'DICTIONARY'>, <TokenType.IPV6: 'IPV6'>, <TokenType.VAR: 'VAR'>, <TokenType.ENUM8: 'ENUM8'>, <TokenType.VARIANT: 'VARIANT'>, <TokenType.AUTO_INCREMENT: 'AUTO_INCREMENT'>, <TokenType.CURRENT_USER: 'CURRENT_USER'>, <TokenType.TIMESTAMP_NS: 'TIMESTAMP_NS'>, <TokenType.INET: 'INET'>, <TokenType.UNNEST: 'UNNEST'>, <TokenType.CURRENT_DATE: 'CURRENT_DATE'>, <TokenType.RECURSIVE: 'RECURSIVE'>, <TokenType.CACHE: 'CACHE'>, <TokenType.NCHAR: 'NCHAR'>, <TokenType.GEOMETRY: 'GEOMETRY'>, <TokenType.TEMPORARY: 'TEMPORARY'>, <TokenType.PRAGMA: 'PRAGMA'>, <TokenType.UNPIVOT: 'UNPIVOT'>, <TokenType.INT4RANGE: 'INT4RANGE'>}
PLACEHOLDER_PARSERS =
{<TokenType.PLACEHOLDER: 'PLACEHOLDER'>: <function Parser.<lambda>>, <TokenType.PARAMETER: 'PARAMETER'>: <function DuckDB.Parser.<lambda>>, <TokenType.COLON: 'COLON'>: <function Parser.<lambda>>}
TYPE_CONVERTER =
{<Type.DECIMAL: 'DECIMAL'>: <function build_default_decimal_type.<locals>._builder>}
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
- AGGREGATE_TYPE_TOKENS
- TYPE_TOKENS
- SIGNED_TO_UNSIGNED_TYPE_TOKEN
- SUBQUERY_PREDICATES
- RESERVED_TOKENS
- DB_CREATABLES
- CREATABLES
- ID_VAR_TOKENS
- INTERVAL_VARS
- ALIAS_TOKENS
- COMMENT_TABLE_ALIAS_TOKENS
- UPDATE_ALIAS_TOKENS
- TRIM_TYPES
- FUNC_TOKENS
- CONJUNCTION
- EQUALITY
- COMPARISON
- TERM
- FACTOR
- EXPONENT
- TIMES
- TIMESTAMPS
- SET_OPERATIONS
- JOIN_METHODS
- JOIN_SIDES
- JOIN_KINDS
- JOIN_HINTS
- LAMBDAS
- COLUMN_OPERATORS
- EXPRESSION_PARSERS
- STATEMENT_PARSERS
- UNARY_PARSERS
- STRING_PARSERS
- NUMERIC_PARSERS
- PRIMARY_PARSERS
- RANGE_PARSERS
- PROPERTY_PARSERS
- CONSTRAINT_PARSERS
- ALTER_PARSERS
- ALTER_ALTER_PARSERS
- SCHEMA_UNNAMED_CONSTRAINTS
- INVALID_FUNC_NAME_TOKENS
- KEY_VALUE_DEFINITIONS
- QUERY_MODIFIER_PARSERS
- SET_PARSERS
- SHOW_PARSERS
- TYPE_LITERAL_PARSERS
- DDL_SELECT_TOKENS
- PRE_VOLATILE_TOKENS
- TRANSACTION_KIND
- TRANSACTION_CHARACTERISTICS
- CONFLICT_ACTIONS
- CREATE_SEQUENCE
- ISOLATED_LOADING_OPTIONS
- USABLES
- CAST_ACTIONS
- INSERT_ALTERNATIVES
- CLONE_KEYWORDS
- HISTORICAL_DATA_KIND
- OPCLASS_FOLLOW_KEYWORDS
- OPTYPE_FOLLOW_TOKENS
- TABLE_INDEX_HINT_TOKENS
- VIEW_ATTRIBUTES
- WINDOW_ALIAS_TOKENS
- WINDOW_BEFORE_PAREN_TOKENS
- WINDOW_SIDES
- JSON_KEY_VALUE_SEPARATOR_TOKENS
- FETCH_TOKENS
- ADD_CONSTRAINT_TOKENS
- DISTINCT_TOKENS
- NULL_TOKENS
- UNNEST_OFFSET_ALIAS_TOKENS
- SELECT_START_TOKENS
- STRICT_CAST
- PREFIXED_PIVOT_COLUMNS
- IDENTIFY_PIVOT_STRINGS
- LOG_DEFAULTS_TO_LN
- ALTER_TABLE_ADD_REQUIRED_FOR_EACH_COLUMN
- TABLESAMPLE_CSV
- DEFAULT_SAMPLING_METHOD
- SET_REQUIRES_ASSIGNMENT_DELIMITER
- TRIM_PATTERN_FIRST
- STRING_ALIASES
- MODIFIERS_ATTACHED_TO_UNION
- UNION_MODIFIERS
- NO_PAREN_IF_COMMANDS
- JSON_ARROWS_REQUIRE_JSON_TYPE
- COLON_IS_JSON_EXTRACT
- VALUES_FOLLOWED_BY_PAREN
- SUPPORTS_IMPLICIT_UNNEST
- INTERVAL_SPANS
- SUPPORTS_PARTITION_SELECTION
- error_level
- error_message_context
- max_errors
- dialect
- reset
- parse
- parse_into
- check_errors
- raise_error
- expression
- validate_expression
- errors
- sql
356 class Generator(generator.Generator): 357 PARAMETER_TOKEN = "$" 358 NAMED_PLACEHOLDER_TOKEN = "$" 359 JOIN_HINTS = False 360 TABLE_HINTS = False 361 QUERY_HINTS = False 362 LIMIT_FETCH = "LIMIT" 363 STRUCT_DELIMITER = ("(", ")") 364 RENAME_TABLE_WITH_DB = False 365 NVL2_SUPPORTED = False 366 SEMI_ANTI_JOIN_WITH_SIDE = False 367 TABLESAMPLE_KEYWORDS = "USING SAMPLE" 368 TABLESAMPLE_SEED_KEYWORD = "REPEATABLE" 369 LAST_DAY_SUPPORTS_DATE_PART = False 370 JSON_KEY_VALUE_PAIR_SEP = "," 371 IGNORE_NULLS_IN_FUNC = True 372 JSON_PATH_BRACKETED_KEY_SUPPORTED = False 373 SUPPORTS_CREATE_TABLE_LIKE = False 374 MULTI_ARG_DISTINCT = False 375 CAN_IMPLEMENT_ARRAY_ANY = True 376 SUPPORTS_TO_NUMBER = False 377 COPY_HAS_INTO_KEYWORD = False 378 STAR_EXCEPT = "EXCLUDE" 379 380 TRANSFORMS = { 381 **generator.Generator.TRANSFORMS, 382 exp.ApproxDistinct: approx_count_distinct_sql, 383 exp.Array: inline_array_unless_query, 384 exp.ArrayContainsAll: rename_func("ARRAY_HAS_ALL"), 385 exp.ArrayFilter: rename_func("LIST_FILTER"), 386 exp.ArraySize: rename_func("ARRAY_LENGTH"), 387 exp.ArgMax: arg_max_or_min_no_count("ARG_MAX"), 388 exp.ArgMin: arg_max_or_min_no_count("ARG_MIN"), 389 exp.ArraySort: _array_sort_sql, 390 exp.ArraySum: rename_func("LIST_SUM"), 391 exp.BitwiseXor: rename_func("XOR"), 392 exp.CommentColumnConstraint: no_comment_column_constraint_sql, 393 exp.CurrentDate: lambda *_: "CURRENT_DATE", 394 exp.CurrentTime: lambda *_: "CURRENT_TIME", 395 exp.CurrentTimestamp: lambda *_: "CURRENT_TIMESTAMP", 396 exp.DayOfMonth: rename_func("DAYOFMONTH"), 397 exp.DayOfWeek: rename_func("DAYOFWEEK"), 398 exp.DayOfYear: rename_func("DAYOFYEAR"), 399 exp.DataType: _datatype_sql, 400 exp.Date: _date_sql, 401 exp.DateAdd: _date_delta_sql, 402 exp.DateFromParts: rename_func("MAKE_DATE"), 403 exp.DateSub: _date_delta_sql, 404 exp.DateDiff: lambda self, e: self.func( 405 "DATE_DIFF", f"'{e.args.get('unit') or 'DAY'}'", e.expression, e.this 406 ), 407 exp.DateStrToDate: datestrtodate_sql, 408 exp.DateToDi: lambda self, 409 e: f"CAST(STRFTIME({self.sql(e, 'this')}, {DuckDB.DATEINT_FORMAT}) AS INT)", 410 exp.Decode: lambda self, e: encode_decode_sql(self, e, "DECODE", replace=False), 411 exp.DiToDate: lambda self, 412 e: f"CAST(STRPTIME(CAST({self.sql(e, 'this')} AS TEXT), {DuckDB.DATEINT_FORMAT}) AS DATE)", 413 exp.Encode: lambda self, e: encode_decode_sql(self, e, "ENCODE", replace=False), 414 exp.Explode: rename_func("UNNEST"), 415 exp.IntDiv: lambda self, e: self.binary(e, "//"), 416 exp.IsInf: rename_func("ISINF"), 417 exp.IsNan: rename_func("ISNAN"), 418 exp.JSONExtract: _arrow_json_extract_sql, 419 exp.JSONExtractScalar: _arrow_json_extract_sql, 420 exp.JSONFormat: _json_format_sql, 421 exp.LogicalOr: rename_func("BOOL_OR"), 422 exp.LogicalAnd: rename_func("BOOL_AND"), 423 exp.MD5Digest: lambda self, e: self.func("UNHEX", self.func("MD5", e.this)), 424 exp.MonthsBetween: lambda self, e: self.func( 425 "DATEDIFF", 426 "'month'", 427 exp.cast(e.expression, exp.DataType.Type.TIMESTAMP, copy=True), 428 exp.cast(e.this, exp.DataType.Type.TIMESTAMP, copy=True), 429 ), 430 exp.ParseJSON: rename_func("JSON"), 431 exp.PercentileCont: rename_func("QUANTILE_CONT"), 432 exp.PercentileDisc: rename_func("QUANTILE_DISC"), 433 # DuckDB doesn't allow qualified columns inside of PIVOT expressions. 434 # See: https://github.com/duckdb/duckdb/blob/671faf92411182f81dce42ac43de8bfb05d9909e/src/planner/binder/tableref/bind_pivot.cpp#L61-L62 435 exp.Pivot: transforms.preprocess([transforms.unqualify_columns]), 436 exp.RegexpExtract: regexp_extract_sql, 437 exp.RegexpReplace: lambda self, e: self.func( 438 "REGEXP_REPLACE", 439 e.this, 440 e.expression, 441 e.args.get("replacement"), 442 e.args.get("modifiers"), 443 ), 444 exp.RegexpLike: rename_func("REGEXP_MATCHES"), 445 exp.RegexpSplit: rename_func("STR_SPLIT_REGEX"), 446 exp.Rand: rename_func("RANDOM"), 447 exp.SafeDivide: no_safe_divide_sql, 448 exp.Split: rename_func("STR_SPLIT"), 449 exp.SortArray: _sort_array_sql, 450 exp.StrPosition: str_position_sql, 451 exp.StrToDate: lambda self, e: f"CAST({str_to_time_sql(self, e)} AS DATE)", 452 exp.StrToTime: str_to_time_sql, 453 exp.StrToUnix: lambda self, e: self.func( 454 "EPOCH", self.func("STRPTIME", e.this, self.format_time(e)) 455 ), 456 exp.Struct: _struct_sql, 457 exp.TimeAdd: _date_delta_sql, 458 exp.Timestamp: no_timestamp_sql, 459 exp.TimestampDiff: lambda self, e: self.func( 460 "DATE_DIFF", exp.Literal.string(e.unit), e.expression, e.this 461 ), 462 exp.TimestampTrunc: timestamptrunc_sql(), 463 exp.TimeStrToDate: lambda self, e: self.sql(exp.cast(e.this, exp.DataType.Type.DATE)), 464 exp.TimeStrToTime: timestrtotime_sql, 465 exp.TimeStrToUnix: lambda self, e: self.func( 466 "EPOCH", exp.cast(e.this, exp.DataType.Type.TIMESTAMP) 467 ), 468 exp.TimeToStr: lambda self, e: self.func("STRFTIME", e.this, self.format_time(e)), 469 exp.TimeToUnix: rename_func("EPOCH"), 470 exp.TsOrDiToDi: lambda self, 471 e: f"CAST(SUBSTR(REPLACE(CAST({self.sql(e, 'this')} AS TEXT), '-', ''), 1, 8) AS INT)", 472 exp.TsOrDsAdd: _ts_or_ds_add_sql, 473 exp.TsOrDsDiff: lambda self, e: self.func( 474 "DATE_DIFF", 475 f"'{e.args.get('unit') or 'DAY'}'", 476 exp.cast(e.expression, exp.DataType.Type.TIMESTAMP), 477 exp.cast(e.this, exp.DataType.Type.TIMESTAMP), 478 ), 479 exp.UnixToStr: lambda self, e: self.func( 480 "STRFTIME", self.func("TO_TIMESTAMP", e.this), self.format_time(e) 481 ), 482 exp.UnixToTime: _unix_to_time_sql, 483 exp.UnixToTimeStr: lambda self, e: f"CAST(TO_TIMESTAMP({self.sql(e, 'this')}) AS TEXT)", 484 exp.VariancePop: rename_func("VAR_POP"), 485 exp.WeekOfYear: rename_func("WEEKOFYEAR"), 486 exp.Xor: bool_xor_sql, 487 } 488 489 SUPPORTED_JSON_PATH_PARTS = { 490 exp.JSONPathKey, 491 exp.JSONPathRoot, 492 exp.JSONPathSubscript, 493 exp.JSONPathWildcard, 494 } 495 496 TYPE_MAPPING = { 497 **generator.Generator.TYPE_MAPPING, 498 exp.DataType.Type.BINARY: "BLOB", 499 exp.DataType.Type.CHAR: "TEXT", 500 exp.DataType.Type.FLOAT: "REAL", 501 exp.DataType.Type.NCHAR: "TEXT", 502 exp.DataType.Type.NVARCHAR: "TEXT", 503 exp.DataType.Type.UINT: "UINTEGER", 504 exp.DataType.Type.VARBINARY: "BLOB", 505 exp.DataType.Type.ROWVERSION: "BLOB", 506 exp.DataType.Type.VARCHAR: "TEXT", 507 exp.DataType.Type.TIMESTAMP_S: "TIMESTAMP_S", 508 exp.DataType.Type.TIMESTAMP_MS: "TIMESTAMP_MS", 509 exp.DataType.Type.TIMESTAMP_NS: "TIMESTAMP_NS", 510 } 511 512 UNWRAPPED_INTERVAL_VALUES = (exp.Literal, exp.Paren) 513 514 # DuckDB doesn't generally support CREATE TABLE .. properties 515 # https://duckdb.org/docs/sql/statements/create_table.html 516 PROPERTIES_LOCATION = { 517 prop: exp.Properties.Location.UNSUPPORTED 518 for prop in generator.Generator.PROPERTIES_LOCATION 519 } 520 521 # There are a few exceptions (e.g. temporary tables) which are supported or 522 # can be transpiled to DuckDB, so we explicitly override them accordingly 523 PROPERTIES_LOCATION[exp.LikeProperty] = exp.Properties.Location.POST_SCHEMA 524 PROPERTIES_LOCATION[exp.TemporaryProperty] = exp.Properties.Location.POST_CREATE 525 526 def timefromparts_sql(self, expression: exp.TimeFromParts) -> str: 527 nano = expression.args.get("nano") 528 if nano is not None: 529 expression.set( 530 "sec", expression.args["sec"] + nano.pop() / exp.Literal.number(1000000000.0) 531 ) 532 533 return rename_func("MAKE_TIME")(self, expression) 534 535 def timestampfromparts_sql(self, expression: exp.TimestampFromParts) -> str: 536 sec = expression.args["sec"] 537 538 milli = expression.args.get("milli") 539 if milli is not None: 540 sec += milli.pop() / exp.Literal.number(1000.0) 541 542 nano = expression.args.get("nano") 543 if nano is not None: 544 sec += nano.pop() / exp.Literal.number(1000000000.0) 545 546 if milli or nano: 547 expression.set("sec", sec) 548 549 return rename_func("MAKE_TIMESTAMP")(self, expression) 550 551 def tablesample_sql( 552 self, 553 expression: exp.TableSample, 554 sep: str = " AS ", 555 tablesample_keyword: t.Optional[str] = None, 556 ) -> str: 557 if not isinstance(expression.parent, exp.Select): 558 # This sample clause only applies to a single source, not the entire resulting relation 559 tablesample_keyword = "TABLESAMPLE" 560 561 if expression.args.get("size"): 562 method = expression.args.get("method") 563 if method and method.name.upper() != "RESERVOIR": 564 self.unsupported( 565 f"Sampling method {method} is not supported with a discrete sample count, " 566 "defaulting to reservoir sampling" 567 ) 568 expression.set("method", exp.var("RESERVOIR")) 569 570 return super().tablesample_sql( 571 expression, sep=sep, tablesample_keyword=tablesample_keyword 572 ) 573 574 def interval_sql(self, expression: exp.Interval) -> str: 575 multiplier: t.Optional[int] = None 576 unit = expression.text("unit").lower() 577 578 if unit.startswith("week"): 579 multiplier = 7 580 if unit.startswith("quarter"): 581 multiplier = 90 582 583 if multiplier: 584 return f"({multiplier} * {super().interval_sql(exp.Interval(this=expression.this, unit=exp.var('DAY')))})" 585 586 return super().interval_sql(expression) 587 588 def columndef_sql(self, expression: exp.ColumnDef, sep: str = " ") -> str: 589 if isinstance(expression.parent, exp.UserDefinedFunction): 590 return self.sql(expression, "this") 591 return super().columndef_sql(expression, sep) 592 593 def join_sql(self, expression: exp.Join) -> str: 594 if ( 595 expression.side == "LEFT" 596 and not expression.args.get("on") 597 and isinstance(expression.this, exp.Unnest) 598 ): 599 # Some dialects support `LEFT JOIN UNNEST(...)` without an explicit ON clause 600 # DuckDB doesn't, but we can just add a dummy ON clause that is always true 601 return super().join_sql(expression.on(exp.true())) 602 603 return super().join_sql(expression) 604 605 def generateseries_sql(self, expression: exp.GenerateSeries) -> str: 606 # GENERATE_SERIES(a, b) -> [a, b], RANGE(a, b) -> [a, b) 607 if expression.args.get("is_end_exclusive"): 608 expression.set("is_end_exclusive", None) 609 return rename_func("RANGE")(self, expression) 610 611 return super().generateseries_sql(expression) 612 613 def bracket_sql(self, expression: exp.Bracket) -> str: 614 this = expression.this 615 if isinstance(this, exp.Array): 616 this.replace(exp.paren(this)) 617 618 bracket = super().bracket_sql(expression) 619 620 if not expression.args.get("returns_list_for_maps"): 621 if not this.type: 622 from sqlglot.optimizer.annotate_types import annotate_types 623 624 this = annotate_types(this) 625 626 if this.is_type(exp.DataType.Type.MAP): 627 bracket = f"({bracket})[1]" 628 629 return bracket 630 631 def withingroup_sql(self, expression: exp.WithinGroup) -> str: 632 expression_sql = self.sql(expression, "expression") 633 634 func = expression.this 635 if isinstance(func, exp.PERCENTILES): 636 # Make the order key the first arg and slide the fraction to the right 637 # https://duckdb.org/docs/sql/aggregates#ordered-set-aggregate-functions 638 order_col = expression.find(exp.Ordered) 639 if order_col: 640 func.set("expression", func.this) 641 func.set("this", order_col.this) 642 643 this = self.sql(expression, "this").rstrip(")") 644 645 return f"{this}{expression_sql})"
Generator converts a given syntax tree to the corresponding SQL string.
Arguments:
- pretty: Whether 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 to normalize identifiers to lowercase. Default: False.
- pad: The pad size in a formatted string. For example, this affects the indentation of a projection in a query, relative to its nesting level. Default: 2.
- indent: The indentation size in a formatted string. For example, this affects the
indentation of subqueries and filters under a
WHERE
clause. Default: 2. - normalize_functions: How to normalize 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: Whether 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 to preserve comments in the output SQL code. Default: True
TRANSFORMS =
{<class 'sqlglot.expressions.JSONPathKey'>: <function <lambda>>, <class 'sqlglot.expressions.JSONPathRoot'>: <function <lambda>>, <class 'sqlglot.expressions.JSONPathSubscript'>: <function <lambda>>, <class 'sqlglot.expressions.JSONPathWildcard'>: <function <lambda>>, <class 'sqlglot.expressions.AllowedValuesProperty'>: <function Generator.<lambda>>, <class 'sqlglot.expressions.AutoRefreshProperty'>: <function Generator.<lambda>>, <class 'sqlglot.expressions.BackupProperty'>: <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.ClusteredColumnConstraint'>: <function Generator.<lambda>>, <class 'sqlglot.expressions.CollateColumnConstraint'>: <function Generator.<lambda>>, <class 'sqlglot.expressions.CommentColumnConstraint'>: <function no_comment_column_constraint_sql>, <class 'sqlglot.expressions.CopyGrantsProperty'>: <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.EphemeralColumnConstraint'>: <function Generator.<lambda>>, <class 'sqlglot.expressions.ExcludeColumnConstraint'>: <function Generator.<lambda>>, <class 'sqlglot.expressions.ExecuteAsProperty'>: <function Generator.<lambda>>, <class 'sqlglot.expressions.ExternalProperty'>: <function Generator.<lambda>>, <class 'sqlglot.expressions.GlobalProperty'>: <function Generator.<lambda>>, <class 'sqlglot.expressions.HeapProperty'>: <function Generator.<lambda>>, <class 'sqlglot.expressions.IcebergProperty'>: <function Generator.<lambda>>, <class 'sqlglot.expressions.InheritsProperty'>: <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.JSONExtract'>: <function _arrow_json_extract_sql>, <class 'sqlglot.expressions.JSONExtractScalar'>: <function _arrow_json_extract_sql>, <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.NonClusteredColumnConstraint'>: <function Generator.<lambda>>, <class 'sqlglot.expressions.NoPrimaryIndexProperty'>: <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.ProjectionPolicyColumnConstraint'>: <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.SetConfigProperty'>: <function Generator.<lambda>>, <class 'sqlglot.expressions.SetProperty'>: <function Generator.<lambda>>, <class 'sqlglot.expressions.SettingsProperty'>: <function Generator.<lambda>>, <class 'sqlglot.expressions.SharingProperty'>: <function Generator.<lambda>>, <class 'sqlglot.expressions.SqlReadWriteProperty'>: <function Generator.<lambda>>, <class 'sqlglot.expressions.SqlSecurityProperty'>: <function Generator.<lambda>>, <class 'sqlglot.expressions.StabilityProperty'>: <function Generator.<lambda>>, <class 'sqlglot.expressions.StrictProperty'>: <function Generator.<lambda>>, <class 'sqlglot.expressions.TemporaryProperty'>: <function Generator.<lambda>>, <class 'sqlglot.expressions.TagColumnConstraint'>: <function Generator.<lambda>>, <class 'sqlglot.expressions.TitleColumnConstraint'>: <function Generator.<lambda>>, <class 'sqlglot.expressions.Timestamp'>: <function no_timestamp_sql>, <class 'sqlglot.expressions.ToMap'>: <function Generator.<lambda>>, <class 'sqlglot.expressions.ToTableProperty'>: <function Generator.<lambda>>, <class 'sqlglot.expressions.TransformModelProperty'>: <function Generator.<lambda>>, <class 'sqlglot.expressions.TransientProperty'>: <function Generator.<lambda>>, <class 'sqlglot.expressions.UppercaseColumnConstraint'>: <function Generator.<lambda>>, <class 'sqlglot.expressions.UnloggedProperty'>: <function Generator.<lambda>>, <class 'sqlglot.expressions.VarMap'>: <function Generator.<lambda>>, <class 'sqlglot.expressions.ViewAttributeProperty'>: <function Generator.<lambda>>, <class 'sqlglot.expressions.VolatileProperty'>: <function Generator.<lambda>>, <class 'sqlglot.expressions.WithJournalTableProperty'>: <function Generator.<lambda>>, <class 'sqlglot.expressions.WithOperator'>: <function Generator.<lambda>>, <class 'sqlglot.expressions.ApproxDistinct'>: <function approx_count_distinct_sql>, <class 'sqlglot.expressions.Array'>: <function inline_array_unless_query>, <class 'sqlglot.expressions.ArrayContainsAll'>: <function rename_func.<locals>.<lambda>>, <class 'sqlglot.expressions.ArrayFilter'>: <function rename_func.<locals>.<lambda>>, <class 'sqlglot.expressions.ArraySize'>: <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.ArraySort'>: <function _array_sort_sql>, <class 'sqlglot.expressions.ArraySum'>: <function rename_func.<locals>.<lambda>>, <class 'sqlglot.expressions.BitwiseXor'>: <function rename_func.<locals>.<lambda>>, <class 'sqlglot.expressions.CurrentDate'>: <function DuckDB.Generator.<lambda>>, <class 'sqlglot.expressions.CurrentTime'>: <function DuckDB.Generator.<lambda>>, <class 'sqlglot.expressions.CurrentTimestamp'>: <function DuckDB.Generator.<lambda>>, <class 'sqlglot.expressions.DayOfMonth'>: <function rename_func.<locals>.<lambda>>, <class 'sqlglot.expressions.DayOfWeek'>: <function rename_func.<locals>.<lambda>>, <class 'sqlglot.expressions.DayOfYear'>: <function rename_func.<locals>.<lambda>>, <class 'sqlglot.expressions.DataType'>: <function _datatype_sql>, <class 'sqlglot.expressions.Date'>: <function _date_sql>, <class 'sqlglot.expressions.DateAdd'>: <function _date_delta_sql>, <class 'sqlglot.expressions.DateFromParts'>: <function rename_func.<locals>.<lambda>>, <class 'sqlglot.expressions.DateSub'>: <function _date_delta_sql>, <class 'sqlglot.expressions.DateDiff'>: <function DuckDB.Generator.<lambda>>, <class 'sqlglot.expressions.DateStrToDate'>: <function datestrtodate_sql>, <class 'sqlglot.expressions.DateToDi'>: <function DuckDB.Generator.<lambda>>, <class 'sqlglot.expressions.Decode'>: <function DuckDB.Generator.<lambda>>, <class 'sqlglot.expressions.DiToDate'>: <function DuckDB.Generator.<lambda>>, <class 'sqlglot.expressions.Encode'>: <function DuckDB.Generator.<lambda>>, <class 'sqlglot.expressions.Explode'>: <function rename_func.<locals>.<lambda>>, <class 'sqlglot.expressions.IntDiv'>: <function DuckDB.Generator.<lambda>>, <class 'sqlglot.expressions.IsInf'>: <function rename_func.<locals>.<lambda>>, <class 'sqlglot.expressions.IsNan'>: <function rename_func.<locals>.<lambda>>, <class 'sqlglot.expressions.JSONFormat'>: <function _json_format_sql>, <class 'sqlglot.expressions.LogicalOr'>: <function rename_func.<locals>.<lambda>>, <class 'sqlglot.expressions.LogicalAnd'>: <function rename_func.<locals>.<lambda>>, <class 'sqlglot.expressions.MD5Digest'>: <function DuckDB.Generator.<lambda>>, <class 'sqlglot.expressions.MonthsBetween'>: <function DuckDB.Generator.<lambda>>, <class 'sqlglot.expressions.ParseJSON'>: <function rename_func.<locals>.<lambda>>, <class 'sqlglot.expressions.PercentileCont'>: <function rename_func.<locals>.<lambda>>, <class 'sqlglot.expressions.PercentileDisc'>: <function rename_func.<locals>.<lambda>>, <class 'sqlglot.expressions.Pivot'>: <function preprocess.<locals>._to_sql>, <class 'sqlglot.expressions.RegexpExtract'>: <function regexp_extract_sql>, <class 'sqlglot.expressions.RegexpReplace'>: <function DuckDB.Generator.<lambda>>, <class 'sqlglot.expressions.RegexpLike'>: <function rename_func.<locals>.<lambda>>, <class 'sqlglot.expressions.RegexpSplit'>: <function rename_func.<locals>.<lambda>>, <class 'sqlglot.expressions.Rand'>: <function rename_func.<locals>.<lambda>>, <class 'sqlglot.expressions.SafeDivide'>: <function no_safe_divide_sql>, <class 'sqlglot.expressions.Split'>: <function rename_func.<locals>.<lambda>>, <class 'sqlglot.expressions.SortArray'>: <function _sort_array_sql>, <class 'sqlglot.expressions.StrPosition'>: <function str_position_sql>, <class 'sqlglot.expressions.StrToDate'>: <function DuckDB.Generator.<lambda>>, <class 'sqlglot.expressions.StrToTime'>: <function str_to_time_sql>, <class 'sqlglot.expressions.StrToUnix'>: <function DuckDB.Generator.<lambda>>, <class 'sqlglot.expressions.Struct'>: <function _struct_sql>, <class 'sqlglot.expressions.TimeAdd'>: <function _date_delta_sql>, <class 'sqlglot.expressions.TimestampDiff'>: <function DuckDB.Generator.<lambda>>, <class 'sqlglot.expressions.TimestampTrunc'>: <function timestamptrunc_sql.<locals>._timestamptrunc_sql>, <class 'sqlglot.expressions.TimeStrToDate'>: <function DuckDB.Generator.<lambda>>, <class 'sqlglot.expressions.TimeStrToTime'>: <function timestrtotime_sql>, <class 'sqlglot.expressions.TimeStrToUnix'>: <function DuckDB.Generator.<lambda>>, <class 'sqlglot.expressions.TimeToStr'>: <function DuckDB.Generator.<lambda>>, <class 'sqlglot.expressions.TimeToUnix'>: <function rename_func.<locals>.<lambda>>, <class 'sqlglot.expressions.TsOrDiToDi'>: <function DuckDB.Generator.<lambda>>, <class 'sqlglot.expressions.TsOrDsAdd'>: <function _ts_or_ds_add_sql>, <class 'sqlglot.expressions.TsOrDsDiff'>: <function DuckDB.Generator.<lambda>>, <class 'sqlglot.expressions.UnixToStr'>: <function DuckDB.Generator.<lambda>>, <class 'sqlglot.expressions.UnixToTime'>: <function _unix_to_time_sql>, <class 'sqlglot.expressions.UnixToTimeStr'>: <function DuckDB.Generator.<lambda>>, <class 'sqlglot.expressions.VariancePop'>: <function rename_func.<locals>.<lambda>>, <class 'sqlglot.expressions.WeekOfYear'>: <function rename_func.<locals>.<lambda>>, <class 'sqlglot.expressions.Xor'>: <function bool_xor_sql>}
SUPPORTED_JSON_PATH_PARTS =
{<class 'sqlglot.expressions.JSONPathRoot'>, <class 'sqlglot.expressions.JSONPathWildcard'>, <class 'sqlglot.expressions.JSONPathSubscript'>, <class 'sqlglot.expressions.JSONPathKey'>}
TYPE_MAPPING =
{<Type.NCHAR: 'NCHAR'>: 'TEXT', <Type.NVARCHAR: 'NVARCHAR'>: 'TEXT', <Type.MEDIUMTEXT: 'MEDIUMTEXT'>: 'TEXT', <Type.LONGTEXT: 'LONGTEXT'>: 'TEXT', <Type.TINYTEXT: 'TINYTEXT'>: 'TEXT', <Type.MEDIUMBLOB: 'MEDIUMBLOB'>: 'BLOB', <Type.LONGBLOB: 'LONGBLOB'>: 'BLOB', <Type.TINYBLOB: 'TINYBLOB'>: 'BLOB', <Type.INET: 'INET'>: 'INET', <Type.ROWVERSION: 'ROWVERSION'>: 'BLOB', <Type.BINARY: 'BINARY'>: 'BLOB', <Type.CHAR: 'CHAR'>: 'TEXT', <Type.FLOAT: 'FLOAT'>: 'REAL', <Type.UINT: 'UINT'>: 'UINTEGER', <Type.VARBINARY: 'VARBINARY'>: 'BLOB', <Type.VARCHAR: 'VARCHAR'>: 'TEXT', <Type.TIMESTAMP_S: 'TIMESTAMP_S'>: 'TIMESTAMP_S', <Type.TIMESTAMP_MS: 'TIMESTAMP_MS'>: 'TIMESTAMP_MS', <Type.TIMESTAMP_NS: 'TIMESTAMP_NS'>: 'TIMESTAMP_NS'}
UNWRAPPED_INTERVAL_VALUES =
(<class 'sqlglot.expressions.Literal'>, <class 'sqlglot.expressions.Paren'>)
PROPERTIES_LOCATION =
{<class 'sqlglot.expressions.AllowedValuesProperty'>: <Location.UNSUPPORTED: 'UNSUPPORTED'>, <class 'sqlglot.expressions.AlgorithmProperty'>: <Location.UNSUPPORTED: 'UNSUPPORTED'>, <class 'sqlglot.expressions.AutoIncrementProperty'>: <Location.UNSUPPORTED: 'UNSUPPORTED'>, <class 'sqlglot.expressions.AutoRefreshProperty'>: <Location.UNSUPPORTED: 'UNSUPPORTED'>, <class 'sqlglot.expressions.BackupProperty'>: <Location.UNSUPPORTED: 'UNSUPPORTED'>, <class 'sqlglot.expressions.BlockCompressionProperty'>: <Location.UNSUPPORTED: 'UNSUPPORTED'>, <class 'sqlglot.expressions.CharacterSetProperty'>: <Location.UNSUPPORTED: 'UNSUPPORTED'>, <class 'sqlglot.expressions.ChecksumProperty'>: <Location.UNSUPPORTED: 'UNSUPPORTED'>, <class 'sqlglot.expressions.CollateProperty'>: <Location.UNSUPPORTED: 'UNSUPPORTED'>, <class 'sqlglot.expressions.CopyGrantsProperty'>: <Location.UNSUPPORTED: 'UNSUPPORTED'>, <class 'sqlglot.expressions.Cluster'>: <Location.UNSUPPORTED: 'UNSUPPORTED'>, <class 'sqlglot.expressions.ClusteredByProperty'>: <Location.UNSUPPORTED: 'UNSUPPORTED'>, <class 'sqlglot.expressions.DataBlocksizeProperty'>: <Location.UNSUPPORTED: 'UNSUPPORTED'>, <class 'sqlglot.expressions.DataDeletionProperty'>: <Location.UNSUPPORTED: 'UNSUPPORTED'>, <class 'sqlglot.expressions.DefinerProperty'>: <Location.UNSUPPORTED: 'UNSUPPORTED'>, <class 'sqlglot.expressions.DictRange'>: <Location.UNSUPPORTED: 'UNSUPPORTED'>, <class 'sqlglot.expressions.DictProperty'>: <Location.UNSUPPORTED: 'UNSUPPORTED'>, <class 'sqlglot.expressions.DistKeyProperty'>: <Location.UNSUPPORTED: 'UNSUPPORTED'>, <class 'sqlglot.expressions.DistStyleProperty'>: <Location.UNSUPPORTED: 'UNSUPPORTED'>, <class 'sqlglot.expressions.EngineProperty'>: <Location.UNSUPPORTED: 'UNSUPPORTED'>, <class 'sqlglot.expressions.ExecuteAsProperty'>: <Location.UNSUPPORTED: 'UNSUPPORTED'>, <class 'sqlglot.expressions.ExternalProperty'>: <Location.UNSUPPORTED: 'UNSUPPORTED'>, <class 'sqlglot.expressions.FallbackProperty'>: <Location.UNSUPPORTED: 'UNSUPPORTED'>, <class 'sqlglot.expressions.FileFormatProperty'>: <Location.UNSUPPORTED: 'UNSUPPORTED'>, <class 'sqlglot.expressions.FreespaceProperty'>: <Location.UNSUPPORTED: 'UNSUPPORTED'>, <class 'sqlglot.expressions.GlobalProperty'>: <Location.UNSUPPORTED: 'UNSUPPORTED'>, <class 'sqlglot.expressions.HeapProperty'>: <Location.UNSUPPORTED: 'UNSUPPORTED'>, <class 'sqlglot.expressions.InheritsProperty'>: <Location.UNSUPPORTED: 'UNSUPPORTED'>, <class 'sqlglot.expressions.IcebergProperty'>: <Location.UNSUPPORTED: 'UNSUPPORTED'>, <class 'sqlglot.expressions.InputModelProperty'>: <Location.UNSUPPORTED: 'UNSUPPORTED'>, <class 'sqlglot.expressions.IsolatedLoadingProperty'>: <Location.UNSUPPORTED: 'UNSUPPORTED'>, <class 'sqlglot.expressions.JournalProperty'>: <Location.UNSUPPORTED: 'UNSUPPORTED'>, <class 'sqlglot.expressions.LanguageProperty'>: <Location.UNSUPPORTED: 'UNSUPPORTED'>, <class 'sqlglot.expressions.LikeProperty'>: <Location.POST_SCHEMA: 'POST_SCHEMA'>, <class 'sqlglot.expressions.LocationProperty'>: <Location.UNSUPPORTED: 'UNSUPPORTED'>, <class 'sqlglot.expressions.LockProperty'>: <Location.UNSUPPORTED: 'UNSUPPORTED'>, <class 'sqlglot.expressions.LockingProperty'>: <Location.UNSUPPORTED: 'UNSUPPORTED'>, <class 'sqlglot.expressions.LogProperty'>: <Location.UNSUPPORTED: 'UNSUPPORTED'>, <class 'sqlglot.expressions.MaterializedProperty'>: <Location.UNSUPPORTED: 'UNSUPPORTED'>, <class 'sqlglot.expressions.MergeBlockRatioProperty'>: <Location.UNSUPPORTED: 'UNSUPPORTED'>, <class 'sqlglot.expressions.NoPrimaryIndexProperty'>: <Location.UNSUPPORTED: 'UNSUPPORTED'>, <class 'sqlglot.expressions.OnProperty'>: <Location.UNSUPPORTED: 'UNSUPPORTED'>, <class 'sqlglot.expressions.OnCommitProperty'>: <Location.UNSUPPORTED: 'UNSUPPORTED'>, <class 'sqlglot.expressions.Order'>: <Location.UNSUPPORTED: 'UNSUPPORTED'>, <class 'sqlglot.expressions.OutputModelProperty'>: <Location.UNSUPPORTED: 'UNSUPPORTED'>, <class 'sqlglot.expressions.PartitionedByProperty'>: <Location.UNSUPPORTED: 'UNSUPPORTED'>, <class 'sqlglot.expressions.PartitionedOfProperty'>: <Location.UNSUPPORTED: 'UNSUPPORTED'>, <class 'sqlglot.expressions.PrimaryKey'>: <Location.UNSUPPORTED: 'UNSUPPORTED'>, <class 'sqlglot.expressions.Property'>: <Location.UNSUPPORTED: 'UNSUPPORTED'>, <class 'sqlglot.expressions.RemoteWithConnectionModelProperty'>: <Location.UNSUPPORTED: 'UNSUPPORTED'>, <class 'sqlglot.expressions.ReturnsProperty'>: <Location.UNSUPPORTED: 'UNSUPPORTED'>, <class 'sqlglot.expressions.RowFormatProperty'>: <Location.UNSUPPORTED: 'UNSUPPORTED'>, <class 'sqlglot.expressions.RowFormatDelimitedProperty'>: <Location.UNSUPPORTED: 'UNSUPPORTED'>, <class 'sqlglot.expressions.RowFormatSerdeProperty'>: <Location.UNSUPPORTED: 'UNSUPPORTED'>, <class 'sqlglot.expressions.SampleProperty'>: <Location.UNSUPPORTED: 'UNSUPPORTED'>, <class 'sqlglot.expressions.SchemaCommentProperty'>: <Location.UNSUPPORTED: 'UNSUPPORTED'>, <class 'sqlglot.expressions.SerdeProperties'>: <Location.UNSUPPORTED: 'UNSUPPORTED'>, <class 'sqlglot.expressions.Set'>: <Location.UNSUPPORTED: 'UNSUPPORTED'>, <class 'sqlglot.expressions.SettingsProperty'>: <Location.UNSUPPORTED: 'UNSUPPORTED'>, <class 'sqlglot.expressions.SetProperty'>: <Location.UNSUPPORTED: 'UNSUPPORTED'>, <class 'sqlglot.expressions.SetConfigProperty'>: <Location.UNSUPPORTED: 'UNSUPPORTED'>, <class 'sqlglot.expressions.SharingProperty'>: <Location.UNSUPPORTED: 'UNSUPPORTED'>, <class 'sqlglot.expressions.SequenceProperties'>: <Location.UNSUPPORTED: 'UNSUPPORTED'>, <class 'sqlglot.expressions.SortKeyProperty'>: <Location.UNSUPPORTED: 'UNSUPPORTED'>, <class 'sqlglot.expressions.SqlReadWriteProperty'>: <Location.UNSUPPORTED: 'UNSUPPORTED'>, <class 'sqlglot.expressions.SqlSecurityProperty'>: <Location.UNSUPPORTED: 'UNSUPPORTED'>, <class 'sqlglot.expressions.StabilityProperty'>: <Location.UNSUPPORTED: 'UNSUPPORTED'>, <class 'sqlglot.expressions.StrictProperty'>: <Location.UNSUPPORTED: 'UNSUPPORTED'>, <class 'sqlglot.expressions.TemporaryProperty'>: <Location.POST_CREATE: 'POST_CREATE'>, <class 'sqlglot.expressions.ToTableProperty'>: <Location.UNSUPPORTED: 'UNSUPPORTED'>, <class 'sqlglot.expressions.TransientProperty'>: <Location.UNSUPPORTED: 'UNSUPPORTED'>, <class 'sqlglot.expressions.TransformModelProperty'>: <Location.UNSUPPORTED: 'UNSUPPORTED'>, <class 'sqlglot.expressions.MergeTreeTTL'>: <Location.UNSUPPORTED: 'UNSUPPORTED'>, <class 'sqlglot.expressions.UnloggedProperty'>: <Location.UNSUPPORTED: 'UNSUPPORTED'>, <class 'sqlglot.expressions.ViewAttributeProperty'>: <Location.UNSUPPORTED: 'UNSUPPORTED'>, <class 'sqlglot.expressions.VolatileProperty'>: <Location.UNSUPPORTED: 'UNSUPPORTED'>, <class 'sqlglot.expressions.WithDataProperty'>: <Location.UNSUPPORTED: 'UNSUPPORTED'>, <class 'sqlglot.expressions.WithJournalTableProperty'>: <Location.UNSUPPORTED: 'UNSUPPORTED'>, <class 'sqlglot.expressions.WithSystemVersioningProperty'>: <Location.UNSUPPORTED: 'UNSUPPORTED'>}
526 def timefromparts_sql(self, expression: exp.TimeFromParts) -> str: 527 nano = expression.args.get("nano") 528 if nano is not None: 529 expression.set( 530 "sec", expression.args["sec"] + nano.pop() / exp.Literal.number(1000000000.0) 531 ) 532 533 return rename_func("MAKE_TIME")(self, expression)
535 def timestampfromparts_sql(self, expression: exp.TimestampFromParts) -> str: 536 sec = expression.args["sec"] 537 538 milli = expression.args.get("milli") 539 if milli is not None: 540 sec += milli.pop() / exp.Literal.number(1000.0) 541 542 nano = expression.args.get("nano") 543 if nano is not None: 544 sec += nano.pop() / exp.Literal.number(1000000000.0) 545 546 if milli or nano: 547 expression.set("sec", sec) 548 549 return rename_func("MAKE_TIMESTAMP")(self, expression)
def
tablesample_sql( self, expression: sqlglot.expressions.TableSample, sep: str = ' AS ', tablesample_keyword: Optional[str] = None) -> str:
551 def tablesample_sql( 552 self, 553 expression: exp.TableSample, 554 sep: str = " AS ", 555 tablesample_keyword: t.Optional[str] = None, 556 ) -> str: 557 if not isinstance(expression.parent, exp.Select): 558 # This sample clause only applies to a single source, not the entire resulting relation 559 tablesample_keyword = "TABLESAMPLE" 560 561 if expression.args.get("size"): 562 method = expression.args.get("method") 563 if method and method.name.upper() != "RESERVOIR": 564 self.unsupported( 565 f"Sampling method {method} is not supported with a discrete sample count, " 566 "defaulting to reservoir sampling" 567 ) 568 expression.set("method", exp.var("RESERVOIR")) 569 570 return super().tablesample_sql( 571 expression, sep=sep, tablesample_keyword=tablesample_keyword 572 )
574 def interval_sql(self, expression: exp.Interval) -> str: 575 multiplier: t.Optional[int] = None 576 unit = expression.text("unit").lower() 577 578 if unit.startswith("week"): 579 multiplier = 7 580 if unit.startswith("quarter"): 581 multiplier = 90 582 583 if multiplier: 584 return f"({multiplier} * {super().interval_sql(exp.Interval(this=expression.this, unit=exp.var('DAY')))})" 585 586 return super().interval_sql(expression)
593 def join_sql(self, expression: exp.Join) -> str: 594 if ( 595 expression.side == "LEFT" 596 and not expression.args.get("on") 597 and isinstance(expression.this, exp.Unnest) 598 ): 599 # Some dialects support `LEFT JOIN UNNEST(...)` without an explicit ON clause 600 # DuckDB doesn't, but we can just add a dummy ON clause that is always true 601 return super().join_sql(expression.on(exp.true())) 602 603 return super().join_sql(expression)
605 def generateseries_sql(self, expression: exp.GenerateSeries) -> str: 606 # GENERATE_SERIES(a, b) -> [a, b], RANGE(a, b) -> [a, b) 607 if expression.args.get("is_end_exclusive"): 608 expression.set("is_end_exclusive", None) 609 return rename_func("RANGE")(self, expression) 610 611 return super().generateseries_sql(expression)
613 def bracket_sql(self, expression: exp.Bracket) -> str: 614 this = expression.this 615 if isinstance(this, exp.Array): 616 this.replace(exp.paren(this)) 617 618 bracket = super().bracket_sql(expression) 619 620 if not expression.args.get("returns_list_for_maps"): 621 if not this.type: 622 from sqlglot.optimizer.annotate_types import annotate_types 623 624 this = annotate_types(this) 625 626 if this.is_type(exp.DataType.Type.MAP): 627 bracket = f"({bracket})[1]" 628 629 return bracket
631 def withingroup_sql(self, expression: exp.WithinGroup) -> str: 632 expression_sql = self.sql(expression, "expression") 633 634 func = expression.this 635 if isinstance(func, exp.PERCENTILES): 636 # Make the order key the first arg and slide the fraction to the right 637 # https://duckdb.org/docs/sql/aggregates#ordered-set-aggregate-functions 638 order_col = expression.find(exp.Ordered) 639 if order_col: 640 func.set("expression", func.this) 641 func.set("this", order_col.this) 642 643 this = self.sql(expression, "this").rstrip(")") 644 645 return f"{this}{expression_sql})"
AFTER_HAVING_MODIFIER_TRANSFORMS =
{'qualify': <function Generator.<lambda>>, 'windows': <function Generator.<lambda>>}
Inherited Members
- sqlglot.generator.Generator
- Generator
- NULL_ORDERING_SUPPORTED
- LOCKING_READS_SUPPORTED
- EXPLICIT_UNION
- WRAP_DERIVED_VALUES
- CREATE_FUNCTION_RETURN_AS
- MATCHED_BY_SOURCE
- SINGLE_STRING_INTERVAL
- INTERVAL_ALLOWS_PLURAL_FORM
- LIMIT_ONLY_LITERALS
- GROUPINGS_SEP
- 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_INCLUDE_COLUMN_KEYWORD
- UNNEST_WITH_ORDINALITY
- AGGREGATE_FILTER_SUPPORTED
- COMPUTED_COLUMN_WITH_TYPE
- SUPPORTS_TABLE_COPY
- TABLESAMPLE_REQUIRES_PARENS
- TABLESAMPLE_SIZE_IS_ROWS
- TABLESAMPLE_WITH_METHOD
- COLLATE_IS_FUNC
- DATA_TYPE_SPECIFIERS_ALLOWED
- ENSURE_BOOLS
- CTE_RECURSIVE_KEYWORD_REQUIRED
- SUPPORTS_SINGLE_ARG_CONCAT
- SUPPORTS_TABLE_ALIAS_COLUMNS
- UNPIVOT_ALIASES_ARE_IDENTIFIERS
- INSERT_OVERWRITE
- SUPPORTS_SELECT_INTO
- SUPPORTS_UNLOGGED_TABLES
- LIKE_PROPERTY_INSIDE_SCHEMA
- JSON_TYPE_REQUIRED_FOR_EXTRACTION
- JSON_PATH_SINGLE_QUOTE_ESCAPE
- OUTER_UNION_MODIFIERS
- COPY_PARAMS_ARE_WRAPPED
- COPY_PARAMS_EQ_REQUIRED
- HEX_FUNC
- WITH_PROPERTIES_PREFIX
- TIME_PART_SINGULARS
- TOKEN_MAPPING
- RESERVED_KEYWORDS
- WITH_SEPARATED_COMMENTS
- EXCLUDE_COMMENTS
- PARAMETERIZABLE_TEXT_TYPES
- EXPRESSIONS_WITHOUT_NESTED_CTES
- SENTINEL_LINE_BREAK
- pretty
- identify
- normalize
- pad
- unsupported_level
- max_unsupported
- leading_comma
- max_text_width
- comments
- dialect
- normalize_functions
- unsupported_messages
- generate
- preprocess
- unsupported
- sep
- seg
- pad_comment
- maybe_comment
- wrap
- no_identify
- normalize_func
- indent
- sql
- uncache_sql
- cache_sql
- characterset_sql
- column_parts
- column_sql
- columnposition_sql
- columnconstraint_sql
- computedcolumnconstraint_sql
- autoincrementcolumnconstraint_sql
- compresscolumnconstraint_sql
- generatedasidentitycolumnconstraint_sql
- generatedasrowcolumnconstraint_sql
- periodforsystemtimeconstraint_sql
- notnullcolumnconstraint_sql
- transformcolumnconstraint_sql
- primarykeycolumnconstraint_sql
- uniquecolumnconstraint_sql
- createable_sql
- create_sql
- sequenceproperties_sql
- clone_sql
- describe_sql
- heredoc_sql
- prepend_ctes
- with_sql
- cte_sql
- tablealias_sql
- bitstring_sql
- hexstring_sql
- bytestring_sql
- unicodestring_sql
- rawstring_sql
- datatypeparam_sql
- datatype_sql
- directory_sql
- delete_sql
- drop_sql
- except_sql
- except_op
- fetch_sql
- filter_sql
- hint_sql
- indexparameters_sql
- index_sql
- identifier_sql
- hex_sql
- lowerhex_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
- historicaldata_sql
- table_parts
- table_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
- lambda_sql
- lateral_op
- 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
- withfill_sql
- cluster_sql
- distribute_sql
- sort_sql
- ordered_sql
- matchrecognizemeasure_sql
- matchrecognize_sql
- query_modifiers
- queryoption_sql
- offset_limit_modifiers
- after_limit_modifiers
- select_sql
- schema_sql
- schema_columns_sql
- star_sql
- parameter_sql
- sessionparameter_sql
- placeholder_sql
- subquery_sql
- qualify_sql
- set_operations
- union_sql
- union_op
- unnest_sql
- prewhere_sql
- where_sql
- window_sql
- partition_by_sql
- windowspec_sql
- between_sql
- bracket_offset_expressions
- all_sql
- any_sql
- exists_sql
- case_sql
- constraint_sql
- nextvaluefor_sql
- extract_sql
- trim_sql
- convert_concat_args
- concat_sql
- concatws_sql
- check_sql
- foreignkey_sql
- primarykey_sql
- if_sql
- matchagainst_sql
- jsonkeyvalue_sql
- jsonpath_sql
- json_path_part
- formatjson_sql
- jsonobject_sql
- jsonobjectagg_sql
- jsonarray_sql
- jsonarrayagg_sql
- jsoncolumndef_sql
- jsonschema_sql
- jsontable_sql
- openjsoncolumndef_sql
- openjson_sql
- in_sql
- in_unnest_op
- return_sql
- reference_sql
- anonymous_sql
- paren_sql
- neg_sql
- not_sql
- alias_sql
- pivotalias_sql
- aliases_sql
- atindex_sql
- attimezone_sql
- fromtimezone_sql
- add_sql
- and_sql
- or_sql
- xor_sql
- connector_sql
- bitwiseand_sql
- bitwiseleftshift_sql
- bitwisenot_sql
- bitwiseor_sql
- bitwiserightshift_sql
- bitwisexor_sql
- cast_sql
- currentdate_sql
- currenttimestamp_sql
- collate_sql
- command_sql
- comment_sql
- mergetreettlaction_sql
- mergetreettl_sql
- transaction_sql
- commit_sql
- rollback_sql
- altercolumn_sql
- alterdiststyle_sql
- altersortkey_sql
- renametable_sql
- renamecolumn_sql
- alterset_sql
- altertable_sql
- add_column_sql
- droppartition_sql
- addconstraint_sql
- distinct_sql
- ignorenulls_sql
- respectnulls_sql
- havingmax_sql
- intdiv_sql
- dpipe_sql
- div_sql
- overlaps_sql
- distance_sql
- dot_sql
- eq_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
- neq_sql
- nullsafeeq_sql
- nullsafeneq_sql
- slice_sql
- sub_sql
- trycast_sql
- try_sql
- log_sql
- use_sql
- binary
- function_fallback_sql
- func
- format_args
- too_wide
- format_time
- expressions
- op_expressions
- naked_property
- tag_sql
- token_sql
- userdefinedfunction_sql
- joinhint_sql
- kwarg_sql
- when_sql
- merge_sql
- tochar_sql
- tonumber_sql
- dictproperty_sql
- dictrange_sql
- dictsubproperty_sql
- oncluster_sql
- clusteredbyproperty_sql
- anyvalue_sql
- querytransform_sql
- indexconstraintoption_sql
- checkcolumnconstraint_sql
- indexcolumnconstraint_sql
- nvl2_sql
- comprehension_sql
- columnprefix_sql
- opclass_sql
- predict_sql
- forin_sql
- refresh_sql
- operator_sql
- toarray_sql
- tsordstotime_sql
- tsordstotimestamp_sql
- tsordstodate_sql
- unixdate_sql
- lastday_sql
- dateadd_sql
- arrayany_sql
- struct_sql
- partitionrange_sql
- truncatetable_sql
- convert_sql
- copyparameter_sql
- credentials_sql
- copy_sql
- semicolon_sql
- datadeletionproperty_sql
- maskingpolicycolumnconstraint_sql