Edit on GitHub

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    NormalizationStrategy,
  9    approx_count_distinct_sql,
 10    arg_max_or_min_no_count,
 11    arrow_json_extract_scalar_sql,
 12    arrow_json_extract_sql,
 13    binary_from_function,
 14    bool_xor_sql,
 15    date_trunc_to_time,
 16    datestrtodate_sql,
 17    encode_decode_sql,
 18    format_time_lambda,
 19    inline_array_sql,
 20    no_comment_column_constraint_sql,
 21    no_properties_sql,
 22    no_safe_divide_sql,
 23    no_timestamp_sql,
 24    pivot_column_names,
 25    prepend_dollar_to_path,
 26    regexp_extract_sql,
 27    rename_func,
 28    str_position_sql,
 29    str_to_time_sql,
 30    timestamptrunc_sql,
 31    timestrtotime_sql,
 32)
 33from sqlglot.helper import flatten, 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    unit = self.sql(expression, "unit").strip("'") or "DAY"
 40    interval = self.sql(exp.Interval(this=expression.expression, unit=unit))
 41    return f"CAST({this} AS {self.sql(expression.return_type)}) + {interval}"
 42
 43
 44def _date_delta_sql(self: DuckDB.Generator, expression: exp.DateAdd | exp.DateSub) -> str:
 45    this = self.sql(expression, "this")
 46    unit = self.sql(expression, "unit").strip("'") or "DAY"
 47    op = "+" if isinstance(expression, exp.DateAdd) else "-"
 48    return f"{this} {op} {self.sql(exp.Interval(this=expression.expression, unit=unit))}"
 49
 50
 51# BigQuery -> DuckDB conversion for the DATE function
 52def _date_sql(self: DuckDB.Generator, expression: exp.Date) -> str:
 53    result = f"CAST({self.sql(expression, 'this')} AS DATE)"
 54    zone = self.sql(expression, "zone")
 55
 56    if zone:
 57        date_str = self.func("STRFTIME", result, "'%d/%m/%Y'")
 58        date_str = f"{date_str} || ' ' || {zone}"
 59
 60        # This will create a TIMESTAMP with time zone information
 61        result = self.func("STRPTIME", date_str, "'%d/%m/%Y %Z'")
 62
 63    return result
 64
 65
 66def _array_sort_sql(self: DuckDB.Generator, expression: exp.ArraySort) -> str:
 67    if expression.expression:
 68        self.unsupported("DUCKDB ARRAY_SORT does not support a comparator")
 69    return f"ARRAY_SORT({self.sql(expression, 'this')})"
 70
 71
 72def _sort_array_sql(self: DuckDB.Generator, expression: exp.SortArray) -> str:
 73    this = self.sql(expression, "this")
 74    if expression.args.get("asc") == exp.false():
 75        return f"ARRAY_REVERSE_SORT({this})"
 76    return f"ARRAY_SORT({this})"
 77
 78
 79def _sort_array_reverse(args: t.List) -> exp.Expression:
 80    return exp.SortArray(this=seq_get(args, 0), asc=exp.false())
 81
 82
 83def _parse_date_diff(args: t.List) -> exp.Expression:
 84    return exp.DateDiff(this=seq_get(args, 2), expression=seq_get(args, 1), unit=seq_get(args, 0))
 85
 86
 87def _parse_make_timestamp(args: t.List) -> exp.Expression:
 88    if len(args) == 1:
 89        return exp.UnixToTime(this=seq_get(args, 0), scale=exp.UnixToTime.MICROS)
 90
 91    return exp.TimestampFromParts(
 92        year=seq_get(args, 0),
 93        month=seq_get(args, 1),
 94        day=seq_get(args, 2),
 95        hour=seq_get(args, 3),
 96        min=seq_get(args, 4),
 97        sec=seq_get(args, 5),
 98    )
 99
100
101def _struct_sql(self: DuckDB.Generator, expression: exp.Struct) -> str:
102    args: t.List[str] = []
103    for expr in expression.expressions:
104        if isinstance(expr, exp.Alias):
105            key = expr.alias
106            value = expr.this
107        else:
108            key = expr.name or expr.this.name
109            if isinstance(expr, exp.Bracket):
110                value = expr.expressions[0]
111            else:
112                value = expr.expression
113
114        args.append(f"{self.sql(exp.Literal.string(key))}: {self.sql(value)}")
115
116    return f"{{{', '.join(args)}}}"
117
118
119def _datatype_sql(self: DuckDB.Generator, expression: exp.DataType) -> str:
120    if expression.is_type("array"):
121        return f"{self.expressions(expression, flat=True)}[]"
122
123    # Type TIMESTAMP / TIME WITH TIME ZONE does not support any modifiers
124    if expression.is_type("timestamptz", "timetz"):
125        return expression.this.value
126
127    return self.datatype_sql(expression)
128
129
130def _json_format_sql(self: DuckDB.Generator, expression: exp.JSONFormat) -> str:
131    sql = self.func("TO_JSON", expression.this, expression.args.get("options"))
132    return f"CAST({sql} AS TEXT)"
133
134
135def _unix_to_time_sql(self: DuckDB.Generator, expression: exp.UnixToTime) -> str:
136    scale = expression.args.get("scale")
137    timestamp = self.sql(expression, "this")
138    if scale in (None, exp.UnixToTime.SECONDS):
139        return f"TO_TIMESTAMP({timestamp})"
140    if scale == exp.UnixToTime.MILLIS:
141        return f"EPOCH_MS({timestamp})"
142    if scale == exp.UnixToTime.MICROS:
143        return f"MAKE_TIMESTAMP({timestamp})"
144
145    return f"TO_TIMESTAMP({timestamp} / POW(10, {scale}))"
146
147
148def _rename_unless_within_group(
149    a: str, b: str
150) -> t.Callable[[DuckDB.Generator, exp.Expression], str]:
151    return lambda self, expression: (
152        self.func(a, *flatten(expression.args.values()))
153        if isinstance(expression.find_ancestor(exp.Select, exp.WithinGroup), exp.WithinGroup)
154        else self.func(b, *flatten(expression.args.values()))
155    )
156
157
158def _parse_struct_pack(args: t.List) -> exp.Struct:
159    args_with_columns_as_identifiers = [
160        exp.PropertyEQ(this=arg.this.this, expression=arg.expression) for arg in args
161    ]
162    return exp.Struct.from_arg_list(args_with_columns_as_identifiers)
163
164
165class DuckDB(Dialect):
166    NULL_ORDERING = "nulls_are_last"
167    SUPPORTS_USER_DEFINED_TYPES = False
168    SAFE_DIVISION = True
169    INDEX_OFFSET = 1
170    CONCAT_COALESCE = True
171
172    # https://duckdb.org/docs/sql/introduction.html#creating-a-new-table
173    NORMALIZATION_STRATEGY = NormalizationStrategy.CASE_INSENSITIVE
174
175    class Tokenizer(tokens.Tokenizer):
176        KEYWORDS = {
177            **tokens.Tokenizer.KEYWORDS,
178            "//": TokenType.DIV,
179            "ATTACH": TokenType.COMMAND,
180            "BINARY": TokenType.VARBINARY,
181            "BITSTRING": TokenType.BIT,
182            "BPCHAR": TokenType.TEXT,
183            "CHAR": TokenType.TEXT,
184            "CHARACTER VARYING": TokenType.TEXT,
185            "EXCLUDE": TokenType.EXCEPT,
186            "LOGICAL": TokenType.BOOLEAN,
187            "PIVOT_WIDER": TokenType.PIVOT,
188            "SIGNED": TokenType.INT,
189            "STRING": TokenType.VARCHAR,
190            "UBIGINT": TokenType.UBIGINT,
191            "UINTEGER": TokenType.UINT,
192            "USMALLINT": TokenType.USMALLINT,
193            "UTINYINT": TokenType.UTINYINT,
194            "TIMESTAMP_S": TokenType.TIMESTAMP_S,
195            "TIMESTAMP_MS": TokenType.TIMESTAMP_MS,
196            "TIMESTAMP_NS": TokenType.TIMESTAMP_NS,
197            "TIMESTAMP_US": TokenType.TIMESTAMP,
198        }
199
200        SINGLE_TOKENS = {
201            **tokens.Tokenizer.SINGLE_TOKENS,
202            "$": TokenType.PARAMETER,
203        }
204
205    class Parser(parser.Parser):
206        BITWISE = {
207            **parser.Parser.BITWISE,
208            TokenType.TILDA: exp.RegexpLike,
209        }
210
211        FUNCTIONS = {
212            **parser.Parser.FUNCTIONS,
213            "ARRAY_HAS": exp.ArrayContains.from_arg_list,
214            "ARRAY_LENGTH": exp.ArraySize.from_arg_list,
215            "ARRAY_SORT": exp.SortArray.from_arg_list,
216            "ARRAY_REVERSE_SORT": _sort_array_reverse,
217            "DATEDIFF": _parse_date_diff,
218            "DATE_DIFF": _parse_date_diff,
219            "DATE_TRUNC": date_trunc_to_time,
220            "DATETRUNC": date_trunc_to_time,
221            "DECODE": lambda args: exp.Decode(
222                this=seq_get(args, 0), charset=exp.Literal.string("utf-8")
223            ),
224            "ENCODE": lambda args: exp.Encode(
225                this=seq_get(args, 0), charset=exp.Literal.string("utf-8")
226            ),
227            "EPOCH": exp.TimeToUnix.from_arg_list,
228            "EPOCH_MS": lambda args: exp.UnixToTime(
229                this=seq_get(args, 0), scale=exp.UnixToTime.MILLIS
230            ),
231            "JSON": exp.ParseJSON.from_arg_list,
232            "LIST_HAS": exp.ArrayContains.from_arg_list,
233            "LIST_REVERSE_SORT": _sort_array_reverse,
234            "LIST_SORT": exp.SortArray.from_arg_list,
235            "LIST_VALUE": exp.Array.from_arg_list,
236            "MAKE_TIME": exp.TimeFromParts.from_arg_list,
237            "MAKE_TIMESTAMP": _parse_make_timestamp,
238            "MEDIAN": lambda args: exp.PercentileCont(
239                this=seq_get(args, 0), expression=exp.Literal.number(0.5)
240            ),
241            "QUANTILE_CONT": exp.PercentileCont.from_arg_list,
242            "QUANTILE_DISC": exp.PercentileDisc.from_arg_list,
243            "REGEXP_EXTRACT": lambda args: exp.RegexpExtract(
244                this=seq_get(args, 0), expression=seq_get(args, 1), group=seq_get(args, 2)
245            ),
246            "REGEXP_MATCHES": exp.RegexpLike.from_arg_list,
247            "REGEXP_REPLACE": lambda args: exp.RegexpReplace(
248                this=seq_get(args, 0),
249                expression=seq_get(args, 1),
250                replacement=seq_get(args, 2),
251                modifiers=seq_get(args, 3),
252            ),
253            "STRFTIME": format_time_lambda(exp.TimeToStr, "duckdb"),
254            "STRING_SPLIT": exp.Split.from_arg_list,
255            "STRING_SPLIT_REGEX": exp.RegexpSplit.from_arg_list,
256            "STRING_TO_ARRAY": exp.Split.from_arg_list,
257            "STRPTIME": format_time_lambda(exp.StrToTime, "duckdb"),
258            "STRUCT_PACK": _parse_struct_pack,
259            "STR_SPLIT": exp.Split.from_arg_list,
260            "STR_SPLIT_REGEX": exp.RegexpSplit.from_arg_list,
261            "TO_TIMESTAMP": exp.UnixToTime.from_arg_list,
262            "UNNEST": exp.Explode.from_arg_list,
263            "XOR": binary_from_function(exp.BitwiseXor),
264        }
265
266        FUNCTION_PARSERS = parser.Parser.FUNCTION_PARSERS.copy()
267        FUNCTION_PARSERS.pop("DECODE", None)
268
269        TABLE_ALIAS_TOKENS = parser.Parser.TABLE_ALIAS_TOKENS - {
270            TokenType.SEMI,
271            TokenType.ANTI,
272        }
273
274        PLACEHOLDER_PARSERS = {
275            **parser.Parser.PLACEHOLDER_PARSERS,
276            TokenType.PARAMETER: lambda self: (
277                self.expression(exp.Placeholder, this=self._prev.text)
278                if self._match(TokenType.NUMBER) or self._match_set(self.ID_VAR_TOKENS)
279                else None
280            ),
281        }
282
283        def _parse_types(
284            self, check_func: bool = False, schema: bool = False, allow_identifiers: bool = True
285        ) -> t.Optional[exp.Expression]:
286            this = super()._parse_types(
287                check_func=check_func, schema=schema, allow_identifiers=allow_identifiers
288            )
289
290            # DuckDB treats NUMERIC and DECIMAL without precision as DECIMAL(18, 3)
291            # See: https://duckdb.org/docs/sql/data_types/numeric
292            if (
293                isinstance(this, exp.DataType)
294                and this.is_type("numeric", "decimal")
295                and not this.expressions
296            ):
297                return exp.DataType.build("DECIMAL(18, 3)")
298
299            return this
300
301        def _parse_struct_types(self, type_required: bool = False) -> t.Optional[exp.Expression]:
302            return self._parse_field_def()
303
304        def _pivot_column_names(self, aggregations: t.List[exp.Expression]) -> t.List[str]:
305            if len(aggregations) == 1:
306                return super()._pivot_column_names(aggregations)
307            return pivot_column_names(aggregations, dialect="duckdb")
308
309    class Generator(generator.Generator):
310        JOIN_HINTS = False
311        TABLE_HINTS = False
312        QUERY_HINTS = False
313        LIMIT_FETCH = "LIMIT"
314        STRUCT_DELIMITER = ("(", ")")
315        RENAME_TABLE_WITH_DB = False
316        NVL2_SUPPORTED = False
317        SEMI_ANTI_JOIN_WITH_SIDE = False
318        TABLESAMPLE_KEYWORDS = "USING SAMPLE"
319        TABLESAMPLE_SEED_KEYWORD = "REPEATABLE"
320        LAST_DAY_SUPPORTS_DATE_PART = False
321        JSON_KEY_VALUE_PAIR_SEP = ","
322
323        TRANSFORMS = {
324            **generator.Generator.TRANSFORMS,
325            exp.ApproxDistinct: approx_count_distinct_sql,
326            exp.Array: lambda self, e: (
327                self.func("ARRAY", e.expressions[0])
328                if e.expressions and e.expressions[0].find(exp.Select)
329                else inline_array_sql(self, e)
330            ),
331            exp.ArraySize: rename_func("ARRAY_LENGTH"),
332            exp.ArgMax: arg_max_or_min_no_count("ARG_MAX"),
333            exp.ArgMin: arg_max_or_min_no_count("ARG_MIN"),
334            exp.ArraySort: _array_sort_sql,
335            exp.ArraySum: rename_func("LIST_SUM"),
336            exp.BitwiseXor: rename_func("XOR"),
337            exp.CommentColumnConstraint: no_comment_column_constraint_sql,
338            exp.CurrentDate: lambda self, e: "CURRENT_DATE",
339            exp.CurrentTime: lambda self, e: "CURRENT_TIME",
340            exp.CurrentTimestamp: lambda self, e: "CURRENT_TIMESTAMP",
341            exp.DayOfMonth: rename_func("DAYOFMONTH"),
342            exp.DayOfWeek: rename_func("DAYOFWEEK"),
343            exp.DayOfYear: rename_func("DAYOFYEAR"),
344            exp.DataType: _datatype_sql,
345            exp.Date: _date_sql,
346            exp.DateAdd: _date_delta_sql,
347            exp.DateFromParts: rename_func("MAKE_DATE"),
348            exp.DateSub: _date_delta_sql,
349            exp.DateDiff: lambda self, e: self.func(
350                "DATE_DIFF", f"'{e.args.get('unit') or 'DAY'}'", e.expression, e.this
351            ),
352            exp.DateStrToDate: datestrtodate_sql,
353            exp.DateToDi: lambda self, e: f"CAST(STRFTIME({self.sql(e, 'this')}, {DuckDB.DATEINT_FORMAT}) AS INT)",
354            exp.Decode: lambda self, e: encode_decode_sql(self, e, "DECODE", replace=False),
355            exp.DiToDate: lambda self, e: f"CAST(STRPTIME(CAST({self.sql(e, 'this')} AS TEXT), {DuckDB.DATEINT_FORMAT}) AS DATE)",
356            exp.Encode: lambda self, e: encode_decode_sql(self, e, "ENCODE", replace=False),
357            exp.Explode: rename_func("UNNEST"),
358            exp.IntDiv: lambda self, e: self.binary(e, "//"),
359            exp.IsInf: rename_func("ISINF"),
360            exp.IsNan: rename_func("ISNAN"),
361            exp.JSONBExtract: arrow_json_extract_sql,
362            exp.JSONBExtractScalar: arrow_json_extract_scalar_sql,
363            exp.JSONExtract: arrow_json_extract_sql,
364            exp.JSONExtractScalar: arrow_json_extract_scalar_sql,
365            exp.JSONFormat: _json_format_sql,
366            exp.LogicalOr: rename_func("BOOL_OR"),
367            exp.LogicalAnd: rename_func("BOOL_AND"),
368            exp.MonthsBetween: lambda self, e: self.func(
369                "DATEDIFF",
370                "'month'",
371                exp.cast(e.expression, "timestamp", copy=True),
372                exp.cast(e.this, "timestamp", copy=True),
373            ),
374            exp.ParseJSON: rename_func("JSON"),
375            exp.PercentileCont: _rename_unless_within_group("PERCENTILE_CONT", "QUANTILE_CONT"),
376            exp.PercentileDisc: _rename_unless_within_group("PERCENTILE_DISC", "QUANTILE_DISC"),
377            # DuckDB doesn't allow qualified columns inside of PIVOT expressions.
378            # See: https://github.com/duckdb/duckdb/blob/671faf92411182f81dce42ac43de8bfb05d9909e/src/planner/binder/tableref/bind_pivot.cpp#L61-L62
379            exp.Pivot: transforms.preprocess([transforms.unqualify_columns]),
380            exp.Properties: no_properties_sql,
381            exp.RegexpExtract: regexp_extract_sql,
382            exp.RegexpReplace: lambda self, e: self.func(
383                "REGEXP_REPLACE",
384                e.this,
385                e.expression,
386                e.args.get("replacement"),
387                e.args.get("modifiers"),
388            ),
389            exp.RegexpLike: rename_func("REGEXP_MATCHES"),
390            exp.RegexpSplit: rename_func("STR_SPLIT_REGEX"),
391            exp.Rand: rename_func("RANDOM"),
392            exp.SafeDivide: no_safe_divide_sql,
393            exp.Split: rename_func("STR_SPLIT"),
394            exp.SortArray: _sort_array_sql,
395            exp.StrPosition: str_position_sql,
396            exp.StrToDate: lambda self, e: f"CAST({str_to_time_sql(self, e)} AS DATE)",
397            exp.StrToTime: str_to_time_sql,
398            exp.StrToUnix: lambda self, e: f"EPOCH(STRPTIME({self.sql(e, 'this')}, {self.format_time(e)}))",
399            exp.Struct: _struct_sql,
400            exp.Timestamp: no_timestamp_sql,
401            exp.TimestampDiff: lambda self, e: self.func(
402                "DATE_DIFF", exp.Literal.string(e.unit), e.expression, e.this
403            ),
404            exp.TimestampTrunc: timestamptrunc_sql,
405            exp.TimeStrToDate: lambda self, e: f"CAST({self.sql(e, 'this')} AS DATE)",
406            exp.TimeStrToTime: timestrtotime_sql,
407            exp.TimeStrToUnix: lambda self, e: f"EPOCH(CAST({self.sql(e, 'this')} AS TIMESTAMP))",
408            exp.TimeToStr: lambda self, e: f"STRFTIME({self.sql(e, 'this')}, {self.format_time(e)})",
409            exp.TimeToUnix: rename_func("EPOCH"),
410            exp.TsOrDiToDi: lambda self, e: f"CAST(SUBSTR(REPLACE(CAST({self.sql(e, 'this')} AS TEXT), '-', ''), 1, 8) AS INT)",
411            exp.TsOrDsAdd: _ts_or_ds_add_sql,
412            exp.TsOrDsDiff: lambda self, e: self.func(
413                "DATE_DIFF",
414                f"'{e.args.get('unit') or 'DAY'}'",
415                exp.cast(e.expression, "TIMESTAMP"),
416                exp.cast(e.this, "TIMESTAMP"),
417            ),
418            exp.UnixToStr: lambda self, e: f"STRFTIME(TO_TIMESTAMP({self.sql(e, 'this')}), {self.format_time(e)})",
419            exp.UnixToTime: _unix_to_time_sql,
420            exp.UnixToTimeStr: lambda self, e: f"CAST(TO_TIMESTAMP({self.sql(e, 'this')}) AS TEXT)",
421            exp.VariancePop: rename_func("VAR_POP"),
422            exp.WeekOfYear: rename_func("WEEKOFYEAR"),
423            exp.Xor: bool_xor_sql,
424        }
425
426        TYPE_MAPPING = {
427            **generator.Generator.TYPE_MAPPING,
428            exp.DataType.Type.BINARY: "BLOB",
429            exp.DataType.Type.CHAR: "TEXT",
430            exp.DataType.Type.FLOAT: "REAL",
431            exp.DataType.Type.NCHAR: "TEXT",
432            exp.DataType.Type.NVARCHAR: "TEXT",
433            exp.DataType.Type.UINT: "UINTEGER",
434            exp.DataType.Type.VARBINARY: "BLOB",
435            exp.DataType.Type.VARCHAR: "TEXT",
436            exp.DataType.Type.TIMESTAMP_S: "TIMESTAMP_S",
437            exp.DataType.Type.TIMESTAMP_MS: "TIMESTAMP_MS",
438            exp.DataType.Type.TIMESTAMP_NS: "TIMESTAMP_NS",
439        }
440
441        STAR_MAPPING = {**generator.Generator.STAR_MAPPING, "except": "EXCLUDE"}
442
443        UNWRAPPED_INTERVAL_VALUES = (exp.Column, exp.Literal, exp.Paren)
444
445        PROPERTIES_LOCATION = {
446            **generator.Generator.PROPERTIES_LOCATION,
447            exp.VolatileProperty: exp.Properties.Location.UNSUPPORTED,
448        }
449
450        def timefromparts_sql(self, expression: exp.TimeFromParts) -> str:
451            nano = expression.args.get("nano")
452            if nano is not None:
453                expression.set(
454                    "sec", expression.args["sec"] + nano.pop() / exp.Literal.number(1000000000.0)
455                )
456
457            return rename_func("MAKE_TIME")(self, expression)
458
459        def timestampfromparts_sql(self, expression: exp.TimestampFromParts) -> str:
460            sec = expression.args["sec"]
461
462            milli = expression.args.get("milli")
463            if milli is not None:
464                sec += milli.pop() / exp.Literal.number(1000.0)
465
466            nano = expression.args.get("nano")
467            if nano is not None:
468                sec += nano.pop() / exp.Literal.number(1000000000.0)
469
470            if milli or nano:
471                expression.set("sec", sec)
472
473            return rename_func("MAKE_TIMESTAMP")(self, expression)
474
475        def tablesample_sql(
476            self,
477            expression: exp.TableSample,
478            sep: str = " AS ",
479            tablesample_keyword: t.Optional[str] = None,
480        ) -> str:
481            if not isinstance(expression.parent, exp.Select):
482                # This sample clause only applies to a single source, not the entire resulting relation
483                tablesample_keyword = "TABLESAMPLE"
484
485            return super().tablesample_sql(
486                expression, sep=sep, tablesample_keyword=tablesample_keyword
487            )
488
489        def getpath_sql(self, expression: exp.GetPath) -> str:
490            expression = prepend_dollar_to_path(expression)
491            return f"{self.sql(expression, 'this')} -> {self.sql(expression, 'expression')}"
492
493        def interval_sql(self, expression: exp.Interval) -> str:
494            multiplier: t.Optional[int] = None
495            unit = expression.text("unit").lower()
496
497            if unit.startswith("week"):
498                multiplier = 7
499            if unit.startswith("quarter"):
500                multiplier = 90
501
502            if multiplier:
503                return f"({multiplier} * {super().interval_sql(exp.Interval(this=expression.this, unit=exp.var('DAY')))})"
504
505            return super().interval_sql(expression)
506
507        def columndef_sql(self, expression: exp.ColumnDef, sep: str = " ") -> str:
508            if isinstance(expression.parent, exp.UserDefinedFunction):
509                return self.sql(expression, "this")
510            return super().columndef_sql(expression, sep)
511
512        def placeholder_sql(self, expression: exp.Placeholder) -> str:
513            return f"${expression.name}" if expression.name else "?"
class DuckDB(sqlglot.dialects.dialect.Dialect):
166class DuckDB(Dialect):
167    NULL_ORDERING = "nulls_are_last"
168    SUPPORTS_USER_DEFINED_TYPES = False
169    SAFE_DIVISION = True
170    INDEX_OFFSET = 1
171    CONCAT_COALESCE = True
172
173    # https://duckdb.org/docs/sql/introduction.html#creating-a-new-table
174    NORMALIZATION_STRATEGY = NormalizationStrategy.CASE_INSENSITIVE
175
176    class Tokenizer(tokens.Tokenizer):
177        KEYWORDS = {
178            **tokens.Tokenizer.KEYWORDS,
179            "//": TokenType.DIV,
180            "ATTACH": TokenType.COMMAND,
181            "BINARY": TokenType.VARBINARY,
182            "BITSTRING": TokenType.BIT,
183            "BPCHAR": TokenType.TEXT,
184            "CHAR": TokenType.TEXT,
185            "CHARACTER VARYING": TokenType.TEXT,
186            "EXCLUDE": TokenType.EXCEPT,
187            "LOGICAL": TokenType.BOOLEAN,
188            "PIVOT_WIDER": TokenType.PIVOT,
189            "SIGNED": TokenType.INT,
190            "STRING": TokenType.VARCHAR,
191            "UBIGINT": TokenType.UBIGINT,
192            "UINTEGER": TokenType.UINT,
193            "USMALLINT": TokenType.USMALLINT,
194            "UTINYINT": TokenType.UTINYINT,
195            "TIMESTAMP_S": TokenType.TIMESTAMP_S,
196            "TIMESTAMP_MS": TokenType.TIMESTAMP_MS,
197            "TIMESTAMP_NS": TokenType.TIMESTAMP_NS,
198            "TIMESTAMP_US": TokenType.TIMESTAMP,
199        }
200
201        SINGLE_TOKENS = {
202            **tokens.Tokenizer.SINGLE_TOKENS,
203            "$": TokenType.PARAMETER,
204        }
205
206    class Parser(parser.Parser):
207        BITWISE = {
208            **parser.Parser.BITWISE,
209            TokenType.TILDA: exp.RegexpLike,
210        }
211
212        FUNCTIONS = {
213            **parser.Parser.FUNCTIONS,
214            "ARRAY_HAS": exp.ArrayContains.from_arg_list,
215            "ARRAY_LENGTH": exp.ArraySize.from_arg_list,
216            "ARRAY_SORT": exp.SortArray.from_arg_list,
217            "ARRAY_REVERSE_SORT": _sort_array_reverse,
218            "DATEDIFF": _parse_date_diff,
219            "DATE_DIFF": _parse_date_diff,
220            "DATE_TRUNC": date_trunc_to_time,
221            "DATETRUNC": date_trunc_to_time,
222            "DECODE": lambda args: exp.Decode(
223                this=seq_get(args, 0), charset=exp.Literal.string("utf-8")
224            ),
225            "ENCODE": lambda args: exp.Encode(
226                this=seq_get(args, 0), charset=exp.Literal.string("utf-8")
227            ),
228            "EPOCH": exp.TimeToUnix.from_arg_list,
229            "EPOCH_MS": lambda args: exp.UnixToTime(
230                this=seq_get(args, 0), scale=exp.UnixToTime.MILLIS
231            ),
232            "JSON": exp.ParseJSON.from_arg_list,
233            "LIST_HAS": exp.ArrayContains.from_arg_list,
234            "LIST_REVERSE_SORT": _sort_array_reverse,
235            "LIST_SORT": exp.SortArray.from_arg_list,
236            "LIST_VALUE": exp.Array.from_arg_list,
237            "MAKE_TIME": exp.TimeFromParts.from_arg_list,
238            "MAKE_TIMESTAMP": _parse_make_timestamp,
239            "MEDIAN": lambda args: exp.PercentileCont(
240                this=seq_get(args, 0), expression=exp.Literal.number(0.5)
241            ),
242            "QUANTILE_CONT": exp.PercentileCont.from_arg_list,
243            "QUANTILE_DISC": exp.PercentileDisc.from_arg_list,
244            "REGEXP_EXTRACT": lambda args: exp.RegexpExtract(
245                this=seq_get(args, 0), expression=seq_get(args, 1), group=seq_get(args, 2)
246            ),
247            "REGEXP_MATCHES": exp.RegexpLike.from_arg_list,
248            "REGEXP_REPLACE": lambda args: exp.RegexpReplace(
249                this=seq_get(args, 0),
250                expression=seq_get(args, 1),
251                replacement=seq_get(args, 2),
252                modifiers=seq_get(args, 3),
253            ),
254            "STRFTIME": format_time_lambda(exp.TimeToStr, "duckdb"),
255            "STRING_SPLIT": exp.Split.from_arg_list,
256            "STRING_SPLIT_REGEX": exp.RegexpSplit.from_arg_list,
257            "STRING_TO_ARRAY": exp.Split.from_arg_list,
258            "STRPTIME": format_time_lambda(exp.StrToTime, "duckdb"),
259            "STRUCT_PACK": _parse_struct_pack,
260            "STR_SPLIT": exp.Split.from_arg_list,
261            "STR_SPLIT_REGEX": exp.RegexpSplit.from_arg_list,
262            "TO_TIMESTAMP": exp.UnixToTime.from_arg_list,
263            "UNNEST": exp.Explode.from_arg_list,
264            "XOR": binary_from_function(exp.BitwiseXor),
265        }
266
267        FUNCTION_PARSERS = parser.Parser.FUNCTION_PARSERS.copy()
268        FUNCTION_PARSERS.pop("DECODE", None)
269
270        TABLE_ALIAS_TOKENS = parser.Parser.TABLE_ALIAS_TOKENS - {
271            TokenType.SEMI,
272            TokenType.ANTI,
273        }
274
275        PLACEHOLDER_PARSERS = {
276            **parser.Parser.PLACEHOLDER_PARSERS,
277            TokenType.PARAMETER: lambda self: (
278                self.expression(exp.Placeholder, this=self._prev.text)
279                if self._match(TokenType.NUMBER) or self._match_set(self.ID_VAR_TOKENS)
280                else None
281            ),
282        }
283
284        def _parse_types(
285            self, check_func: bool = False, schema: bool = False, allow_identifiers: bool = True
286        ) -> t.Optional[exp.Expression]:
287            this = super()._parse_types(
288                check_func=check_func, schema=schema, allow_identifiers=allow_identifiers
289            )
290
291            # DuckDB treats NUMERIC and DECIMAL without precision as DECIMAL(18, 3)
292            # See: https://duckdb.org/docs/sql/data_types/numeric
293            if (
294                isinstance(this, exp.DataType)
295                and this.is_type("numeric", "decimal")
296                and not this.expressions
297            ):
298                return exp.DataType.build("DECIMAL(18, 3)")
299
300            return this
301
302        def _parse_struct_types(self, type_required: bool = False) -> t.Optional[exp.Expression]:
303            return self._parse_field_def()
304
305        def _pivot_column_names(self, aggregations: t.List[exp.Expression]) -> t.List[str]:
306            if len(aggregations) == 1:
307                return super()._pivot_column_names(aggregations)
308            return pivot_column_names(aggregations, dialect="duckdb")
309
310    class Generator(generator.Generator):
311        JOIN_HINTS = False
312        TABLE_HINTS = False
313        QUERY_HINTS = False
314        LIMIT_FETCH = "LIMIT"
315        STRUCT_DELIMITER = ("(", ")")
316        RENAME_TABLE_WITH_DB = False
317        NVL2_SUPPORTED = False
318        SEMI_ANTI_JOIN_WITH_SIDE = False
319        TABLESAMPLE_KEYWORDS = "USING SAMPLE"
320        TABLESAMPLE_SEED_KEYWORD = "REPEATABLE"
321        LAST_DAY_SUPPORTS_DATE_PART = False
322        JSON_KEY_VALUE_PAIR_SEP = ","
323
324        TRANSFORMS = {
325            **generator.Generator.TRANSFORMS,
326            exp.ApproxDistinct: approx_count_distinct_sql,
327            exp.Array: lambda self, e: (
328                self.func("ARRAY", e.expressions[0])
329                if e.expressions and e.expressions[0].find(exp.Select)
330                else inline_array_sql(self, e)
331            ),
332            exp.ArraySize: rename_func("ARRAY_LENGTH"),
333            exp.ArgMax: arg_max_or_min_no_count("ARG_MAX"),
334            exp.ArgMin: arg_max_or_min_no_count("ARG_MIN"),
335            exp.ArraySort: _array_sort_sql,
336            exp.ArraySum: rename_func("LIST_SUM"),
337            exp.BitwiseXor: rename_func("XOR"),
338            exp.CommentColumnConstraint: no_comment_column_constraint_sql,
339            exp.CurrentDate: lambda self, e: "CURRENT_DATE",
340            exp.CurrentTime: lambda self, e: "CURRENT_TIME",
341            exp.CurrentTimestamp: lambda self, e: "CURRENT_TIMESTAMP",
342            exp.DayOfMonth: rename_func("DAYOFMONTH"),
343            exp.DayOfWeek: rename_func("DAYOFWEEK"),
344            exp.DayOfYear: rename_func("DAYOFYEAR"),
345            exp.DataType: _datatype_sql,
346            exp.Date: _date_sql,
347            exp.DateAdd: _date_delta_sql,
348            exp.DateFromParts: rename_func("MAKE_DATE"),
349            exp.DateSub: _date_delta_sql,
350            exp.DateDiff: lambda self, e: self.func(
351                "DATE_DIFF", f"'{e.args.get('unit') or 'DAY'}'", e.expression, e.this
352            ),
353            exp.DateStrToDate: datestrtodate_sql,
354            exp.DateToDi: lambda self, e: f"CAST(STRFTIME({self.sql(e, 'this')}, {DuckDB.DATEINT_FORMAT}) AS INT)",
355            exp.Decode: lambda self, e: encode_decode_sql(self, e, "DECODE", replace=False),
356            exp.DiToDate: lambda self, e: f"CAST(STRPTIME(CAST({self.sql(e, 'this')} AS TEXT), {DuckDB.DATEINT_FORMAT}) AS DATE)",
357            exp.Encode: lambda self, e: encode_decode_sql(self, e, "ENCODE", replace=False),
358            exp.Explode: rename_func("UNNEST"),
359            exp.IntDiv: lambda self, e: self.binary(e, "//"),
360            exp.IsInf: rename_func("ISINF"),
361            exp.IsNan: rename_func("ISNAN"),
362            exp.JSONBExtract: arrow_json_extract_sql,
363            exp.JSONBExtractScalar: arrow_json_extract_scalar_sql,
364            exp.JSONExtract: arrow_json_extract_sql,
365            exp.JSONExtractScalar: arrow_json_extract_scalar_sql,
366            exp.JSONFormat: _json_format_sql,
367            exp.LogicalOr: rename_func("BOOL_OR"),
368            exp.LogicalAnd: rename_func("BOOL_AND"),
369            exp.MonthsBetween: lambda self, e: self.func(
370                "DATEDIFF",
371                "'month'",
372                exp.cast(e.expression, "timestamp", copy=True),
373                exp.cast(e.this, "timestamp", copy=True),
374            ),
375            exp.ParseJSON: rename_func("JSON"),
376            exp.PercentileCont: _rename_unless_within_group("PERCENTILE_CONT", "QUANTILE_CONT"),
377            exp.PercentileDisc: _rename_unless_within_group("PERCENTILE_DISC", "QUANTILE_DISC"),
378            # DuckDB doesn't allow qualified columns inside of PIVOT expressions.
379            # See: https://github.com/duckdb/duckdb/blob/671faf92411182f81dce42ac43de8bfb05d9909e/src/planner/binder/tableref/bind_pivot.cpp#L61-L62
380            exp.Pivot: transforms.preprocess([transforms.unqualify_columns]),
381            exp.Properties: no_properties_sql,
382            exp.RegexpExtract: regexp_extract_sql,
383            exp.RegexpReplace: lambda self, e: self.func(
384                "REGEXP_REPLACE",
385                e.this,
386                e.expression,
387                e.args.get("replacement"),
388                e.args.get("modifiers"),
389            ),
390            exp.RegexpLike: rename_func("REGEXP_MATCHES"),
391            exp.RegexpSplit: rename_func("STR_SPLIT_REGEX"),
392            exp.Rand: rename_func("RANDOM"),
393            exp.SafeDivide: no_safe_divide_sql,
394            exp.Split: rename_func("STR_SPLIT"),
395            exp.SortArray: _sort_array_sql,
396            exp.StrPosition: str_position_sql,
397            exp.StrToDate: lambda self, e: f"CAST({str_to_time_sql(self, e)} AS DATE)",
398            exp.StrToTime: str_to_time_sql,
399            exp.StrToUnix: lambda self, e: f"EPOCH(STRPTIME({self.sql(e, 'this')}, {self.format_time(e)}))",
400            exp.Struct: _struct_sql,
401            exp.Timestamp: no_timestamp_sql,
402            exp.TimestampDiff: lambda self, e: self.func(
403                "DATE_DIFF", exp.Literal.string(e.unit), e.expression, e.this
404            ),
405            exp.TimestampTrunc: timestamptrunc_sql,
406            exp.TimeStrToDate: lambda self, e: f"CAST({self.sql(e, 'this')} AS DATE)",
407            exp.TimeStrToTime: timestrtotime_sql,
408            exp.TimeStrToUnix: lambda self, e: f"EPOCH(CAST({self.sql(e, 'this')} AS TIMESTAMP))",
409            exp.TimeToStr: lambda self, e: f"STRFTIME({self.sql(e, 'this')}, {self.format_time(e)})",
410            exp.TimeToUnix: rename_func("EPOCH"),
411            exp.TsOrDiToDi: lambda self, e: f"CAST(SUBSTR(REPLACE(CAST({self.sql(e, 'this')} AS TEXT), '-', ''), 1, 8) AS INT)",
412            exp.TsOrDsAdd: _ts_or_ds_add_sql,
413            exp.TsOrDsDiff: lambda self, e: self.func(
414                "DATE_DIFF",
415                f"'{e.args.get('unit') or 'DAY'}'",
416                exp.cast(e.expression, "TIMESTAMP"),
417                exp.cast(e.this, "TIMESTAMP"),
418            ),
419            exp.UnixToStr: lambda self, e: f"STRFTIME(TO_TIMESTAMP({self.sql(e, 'this')}), {self.format_time(e)})",
420            exp.UnixToTime: _unix_to_time_sql,
421            exp.UnixToTimeStr: lambda self, e: f"CAST(TO_TIMESTAMP({self.sql(e, 'this')}) AS TEXT)",
422            exp.VariancePop: rename_func("VAR_POP"),
423            exp.WeekOfYear: rename_func("WEEKOFYEAR"),
424            exp.Xor: bool_xor_sql,
425        }
426
427        TYPE_MAPPING = {
428            **generator.Generator.TYPE_MAPPING,
429            exp.DataType.Type.BINARY: "BLOB",
430            exp.DataType.Type.CHAR: "TEXT",
431            exp.DataType.Type.FLOAT: "REAL",
432            exp.DataType.Type.NCHAR: "TEXT",
433            exp.DataType.Type.NVARCHAR: "TEXT",
434            exp.DataType.Type.UINT: "UINTEGER",
435            exp.DataType.Type.VARBINARY: "BLOB",
436            exp.DataType.Type.VARCHAR: "TEXT",
437            exp.DataType.Type.TIMESTAMP_S: "TIMESTAMP_S",
438            exp.DataType.Type.TIMESTAMP_MS: "TIMESTAMP_MS",
439            exp.DataType.Type.TIMESTAMP_NS: "TIMESTAMP_NS",
440        }
441
442        STAR_MAPPING = {**generator.Generator.STAR_MAPPING, "except": "EXCLUDE"}
443
444        UNWRAPPED_INTERVAL_VALUES = (exp.Column, exp.Literal, exp.Paren)
445
446        PROPERTIES_LOCATION = {
447            **generator.Generator.PROPERTIES_LOCATION,
448            exp.VolatileProperty: exp.Properties.Location.UNSUPPORTED,
449        }
450
451        def timefromparts_sql(self, expression: exp.TimeFromParts) -> str:
452            nano = expression.args.get("nano")
453            if nano is not None:
454                expression.set(
455                    "sec", expression.args["sec"] + nano.pop() / exp.Literal.number(1000000000.0)
456                )
457
458            return rename_func("MAKE_TIME")(self, expression)
459
460        def timestampfromparts_sql(self, expression: exp.TimestampFromParts) -> str:
461            sec = expression.args["sec"]
462
463            milli = expression.args.get("milli")
464            if milli is not None:
465                sec += milli.pop() / exp.Literal.number(1000.0)
466
467            nano = expression.args.get("nano")
468            if nano is not None:
469                sec += nano.pop() / exp.Literal.number(1000000000.0)
470
471            if milli or nano:
472                expression.set("sec", sec)
473
474            return rename_func("MAKE_TIMESTAMP")(self, expression)
475
476        def tablesample_sql(
477            self,
478            expression: exp.TableSample,
479            sep: str = " AS ",
480            tablesample_keyword: t.Optional[str] = None,
481        ) -> str:
482            if not isinstance(expression.parent, exp.Select):
483                # This sample clause only applies to a single source, not the entire resulting relation
484                tablesample_keyword = "TABLESAMPLE"
485
486            return super().tablesample_sql(
487                expression, sep=sep, tablesample_keyword=tablesample_keyword
488            )
489
490        def getpath_sql(self, expression: exp.GetPath) -> str:
491            expression = prepend_dollar_to_path(expression)
492            return f"{self.sql(expression, 'this')} -> {self.sql(expression, 'expression')}"
493
494        def interval_sql(self, expression: exp.Interval) -> str:
495            multiplier: t.Optional[int] = None
496            unit = expression.text("unit").lower()
497
498            if unit.startswith("week"):
499                multiplier = 7
500            if unit.startswith("quarter"):
501                multiplier = 90
502
503            if multiplier:
504                return f"({multiplier} * {super().interval_sql(exp.Interval(this=expression.this, unit=exp.var('DAY')))})"
505
506            return super().interval_sql(expression)
507
508        def columndef_sql(self, expression: exp.ColumnDef, sep: str = " ") -> str:
509            if isinstance(expression.parent, exp.UserDefinedFunction):
510                return self.sql(expression, "this")
511            return super().columndef_sql(expression, sep)
512
513        def placeholder_sql(self, expression: exp.Placeholder) -> str:
514            return f"${expression.name}" if expression.name else "?"
NULL_ORDERING = 'nulls_are_last'

Indicates the default NULL ordering method to use if not explicitly set. Possible values: "nulls_are_small", "nulls_are_large", "nulls_are_last"

SUPPORTS_USER_DEFINED_TYPES = False

Determines whether or not user-defined data types are supported.

SAFE_DIVISION = True

Determines whether division by zero throws an error (False) or returns NULL (True).

INDEX_OFFSET = 1

Determines the base index offset for arrays.

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.

tokenizer_class = <class 'DuckDB.Tokenizer'>
parser_class = <class 'DuckDB.Parser'>
generator_class = <class 'DuckDB.Generator'>
TIME_TRIE: Dict = {}
FORMAT_TRIE: Dict = {}
INVERSE_TIME_MAPPING: Dict[str, str] = {}
INVERSE_TIME_TRIE: Dict = {}
INVERSE_ESCAPE_SEQUENCES: Dict[str, str] = {}
QUOTE_START = "'"
QUOTE_END = "'"
IDENTIFIER_START = '"'
IDENTIFIER_END = '"'
BIT_START: Optional[str] = None
BIT_END: Optional[str] = None
HEX_START: Optional[str] = None
HEX_END: Optional[str] = None
BYTE_START: Optional[str] = None
BYTE_END: Optional[str] = None
UNICODE_START: Optional[str] = None
UNICODE_END: Optional[str] = None
class DuckDB.Tokenizer(sqlglot.tokens.Tokenizer):
176    class Tokenizer(tokens.Tokenizer):
177        KEYWORDS = {
178            **tokens.Tokenizer.KEYWORDS,
179            "//": TokenType.DIV,
180            "ATTACH": TokenType.COMMAND,
181            "BINARY": TokenType.VARBINARY,
182            "BITSTRING": TokenType.BIT,
183            "BPCHAR": TokenType.TEXT,
184            "CHAR": TokenType.TEXT,
185            "CHARACTER VARYING": TokenType.TEXT,
186            "EXCLUDE": TokenType.EXCEPT,
187            "LOGICAL": TokenType.BOOLEAN,
188            "PIVOT_WIDER": TokenType.PIVOT,
189            "SIGNED": TokenType.INT,
190            "STRING": TokenType.VARCHAR,
191            "UBIGINT": TokenType.UBIGINT,
192            "UINTEGER": TokenType.UINT,
193            "USMALLINT": TokenType.USMALLINT,
194            "UTINYINT": TokenType.UTINYINT,
195            "TIMESTAMP_S": TokenType.TIMESTAMP_S,
196            "TIMESTAMP_MS": TokenType.TIMESTAMP_MS,
197            "TIMESTAMP_NS": TokenType.TIMESTAMP_NS,
198            "TIMESTAMP_US": TokenType.TIMESTAMP,
199        }
200
201        SINGLE_TOKENS = {
202            **tokens.Tokenizer.SINGLE_TOKENS,
203            "$": TokenType.PARAMETER,
204        }
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'>, 'CREATE': <TokenType.CREATE: 'CREATE'>, 'CROSS': <TokenType.CROSS: 'CROSS'>, 'CUBE': <TokenType.CUBE: 'CUBE'>, 'CURRENT_DATE': <TokenType.CURRENT_DATE: 'CURRENT_DATE'>, 'CURRENT_TIME': <TokenType.CURRENT_TIME: 'CURRENT_TIME'>, 'CURRENT_TIMESTAMP': <TokenType.CURRENT_TIMESTAMP: 'CURRENT_TIMESTAMP'>, 'CURRENT_USER': <TokenType.CURRENT_USER: 'CURRENT_USER'>, 'DATABASE': <TokenType.DATABASE: 'DATABASE'>, 'DEFAULT': <TokenType.DEFAULT: 'DEFAULT'>, 'DELETE': <TokenType.DELETE: 'DELETE'>, 'DESC': <TokenType.DESC: 'DESC'>, 'DESCRIBE': <TokenType.DESCRIBE: 'DESCRIBE'>, 'DISTINCT': <TokenType.DISTINCT: 'DISTINCT'>, 'DISTRIBUTE BY': <TokenType.DISTRIBUTE_BY: 'DISTRIBUTE_BY'>, 'DIV': <TokenType.DIV: 'DIV'>, 'DROP': <TokenType.DROP: 'DROP'>, 'ELSE': <TokenType.ELSE: 'ELSE'>, 'END': <TokenType.END: 'END'>, 'ESCAPE': <TokenType.ESCAPE: 'ESCAPE'>, 'EXCEPT': <TokenType.EXCEPT: 'EXCEPT'>, 'EXECUTE': <TokenType.EXECUTE: 'EXECUTE'>, 'EXISTS': <TokenType.EXISTS: 'EXISTS'>, 'FALSE': <TokenType.FALSE: 'FALSE'>, 'FETCH': <TokenType.FETCH: 'FETCH'>, 'FILTER': <TokenType.FILTER: 'FILTER'>, 'FIRST': <TokenType.FIRST: 'FIRST'>, 'FULL': <TokenType.FULL: 'FULL'>, 'FUNCTION': <TokenType.FUNCTION: 'FUNCTION'>, 'FOR': <TokenType.FOR: 'FOR'>, 'FOREIGN KEY': <TokenType.FOREIGN_KEY: 'FOREIGN_KEY'>, 'FORMAT': <TokenType.FORMAT: 'FORMAT'>, 'FROM': <TokenType.FROM: 'FROM'>, 'GEOGRAPHY': <TokenType.GEOGRAPHY: 'GEOGRAPHY'>, 'GEOMETRY': <TokenType.GEOMETRY: 'GEOMETRY'>, 'GLOB': <TokenType.GLOB: 'GLOB'>, 'GROUP BY': <TokenType.GROUP_BY: 'GROUP_BY'>, 'GROUPING SETS': <TokenType.GROUPING_SETS: 'GROUPING_SETS'>, 'HAVING': <TokenType.HAVING: 'HAVING'>, 'ILIKE': <TokenType.ILIKE: 'ILIKE'>, 'IN': <TokenType.IN: 'IN'>, 'INDEX': <TokenType.INDEX: 'INDEX'>, 'INET': <TokenType.INET: 'INET'>, 'INNER': <TokenType.INNER: 'INNER'>, 'INSERT': <TokenType.INSERT: 'INSERT'>, 'INTERVAL': <TokenType.INTERVAL: 'INTERVAL'>, 'INTERSECT': <TokenType.INTERSECT: 'INTERSECT'>, 'INTO': <TokenType.INTO: 'INTO'>, 'IS': <TokenType.IS: 'IS'>, 'ISNULL': <TokenType.ISNULL: 'ISNULL'>, 'JOIN': <TokenType.JOIN: 'JOIN'>, 'KEEP': <TokenType.KEEP: 'KEEP'>, 'KILL': <TokenType.KILL: 'KILL'>, 'LATERAL': <TokenType.LATERAL: 'LATERAL'>, 'LEFT': <TokenType.LEFT: 'LEFT'>, 'LIKE': <TokenType.LIKE: 'LIKE'>, 'LIMIT': <TokenType.LIMIT: 'LIMIT'>, 'LOAD': <TokenType.LOAD: 'LOAD'>, 'LOCK': <TokenType.LOCK: 'LOCK'>, 'MERGE': <TokenType.MERGE: 'MERGE'>, 'NATURAL': <TokenType.NATURAL: 'NATURAL'>, 'NEXT': <TokenType.NEXT: 'NEXT'>, 'NOT': <TokenType.NOT: 'NOT'>, 'NOTNULL': <TokenType.NOTNULL: 'NOTNULL'>, 'NULL': <TokenType.NULL: 'NULL'>, 'OBJECT': <TokenType.OBJECT: 'OBJECT'>, 'OFFSET': <TokenType.OFFSET: 'OFFSET'>, 'ON': <TokenType.ON: 'ON'>, 'OR': <TokenType.OR: 'OR'>, 'XOR': <TokenType.XOR: 'XOR'>, 'ORDER BY': <TokenType.ORDER_BY: 'ORDER_BY'>, 'ORDINALITY': <TokenType.ORDINALITY: 'ORDINALITY'>, 'OUTER': <TokenType.OUTER: 'OUTER'>, 'OVER': <TokenType.OVER: 'OVER'>, 'OVERLAPS': <TokenType.OVERLAPS: 'OVERLAPS'>, 'OVERWRITE': <TokenType.OVERWRITE: 'OVERWRITE'>, 'PARTITION': <TokenType.PARTITION: 'PARTITION'>, 'PARTITION BY': <TokenType.PARTITION_BY: 'PARTITION_BY'>, 'PARTITIONED BY': <TokenType.PARTITION_BY: 'PARTITION_BY'>, 'PARTITIONED_BY': <TokenType.PARTITION_BY: 'PARTITION_BY'>, 'PERCENT': <TokenType.PERCENT: 'PERCENT'>, 'PIVOT': <TokenType.PIVOT: 'PIVOT'>, 'PRAGMA': <TokenType.PRAGMA: 'PRAGMA'>, 'PRIMARY KEY': <TokenType.PRIMARY_KEY: 'PRIMARY_KEY'>, 'PROCEDURE': <TokenType.PROCEDURE: 'PROCEDURE'>, 'QUALIFY': <TokenType.QUALIFY: 'QUALIFY'>, 'RANGE': <TokenType.RANGE: 'RANGE'>, 'RECURSIVE': <TokenType.RECURSIVE: 'RECURSIVE'>, 'REGEXP': <TokenType.RLIKE: 'RLIKE'>, 'REPLACE': <TokenType.REPLACE: 'REPLACE'>, 'RETURNING': <TokenType.RETURNING: 'RETURNING'>, 'REFERENCES': <TokenType.REFERENCES: 'REFERENCES'>, 'RIGHT': <TokenType.RIGHT: 'RIGHT'>, 'RLIKE': <TokenType.RLIKE: 'RLIKE'>, 'ROLLBACK': <TokenType.ROLLBACK: 'ROLLBACK'>, 'ROLLUP': <TokenType.ROLLUP: 'ROLLUP'>, 'ROW': <TokenType.ROW: 'ROW'>, 'ROWS': <TokenType.ROWS: 'ROWS'>, 'SCHEMA': <TokenType.SCHEMA: 'SCHEMA'>, 'SELECT': <TokenType.SELECT: 'SELECT'>, 'SEMI': <TokenType.SEMI: 'SEMI'>, 'SET': <TokenType.SET: 'SET'>, 'SETTINGS': <TokenType.SETTINGS: 'SETTINGS'>, 'SHOW': <TokenType.SHOW: 'SHOW'>, 'SIMILAR TO': <TokenType.SIMILAR_TO: 'SIMILAR_TO'>, 'SOME': <TokenType.SOME: 'SOME'>, 'SORT BY': <TokenType.SORT_BY: 'SORT_BY'>, 'START WITH': <TokenType.START_WITH: 'START_WITH'>, 'TABLE': <TokenType.TABLE: 'TABLE'>, 'TABLESAMPLE': <TokenType.TABLE_SAMPLE: 'TABLE_SAMPLE'>, 'TEMP': <TokenType.TEMPORARY: 'TEMPORARY'>, 'TEMPORARY': <TokenType.TEMPORARY: 'TEMPORARY'>, 'THEN': <TokenType.THEN: 'THEN'>, 'TRUE': <TokenType.TRUE: 'TRUE'>, 'UNION': <TokenType.UNION: 'UNION'>, 'UNKNOWN': <TokenType.UNKNOWN: 'UNKNOWN'>, 'UNNEST': <TokenType.UNNEST: 'UNNEST'>, 'UNPIVOT': <TokenType.UNPIVOT: 'UNPIVOT'>, 'UPDATE': <TokenType.UPDATE: 'UPDATE'>, 'USE': <TokenType.USE: 'USE'>, 'USING': <TokenType.USING: 'USING'>, 'UUID': <TokenType.UUID: 'UUID'>, 'VALUES': <TokenType.VALUES: 'VALUES'>, 'VIEW': <TokenType.VIEW: 'VIEW'>, 'VOLATILE': <TokenType.VOLATILE: 'VOLATILE'>, 'WHEN': <TokenType.WHEN: 'WHEN'>, 'WHERE': <TokenType.WHERE: 'WHERE'>, 'WINDOW': <TokenType.WINDOW: 'WINDOW'>, 'WITH': <TokenType.WITH: 'WITH'>, 'APPLY': <TokenType.APPLY: 'APPLY'>, 'ARRAY': <TokenType.ARRAY: 'ARRAY'>, 'BIT': <TokenType.BIT: 'BIT'>, 'BOOL': <TokenType.BOOLEAN: 'BOOLEAN'>, 'BOOLEAN': <TokenType.BOOLEAN: 'BOOLEAN'>, 'BYTE': <TokenType.TINYINT: 'TINYINT'>, 'MEDIUMINT': <TokenType.MEDIUMINT: 'MEDIUMINT'>, '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'>, 'DEC': <TokenType.DECIMAL: 'DECIMAL'>, 'DECIMAL': <TokenType.DECIMAL: 'DECIMAL'>, 'BIGDECIMAL': <TokenType.BIGDECIMAL: 'BIGDECIMAL'>, 'BIGNUMERIC': <TokenType.BIGDECIMAL: 'BIGDECIMAL'>, 'MAP': <TokenType.MAP: 'MAP'>, 'NULLABLE': <TokenType.NULLABLE: 'NULLABLE'>, 'NUMBER': <TokenType.DECIMAL: 'DECIMAL'>, 'NUMERIC': <TokenType.DECIMAL: 'DECIMAL'>, 'FIXED': <TokenType.DECIMAL: 'DECIMAL'>, 'REAL': <TokenType.FLOAT: 'FLOAT'>, 'FLOAT': <TokenType.FLOAT: 'FLOAT'>, 'FLOAT4': <TokenType.FLOAT: 'FLOAT'>, 'FLOAT8': <TokenType.DOUBLE: 'DOUBLE'>, 'DOUBLE': <TokenType.DOUBLE: 'DOUBLE'>, 'DOUBLE PRECISION': <TokenType.DOUBLE: 'DOUBLE'>, 'JSON': <TokenType.JSON: 'JSON'>, 'CHAR': <TokenType.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'>, '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'>, 'DATE': <TokenType.DATE: 'DATE'>, 'DATETIME': <TokenType.DATETIME: 'DATETIME'>, 'INT4RANGE': <TokenType.INT4RANGE: 'INT4RANGE'>, 'INT4MULTIRANGE': <TokenType.INT4MULTIRANGE: 'INT4MULTIRANGE'>, 'INT8RANGE': <TokenType.INT8RANGE: 'INT8RANGE'>, 'INT8MULTIRANGE': <TokenType.INT8MULTIRANGE: 'INT8MULTIRANGE'>, 'NUMRANGE': <TokenType.NUMRANGE: 'NUMRANGE'>, 'NUMMULTIRANGE': <TokenType.NUMMULTIRANGE: 'NUMMULTIRANGE'>, 'TSRANGE': <TokenType.TSRANGE: 'TSRANGE'>, 'TSMULTIRANGE': <TokenType.TSMULTIRANGE: 'TSMULTIRANGE'>, 'TSTZRANGE': <TokenType.TSTZRANGE: 'TSTZRANGE'>, 'TSTZMULTIRANGE': <TokenType.TSTZMULTIRANGE: 'TSTZMULTIRANGE'>, 'DATERANGE': <TokenType.DATERANGE: 'DATERANGE'>, 'DATEMULTIRANGE': <TokenType.DATEMULTIRANGE: 'DATEMULTIRANGE'>, 'UNIQUE': <TokenType.UNIQUE: 'UNIQUE'>, 'STRUCT': <TokenType.STRUCT: 'STRUCT'>, 'VARIANT': <TokenType.VARIANT: 'VARIANT'>, 'ALTER': <TokenType.ALTER: 'ALTER'>, 'ANALYZE': <TokenType.COMMAND: 'COMMAND'>, 'CALL': <TokenType.COMMAND: 'COMMAND'>, 'COMMENT': <TokenType.COMMENT: 'COMMENT'>, 'COPY': <TokenType.COMMAND: 'COMMAND'>, 'EXPLAIN': <TokenType.COMMAND: 'COMMAND'>, 'GRANT': <TokenType.COMMAND: 'COMMAND'>, 'OPTIMIZE': <TokenType.COMMAND: 'COMMAND'>, 'PREPARE': <TokenType.COMMAND: 'COMMAND'>, 'TRUNCATE': <TokenType.COMMAND: 'COMMAND'>, 'VACUUM': <TokenType.COMMAND: 'COMMAND'>, 'USER-DEFINED': <TokenType.USERDEFINED: 'USERDEFINED'>, 'FOR VERSION': <TokenType.VERSION_SNAPSHOT: 'VERSION_SNAPSHOT'>, 'FOR TIMESTAMP': <TokenType.TIMESTAMP_SNAPSHOT: 'TIMESTAMP_SNAPSHOT'>, '//': <TokenType.DIV: 'DIV'>, 'ATTACH': <TokenType.COMMAND: 'COMMAND'>, 'BITSTRING': <TokenType.BIT: 'BIT'>, 'BPCHAR': <TokenType.TEXT: 'TEXT'>, 'CHARACTER VARYING': <TokenType.TEXT: 'TEXT'>, 'EXCLUDE': <TokenType.EXCEPT: 'EXCEPT'>, 'LOGICAL': <TokenType.BOOLEAN: 'BOOLEAN'>, 'PIVOT_WIDER': <TokenType.PIVOT: 'PIVOT'>, '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.QUOTE: 'QUOTE'>, '`': <TokenType.IDENTIFIER: 'IDENTIFIER'>, '"': <TokenType.IDENTIFIER: 'IDENTIFIER'>, '#': <TokenType.HASH: 'HASH'>, '$': <TokenType.PARAMETER: 'PARAMETER'>}
class DuckDB.Parser(sqlglot.parser.Parser):
206    class Parser(parser.Parser):
207        BITWISE = {
208            **parser.Parser.BITWISE,
209            TokenType.TILDA: exp.RegexpLike,
210        }
211
212        FUNCTIONS = {
213            **parser.Parser.FUNCTIONS,
214            "ARRAY_HAS": exp.ArrayContains.from_arg_list,
215            "ARRAY_LENGTH": exp.ArraySize.from_arg_list,
216            "ARRAY_SORT": exp.SortArray.from_arg_list,
217            "ARRAY_REVERSE_SORT": _sort_array_reverse,
218            "DATEDIFF": _parse_date_diff,
219            "DATE_DIFF": _parse_date_diff,
220            "DATE_TRUNC": date_trunc_to_time,
221            "DATETRUNC": date_trunc_to_time,
222            "DECODE": lambda args: exp.Decode(
223                this=seq_get(args, 0), charset=exp.Literal.string("utf-8")
224            ),
225            "ENCODE": lambda args: exp.Encode(
226                this=seq_get(args, 0), charset=exp.Literal.string("utf-8")
227            ),
228            "EPOCH": exp.TimeToUnix.from_arg_list,
229            "EPOCH_MS": lambda args: exp.UnixToTime(
230                this=seq_get(args, 0), scale=exp.UnixToTime.MILLIS
231            ),
232            "JSON": exp.ParseJSON.from_arg_list,
233            "LIST_HAS": exp.ArrayContains.from_arg_list,
234            "LIST_REVERSE_SORT": _sort_array_reverse,
235            "LIST_SORT": exp.SortArray.from_arg_list,
236            "LIST_VALUE": exp.Array.from_arg_list,
237            "MAKE_TIME": exp.TimeFromParts.from_arg_list,
238            "MAKE_TIMESTAMP": _parse_make_timestamp,
239            "MEDIAN": lambda args: exp.PercentileCont(
240                this=seq_get(args, 0), expression=exp.Literal.number(0.5)
241            ),
242            "QUANTILE_CONT": exp.PercentileCont.from_arg_list,
243            "QUANTILE_DISC": exp.PercentileDisc.from_arg_list,
244            "REGEXP_EXTRACT": lambda args: exp.RegexpExtract(
245                this=seq_get(args, 0), expression=seq_get(args, 1), group=seq_get(args, 2)
246            ),
247            "REGEXP_MATCHES": exp.RegexpLike.from_arg_list,
248            "REGEXP_REPLACE": lambda args: exp.RegexpReplace(
249                this=seq_get(args, 0),
250                expression=seq_get(args, 1),
251                replacement=seq_get(args, 2),
252                modifiers=seq_get(args, 3),
253            ),
254            "STRFTIME": format_time_lambda(exp.TimeToStr, "duckdb"),
255            "STRING_SPLIT": exp.Split.from_arg_list,
256            "STRING_SPLIT_REGEX": exp.RegexpSplit.from_arg_list,
257            "STRING_TO_ARRAY": exp.Split.from_arg_list,
258            "STRPTIME": format_time_lambda(exp.StrToTime, "duckdb"),
259            "STRUCT_PACK": _parse_struct_pack,
260            "STR_SPLIT": exp.Split.from_arg_list,
261            "STR_SPLIT_REGEX": exp.RegexpSplit.from_arg_list,
262            "TO_TIMESTAMP": exp.UnixToTime.from_arg_list,
263            "UNNEST": exp.Explode.from_arg_list,
264            "XOR": binary_from_function(exp.BitwiseXor),
265        }
266
267        FUNCTION_PARSERS = parser.Parser.FUNCTION_PARSERS.copy()
268        FUNCTION_PARSERS.pop("DECODE", None)
269
270        TABLE_ALIAS_TOKENS = parser.Parser.TABLE_ALIAS_TOKENS - {
271            TokenType.SEMI,
272            TokenType.ANTI,
273        }
274
275        PLACEHOLDER_PARSERS = {
276            **parser.Parser.PLACEHOLDER_PARSERS,
277            TokenType.PARAMETER: lambda self: (
278                self.expression(exp.Placeholder, this=self._prev.text)
279                if self._match(TokenType.NUMBER) or self._match_set(self.ID_VAR_TOKENS)
280                else None
281            ),
282        }
283
284        def _parse_types(
285            self, check_func: bool = False, schema: bool = False, allow_identifiers: bool = True
286        ) -> t.Optional[exp.Expression]:
287            this = super()._parse_types(
288                check_func=check_func, schema=schema, allow_identifiers=allow_identifiers
289            )
290
291            # DuckDB treats NUMERIC and DECIMAL without precision as DECIMAL(18, 3)
292            # See: https://duckdb.org/docs/sql/data_types/numeric
293            if (
294                isinstance(this, exp.DataType)
295                and this.is_type("numeric", "decimal")
296                and not this.expressions
297            ):
298                return exp.DataType.build("DECIMAL(18, 3)")
299
300            return this
301
302        def _parse_struct_types(self, type_required: bool = False) -> t.Optional[exp.Expression]:
303            return self._parse_field_def()
304
305        def _pivot_column_names(self, aggregations: t.List[exp.Expression]) -> t.List[str]:
306            if len(aggregations) == 1:
307                return super()._pivot_column_names(aggregations)
308            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: Determines the amount of context to capture from a query string when displaying the error message (in number of characters). Default: 100
  • max_errors: Maximum number of error messages to include in a raised ParseError. This is only relevant if error_level is ErrorLevel.RAISE. Default: 3
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'>>, '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_CONTAINS': <bound method Func.from_arg_list of <class 'sqlglot.expressions.ArrayContains'>>, 'FILTER': <bound method Func.from_arg_list of <class 'sqlglot.expressions.ArrayFilter'>>, 'ARRAY_FILTER': <bound method Func.from_arg_list of <class 'sqlglot.expressions.ArrayFilter'>>, 'ARRAY_JOIN': <bound method Func.from_arg_list of <class 'sqlglot.expressions.ArrayJoin'>>, 'ARRAY_SIZE': <bound method Func.from_arg_list of <class 'sqlglot.expressions.ArraySize'>>, 'ARRAY_SORT': <bound method Func.from_arg_list of <class 'sqlglot.expressions.SortArray'>>, 'ARRAY_SUM': <bound method Func.from_arg_list of <class 'sqlglot.expressions.ArraySum'>>, 'ARRAY_UNION_AGG': <bound method Func.from_arg_list of <class 'sqlglot.expressions.ArrayUnionAgg'>>, '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'>>, '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>>, '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'>>, '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 _parse_date_diff>, 'DATE_DIFF': <function _parse_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_SUB': <bound method Func.from_arg_list of <class 'sqlglot.expressions.DateSub'>>, 'DATE_TO_DATE_STR': <function Parser.<lambda>>, 'DATE_TO_DI': <bound method Func.from_arg_list of <class 'sqlglot.expressions.DateToDi'>>, 'DATE_TRUNC': <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'>>, 'FLATTEN': <bound method Func.from_arg_list of <class 'sqlglot.expressions.Flatten'>>, 'FLOOR': <bound method Func.from_arg_list of <class 'sqlglot.expressions.Floor'>>, 'FROM_BASE': <bound method Func.from_arg_list of <class 'sqlglot.expressions.FromBase'>>, 'FROM_BASE64': <bound method Func.from_arg_list of <class 'sqlglot.expressions.FromBase64'>>, 'GENERATE_SERIES': <bound method Func.from_arg_list of <class 'sqlglot.expressions.GenerateSeries'>>, 'GET_PATH': <bound method Func.from_arg_list of <class 'sqlglot.expressions.GetPath'>>, 'GREATEST': <bound method Func.from_arg_list of <class 'sqlglot.expressions.Greatest'>>, 'GROUP_CONCAT': <bound method Func.from_arg_list of <class 'sqlglot.expressions.GroupConcat'>>, 'HEX': <bound method Func.from_arg_list of <class 'sqlglot.expressions.Hex'>>, 'HLL': <bound method Func.from_arg_list of <class 'sqlglot.expressions.Hll'>>, 'IF': <bound method Func.from_arg_list of <class 'sqlglot.expressions.If'>>, 'INITCAP': <bound method Func.from_arg_list of <class 'sqlglot.expressions.Initcap'>>, 'IS_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': <bound method Func.from_arg_list of <class 'sqlglot.expressions.JSONExtract'>>, 'JSON_EXTRACT_SCALAR': <bound method Func.from_arg_list of <class 'sqlglot.expressions.JSONExtractScalar'>>, 'JSON_FORMAT': <bound method Func.from_arg_list of <class 'sqlglot.expressions.JSONFormat'>>, 'J_S_O_N_OBJECT': <bound method Func.from_arg_list of <class 'sqlglot.expressions.JSONObject'>>, 'J_S_O_N_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'>>, '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'>>, '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 parse_logarithm>, 'LOG10': <bound method Func.from_arg_list of <class 'sqlglot.expressions.Log10'>>, 'LOG2': <bound method Func.from_arg_list of <class 'sqlglot.expressions.Log2'>>, 'LOGICAL_AND': <bound method Func.from_arg_list of <class 'sqlglot.expressions.LogicalAnd'>>, 'BOOL_AND': <bound method Func.from_arg_list of <class 'sqlglot.expressions.LogicalAnd'>>, 'BOOLAND_AGG': <bound method Func.from_arg_list of <class 'sqlglot.expressions.LogicalAnd'>>, 'LOGICAL_OR': <bound method Func.from_arg_list of <class 'sqlglot.expressions.LogicalOr'>>, 'BOOL_OR': <bound method Func.from_arg_list of <class 'sqlglot.expressions.LogicalOr'>>, 'BOOLOR_AGG': <bound method Func.from_arg_list of <class 'sqlglot.expressions.LogicalOr'>>, 'LOWER': <bound method Func.from_arg_list of <class 'sqlglot.expressions.Lower'>>, 'LCASE': <bound method Func.from_arg_list of <class 'sqlglot.expressions.Lower'>>, 'MD5': <bound method Func.from_arg_list of <class 'sqlglot.expressions.MD5'>>, 'MD5_DIGEST': <bound method Func.from_arg_list of <class 'sqlglot.expressions.MD5Digest'>>, 'MAP': <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'>>, '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'>>, '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'>>, 'SORT_ARRAY': <bound method Func.from_arg_list of <class 'sqlglot.expressions.SortArray'>>, 'SPLIT': <bound method Func.from_arg_list of <class 'sqlglot.expressions.Split'>>, 'SQRT': <bound method Func.from_arg_list of <class 'sqlglot.expressions.Sqrt'>>, 'STANDARD_HASH': <bound method Func.from_arg_list of <class 'sqlglot.expressions.StandardHash'>>, 'STAR_MAP': <bound method Func.from_arg_list of <class 'sqlglot.expressions.StarMap'>>, 'STARTS_WITH': <bound method Func.from_arg_list of <class 'sqlglot.expressions.StartsWith'>>, 'STARTSWITH': <bound method Func.from_arg_list of <class 'sqlglot.expressions.StartsWith'>>, 'STDDEV': <bound method Func.from_arg_list of <class 'sqlglot.expressions.Stddev'>>, 'STDDEV_POP': <bound method Func.from_arg_list of <class 'sqlglot.expressions.StddevPop'>>, 'STDDEV_SAMP': <bound method Func.from_arg_list of <class 'sqlglot.expressions.StddevSamp'>>, 'STR_POSITION': <bound method Func.from_arg_list of <class 'sqlglot.expressions.StrPosition'>>, 'STR_TO_DATE': <bound method Func.from_arg_list of <class 'sqlglot.expressions.StrToDate'>>, 'STR_TO_MAP': <bound method Func.from_arg_list of <class 'sqlglot.expressions.StrToMap'>>, 'STR_TO_TIME': <bound method Func.from_arg_list of <class 'sqlglot.expressions.StrToTime'>>, 'STR_TO_UNIX': <bound method Func.from_arg_list of <class 'sqlglot.expressions.StrToUnix'>>, 'STRUCT': <bound method Func.from_arg_list of <class 'sqlglot.expressions.Struct'>>, 'STRUCT_EXTRACT': <bound method Func.from_arg_list of <class 'sqlglot.expressions.StructExtract'>>, 'STUFF': <bound method Func.from_arg_list of <class 'sqlglot.expressions.Stuff'>>, 'INSERT': <bound method Func.from_arg_list of <class 'sqlglot.expressions.Stuff'>>, 'SUBSTRING': <bound method Func.from_arg_list of <class 'sqlglot.expressions.Substring'>>, 'SUM': <bound method Func.from_arg_list of <class 'sqlglot.expressions.Sum'>>, 'TIME_ADD': <bound method Func.from_arg_list of <class 'sqlglot.expressions.TimeAdd'>>, 'TIME_DIFF': <bound method Func.from_arg_list of <class 'sqlglot.expressions.TimeDiff'>>, 'TIME_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'>>, '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'>>, 'TRANSFORM': <bound method Func.from_arg_list of <class 'sqlglot.expressions.Transform'>>, 'TRIM': <bound method Func.from_arg_list of <class 'sqlglot.expressions.Trim'>>, 'TRY_CAST': <bound method Func.from_arg_list of <class 'sqlglot.expressions.TryCast'>>, 'TS_OR_DI_TO_DI': <bound method Func.from_arg_list of <class 'sqlglot.expressions.TsOrDiToDi'>>, 'TS_OR_DS_ADD': <bound method Func.from_arg_list of <class 'sqlglot.expressions.TsOrDsAdd'>>, 'TS_OR_DS_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'>>, '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': <bound method Func.from_arg_list of <class 'sqlglot.expressions.Upper'>>, 'UCASE': <bound method Func.from_arg_list of <class 'sqlglot.expressions.Upper'>>, 'VAR_MAP': <function parse_var_map>, 'VARIANCE': <bound method Func.from_arg_list of <class 'sqlglot.expressions.Variance'>>, 'VARIANCE_SAMP': <bound method Func.from_arg_list of <class 'sqlglot.expressions.Variance'>>, 'VAR_SAMP': <bound method Func.from_arg_list of <class 'sqlglot.expressions.Variance'>>, 'VARIANCE_POP': <bound method Func.from_arg_list of <class 'sqlglot.expressions.VariancePop'>>, 'VAR_POP': <bound method Func.from_arg_list of <class 'sqlglot.expressions.VariancePop'>>, 'WEEK': <bound method Func.from_arg_list of <class 'sqlglot.expressions.Week'>>, 'WEEK_OF_YEAR': <bound method Func.from_arg_list of <class 'sqlglot.expressions.WeekOfYear'>>, 'WEEKOFYEAR': <bound method Func.from_arg_list of <class 'sqlglot.expressions.WeekOfYear'>>, 'WHEN': <bound method Func.from_arg_list of <class 'sqlglot.expressions.When'>>, 'X_M_L_TABLE': <bound method Func.from_arg_list of <class 'sqlglot.expressions.XMLTable'>>, 'XOR': <function binary_from_function.<locals>.<lambda>>, 'YEAR': <bound method Func.from_arg_list of <class 'sqlglot.expressions.Year'>>, 'GLOB': <function Parser.<lambda>>, 'LIKE': <function parse_like>, 'ARRAY_HAS': <bound method Func.from_arg_list of <class 'sqlglot.expressions.ArrayContains'>>, 'ARRAY_LENGTH': <bound method Func.from_arg_list of <class 'sqlglot.expressions.ArraySize'>>, 'ARRAY_REVERSE_SORT': <function _sort_array_reverse>, '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'>>, 'LIST_HAS': <bound method Func.from_arg_list of <class 'sqlglot.expressions.ArrayContains'>>, 'LIST_REVERSE_SORT': <function _sort_array_reverse>, '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 _parse_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 format_time_lambda.<locals>._format_time>, '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'>>, 'STRING_TO_ARRAY': <bound method Func.from_arg_list of <class 'sqlglot.expressions.Split'>>, 'STRPTIME': <function format_time_lambda.<locals>._format_time>, 'STRUCT_PACK': <function _parse_struct_pack>, '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'>>}
FUNCTION_PARSERS = {'ANY_VALUE': <function Parser.<lambda>>, '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>>}
TABLE_ALIAS_TOKENS = {<TokenType.TOP: 'TOP'>, <TokenType.PIVOT: 'PIVOT'>, <TokenType.BINARY: 'BINARY'>, <TokenType.UTINYINT: 'UTINYINT'>, <TokenType.BEGIN: 'BEGIN'>, <TokenType.CURRENT_DATETIME: 'CURRENT_DATETIME'>, <TokenType.OBJECT: 'OBJECT'>, <TokenType.VAR: 'VAR'>, <TokenType.OVERWRITE: 'OVERWRITE'>, <TokenType.PARTITION: 'PARTITION'>, <TokenType.EXISTS: 'EXISTS'>, <TokenType.DIV: 'DIV'>, <TokenType.TSMULTIRANGE: 'TSMULTIRANGE'>, <TokenType.PSEUDO_TYPE: 'PSEUDO_TYPE'>, <TokenType.VARIANT: 'VARIANT'>, <TokenType.FALSE: 'FALSE'>, <TokenType.UMEDIUMINT: 'UMEDIUMINT'>, <TokenType.MERGE: 'MERGE'>, <TokenType.DATETIME64: 'DATETIME64'>, <TokenType.COMMENT: 'COMMENT'>, <TokenType.FLOAT: 'FLOAT'>, <TokenType.MEDIUMBLOB: 'MEDIUMBLOB'>, <TokenType.TIMETZ: 'TIMETZ'>, <TokenType.MODEL: 'MODEL'>, <TokenType.BIT: 'BIT'>, <TokenType.UUID: 'UUID'>, <TokenType.DATE32: 'DATE32'>, <TokenType.ESCAPE: 'ESCAPE'>, <TokenType.MEDIUMTEXT: 'MEDIUMTEXT'>, <TokenType.LOWCARDINALITY: 'LOWCARDINALITY'>, <TokenType.COLUMN: 'COLUMN'>, <TokenType.IPV4: 'IPV4'>, <TokenType.COMMIT: 'COMMIT'>, <TokenType.UINT: 'UINT'>, <TokenType.ALL: 'ALL'>, <TokenType.UDECIMAL: 'UDECIMAL'>, <TokenType.TSTZMULTIRANGE: 'TSTZMULTIRANGE'>, <TokenType.USMALLINT: 'USMALLINT'>, <TokenType.NULL: 'NULL'>, <TokenType.VARCHAR: 'VARCHAR'>, <TokenType.FIRST: 'FIRST'>, <TokenType.PRAGMA: 'PRAGMA'>, <TokenType.EXECUTE: 'EXECUTE'>, <TokenType.TINYTEXT: 'TINYTEXT'>, <TokenType.INTERVAL: 'INTERVAL'>, <TokenType.NEXT: 'NEXT'>, <TokenType.UINT128: 'UINT128'>, <TokenType.DATEMULTIRANGE: 'DATEMULTIRANGE'>, <TokenType.ROW: 'ROW'>, <TokenType.DEFAULT: 'DEFAULT'>, <TokenType.KEEP: 'KEEP'>, <TokenType.TIMESTAMP_S: 'TIMESTAMP_S'>, <TokenType.ENUM: 'ENUM'>, <TokenType.TIMESTAMP_NS: 'TIMESTAMP_NS'>, <TokenType.NCHAR: 'NCHAR'>, <TokenType.TABLE: 'TABLE'>, <TokenType.MEDIUMINT: 'MEDIUMINT'>, <TokenType.TIMESTAMPTZ: 'TIMESTAMPTZ'>, <TokenType.SUPER: 'SUPER'>, <TokenType.AGGREGATEFUNCTION: 'AGGREGATEFUNCTION'>, <TokenType.FOREIGN_KEY: 'FOREIGN_KEY'>, <TokenType.IPADDRESS: 'IPADDRESS'>, <TokenType.INT4MULTIRANGE: 'INT4MULTIRANGE'>, <TokenType.UBIGINT: 'UBIGINT'>, <TokenType.VIEW: 'VIEW'>, <TokenType.BIGINT: 'BIGINT'>, <TokenType.XML: 'XML'>, <TokenType.COLLATE: 'COLLATE'>, <TokenType.FORMAT: 'FORMAT'>, <TokenType.JSON: 'JSON'>, <TokenType.HSTORE: 'HSTORE'>, <TokenType.SOME: 'SOME'>, <TokenType.ISNULL: 'ISNULL'>, <TokenType.UNPIVOT: 'UNPIVOT'>, <TokenType.SIMPLEAGGREGATEFUNCTION: 'SIMPLEAGGREGATEFUNCTION'>, <TokenType.TSRANGE: 'TSRANGE'>, <TokenType.IMAGE: 'IMAGE'>, <TokenType.IPPREFIX: 'IPPREFIX'>, <TokenType.CASE: 'CASE'>, <TokenType.CONSTRAINT: 'CONSTRAINT'>, <TokenType.TSTZRANGE: 'TSTZRANGE'>, <TokenType.TRUE: 'TRUE'>, <TokenType.DATERANGE: 'DATERANGE'>, <TokenType.INT8RANGE: 'INT8RANGE'>, <TokenType.INT128: 'INT128'>, <TokenType.REFRESH: 'REFRESH'>, <TokenType.INT: 'INT'>, <TokenType.TEXT: 'TEXT'>, <TokenType.FIXEDSTRING: 'FIXEDSTRING'>, <TokenType.TIMESTAMP: 'TIMESTAMP'>, <TokenType.END: 'END'>, <TokenType.SMALLINT: 'SMALLINT'>, <TokenType.HLLSKETCH: 'HLLSKETCH'>, <TokenType.CURRENT_DATE: 'CURRENT_DATE'>, <TokenType.ENUM16: 'ENUM16'>, <TokenType.IPV6: 'IPV6'>, <TokenType.STRUCT: 'STRUCT'>, <TokenType.NULLABLE: 'NULLABLE'>, <TokenType.RANGE: 'RANGE'>, <TokenType.FUNCTION: 'FUNCTION'>, <TokenType.SCHEMA: 'SCHEMA'>, <TokenType.ASC: 'ASC'>, <TokenType.DESCRIBE: 'DESCRIBE'>, <TokenType.UPDATE: 'UPDATE'>, <TokenType.NUMMULTIRANGE: 'NUMMULTIRANGE'>, <TokenType.ROWS: 'ROWS'>, <TokenType.COMMAND: 'COMMAND'>, <TokenType.INT8MULTIRANGE: 'INT8MULTIRANGE'>, <TokenType.BIGDECIMAL: 'BIGDECIMAL'>, <TokenType.LOAD: 'LOAD'>, <TokenType.USE: 'USE'>, <TokenType.UNIQUEIDENTIFIER: 'UNIQUEIDENTIFIER'>, <TokenType.INT4RANGE: 'INT4RANGE'>, <TokenType.SMALLSERIAL: 'SMALLSERIAL'>, <TokenType.ROWVERSION: 'ROWVERSION'>, <TokenType.ARRAY: 'ARRAY'>, <TokenType.INDEX: 'INDEX'>, <TokenType.NUMRANGE: 'NUMRANGE'>, <TokenType.PROCEDURE: 'PROCEDURE'>, <TokenType.CACHE: 'CACHE'>, <TokenType.UNKNOWN: 'UNKNOWN'>, <TokenType.KILL: 'KILL'>, <TokenType.SET: 'SET'>, <TokenType.JSONB: 'JSONB'>, <TokenType.PERCENT: 'PERCENT'>, <TokenType.SETTINGS: 'SETTINGS'>, <TokenType.REFERENCES: 'REFERENCES'>, <TokenType.ENUM8: 'ENUM8'>, <TokenType.NESTED: 'NESTED'>, <TokenType.TIME: 'TIME'>, <TokenType.BOOLEAN: 'BOOLEAN'>, <TokenType.DATETIME: 'DATETIME'>, <TokenType.DELETE: 'DELETE'>, <TokenType.INET: 'INET'>, <TokenType.CURRENT_USER: 'CURRENT_USER'>, <TokenType.DESC: 'DESC'>, <TokenType.TIMESTAMP_MS: 'TIMESTAMP_MS'>, <TokenType.SMALLMONEY: 'SMALLMONEY'>, <TokenType.DATE: 'DATE'>, <TokenType.ORDINALITY: 'ORDINALITY'>, <TokenType.VARBINARY: 'VARBINARY'>, <TokenType.BIGSERIAL: 'BIGSERIAL'>, <TokenType.OPERATOR: 'OPERATOR'>, <TokenType.TIMESTAMPLTZ: 'TIMESTAMPLTZ'>, <TokenType.GEOMETRY: 'GEOMETRY'>, <TokenType.MAP: 'MAP'>, <TokenType.FINAL: 'FINAL'>, <TokenType.DECIMAL: 'DECIMAL'>, <TokenType.UINT256: 'UINT256'>, <TokenType.IS: 'IS'>, <TokenType.NVARCHAR: 'NVARCHAR'>, <TokenType.INT256: 'INT256'>, <TokenType.TINYBLOB: 'TINYBLOB'>, <TokenType.MONEY: 'MONEY'>, <TokenType.REPLACE: 'REPLACE'>, <TokenType.DATABASE: 'DATABASE'>, <TokenType.ANY: 'ANY'>, <TokenType.UNIQUE: 'UNIQUE'>, <TokenType.OBJECT_IDENTIFIER: 'OBJECT_IDENTIFIER'>, <TokenType.YEAR: 'YEAR'>, <TokenType.CHAR: 'CHAR'>, <TokenType.DICTIONARY: 'DICTIONARY'>, <TokenType.USERDEFINED: 'USERDEFINED'>, <TokenType.OVERLAPS: 'OVERLAPS'>, <TokenType.SERIAL: 'SERIAL'>, <TokenType.FILTER: 'FILTER'>, <TokenType.DOUBLE: 'DOUBLE'>, <TokenType.AUTO_INCREMENT: 'AUTO_INCREMENT'>, <TokenType.RECURSIVE: 'RECURSIVE'>, <TokenType.VOLATILE: 'VOLATILE'>, <TokenType.SHOW: 'SHOW'>, <TokenType.LONGTEXT: 'LONGTEXT'>, <TokenType.CURRENT_TIME: 'CURRENT_TIME'>, <TokenType.TINYINT: 'TINYINT'>, <TokenType.TEMPORARY: 'TEMPORARY'>, <TokenType.GEOGRAPHY: 'GEOGRAPHY'>, <TokenType.LONGBLOB: 'LONGBLOB'>, <TokenType.CURRENT_TIMESTAMP: 'CURRENT_TIMESTAMP'>}
PLACEHOLDER_PARSERS = {<TokenType.PLACEHOLDER: 'PLACEHOLDER'>: <function Parser.<lambda>>, <TokenType.PARAMETER: 'PARAMETER'>: <function DuckDB.Parser.<lambda>>, <TokenType.COLON: 'COLON'>: <function Parser.<lambda>>}
SHOW_TRIE: Dict = {}
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
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
PRIMARY_PARSERS
RANGE_PARSERS
PROPERTY_PARSERS
CONSTRAINT_PARSERS
ALTER_PARSERS
SCHEMA_UNNAMED_CONSTRAINTS
NO_PAREN_FUNCTION_PARSERS
INVALID_FUNC_NAME_TOKENS
FUNCTIONS_WITH_ALIASED_ARGS
QUERY_MODIFIER_PARSERS
SET_PARSERS
SHOW_PARSERS
TYPE_LITERAL_PARSERS
MODIFIABLES
DDL_SELECT_TOKENS
PRE_VOLATILE_TOKENS
TRANSACTION_KIND
TRANSACTION_CHARACTERISTICS
INSERT_ALTERNATIVES
CLONE_KEYWORDS
HISTORICAL_DATA_KIND
OPCLASS_FOLLOW_KEYWORDS
OPTYPE_FOLLOW_TOKENS
TABLE_INDEX_HINT_TOKENS
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
STRICT_CAST
PREFIXED_PIVOT_COLUMNS
IDENTIFY_PIVOT_STRINGS
LOG_DEFAULTS_TO_LN
ALTER_TABLE_ADD_REQUIRED_FOR_EACH_COLUMN
TABLESAMPLE_CSV
SET_REQUIRES_ASSIGNMENT_DELIMITER
TRIM_PATTERN_FIRST
STRING_ALIASES
MODIFIERS_ATTACHED_TO_UNION
UNION_MODIFIERS
NO_PAREN_IF_COMMANDS
error_level
error_message_context
max_errors
dialect
reset
parse
parse_into
check_errors
raise_error
expression
validate_expression
errors
sql
class DuckDB.Generator(sqlglot.generator.Generator):
310    class Generator(generator.Generator):
311        JOIN_HINTS = False
312        TABLE_HINTS = False
313        QUERY_HINTS = False
314        LIMIT_FETCH = "LIMIT"
315        STRUCT_DELIMITER = ("(", ")")
316        RENAME_TABLE_WITH_DB = False
317        NVL2_SUPPORTED = False
318        SEMI_ANTI_JOIN_WITH_SIDE = False
319        TABLESAMPLE_KEYWORDS = "USING SAMPLE"
320        TABLESAMPLE_SEED_KEYWORD = "REPEATABLE"
321        LAST_DAY_SUPPORTS_DATE_PART = False
322        JSON_KEY_VALUE_PAIR_SEP = ","
323
324        TRANSFORMS = {
325            **generator.Generator.TRANSFORMS,
326            exp.ApproxDistinct: approx_count_distinct_sql,
327            exp.Array: lambda self, e: (
328                self.func("ARRAY", e.expressions[0])
329                if e.expressions and e.expressions[0].find(exp.Select)
330                else inline_array_sql(self, e)
331            ),
332            exp.ArraySize: rename_func("ARRAY_LENGTH"),
333            exp.ArgMax: arg_max_or_min_no_count("ARG_MAX"),
334            exp.ArgMin: arg_max_or_min_no_count("ARG_MIN"),
335            exp.ArraySort: _array_sort_sql,
336            exp.ArraySum: rename_func("LIST_SUM"),
337            exp.BitwiseXor: rename_func("XOR"),
338            exp.CommentColumnConstraint: no_comment_column_constraint_sql,
339            exp.CurrentDate: lambda self, e: "CURRENT_DATE",
340            exp.CurrentTime: lambda self, e: "CURRENT_TIME",
341            exp.CurrentTimestamp: lambda self, e: "CURRENT_TIMESTAMP",
342            exp.DayOfMonth: rename_func("DAYOFMONTH"),
343            exp.DayOfWeek: rename_func("DAYOFWEEK"),
344            exp.DayOfYear: rename_func("DAYOFYEAR"),
345            exp.DataType: _datatype_sql,
346            exp.Date: _date_sql,
347            exp.DateAdd: _date_delta_sql,
348            exp.DateFromParts: rename_func("MAKE_DATE"),
349            exp.DateSub: _date_delta_sql,
350            exp.DateDiff: lambda self, e: self.func(
351                "DATE_DIFF", f"'{e.args.get('unit') or 'DAY'}'", e.expression, e.this
352            ),
353            exp.DateStrToDate: datestrtodate_sql,
354            exp.DateToDi: lambda self, e: f"CAST(STRFTIME({self.sql(e, 'this')}, {DuckDB.DATEINT_FORMAT}) AS INT)",
355            exp.Decode: lambda self, e: encode_decode_sql(self, e, "DECODE", replace=False),
356            exp.DiToDate: lambda self, e: f"CAST(STRPTIME(CAST({self.sql(e, 'this')} AS TEXT), {DuckDB.DATEINT_FORMAT}) AS DATE)",
357            exp.Encode: lambda self, e: encode_decode_sql(self, e, "ENCODE", replace=False),
358            exp.Explode: rename_func("UNNEST"),
359            exp.IntDiv: lambda self, e: self.binary(e, "//"),
360            exp.IsInf: rename_func("ISINF"),
361            exp.IsNan: rename_func("ISNAN"),
362            exp.JSONBExtract: arrow_json_extract_sql,
363            exp.JSONBExtractScalar: arrow_json_extract_scalar_sql,
364            exp.JSONExtract: arrow_json_extract_sql,
365            exp.JSONExtractScalar: arrow_json_extract_scalar_sql,
366            exp.JSONFormat: _json_format_sql,
367            exp.LogicalOr: rename_func("BOOL_OR"),
368            exp.LogicalAnd: rename_func("BOOL_AND"),
369            exp.MonthsBetween: lambda self, e: self.func(
370                "DATEDIFF",
371                "'month'",
372                exp.cast(e.expression, "timestamp", copy=True),
373                exp.cast(e.this, "timestamp", copy=True),
374            ),
375            exp.ParseJSON: rename_func("JSON"),
376            exp.PercentileCont: _rename_unless_within_group("PERCENTILE_CONT", "QUANTILE_CONT"),
377            exp.PercentileDisc: _rename_unless_within_group("PERCENTILE_DISC", "QUANTILE_DISC"),
378            # DuckDB doesn't allow qualified columns inside of PIVOT expressions.
379            # See: https://github.com/duckdb/duckdb/blob/671faf92411182f81dce42ac43de8bfb05d9909e/src/planner/binder/tableref/bind_pivot.cpp#L61-L62
380            exp.Pivot: transforms.preprocess([transforms.unqualify_columns]),
381            exp.Properties: no_properties_sql,
382            exp.RegexpExtract: regexp_extract_sql,
383            exp.RegexpReplace: lambda self, e: self.func(
384                "REGEXP_REPLACE",
385                e.this,
386                e.expression,
387                e.args.get("replacement"),
388                e.args.get("modifiers"),
389            ),
390            exp.RegexpLike: rename_func("REGEXP_MATCHES"),
391            exp.RegexpSplit: rename_func("STR_SPLIT_REGEX"),
392            exp.Rand: rename_func("RANDOM"),
393            exp.SafeDivide: no_safe_divide_sql,
394            exp.Split: rename_func("STR_SPLIT"),
395            exp.SortArray: _sort_array_sql,
396            exp.StrPosition: str_position_sql,
397            exp.StrToDate: lambda self, e: f"CAST({str_to_time_sql(self, e)} AS DATE)",
398            exp.StrToTime: str_to_time_sql,
399            exp.StrToUnix: lambda self, e: f"EPOCH(STRPTIME({self.sql(e, 'this')}, {self.format_time(e)}))",
400            exp.Struct: _struct_sql,
401            exp.Timestamp: no_timestamp_sql,
402            exp.TimestampDiff: lambda self, e: self.func(
403                "DATE_DIFF", exp.Literal.string(e.unit), e.expression, e.this
404            ),
405            exp.TimestampTrunc: timestamptrunc_sql,
406            exp.TimeStrToDate: lambda self, e: f"CAST({self.sql(e, 'this')} AS DATE)",
407            exp.TimeStrToTime: timestrtotime_sql,
408            exp.TimeStrToUnix: lambda self, e: f"EPOCH(CAST({self.sql(e, 'this')} AS TIMESTAMP))",
409            exp.TimeToStr: lambda self, e: f"STRFTIME({self.sql(e, 'this')}, {self.format_time(e)})",
410            exp.TimeToUnix: rename_func("EPOCH"),
411            exp.TsOrDiToDi: lambda self, e: f"CAST(SUBSTR(REPLACE(CAST({self.sql(e, 'this')} AS TEXT), '-', ''), 1, 8) AS INT)",
412            exp.TsOrDsAdd: _ts_or_ds_add_sql,
413            exp.TsOrDsDiff: lambda self, e: self.func(
414                "DATE_DIFF",
415                f"'{e.args.get('unit') or 'DAY'}'",
416                exp.cast(e.expression, "TIMESTAMP"),
417                exp.cast(e.this, "TIMESTAMP"),
418            ),
419            exp.UnixToStr: lambda self, e: f"STRFTIME(TO_TIMESTAMP({self.sql(e, 'this')}), {self.format_time(e)})",
420            exp.UnixToTime: _unix_to_time_sql,
421            exp.UnixToTimeStr: lambda self, e: f"CAST(TO_TIMESTAMP({self.sql(e, 'this')}) AS TEXT)",
422            exp.VariancePop: rename_func("VAR_POP"),
423            exp.WeekOfYear: rename_func("WEEKOFYEAR"),
424            exp.Xor: bool_xor_sql,
425        }
426
427        TYPE_MAPPING = {
428            **generator.Generator.TYPE_MAPPING,
429            exp.DataType.Type.BINARY: "BLOB",
430            exp.DataType.Type.CHAR: "TEXT",
431            exp.DataType.Type.FLOAT: "REAL",
432            exp.DataType.Type.NCHAR: "TEXT",
433            exp.DataType.Type.NVARCHAR: "TEXT",
434            exp.DataType.Type.UINT: "UINTEGER",
435            exp.DataType.Type.VARBINARY: "BLOB",
436            exp.DataType.Type.VARCHAR: "TEXT",
437            exp.DataType.Type.TIMESTAMP_S: "TIMESTAMP_S",
438            exp.DataType.Type.TIMESTAMP_MS: "TIMESTAMP_MS",
439            exp.DataType.Type.TIMESTAMP_NS: "TIMESTAMP_NS",
440        }
441
442        STAR_MAPPING = {**generator.Generator.STAR_MAPPING, "except": "EXCLUDE"}
443
444        UNWRAPPED_INTERVAL_VALUES = (exp.Column, exp.Literal, exp.Paren)
445
446        PROPERTIES_LOCATION = {
447            **generator.Generator.PROPERTIES_LOCATION,
448            exp.VolatileProperty: exp.Properties.Location.UNSUPPORTED,
449        }
450
451        def timefromparts_sql(self, expression: exp.TimeFromParts) -> str:
452            nano = expression.args.get("nano")
453            if nano is not None:
454                expression.set(
455                    "sec", expression.args["sec"] + nano.pop() / exp.Literal.number(1000000000.0)
456                )
457
458            return rename_func("MAKE_TIME")(self, expression)
459
460        def timestampfromparts_sql(self, expression: exp.TimestampFromParts) -> str:
461            sec = expression.args["sec"]
462
463            milli = expression.args.get("milli")
464            if milli is not None:
465                sec += milli.pop() / exp.Literal.number(1000.0)
466
467            nano = expression.args.get("nano")
468            if nano is not None:
469                sec += nano.pop() / exp.Literal.number(1000000000.0)
470
471            if milli or nano:
472                expression.set("sec", sec)
473
474            return rename_func("MAKE_TIMESTAMP")(self, expression)
475
476        def tablesample_sql(
477            self,
478            expression: exp.TableSample,
479            sep: str = " AS ",
480            tablesample_keyword: t.Optional[str] = None,
481        ) -> str:
482            if not isinstance(expression.parent, exp.Select):
483                # This sample clause only applies to a single source, not the entire resulting relation
484                tablesample_keyword = "TABLESAMPLE"
485
486            return super().tablesample_sql(
487                expression, sep=sep, tablesample_keyword=tablesample_keyword
488            )
489
490        def getpath_sql(self, expression: exp.GetPath) -> str:
491            expression = prepend_dollar_to_path(expression)
492            return f"{self.sql(expression, 'this')} -> {self.sql(expression, 'expression')}"
493
494        def interval_sql(self, expression: exp.Interval) -> str:
495            multiplier: t.Optional[int] = None
496            unit = expression.text("unit").lower()
497
498            if unit.startswith("week"):
499                multiplier = 7
500            if unit.startswith("quarter"):
501                multiplier = 90
502
503            if multiplier:
504                return f"({multiplier} * {super().interval_sql(exp.Interval(this=expression.this, unit=exp.var('DAY')))})"
505
506            return super().interval_sql(expression)
507
508        def columndef_sql(self, expression: exp.ColumnDef, sep: str = " ") -> str:
509            if isinstance(expression.parent, exp.UserDefinedFunction):
510                return self.sql(expression, "this")
511            return super().columndef_sql(expression, sep)
512
513        def placeholder_sql(self, expression: exp.Placeholder) -> str:
514            return f"${expression.name}" if expression.name else "?"

Generator converts a given syntax tree to the corresponding SQL string.

Arguments:
  • pretty: Whether or not to format the produced SQL string. Default: False.
  • identify: Determines when an identifier should be quoted. Possible values are: False (default): Never quote, except in cases where it's mandatory by the dialect. True or 'always': Always quote. 'safe': Only quote identifiers that are case insensitive.
  • normalize: Whether or not to normalize identifiers to lowercase. Default: False.
  • pad: Determines the pad size in a formatted string. Default: 2.
  • indent: Determines the indentation size in a formatted string. Default: 2.
  • normalize_functions: Whether or not to normalize all function names. Possible values are: "upper" or True (default): Convert names to uppercase. "lower": Convert names to lowercase. False: Disables function name normalization.
  • unsupported_level: Determines the generator's behavior when it encounters unsupported expressions. Default ErrorLevel.WARN.
  • max_unsupported: Maximum number of unsupported messages to include in a raised UnsupportedError. This is only relevant if unsupported_level is ErrorLevel.RAISE. Default: 3
  • leading_comma: Determines whether or not the comma is leading or trailing in select expressions. This is only relevant when generating in pretty mode. Default: False
  • max_text_width: The max number of characters in a segment before creating new lines in pretty mode. The default is on the smaller end because the length only represents a segment and not the true line length. Default: 80
  • comments: Whether or not to preserve comments in the output SQL code. Default: True
JOIN_HINTS = False
TABLE_HINTS = False
QUERY_HINTS = False
LIMIT_FETCH = 'LIMIT'
STRUCT_DELIMITER = ('(', ')')
RENAME_TABLE_WITH_DB = False
NVL2_SUPPORTED = False
SEMI_ANTI_JOIN_WITH_SIDE = False
TABLESAMPLE_KEYWORDS = 'USING SAMPLE'
TABLESAMPLE_SEED_KEYWORD = 'REPEATABLE'
LAST_DAY_SUPPORTS_DATE_PART = False
JSON_KEY_VALUE_PAIR_SEP = ','
TRANSFORMS = {<class 'sqlglot.expressions.DateAdd'>: <function _date_delta_sql>, <class 'sqlglot.expressions.CaseSpecificColumnConstraint'>: <function Generator.<lambda>>, <class 'sqlglot.expressions.CharacterSetColumnConstraint'>: <function Generator.<lambda>>, <class 'sqlglot.expressions.CharacterSetProperty'>: <function Generator.<lambda>>, <class 'sqlglot.expressions.CheckColumnConstraint'>: <function Generator.<lambda>>, <class 'sqlglot.expressions.ClusteredColumnConstraint'>: <function Generator.<lambda>>, <class 'sqlglot.expressions.CollateColumnConstraint'>: <function Generator.<lambda>>, <class 'sqlglot.expressions.AutoRefreshProperty'>: <function Generator.<lambda>>, <class 'sqlglot.expressions.CopyGrantsProperty'>: <function Generator.<lambda>>, <class 'sqlglot.expressions.CommentColumnConstraint'>: <function no_comment_column_constraint_sql>, <class 'sqlglot.expressions.DateFormatColumnConstraint'>: <function Generator.<lambda>>, <class 'sqlglot.expressions.DefaultColumnConstraint'>: <function Generator.<lambda>>, <class 'sqlglot.expressions.EncodeColumnConstraint'>: <function Generator.<lambda>>, <class 'sqlglot.expressions.ExecuteAsProperty'>: <function Generator.<lambda>>, <class 'sqlglot.expressions.ExternalProperty'>: <function Generator.<lambda>>, <class 'sqlglot.expressions.HeapProperty'>: <function Generator.<lambda>>, <class 'sqlglot.expressions.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.LanguageProperty'>: <function Generator.<lambda>>, <class 'sqlglot.expressions.LocationProperty'>: <function Generator.<lambda>>, <class 'sqlglot.expressions.LogProperty'>: <function Generator.<lambda>>, <class 'sqlglot.expressions.MaterializedProperty'>: <function Generator.<lambda>>, <class 'sqlglot.expressions.NoPrimaryIndexProperty'>: <function Generator.<lambda>>, <class 'sqlglot.expressions.NonClusteredColumnConstraint'>: <function Generator.<lambda>>, <class 'sqlglot.expressions.NotForReplicationColumnConstraint'>: <function Generator.<lambda>>, <class 'sqlglot.expressions.OnCommitProperty'>: <function Generator.<lambda>>, <class 'sqlglot.expressions.OnProperty'>: <function Generator.<lambda>>, <class 'sqlglot.expressions.OnUpdateColumnConstraint'>: <function Generator.<lambda>>, <class 'sqlglot.expressions.OutputModelProperty'>: <function Generator.<lambda>>, <class 'sqlglot.expressions.PathColumnConstraint'>: <function Generator.<lambda>>, <class 'sqlglot.expressions.RemoteWithConnectionModelProperty'>: <function Generator.<lambda>>, <class 'sqlglot.expressions.ReturnsProperty'>: <function Generator.<lambda>>, <class 'sqlglot.expressions.SampleProperty'>: <function Generator.<lambda>>, <class 'sqlglot.expressions.SetProperty'>: <function Generator.<lambda>>, <class 'sqlglot.expressions.SetConfigProperty'>: <function Generator.<lambda>>, <class 'sqlglot.expressions.SettingsProperty'>: <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.TemporaryProperty'>: <function Generator.<lambda>>, <class 'sqlglot.expressions.ToTableProperty'>: <function Generator.<lambda>>, <class 'sqlglot.expressions.TransientProperty'>: <function Generator.<lambda>>, <class 'sqlglot.expressions.TransformModelProperty'>: <function Generator.<lambda>>, <class 'sqlglot.expressions.TitleColumnConstraint'>: <function Generator.<lambda>>, <class 'sqlglot.expressions.UppercaseColumnConstraint'>: <function Generator.<lambda>>, <class 'sqlglot.expressions.VarMap'>: <function Generator.<lambda>>, <class 'sqlglot.expressions.VolatileProperty'>: <function Generator.<lambda>>, <class 'sqlglot.expressions.WithJournalTableProperty'>: <function Generator.<lambda>>, <class 'sqlglot.expressions.ApproxDistinct'>: <function approx_count_distinct_sql>, <class 'sqlglot.expressions.Array'>: <function DuckDB.Generator.<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.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.JSONBExtract'>: <function arrow_json_extract_sql>, <class 'sqlglot.expressions.JSONBExtractScalar'>: <function arrow_json_extract_scalar_sql>, <class 'sqlglot.expressions.JSONExtract'>: <function arrow_json_extract_sql>, <class 'sqlglot.expressions.JSONExtractScalar'>: <function arrow_json_extract_scalar_sql>, <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.MonthsBetween'>: <function DuckDB.Generator.<lambda>>, <class 'sqlglot.expressions.ParseJSON'>: <function rename_func.<locals>.<lambda>>, <class 'sqlglot.expressions.PercentileCont'>: <function _rename_unless_within_group.<locals>.<lambda>>, <class 'sqlglot.expressions.PercentileDisc'>: <function _rename_unless_within_group.<locals>.<lambda>>, <class 'sqlglot.expressions.Pivot'>: <function preprocess.<locals>._to_sql>, <class 'sqlglot.expressions.Properties'>: <function no_properties_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.Timestamp'>: <function no_timestamp_sql>, <class 'sqlglot.expressions.TimestampDiff'>: <function DuckDB.Generator.<lambda>>, <class 'sqlglot.expressions.TimestampTrunc'>: <function 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>}
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.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'}
STAR_MAPPING = {'except': 'EXCLUDE', 'replace': 'REPLACE'}
UNWRAPPED_INTERVAL_VALUES = (<class 'sqlglot.expressions.Column'>, <class 'sqlglot.expressions.Literal'>, <class 'sqlglot.expressions.Paren'>)
PROPERTIES_LOCATION = {<class 'sqlglot.expressions.AlgorithmProperty'>: <Location.POST_CREATE: 'POST_CREATE'>, <class 'sqlglot.expressions.AutoIncrementProperty'>: <Location.POST_SCHEMA: 'POST_SCHEMA'>, <class 'sqlglot.expressions.AutoRefreshProperty'>: <Location.POST_SCHEMA: 'POST_SCHEMA'>, <class 'sqlglot.expressions.BlockCompressionProperty'>: <Location.POST_NAME: 'POST_NAME'>, <class 'sqlglot.expressions.CharacterSetProperty'>: <Location.POST_SCHEMA: 'POST_SCHEMA'>, <class 'sqlglot.expressions.ChecksumProperty'>: <Location.POST_NAME: 'POST_NAME'>, <class 'sqlglot.expressions.CollateProperty'>: <Location.POST_SCHEMA: 'POST_SCHEMA'>, <class 'sqlglot.expressions.CopyGrantsProperty'>: <Location.POST_SCHEMA: 'POST_SCHEMA'>, <class 'sqlglot.expressions.Cluster'>: <Location.POST_SCHEMA: 'POST_SCHEMA'>, <class 'sqlglot.expressions.ClusteredByProperty'>: <Location.POST_SCHEMA: 'POST_SCHEMA'>, <class 'sqlglot.expressions.DataBlocksizeProperty'>: <Location.POST_NAME: 'POST_NAME'>, <class 'sqlglot.expressions.DefinerProperty'>: <Location.POST_CREATE: 'POST_CREATE'>, <class 'sqlglot.expressions.DictRange'>: <Location.POST_SCHEMA: 'POST_SCHEMA'>, <class 'sqlglot.expressions.DictProperty'>: <Location.POST_SCHEMA: 'POST_SCHEMA'>, <class 'sqlglot.expressions.DistKeyProperty'>: <Location.POST_SCHEMA: 'POST_SCHEMA'>, <class 'sqlglot.expressions.DistStyleProperty'>: <Location.POST_SCHEMA: 'POST_SCHEMA'>, <class 'sqlglot.expressions.EngineProperty'>: <Location.POST_SCHEMA: 'POST_SCHEMA'>, <class 'sqlglot.expressions.ExecuteAsProperty'>: <Location.POST_SCHEMA: 'POST_SCHEMA'>, <class 'sqlglot.expressions.ExternalProperty'>: <Location.POST_CREATE: 'POST_CREATE'>, <class 'sqlglot.expressions.FallbackProperty'>: <Location.POST_NAME: 'POST_NAME'>, <class 'sqlglot.expressions.FileFormatProperty'>: <Location.POST_WITH: 'POST_WITH'>, <class 'sqlglot.expressions.FreespaceProperty'>: <Location.POST_NAME: 'POST_NAME'>, <class 'sqlglot.expressions.HeapProperty'>: <Location.POST_WITH: 'POST_WITH'>, <class 'sqlglot.expressions.InheritsProperty'>: <Location.POST_SCHEMA: 'POST_SCHEMA'>, <class 'sqlglot.expressions.InputModelProperty'>: <Location.POST_SCHEMA: 'POST_SCHEMA'>, <class 'sqlglot.expressions.IsolatedLoadingProperty'>: <Location.POST_NAME: 'POST_NAME'>, <class 'sqlglot.expressions.JournalProperty'>: <Location.POST_NAME: 'POST_NAME'>, <class 'sqlglot.expressions.LanguageProperty'>: <Location.POST_SCHEMA: 'POST_SCHEMA'>, <class 'sqlglot.expressions.LikeProperty'>: <Location.POST_SCHEMA: 'POST_SCHEMA'>, <class 'sqlglot.expressions.LocationProperty'>: <Location.POST_SCHEMA: 'POST_SCHEMA'>, <class 'sqlglot.expressions.LockingProperty'>: <Location.POST_ALIAS: 'POST_ALIAS'>, <class 'sqlglot.expressions.LogProperty'>: <Location.POST_NAME: 'POST_NAME'>, <class 'sqlglot.expressions.MaterializedProperty'>: <Location.POST_CREATE: 'POST_CREATE'>, <class 'sqlglot.expressions.MergeBlockRatioProperty'>: <Location.POST_NAME: 'POST_NAME'>, <class 'sqlglot.expressions.NoPrimaryIndexProperty'>: <Location.POST_EXPRESSION: 'POST_EXPRESSION'>, <class 'sqlglot.expressions.OnProperty'>: <Location.POST_SCHEMA: 'POST_SCHEMA'>, <class 'sqlglot.expressions.OnCommitProperty'>: <Location.POST_EXPRESSION: 'POST_EXPRESSION'>, <class 'sqlglot.expressions.Order'>: <Location.POST_SCHEMA: 'POST_SCHEMA'>, <class 'sqlglot.expressions.OutputModelProperty'>: <Location.POST_SCHEMA: 'POST_SCHEMA'>, <class 'sqlglot.expressions.PartitionedByProperty'>: <Location.POST_WITH: 'POST_WITH'>, <class 'sqlglot.expressions.PartitionedOfProperty'>: <Location.POST_SCHEMA: 'POST_SCHEMA'>, <class 'sqlglot.expressions.PrimaryKey'>: <Location.POST_SCHEMA: 'POST_SCHEMA'>, <class 'sqlglot.expressions.Property'>: <Location.POST_WITH: 'POST_WITH'>, <class 'sqlglot.expressions.RemoteWithConnectionModelProperty'>: <Location.POST_SCHEMA: 'POST_SCHEMA'>, <class 'sqlglot.expressions.ReturnsProperty'>: <Location.POST_SCHEMA: 'POST_SCHEMA'>, <class 'sqlglot.expressions.RowFormatProperty'>: <Location.POST_SCHEMA: 'POST_SCHEMA'>, <class 'sqlglot.expressions.RowFormatDelimitedProperty'>: <Location.POST_SCHEMA: 'POST_SCHEMA'>, <class 'sqlglot.expressions.RowFormatSerdeProperty'>: <Location.POST_SCHEMA: 'POST_SCHEMA'>, <class 'sqlglot.expressions.SampleProperty'>: <Location.POST_SCHEMA: 'POST_SCHEMA'>, <class 'sqlglot.expressions.SchemaCommentProperty'>: <Location.POST_SCHEMA: 'POST_SCHEMA'>, <class 'sqlglot.expressions.SerdeProperties'>: <Location.POST_SCHEMA: 'POST_SCHEMA'>, <class 'sqlglot.expressions.Set'>: <Location.POST_SCHEMA: 'POST_SCHEMA'>, <class 'sqlglot.expressions.SettingsProperty'>: <Location.POST_SCHEMA: 'POST_SCHEMA'>, <class 'sqlglot.expressions.SetProperty'>: <Location.POST_CREATE: 'POST_CREATE'>, <class 'sqlglot.expressions.SetConfigProperty'>: <Location.POST_SCHEMA: 'POST_SCHEMA'>, <class 'sqlglot.expressions.SortKeyProperty'>: <Location.POST_SCHEMA: 'POST_SCHEMA'>, <class 'sqlglot.expressions.SqlReadWriteProperty'>: <Location.POST_SCHEMA: 'POST_SCHEMA'>, <class 'sqlglot.expressions.SqlSecurityProperty'>: <Location.POST_CREATE: 'POST_CREATE'>, <class 'sqlglot.expressions.StabilityProperty'>: <Location.POST_SCHEMA: 'POST_SCHEMA'>, <class 'sqlglot.expressions.TemporaryProperty'>: <Location.POST_CREATE: 'POST_CREATE'>, <class 'sqlglot.expressions.ToTableProperty'>: <Location.POST_SCHEMA: 'POST_SCHEMA'>, <class 'sqlglot.expressions.TransientProperty'>: <Location.POST_CREATE: 'POST_CREATE'>, <class 'sqlglot.expressions.TransformModelProperty'>: <Location.POST_SCHEMA: 'POST_SCHEMA'>, <class 'sqlglot.expressions.MergeTreeTTL'>: <Location.POST_SCHEMA: 'POST_SCHEMA'>, <class 'sqlglot.expressions.VolatileProperty'>: <Location.UNSUPPORTED: 'UNSUPPORTED'>, <class 'sqlglot.expressions.WithDataProperty'>: <Location.POST_EXPRESSION: 'POST_EXPRESSION'>, <class 'sqlglot.expressions.WithJournalTableProperty'>: <Location.POST_NAME: 'POST_NAME'>, <class 'sqlglot.expressions.WithSystemVersioningProperty'>: <Location.POST_SCHEMA: 'POST_SCHEMA'>}
def timefromparts_sql(self, expression: sqlglot.expressions.TimeFromParts) -> str:
451        def timefromparts_sql(self, expression: exp.TimeFromParts) -> str:
452            nano = expression.args.get("nano")
453            if nano is not None:
454                expression.set(
455                    "sec", expression.args["sec"] + nano.pop() / exp.Literal.number(1000000000.0)
456                )
457
458            return rename_func("MAKE_TIME")(self, expression)
def timestampfromparts_sql(self, expression: sqlglot.expressions.TimestampFromParts) -> str:
460        def timestampfromparts_sql(self, expression: exp.TimestampFromParts) -> str:
461            sec = expression.args["sec"]
462
463            milli = expression.args.get("milli")
464            if milli is not None:
465                sec += milli.pop() / exp.Literal.number(1000.0)
466
467            nano = expression.args.get("nano")
468            if nano is not None:
469                sec += nano.pop() / exp.Literal.number(1000000000.0)
470
471            if milli or nano:
472                expression.set("sec", sec)
473
474            return rename_func("MAKE_TIMESTAMP")(self, expression)
def tablesample_sql( self, expression: sqlglot.expressions.TableSample, sep: str = ' AS ', tablesample_keyword: Optional[str] = None) -> str:
476        def tablesample_sql(
477            self,
478            expression: exp.TableSample,
479            sep: str = " AS ",
480            tablesample_keyword: t.Optional[str] = None,
481        ) -> str:
482            if not isinstance(expression.parent, exp.Select):
483                # This sample clause only applies to a single source, not the entire resulting relation
484                tablesample_keyword = "TABLESAMPLE"
485
486            return super().tablesample_sql(
487                expression, sep=sep, tablesample_keyword=tablesample_keyword
488            )
def getpath_sql(self, expression: sqlglot.expressions.GetPath) -> str:
490        def getpath_sql(self, expression: exp.GetPath) -> str:
491            expression = prepend_dollar_to_path(expression)
492            return f"{self.sql(expression, 'this')} -> {self.sql(expression, 'expression')}"
def interval_sql(self, expression: sqlglot.expressions.Interval) -> str:
494        def interval_sql(self, expression: exp.Interval) -> str:
495            multiplier: t.Optional[int] = None
496            unit = expression.text("unit").lower()
497
498            if unit.startswith("week"):
499                multiplier = 7
500            if unit.startswith("quarter"):
501                multiplier = 90
502
503            if multiplier:
504                return f"({multiplier} * {super().interval_sql(exp.Interval(this=expression.this, unit=exp.var('DAY')))})"
505
506            return super().interval_sql(expression)
def columndef_sql(self, expression: sqlglot.expressions.ColumnDef, sep: str = ' ') -> str:
508        def columndef_sql(self, expression: exp.ColumnDef, sep: str = " ") -> str:
509            if isinstance(expression.parent, exp.UserDefinedFunction):
510                return self.sql(expression, "this")
511            return super().columndef_sql(expression, sep)
def placeholder_sql(self, expression: sqlglot.expressions.Placeholder) -> str:
513        def placeholder_sql(self, expression: exp.Placeholder) -> str:
514            return f"${expression.name}" if expression.name else "?"
SELECT_KINDS: Tuple[str, ...] = ()
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
TIME_PART_SINGULARS
TOKEN_MAPPING
PARAMETER_TOKEN
RESERVED_KEYWORDS
WITH_SEPARATED_COMMENTS
EXCLUDE_COMMENTS
EXPRESSIONS_WITHOUT_NESTED_CTES
KEY_VALUE_DEFINITIONS
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_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
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
index_sql
identifier_sql
inputoutputformat_sql
national_sql
partition_sql
properties_sql
root_properties
properties
with_properties
locate_properties
property_name
property_sql
likeproperty_sql
fallbackproperty_sql
journalproperty_sql
freespaceproperty_sql
checksumproperty_sql
mergeblockratioproperty_sql
datablocksizeproperty_sql
blockcompressionproperty_sql
isolatedloadingproperty_sql
partitionboundspec_sql
partitionedofproperty_sql
lockingproperty_sql
withdataproperty_sql
withsystemversioningproperty_sql
insert_sql
intersect_sql
intersect_op
introducer_sql
kill_sql
pseudotype_sql
objectidentifier_sql
onconflict_sql
returning_sql
rowformatdelimitedproperty_sql
withtablehint_sql
indextablehint_sql
historicaldata_sql
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
join_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
matchrecognize_sql
query_modifiers
offset_limit_modifiers
after_having_modifiers
after_limit_modifiers
select_sql
schema_sql
schema_columns_sql
star_sql
parameter_sql
sessionparameter_sql
subquery_sql
qualify_sql
union_sql
union_op
unnest_sql
where_sql
window_sql
partition_by_sql
windowspec_sql
withingroup_sql
between_sql
bracket_sql
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
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
xor_sql
connector_sql
bitwiseand_sql
bitwiseleftshift_sql
bitwisenot_sql
bitwiseor_sql
bitwiserightshift_sql
bitwisexor_sql
cast_sql
currentdate_sql
collate_sql
command_sql
comment_sql
mergetreettlaction_sql
mergetreettl_sql
transaction_sql
commit_sql
rollback_sql
altercolumn_sql
renametable_sql
renamecolumn_sql
altertable_sql
add_column_sql
droppartition_sql
addconstraint_sql
distinct_sql
ignorenulls_sql
respectnulls_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
or_sql
slice_sql
sub_sql
trycast_sql
log_sql
use_sql
binary
function_fallback_sql
func
format_args
text_width
format_time
expressions
op_expressions
naked_property
set_operation
tag_sql
token_sql
userdefinedfunction_sql
joinhint_sql
kwarg_sql
when_sql
merge_sql
tochar_sql
dictproperty_sql
dictrange_sql
dictsubproperty_sql
oncluster_sql
clusteredbyproperty_sql
anyvalue_sql
querytransform_sql
indexconstraintoption_sql
indexcolumnconstraint_sql
nvl2_sql
comprehension_sql
columnprefix_sql
opclass_sql
predict_sql
forin_sql
refresh_sql
operator_sql
toarray_sql
tsordstotime_sql
tsordstodate_sql
unixdate_sql
lastday_sql