Edit on GitHub

sqlglot.dialects.hive

  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    approx_count_distinct_sql,
  9    arg_max_or_min_no_count,
 10    create_with_partitions_sql,
 11    format_time_lambda,
 12    if_sql,
 13    is_parse_json,
 14    left_to_substring_sql,
 15    locate_to_strposition,
 16    max_or_greatest,
 17    min_or_least,
 18    no_ilike_sql,
 19    no_recursive_cte_sql,
 20    no_safe_divide_sql,
 21    no_trycast_sql,
 22    regexp_extract_sql,
 23    regexp_replace_sql,
 24    rename_func,
 25    right_to_substring_sql,
 26    strposition_to_locate_sql,
 27    struct_extract_sql,
 28    time_format,
 29    timestrtotime_sql,
 30    var_map_sql,
 31)
 32from sqlglot.helper import seq_get
 33from sqlglot.parser import parse_var_map
 34from sqlglot.tokens import TokenType
 35
 36# (FuncType, Multiplier)
 37DATE_DELTA_INTERVAL = {
 38    "YEAR": ("ADD_MONTHS", 12),
 39    "MONTH": ("ADD_MONTHS", 1),
 40    "QUARTER": ("ADD_MONTHS", 3),
 41    "WEEK": ("DATE_ADD", 7),
 42    "DAY": ("DATE_ADD", 1),
 43}
 44
 45TIME_DIFF_FACTOR = {
 46    "MILLISECOND": " * 1000",
 47    "SECOND": "",
 48    "MINUTE": " / 60",
 49    "HOUR": " / 3600",
 50}
 51
 52DIFF_MONTH_SWITCH = ("YEAR", "QUARTER", "MONTH")
 53
 54
 55def _create_sql(self, expression: exp.Create) -> str:
 56    # remove UNIQUE column constraints
 57    for constraint in expression.find_all(exp.UniqueColumnConstraint):
 58        if constraint.parent:
 59            constraint.parent.pop()
 60
 61    properties = expression.args.get("properties")
 62    temporary = any(
 63        isinstance(prop, exp.TemporaryProperty)
 64        for prop in (properties.expressions if properties else [])
 65    )
 66
 67    # CTAS with temp tables map to CREATE TEMPORARY VIEW
 68    kind = expression.args["kind"]
 69    if kind.upper() == "TABLE" and temporary:
 70        if expression.expression:
 71            return f"CREATE TEMPORARY VIEW {self.sql(expression, 'this')} AS {self.sql(expression, 'expression')}"
 72        else:
 73            # CREATE TEMPORARY TABLE may require storage provider
 74            expression = self.temporary_storage_provider(expression)
 75
 76    return create_with_partitions_sql(self, expression)
 77
 78
 79def _add_date_sql(self: Hive.Generator, expression: exp.DateAdd | exp.DateSub) -> str:
 80    unit = expression.text("unit").upper()
 81    func, multiplier = DATE_DELTA_INTERVAL.get(unit, ("DATE_ADD", 1))
 82
 83    if isinstance(expression, exp.DateSub):
 84        multiplier *= -1
 85
 86    if expression.expression.is_number:
 87        modified_increment = exp.Literal.number(int(expression.text("expression")) * multiplier)
 88    else:
 89        modified_increment = expression.expression
 90        if multiplier != 1:
 91            modified_increment = exp.Mul(  # type: ignore
 92                this=modified_increment, expression=exp.Literal.number(multiplier)
 93            )
 94
 95    return self.func(func, expression.this, modified_increment)
 96
 97
 98def _date_diff_sql(self: Hive.Generator, expression: exp.DateDiff) -> str:
 99    unit = expression.text("unit").upper()
100
101    factor = TIME_DIFF_FACTOR.get(unit)
102    if factor is not None:
103        left = self.sql(expression, "this")
104        right = self.sql(expression, "expression")
105        sec_diff = f"UNIX_TIMESTAMP({left}) - UNIX_TIMESTAMP({right})"
106        return f"({sec_diff}){factor}" if factor else sec_diff
107
108    months_between = unit in DIFF_MONTH_SWITCH
109    sql_func = "MONTHS_BETWEEN" if months_between else "DATEDIFF"
110    _, multiplier = DATE_DELTA_INTERVAL.get(unit, ("", 1))
111    multiplier_sql = f" / {multiplier}" if multiplier > 1 else ""
112    diff_sql = f"{sql_func}({self.format_args(expression.this, expression.expression)})"
113
114    if months_between:
115        # MONTHS_BETWEEN returns a float, so we need to truncate the fractional part
116        diff_sql = f"CAST({diff_sql} AS INT)"
117
118    return f"{diff_sql}{multiplier_sql}"
119
120
121def _json_format_sql(self: Hive.Generator, expression: exp.JSONFormat) -> str:
122    this = expression.this
123
124    if is_parse_json(this):
125        if this.this.is_string:
126            # Since FROM_JSON requires a nested type, we always wrap the json string with
127            # an array to ensure that "naked" strings like "'a'" will be handled correctly
128            wrapped_json = exp.Literal.string(f"[{this.this.name}]")
129
130            from_json = self.func(
131                "FROM_JSON", wrapped_json, self.func("SCHEMA_OF_JSON", wrapped_json)
132            )
133            to_json = self.func("TO_JSON", from_json)
134
135            # This strips the [, ] delimiters of the dummy array printed by TO_JSON
136            return self.func("REGEXP_EXTRACT", to_json, "'^.(.*).$'", "1")
137        return self.sql(this)
138
139    return self.func("TO_JSON", this, expression.args.get("options"))
140
141
142def _array_sort_sql(self: Hive.Generator, expression: exp.ArraySort) -> str:
143    if expression.expression:
144        self.unsupported("Hive SORT_ARRAY does not support a comparator")
145    return f"SORT_ARRAY({self.sql(expression, 'this')})"
146
147
148def _property_sql(self: Hive.Generator, expression: exp.Property) -> str:
149    return f"{self.property_name(expression, string_key=True)}={self.sql(expression, 'value')}"
150
151
152def _str_to_unix_sql(self: Hive.Generator, expression: exp.StrToUnix) -> str:
153    return self.func("UNIX_TIMESTAMP", expression.this, time_format("hive")(self, expression))
154
155
156def _str_to_date_sql(self: Hive.Generator, expression: exp.StrToDate) -> str:
157    this = self.sql(expression, "this")
158    time_format = self.format_time(expression)
159    if time_format not in (Hive.TIME_FORMAT, Hive.DATE_FORMAT):
160        this = f"FROM_UNIXTIME(UNIX_TIMESTAMP({this}, {time_format}))"
161    return f"CAST({this} AS DATE)"
162
163
164def _str_to_time_sql(self: Hive.Generator, expression: exp.StrToTime) -> str:
165    this = self.sql(expression, "this")
166    time_format = self.format_time(expression)
167    if time_format not in (Hive.TIME_FORMAT, Hive.DATE_FORMAT):
168        this = f"FROM_UNIXTIME(UNIX_TIMESTAMP({this}, {time_format}))"
169    return f"CAST({this} AS TIMESTAMP)"
170
171
172def _time_to_str(self: Hive.Generator, expression: exp.TimeToStr) -> str:
173    this = self.sql(expression, "this")
174    time_format = self.format_time(expression)
175    return f"DATE_FORMAT({this}, {time_format})"
176
177
178def _to_date_sql(self: Hive.Generator, expression: exp.TsOrDsToDate) -> str:
179    this = self.sql(expression, "this")
180    time_format = self.format_time(expression)
181    if time_format and time_format not in (Hive.TIME_FORMAT, Hive.DATE_FORMAT):
182        return f"TO_DATE({this}, {time_format})"
183    return f"TO_DATE({this})"
184
185
186class Hive(Dialect):
187    ALIAS_POST_TABLESAMPLE = True
188    IDENTIFIERS_CAN_START_WITH_DIGIT = True
189    SUPPORTS_USER_DEFINED_TYPES = False
190    SAFE_DIVISION = True
191
192    # https://spark.apache.org/docs/latest/sql-ref-identifier.html#description
193    RESOLVES_IDENTIFIERS_AS_UPPERCASE = None
194
195    TIME_MAPPING = {
196        "y": "%Y",
197        "Y": "%Y",
198        "YYYY": "%Y",
199        "yyyy": "%Y",
200        "YY": "%y",
201        "yy": "%y",
202        "MMMM": "%B",
203        "MMM": "%b",
204        "MM": "%m",
205        "M": "%-m",
206        "dd": "%d",
207        "d": "%-d",
208        "HH": "%H",
209        "H": "%-H",
210        "hh": "%I",
211        "h": "%-I",
212        "mm": "%M",
213        "m": "%-M",
214        "ss": "%S",
215        "s": "%-S",
216        "SSSSSS": "%f",
217        "a": "%p",
218        "DD": "%j",
219        "D": "%-j",
220        "E": "%a",
221        "EE": "%a",
222        "EEE": "%a",
223        "EEEE": "%A",
224    }
225
226    DATE_FORMAT = "'yyyy-MM-dd'"
227    DATEINT_FORMAT = "'yyyyMMdd'"
228    TIME_FORMAT = "'yyyy-MM-dd HH:mm:ss'"
229
230    class Tokenizer(tokens.Tokenizer):
231        QUOTES = ["'", '"']
232        IDENTIFIERS = ["`"]
233        STRING_ESCAPES = ["\\"]
234        ENCODE = "utf-8"
235
236        SINGLE_TOKENS = {
237            **tokens.Tokenizer.SINGLE_TOKENS,
238            "$": TokenType.PARAMETER,
239        }
240
241        KEYWORDS = {
242            **tokens.Tokenizer.KEYWORDS,
243            "ADD ARCHIVE": TokenType.COMMAND,
244            "ADD ARCHIVES": TokenType.COMMAND,
245            "ADD FILE": TokenType.COMMAND,
246            "ADD FILES": TokenType.COMMAND,
247            "ADD JAR": TokenType.COMMAND,
248            "ADD JARS": TokenType.COMMAND,
249            "MSCK REPAIR": TokenType.COMMAND,
250            "REFRESH": TokenType.REFRESH,
251            "TIMESTAMP AS OF": TokenType.TIMESTAMP_SNAPSHOT,
252            "VERSION AS OF": TokenType.VERSION_SNAPSHOT,
253            "WITH SERDEPROPERTIES": TokenType.SERDE_PROPERTIES,
254        }
255
256        NUMERIC_LITERALS = {
257            "L": "BIGINT",
258            "S": "SMALLINT",
259            "Y": "TINYINT",
260            "D": "DOUBLE",
261            "F": "FLOAT",
262            "BD": "DECIMAL",
263        }
264
265    class Parser(parser.Parser):
266        LOG_DEFAULTS_TO_LN = True
267        STRICT_CAST = False
268
269        FUNCTIONS = {
270            **parser.Parser.FUNCTIONS,
271            "BASE64": exp.ToBase64.from_arg_list,
272            "COLLECT_LIST": exp.ArrayAgg.from_arg_list,
273            "COLLECT_SET": exp.SetAgg.from_arg_list,
274            "DATE_ADD": lambda args: exp.TsOrDsAdd(
275                this=seq_get(args, 0), expression=seq_get(args, 1), unit=exp.Literal.string("DAY")
276            ),
277            "DATE_FORMAT": lambda args: format_time_lambda(exp.TimeToStr, "hive")(
278                [
279                    exp.TimeStrToTime(this=seq_get(args, 0)),
280                    seq_get(args, 1),
281                ]
282            ),
283            "DATE_SUB": lambda args: exp.TsOrDsAdd(
284                this=seq_get(args, 0),
285                expression=exp.Mul(this=seq_get(args, 1), expression=exp.Literal.number(-1)),
286                unit=exp.Literal.string("DAY"),
287            ),
288            "DATEDIFF": lambda args: exp.DateDiff(
289                this=exp.TsOrDsToDate(this=seq_get(args, 0)),
290                expression=exp.TsOrDsToDate(this=seq_get(args, 1)),
291            ),
292            "DAY": lambda args: exp.Day(this=exp.TsOrDsToDate(this=seq_get(args, 0))),
293            "FROM_UNIXTIME": format_time_lambda(exp.UnixToStr, "hive", True),
294            "GET_JSON_OBJECT": exp.JSONExtractScalar.from_arg_list,
295            "LOCATE": locate_to_strposition,
296            "MAP": parse_var_map,
297            "MONTH": lambda args: exp.Month(this=exp.TsOrDsToDate.from_arg_list(args)),
298            "PERCENTILE": exp.Quantile.from_arg_list,
299            "PERCENTILE_APPROX": exp.ApproxQuantile.from_arg_list,
300            "REGEXP_EXTRACT": lambda args: exp.RegexpExtract(
301                this=seq_get(args, 0), expression=seq_get(args, 1), group=seq_get(args, 2)
302            ),
303            "SIZE": exp.ArraySize.from_arg_list,
304            "SPLIT": exp.RegexpSplit.from_arg_list,
305            "STR_TO_MAP": lambda args: exp.StrToMap(
306                this=seq_get(args, 0),
307                pair_delim=seq_get(args, 1) or exp.Literal.string(","),
308                key_value_delim=seq_get(args, 2) or exp.Literal.string(":"),
309            ),
310            "TO_DATE": format_time_lambda(exp.TsOrDsToDate, "hive"),
311            "TO_JSON": exp.JSONFormat.from_arg_list,
312            "UNBASE64": exp.FromBase64.from_arg_list,
313            "UNIX_TIMESTAMP": format_time_lambda(exp.StrToUnix, "hive", True),
314            "YEAR": lambda args: exp.Year(this=exp.TsOrDsToDate.from_arg_list(args)),
315        }
316
317        NO_PAREN_FUNCTION_PARSERS = {
318            **parser.Parser.NO_PAREN_FUNCTION_PARSERS,
319            "TRANSFORM": lambda self: self._parse_transform(),
320        }
321
322        PROPERTY_PARSERS = {
323            **parser.Parser.PROPERTY_PARSERS,
324            "WITH SERDEPROPERTIES": lambda self: exp.SerdeProperties(
325                expressions=self._parse_wrapped_csv(self._parse_property)
326            ),
327        }
328
329        def _parse_transform(self) -> t.Optional[exp.Transform | exp.QueryTransform]:
330            if not self._match(TokenType.L_PAREN, advance=False):
331                self._retreat(self._index - 1)
332                return None
333
334            args = self._parse_wrapped_csv(self._parse_lambda)
335            row_format_before = self._parse_row_format(match_row=True)
336
337            record_writer = None
338            if self._match_text_seq("RECORDWRITER"):
339                record_writer = self._parse_string()
340
341            if not self._match(TokenType.USING):
342                return exp.Transform.from_arg_list(args)
343
344            command_script = self._parse_string()
345
346            self._match(TokenType.ALIAS)
347            schema = self._parse_schema()
348
349            row_format_after = self._parse_row_format(match_row=True)
350            record_reader = None
351            if self._match_text_seq("RECORDREADER"):
352                record_reader = self._parse_string()
353
354            return self.expression(
355                exp.QueryTransform,
356                expressions=args,
357                command_script=command_script,
358                schema=schema,
359                row_format_before=row_format_before,
360                record_writer=record_writer,
361                row_format_after=row_format_after,
362                record_reader=record_reader,
363            )
364
365        def _parse_types(
366            self, check_func: bool = False, schema: bool = False, allow_identifiers: bool = True
367        ) -> t.Optional[exp.Expression]:
368            """
369            Spark (and most likely Hive) treats casts to CHAR(length) and VARCHAR(length) as casts to
370            STRING in all contexts except for schema definitions. For example, this is in Spark v3.4.0:
371
372                spark-sql (default)> select cast(1234 as varchar(2));
373                23/06/06 15:51:18 WARN CharVarcharUtils: The Spark cast operator does not support
374                char/varchar type and simply treats them as string type. Please use string type
375                directly to avoid confusion. Otherwise, you can set spark.sql.legacy.charVarcharAsString
376                to true, so that Spark treat them as string type as same as Spark 3.0 and earlier
377
378                1234
379                Time taken: 4.265 seconds, Fetched 1 row(s)
380
381            This shows that Spark doesn't truncate the value into '12', which is inconsistent with
382            what other dialects (e.g. postgres) do, so we need to drop the length to transpile correctly.
383
384            Reference: https://spark.apache.org/docs/latest/sql-ref-datatypes.html
385            """
386            this = super()._parse_types(
387                check_func=check_func, schema=schema, allow_identifiers=allow_identifiers
388            )
389
390            if this and not schema:
391                return this.transform(
392                    lambda node: node.replace(exp.DataType.build("text"))
393                    if isinstance(node, exp.DataType) and node.is_type("char", "varchar")
394                    else node,
395                    copy=False,
396                )
397
398            return this
399
400        def _parse_partition_and_order(
401            self,
402        ) -> t.Tuple[t.List[exp.Expression], t.Optional[exp.Expression]]:
403            return (
404                self._parse_csv(self._parse_conjunction)
405                if self._match_set({TokenType.PARTITION_BY, TokenType.DISTRIBUTE_BY})
406                else [],
407                super()._parse_order(skip_order_token=self._match(TokenType.SORT_BY)),
408            )
409
410    class Generator(generator.Generator):
411        LIMIT_FETCH = "LIMIT"
412        TABLESAMPLE_WITH_METHOD = False
413        TABLESAMPLE_SIZE_IS_PERCENT = True
414        JOIN_HINTS = False
415        TABLE_HINTS = False
416        QUERY_HINTS = False
417        INDEX_ON = "ON TABLE"
418        EXTRACT_ALLOWS_QUOTES = False
419        NVL2_SUPPORTED = False
420
421        EXPRESSIONS_WITHOUT_NESTED_CTES = {
422            exp.Insert,
423            exp.Select,
424            exp.Subquery,
425            exp.Union,
426        }
427
428        TYPE_MAPPING = {
429            **generator.Generator.TYPE_MAPPING,
430            exp.DataType.Type.BIT: "BOOLEAN",
431            exp.DataType.Type.DATETIME: "TIMESTAMP",
432            exp.DataType.Type.TEXT: "STRING",
433            exp.DataType.Type.TIME: "TIMESTAMP",
434            exp.DataType.Type.TIMESTAMPTZ: "TIMESTAMP",
435            exp.DataType.Type.VARBINARY: "BINARY",
436        }
437
438        TRANSFORMS = {
439            **generator.Generator.TRANSFORMS,
440            exp.Group: transforms.preprocess([transforms.unalias_group]),
441            exp.Select: transforms.preprocess(
442                [
443                    transforms.eliminate_qualify,
444                    transforms.eliminate_distinct_on,
445                    transforms.unnest_to_explode,
446                ]
447            ),
448            exp.Property: _property_sql,
449            exp.AnyValue: rename_func("FIRST"),
450            exp.ApproxDistinct: approx_count_distinct_sql,
451            exp.ArgMax: arg_max_or_min_no_count("MAX_BY"),
452            exp.ArgMin: arg_max_or_min_no_count("MIN_BY"),
453            exp.ArrayConcat: rename_func("CONCAT"),
454            exp.ArrayJoin: lambda self, e: self.func("CONCAT_WS", e.expression, e.this),
455            exp.ArraySize: rename_func("SIZE"),
456            exp.ArraySort: _array_sort_sql,
457            exp.With: no_recursive_cte_sql,
458            exp.DateAdd: _add_date_sql,
459            exp.DateDiff: _date_diff_sql,
460            exp.DateStrToDate: rename_func("TO_DATE"),
461            exp.DateSub: _add_date_sql,
462            exp.DateToDi: lambda self, e: f"CAST(DATE_FORMAT({self.sql(e, 'this')}, {Hive.DATEINT_FORMAT}) AS INT)",
463            exp.DiToDate: lambda self, e: f"TO_DATE(CAST({self.sql(e, 'this')} AS STRING), {Hive.DATEINT_FORMAT})",
464            exp.FileFormatProperty: lambda self, e: f"STORED AS {self.sql(e, 'this') if isinstance(e.this, exp.InputOutputFormat) else e.name.upper()}",
465            exp.FromBase64: rename_func("UNBASE64"),
466            exp.If: if_sql(),
467            exp.ILike: no_ilike_sql,
468            exp.IsNan: rename_func("ISNAN"),
469            exp.JSONExtract: rename_func("GET_JSON_OBJECT"),
470            exp.JSONExtractScalar: rename_func("GET_JSON_OBJECT"),
471            exp.JSONFormat: _json_format_sql,
472            exp.Left: left_to_substring_sql,
473            exp.Map: var_map_sql,
474            exp.Max: max_or_greatest,
475            exp.MD5Digest: lambda self, e: self.func("UNHEX", self.func("MD5", e.this)),
476            exp.Min: min_or_least,
477            exp.MonthsBetween: lambda self, e: self.func("MONTHS_BETWEEN", e.this, e.expression),
478            exp.NotNullColumnConstraint: lambda self, e: ""
479            if e.args.get("allow_null")
480            else "NOT NULL",
481            exp.VarMap: var_map_sql,
482            exp.Create: _create_sql,
483            exp.Quantile: rename_func("PERCENTILE"),
484            exp.ApproxQuantile: rename_func("PERCENTILE_APPROX"),
485            exp.RegexpExtract: regexp_extract_sql,
486            exp.RegexpReplace: regexp_replace_sql,
487            exp.RegexpLike: lambda self, e: self.binary(e, "RLIKE"),
488            exp.RegexpSplit: rename_func("SPLIT"),
489            exp.Right: right_to_substring_sql,
490            exp.SafeDivide: no_safe_divide_sql,
491            exp.SchemaCommentProperty: lambda self, e: self.naked_property(e),
492            exp.SetAgg: rename_func("COLLECT_SET"),
493            exp.Split: lambda self, e: f"SPLIT({self.sql(e, 'this')}, CONCAT('\\\\Q', {self.sql(e, 'expression')}))",
494            exp.StrPosition: strposition_to_locate_sql,
495            exp.StrToDate: _str_to_date_sql,
496            exp.StrToTime: _str_to_time_sql,
497            exp.StrToUnix: _str_to_unix_sql,
498            exp.StructExtract: struct_extract_sql,
499            exp.TimeStrToDate: rename_func("TO_DATE"),
500            exp.TimeStrToTime: timestrtotime_sql,
501            exp.TimeStrToUnix: rename_func("UNIX_TIMESTAMP"),
502            exp.TimeToStr: _time_to_str,
503            exp.TimeToUnix: rename_func("UNIX_TIMESTAMP"),
504            exp.ToBase64: rename_func("BASE64"),
505            exp.TsOrDiToDi: lambda self, e: f"CAST(SUBSTR(REPLACE(CAST({self.sql(e, 'this')} AS STRING), '-', ''), 1, 8) AS INT)",
506            exp.TsOrDsAdd: lambda self, e: f"DATE_ADD({self.sql(e, 'this')}, {self.sql(e, 'expression')})",
507            exp.TsOrDsToDate: _to_date_sql,
508            exp.TryCast: no_trycast_sql,
509            exp.UnixToStr: lambda self, e: self.func(
510                "FROM_UNIXTIME", e.this, time_format("hive")(self, e)
511            ),
512            exp.UnixToTime: rename_func("FROM_UNIXTIME"),
513            exp.UnixToTimeStr: rename_func("FROM_UNIXTIME"),
514            exp.PartitionedByProperty: lambda self, e: f"PARTITIONED BY {self.sql(e, 'this')}",
515            exp.SerdeProperties: lambda self, e: self.properties(e, prefix="WITH SERDEPROPERTIES"),
516            exp.NumberToStr: rename_func("FORMAT_NUMBER"),
517            exp.LastDateOfMonth: rename_func("LAST_DAY"),
518            exp.National: lambda self, e: self.national_sql(e, prefix=""),
519            exp.ClusteredColumnConstraint: lambda self, e: f"({self.expressions(e, 'this', indent=False)})",
520            exp.NonClusteredColumnConstraint: lambda self, e: f"({self.expressions(e, 'this', indent=False)})",
521            exp.NotForReplicationColumnConstraint: lambda self, e: "",
522            exp.OnProperty: lambda self, e: "",
523            exp.PrimaryKeyColumnConstraint: lambda self, e: "PRIMARY KEY",
524        }
525
526        PROPERTIES_LOCATION = {
527            **generator.Generator.PROPERTIES_LOCATION,
528            exp.FileFormatProperty: exp.Properties.Location.POST_SCHEMA,
529            exp.PartitionedByProperty: exp.Properties.Location.POST_SCHEMA,
530            exp.VolatileProperty: exp.Properties.Location.UNSUPPORTED,
531            exp.WithDataProperty: exp.Properties.Location.UNSUPPORTED,
532        }
533
534        def temporary_storage_provider(self, expression: exp.Create) -> exp.Create:
535            # Hive has no temporary storage provider (there are hive settings though)
536            return expression
537
538        def parameter_sql(self, expression: exp.Parameter) -> str:
539            this = self.sql(expression, "this")
540            expression_sql = self.sql(expression, "expression")
541
542            parent = expression.parent
543            this = f"{this}:{expression_sql}" if expression_sql else this
544
545            if isinstance(parent, exp.EQ) and isinstance(parent.parent, exp.SetItem):
546                # We need to produce SET key = value instead of SET ${key} = value
547                return this
548
549            return f"${{{this}}}"
550
551        def schema_sql(self, expression: exp.Schema) -> str:
552            for ordered in expression.find_all(exp.Ordered):
553                if ordered.args.get("desc") is False:
554                    ordered.set("desc", None)
555
556            return super().schema_sql(expression)
557
558        def constraint_sql(self, expression: exp.Constraint) -> str:
559            for prop in list(expression.find_all(exp.Properties)):
560                prop.pop()
561
562            this = self.sql(expression, "this")
563            expressions = self.expressions(expression, sep=" ", flat=True)
564            return f"CONSTRAINT {this} {expressions}"
565
566        def rowformatserdeproperty_sql(self, expression: exp.RowFormatSerdeProperty) -> str:
567            serde_props = self.sql(expression, "serde_properties")
568            serde_props = f" {serde_props}" if serde_props else ""
569            return f"ROW FORMAT SERDE {self.sql(expression, 'this')}{serde_props}"
570
571        def arrayagg_sql(self, expression: exp.ArrayAgg) -> str:
572            return self.func(
573                "COLLECT_LIST",
574                expression.this.this if isinstance(expression.this, exp.Order) else expression.this,
575            )
576
577        def with_properties(self, properties: exp.Properties) -> str:
578            return self.properties(properties, prefix=self.seg("TBLPROPERTIES"))
579
580        def datatype_sql(self, expression: exp.DataType) -> str:
581            if (
582                expression.this in (exp.DataType.Type.VARCHAR, exp.DataType.Type.NVARCHAR)
583                and not expression.expressions
584            ):
585                expression = exp.DataType.build("text")
586            elif expression.this in exp.DataType.TEMPORAL_TYPES:
587                expression = exp.DataType.build(expression.this)
588            elif expression.is_type("float"):
589                size_expression = expression.find(exp.DataTypeParam)
590                if size_expression:
591                    size = int(size_expression.name)
592                    expression = (
593                        exp.DataType.build("float") if size <= 32 else exp.DataType.build("double")
594                    )
595
596            return super().datatype_sql(expression)
597
598        def version_sql(self, expression: exp.Version) -> str:
599            sql = super().version_sql(expression)
600            return sql.replace("FOR ", "", 1)
DATE_DELTA_INTERVAL = {'YEAR': ('ADD_MONTHS', 12), 'MONTH': ('ADD_MONTHS', 1), 'QUARTER': ('ADD_MONTHS', 3), 'WEEK': ('DATE_ADD', 7), 'DAY': ('DATE_ADD', 1)}
TIME_DIFF_FACTOR = {'MILLISECOND': ' * 1000', 'SECOND': '', 'MINUTE': ' / 60', 'HOUR': ' / 3600'}
DIFF_MONTH_SWITCH = ('YEAR', 'QUARTER', 'MONTH')
class Hive(sqlglot.dialects.dialect.Dialect):
187class Hive(Dialect):
188    ALIAS_POST_TABLESAMPLE = True
189    IDENTIFIERS_CAN_START_WITH_DIGIT = True
190    SUPPORTS_USER_DEFINED_TYPES = False
191    SAFE_DIVISION = True
192
193    # https://spark.apache.org/docs/latest/sql-ref-identifier.html#description
194    RESOLVES_IDENTIFIERS_AS_UPPERCASE = None
195
196    TIME_MAPPING = {
197        "y": "%Y",
198        "Y": "%Y",
199        "YYYY": "%Y",
200        "yyyy": "%Y",
201        "YY": "%y",
202        "yy": "%y",
203        "MMMM": "%B",
204        "MMM": "%b",
205        "MM": "%m",
206        "M": "%-m",
207        "dd": "%d",
208        "d": "%-d",
209        "HH": "%H",
210        "H": "%-H",
211        "hh": "%I",
212        "h": "%-I",
213        "mm": "%M",
214        "m": "%-M",
215        "ss": "%S",
216        "s": "%-S",
217        "SSSSSS": "%f",
218        "a": "%p",
219        "DD": "%j",
220        "D": "%-j",
221        "E": "%a",
222        "EE": "%a",
223        "EEE": "%a",
224        "EEEE": "%A",
225    }
226
227    DATE_FORMAT = "'yyyy-MM-dd'"
228    DATEINT_FORMAT = "'yyyyMMdd'"
229    TIME_FORMAT = "'yyyy-MM-dd HH:mm:ss'"
230
231    class Tokenizer(tokens.Tokenizer):
232        QUOTES = ["'", '"']
233        IDENTIFIERS = ["`"]
234        STRING_ESCAPES = ["\\"]
235        ENCODE = "utf-8"
236
237        SINGLE_TOKENS = {
238            **tokens.Tokenizer.SINGLE_TOKENS,
239            "$": TokenType.PARAMETER,
240        }
241
242        KEYWORDS = {
243            **tokens.Tokenizer.KEYWORDS,
244            "ADD ARCHIVE": TokenType.COMMAND,
245            "ADD ARCHIVES": TokenType.COMMAND,
246            "ADD FILE": TokenType.COMMAND,
247            "ADD FILES": TokenType.COMMAND,
248            "ADD JAR": TokenType.COMMAND,
249            "ADD JARS": TokenType.COMMAND,
250            "MSCK REPAIR": TokenType.COMMAND,
251            "REFRESH": TokenType.REFRESH,
252            "TIMESTAMP AS OF": TokenType.TIMESTAMP_SNAPSHOT,
253            "VERSION AS OF": TokenType.VERSION_SNAPSHOT,
254            "WITH SERDEPROPERTIES": TokenType.SERDE_PROPERTIES,
255        }
256
257        NUMERIC_LITERALS = {
258            "L": "BIGINT",
259            "S": "SMALLINT",
260            "Y": "TINYINT",
261            "D": "DOUBLE",
262            "F": "FLOAT",
263            "BD": "DECIMAL",
264        }
265
266    class Parser(parser.Parser):
267        LOG_DEFAULTS_TO_LN = True
268        STRICT_CAST = False
269
270        FUNCTIONS = {
271            **parser.Parser.FUNCTIONS,
272            "BASE64": exp.ToBase64.from_arg_list,
273            "COLLECT_LIST": exp.ArrayAgg.from_arg_list,
274            "COLLECT_SET": exp.SetAgg.from_arg_list,
275            "DATE_ADD": lambda args: exp.TsOrDsAdd(
276                this=seq_get(args, 0), expression=seq_get(args, 1), unit=exp.Literal.string("DAY")
277            ),
278            "DATE_FORMAT": lambda args: format_time_lambda(exp.TimeToStr, "hive")(
279                [
280                    exp.TimeStrToTime(this=seq_get(args, 0)),
281                    seq_get(args, 1),
282                ]
283            ),
284            "DATE_SUB": lambda args: exp.TsOrDsAdd(
285                this=seq_get(args, 0),
286                expression=exp.Mul(this=seq_get(args, 1), expression=exp.Literal.number(-1)),
287                unit=exp.Literal.string("DAY"),
288            ),
289            "DATEDIFF": lambda args: exp.DateDiff(
290                this=exp.TsOrDsToDate(this=seq_get(args, 0)),
291                expression=exp.TsOrDsToDate(this=seq_get(args, 1)),
292            ),
293            "DAY": lambda args: exp.Day(this=exp.TsOrDsToDate(this=seq_get(args, 0))),
294            "FROM_UNIXTIME": format_time_lambda(exp.UnixToStr, "hive", True),
295            "GET_JSON_OBJECT": exp.JSONExtractScalar.from_arg_list,
296            "LOCATE": locate_to_strposition,
297            "MAP": parse_var_map,
298            "MONTH": lambda args: exp.Month(this=exp.TsOrDsToDate.from_arg_list(args)),
299            "PERCENTILE": exp.Quantile.from_arg_list,
300            "PERCENTILE_APPROX": exp.ApproxQuantile.from_arg_list,
301            "REGEXP_EXTRACT": lambda args: exp.RegexpExtract(
302                this=seq_get(args, 0), expression=seq_get(args, 1), group=seq_get(args, 2)
303            ),
304            "SIZE": exp.ArraySize.from_arg_list,
305            "SPLIT": exp.RegexpSplit.from_arg_list,
306            "STR_TO_MAP": lambda args: exp.StrToMap(
307                this=seq_get(args, 0),
308                pair_delim=seq_get(args, 1) or exp.Literal.string(","),
309                key_value_delim=seq_get(args, 2) or exp.Literal.string(":"),
310            ),
311            "TO_DATE": format_time_lambda(exp.TsOrDsToDate, "hive"),
312            "TO_JSON": exp.JSONFormat.from_arg_list,
313            "UNBASE64": exp.FromBase64.from_arg_list,
314            "UNIX_TIMESTAMP": format_time_lambda(exp.StrToUnix, "hive", True),
315            "YEAR": lambda args: exp.Year(this=exp.TsOrDsToDate.from_arg_list(args)),
316        }
317
318        NO_PAREN_FUNCTION_PARSERS = {
319            **parser.Parser.NO_PAREN_FUNCTION_PARSERS,
320            "TRANSFORM": lambda self: self._parse_transform(),
321        }
322
323        PROPERTY_PARSERS = {
324            **parser.Parser.PROPERTY_PARSERS,
325            "WITH SERDEPROPERTIES": lambda self: exp.SerdeProperties(
326                expressions=self._parse_wrapped_csv(self._parse_property)
327            ),
328        }
329
330        def _parse_transform(self) -> t.Optional[exp.Transform | exp.QueryTransform]:
331            if not self._match(TokenType.L_PAREN, advance=False):
332                self._retreat(self._index - 1)
333                return None
334
335            args = self._parse_wrapped_csv(self._parse_lambda)
336            row_format_before = self._parse_row_format(match_row=True)
337
338            record_writer = None
339            if self._match_text_seq("RECORDWRITER"):
340                record_writer = self._parse_string()
341
342            if not self._match(TokenType.USING):
343                return exp.Transform.from_arg_list(args)
344
345            command_script = self._parse_string()
346
347            self._match(TokenType.ALIAS)
348            schema = self._parse_schema()
349
350            row_format_after = self._parse_row_format(match_row=True)
351            record_reader = None
352            if self._match_text_seq("RECORDREADER"):
353                record_reader = self._parse_string()
354
355            return self.expression(
356                exp.QueryTransform,
357                expressions=args,
358                command_script=command_script,
359                schema=schema,
360                row_format_before=row_format_before,
361                record_writer=record_writer,
362                row_format_after=row_format_after,
363                record_reader=record_reader,
364            )
365
366        def _parse_types(
367            self, check_func: bool = False, schema: bool = False, allow_identifiers: bool = True
368        ) -> t.Optional[exp.Expression]:
369            """
370            Spark (and most likely Hive) treats casts to CHAR(length) and VARCHAR(length) as casts to
371            STRING in all contexts except for schema definitions. For example, this is in Spark v3.4.0:
372
373                spark-sql (default)> select cast(1234 as varchar(2));
374                23/06/06 15:51:18 WARN CharVarcharUtils: The Spark cast operator does not support
375                char/varchar type and simply treats them as string type. Please use string type
376                directly to avoid confusion. Otherwise, you can set spark.sql.legacy.charVarcharAsString
377                to true, so that Spark treat them as string type as same as Spark 3.0 and earlier
378
379                1234
380                Time taken: 4.265 seconds, Fetched 1 row(s)
381
382            This shows that Spark doesn't truncate the value into '12', which is inconsistent with
383            what other dialects (e.g. postgres) do, so we need to drop the length to transpile correctly.
384
385            Reference: https://spark.apache.org/docs/latest/sql-ref-datatypes.html
386            """
387            this = super()._parse_types(
388                check_func=check_func, schema=schema, allow_identifiers=allow_identifiers
389            )
390
391            if this and not schema:
392                return this.transform(
393                    lambda node: node.replace(exp.DataType.build("text"))
394                    if isinstance(node, exp.DataType) and node.is_type("char", "varchar")
395                    else node,
396                    copy=False,
397                )
398
399            return this
400
401        def _parse_partition_and_order(
402            self,
403        ) -> t.Tuple[t.List[exp.Expression], t.Optional[exp.Expression]]:
404            return (
405                self._parse_csv(self._parse_conjunction)
406                if self._match_set({TokenType.PARTITION_BY, TokenType.DISTRIBUTE_BY})
407                else [],
408                super()._parse_order(skip_order_token=self._match(TokenType.SORT_BY)),
409            )
410
411    class Generator(generator.Generator):
412        LIMIT_FETCH = "LIMIT"
413        TABLESAMPLE_WITH_METHOD = False
414        TABLESAMPLE_SIZE_IS_PERCENT = True
415        JOIN_HINTS = False
416        TABLE_HINTS = False
417        QUERY_HINTS = False
418        INDEX_ON = "ON TABLE"
419        EXTRACT_ALLOWS_QUOTES = False
420        NVL2_SUPPORTED = False
421
422        EXPRESSIONS_WITHOUT_NESTED_CTES = {
423            exp.Insert,
424            exp.Select,
425            exp.Subquery,
426            exp.Union,
427        }
428
429        TYPE_MAPPING = {
430            **generator.Generator.TYPE_MAPPING,
431            exp.DataType.Type.BIT: "BOOLEAN",
432            exp.DataType.Type.DATETIME: "TIMESTAMP",
433            exp.DataType.Type.TEXT: "STRING",
434            exp.DataType.Type.TIME: "TIMESTAMP",
435            exp.DataType.Type.TIMESTAMPTZ: "TIMESTAMP",
436            exp.DataType.Type.VARBINARY: "BINARY",
437        }
438
439        TRANSFORMS = {
440            **generator.Generator.TRANSFORMS,
441            exp.Group: transforms.preprocess([transforms.unalias_group]),
442            exp.Select: transforms.preprocess(
443                [
444                    transforms.eliminate_qualify,
445                    transforms.eliminate_distinct_on,
446                    transforms.unnest_to_explode,
447                ]
448            ),
449            exp.Property: _property_sql,
450            exp.AnyValue: rename_func("FIRST"),
451            exp.ApproxDistinct: approx_count_distinct_sql,
452            exp.ArgMax: arg_max_or_min_no_count("MAX_BY"),
453            exp.ArgMin: arg_max_or_min_no_count("MIN_BY"),
454            exp.ArrayConcat: rename_func("CONCAT"),
455            exp.ArrayJoin: lambda self, e: self.func("CONCAT_WS", e.expression, e.this),
456            exp.ArraySize: rename_func("SIZE"),
457            exp.ArraySort: _array_sort_sql,
458            exp.With: no_recursive_cte_sql,
459            exp.DateAdd: _add_date_sql,
460            exp.DateDiff: _date_diff_sql,
461            exp.DateStrToDate: rename_func("TO_DATE"),
462            exp.DateSub: _add_date_sql,
463            exp.DateToDi: lambda self, e: f"CAST(DATE_FORMAT({self.sql(e, 'this')}, {Hive.DATEINT_FORMAT}) AS INT)",
464            exp.DiToDate: lambda self, e: f"TO_DATE(CAST({self.sql(e, 'this')} AS STRING), {Hive.DATEINT_FORMAT})",
465            exp.FileFormatProperty: lambda self, e: f"STORED AS {self.sql(e, 'this') if isinstance(e.this, exp.InputOutputFormat) else e.name.upper()}",
466            exp.FromBase64: rename_func("UNBASE64"),
467            exp.If: if_sql(),
468            exp.ILike: no_ilike_sql,
469            exp.IsNan: rename_func("ISNAN"),
470            exp.JSONExtract: rename_func("GET_JSON_OBJECT"),
471            exp.JSONExtractScalar: rename_func("GET_JSON_OBJECT"),
472            exp.JSONFormat: _json_format_sql,
473            exp.Left: left_to_substring_sql,
474            exp.Map: var_map_sql,
475            exp.Max: max_or_greatest,
476            exp.MD5Digest: lambda self, e: self.func("UNHEX", self.func("MD5", e.this)),
477            exp.Min: min_or_least,
478            exp.MonthsBetween: lambda self, e: self.func("MONTHS_BETWEEN", e.this, e.expression),
479            exp.NotNullColumnConstraint: lambda self, e: ""
480            if e.args.get("allow_null")
481            else "NOT NULL",
482            exp.VarMap: var_map_sql,
483            exp.Create: _create_sql,
484            exp.Quantile: rename_func("PERCENTILE"),
485            exp.ApproxQuantile: rename_func("PERCENTILE_APPROX"),
486            exp.RegexpExtract: regexp_extract_sql,
487            exp.RegexpReplace: regexp_replace_sql,
488            exp.RegexpLike: lambda self, e: self.binary(e, "RLIKE"),
489            exp.RegexpSplit: rename_func("SPLIT"),
490            exp.Right: right_to_substring_sql,
491            exp.SafeDivide: no_safe_divide_sql,
492            exp.SchemaCommentProperty: lambda self, e: self.naked_property(e),
493            exp.SetAgg: rename_func("COLLECT_SET"),
494            exp.Split: lambda self, e: f"SPLIT({self.sql(e, 'this')}, CONCAT('\\\\Q', {self.sql(e, 'expression')}))",
495            exp.StrPosition: strposition_to_locate_sql,
496            exp.StrToDate: _str_to_date_sql,
497            exp.StrToTime: _str_to_time_sql,
498            exp.StrToUnix: _str_to_unix_sql,
499            exp.StructExtract: struct_extract_sql,
500            exp.TimeStrToDate: rename_func("TO_DATE"),
501            exp.TimeStrToTime: timestrtotime_sql,
502            exp.TimeStrToUnix: rename_func("UNIX_TIMESTAMP"),
503            exp.TimeToStr: _time_to_str,
504            exp.TimeToUnix: rename_func("UNIX_TIMESTAMP"),
505            exp.ToBase64: rename_func("BASE64"),
506            exp.TsOrDiToDi: lambda self, e: f"CAST(SUBSTR(REPLACE(CAST({self.sql(e, 'this')} AS STRING), '-', ''), 1, 8) AS INT)",
507            exp.TsOrDsAdd: lambda self, e: f"DATE_ADD({self.sql(e, 'this')}, {self.sql(e, 'expression')})",
508            exp.TsOrDsToDate: _to_date_sql,
509            exp.TryCast: no_trycast_sql,
510            exp.UnixToStr: lambda self, e: self.func(
511                "FROM_UNIXTIME", e.this, time_format("hive")(self, e)
512            ),
513            exp.UnixToTime: rename_func("FROM_UNIXTIME"),
514            exp.UnixToTimeStr: rename_func("FROM_UNIXTIME"),
515            exp.PartitionedByProperty: lambda self, e: f"PARTITIONED BY {self.sql(e, 'this')}",
516            exp.SerdeProperties: lambda self, e: self.properties(e, prefix="WITH SERDEPROPERTIES"),
517            exp.NumberToStr: rename_func("FORMAT_NUMBER"),
518            exp.LastDateOfMonth: rename_func("LAST_DAY"),
519            exp.National: lambda self, e: self.national_sql(e, prefix=""),
520            exp.ClusteredColumnConstraint: lambda self, e: f"({self.expressions(e, 'this', indent=False)})",
521            exp.NonClusteredColumnConstraint: lambda self, e: f"({self.expressions(e, 'this', indent=False)})",
522            exp.NotForReplicationColumnConstraint: lambda self, e: "",
523            exp.OnProperty: lambda self, e: "",
524            exp.PrimaryKeyColumnConstraint: lambda self, e: "PRIMARY KEY",
525        }
526
527        PROPERTIES_LOCATION = {
528            **generator.Generator.PROPERTIES_LOCATION,
529            exp.FileFormatProperty: exp.Properties.Location.POST_SCHEMA,
530            exp.PartitionedByProperty: exp.Properties.Location.POST_SCHEMA,
531            exp.VolatileProperty: exp.Properties.Location.UNSUPPORTED,
532            exp.WithDataProperty: exp.Properties.Location.UNSUPPORTED,
533        }
534
535        def temporary_storage_provider(self, expression: exp.Create) -> exp.Create:
536            # Hive has no temporary storage provider (there are hive settings though)
537            return expression
538
539        def parameter_sql(self, expression: exp.Parameter) -> str:
540            this = self.sql(expression, "this")
541            expression_sql = self.sql(expression, "expression")
542
543            parent = expression.parent
544            this = f"{this}:{expression_sql}" if expression_sql else this
545
546            if isinstance(parent, exp.EQ) and isinstance(parent.parent, exp.SetItem):
547                # We need to produce SET key = value instead of SET ${key} = value
548                return this
549
550            return f"${{{this}}}"
551
552        def schema_sql(self, expression: exp.Schema) -> str:
553            for ordered in expression.find_all(exp.Ordered):
554                if ordered.args.get("desc") is False:
555                    ordered.set("desc", None)
556
557            return super().schema_sql(expression)
558
559        def constraint_sql(self, expression: exp.Constraint) -> str:
560            for prop in list(expression.find_all(exp.Properties)):
561                prop.pop()
562
563            this = self.sql(expression, "this")
564            expressions = self.expressions(expression, sep=" ", flat=True)
565            return f"CONSTRAINT {this} {expressions}"
566
567        def rowformatserdeproperty_sql(self, expression: exp.RowFormatSerdeProperty) -> str:
568            serde_props = self.sql(expression, "serde_properties")
569            serde_props = f" {serde_props}" if serde_props else ""
570            return f"ROW FORMAT SERDE {self.sql(expression, 'this')}{serde_props}"
571
572        def arrayagg_sql(self, expression: exp.ArrayAgg) -> str:
573            return self.func(
574                "COLLECT_LIST",
575                expression.this.this if isinstance(expression.this, exp.Order) else expression.this,
576            )
577
578        def with_properties(self, properties: exp.Properties) -> str:
579            return self.properties(properties, prefix=self.seg("TBLPROPERTIES"))
580
581        def datatype_sql(self, expression: exp.DataType) -> str:
582            if (
583                expression.this in (exp.DataType.Type.VARCHAR, exp.DataType.Type.NVARCHAR)
584                and not expression.expressions
585            ):
586                expression = exp.DataType.build("text")
587            elif expression.this in exp.DataType.TEMPORAL_TYPES:
588                expression = exp.DataType.build(expression.this)
589            elif expression.is_type("float"):
590                size_expression = expression.find(exp.DataTypeParam)
591                if size_expression:
592                    size = int(size_expression.name)
593                    expression = (
594                        exp.DataType.build("float") if size <= 32 else exp.DataType.build("double")
595                    )
596
597            return super().datatype_sql(expression)
598
599        def version_sql(self, expression: exp.Version) -> str:
600            sql = super().version_sql(expression)
601            return sql.replace("FOR ", "", 1)
ALIAS_POST_TABLESAMPLE = True
IDENTIFIERS_CAN_START_WITH_DIGIT = True
SUPPORTS_USER_DEFINED_TYPES = False
SAFE_DIVISION = True
RESOLVES_IDENTIFIERS_AS_UPPERCASE: Optional[bool] = None
TIME_MAPPING: Dict[str, str] = {'y': '%Y', 'Y': '%Y', 'YYYY': '%Y', 'yyyy': '%Y', 'YY': '%y', 'yy': '%y', 'MMMM': '%B', 'MMM': '%b', 'MM': '%m', 'M': '%-m', 'dd': '%d', 'd': '%-d', 'HH': '%H', 'H': '%-H', 'hh': '%I', 'h': '%-I', 'mm': '%M', 'm': '%-M', 'ss': '%S', 's': '%-S', 'SSSSSS': '%f', 'a': '%p', 'DD': '%j', 'D': '%-j', 'E': '%a', 'EE': '%a', 'EEE': '%a', 'EEEE': '%A'}
DATE_FORMAT = "'yyyy-MM-dd'"
DATEINT_FORMAT = "'yyyyMMdd'"
TIME_FORMAT = "'yyyy-MM-dd HH:mm:ss'"
tokenizer_class = <class 'Hive.Tokenizer'>
parser_class = <class 'Hive.Parser'>
generator_class = <class 'Hive.Generator'>
TIME_TRIE: Dict = {'y': {0: True, 'y': {'y': {'y': {0: True}}, 0: True}}, 'Y': {0: True, 'Y': {'Y': {'Y': {0: True}}, 0: True}}, 'M': {'M': {'M': {'M': {0: True}, 0: True}, 0: True}, 0: True}, 'd': {'d': {0: True}, 0: True}, 'H': {'H': {0: True}, 0: True}, 'h': {'h': {0: True}, 0: True}, 'm': {'m': {0: True}, 0: True}, 's': {'s': {0: True}, 0: True}, 'S': {'S': {'S': {'S': {'S': {'S': {0: True}}}}}}, 'a': {0: True}, 'D': {'D': {0: True}, 0: True}, 'E': {0: True, 'E': {0: True, 'E': {0: True, 'E': {0: True}}}}}
FORMAT_TRIE: Dict = {'y': {0: True, 'y': {'y': {'y': {0: True}}, 0: True}}, 'Y': {0: True, 'Y': {'Y': {'Y': {0: True}}, 0: True}}, 'M': {'M': {'M': {'M': {0: True}, 0: True}, 0: True}, 0: True}, 'd': {'d': {0: True}, 0: True}, 'H': {'H': {0: True}, 0: True}, 'h': {'h': {0: True}, 0: True}, 'm': {'m': {0: True}, 0: True}, 's': {'s': {0: True}, 0: True}, 'S': {'S': {'S': {'S': {'S': {'S': {0: True}}}}}}, 'a': {0: True}, 'D': {'D': {0: True}, 0: True}, 'E': {0: True, 'E': {0: True, 'E': {0: True, 'E': {0: True}}}}}
INVERSE_TIME_MAPPING: Dict[str, str] = {'%Y': 'yyyy', '%y': 'yy', '%B': 'MMMM', '%b': 'MMM', '%m': 'MM', '%-m': 'M', '%d': 'dd', '%-d': 'd', '%H': 'HH', '%-H': 'H', '%I': 'hh', '%-I': 'h', '%M': 'mm', '%-M': 'm', '%S': 'ss', '%-S': 's', '%f': 'SSSSSS', '%p': 'a', '%j': 'DD', '%-j': 'D', '%a': 'EEE', '%A': 'EEEE'}
INVERSE_TIME_TRIE: Dict = {'%': {'Y': {0: True}, 'y': {0: True}, 'B': {0: True}, 'b': {0: True}, 'm': {0: True}, '-': {'m': {0: True}, 'd': {0: True}, 'H': {0: True}, 'I': {0: True}, 'M': {0: True}, 'S': {0: True}, 'j': {0: True}}, 'd': {0: True}, 'H': {0: True}, 'I': {0: True}, 'M': {0: True}, 'S': {0: True}, 'f': {0: True}, 'p': {0: True}, 'j': {0: True}, 'a': {0: True}, 'A': {0: True}}}
INVERSE_ESCAPE_SEQUENCES: Dict[str, str] = {}
QUOTE_START = "'"
QUOTE_END = "'"
IDENTIFIER_START = '`'
IDENTIFIER_END = '`'
BIT_START = None
BIT_END = None
HEX_START = None
HEX_END = None
BYTE_START = None
BYTE_END = None
class Hive.Tokenizer(sqlglot.tokens.Tokenizer):
231    class Tokenizer(tokens.Tokenizer):
232        QUOTES = ["'", '"']
233        IDENTIFIERS = ["`"]
234        STRING_ESCAPES = ["\\"]
235        ENCODE = "utf-8"
236
237        SINGLE_TOKENS = {
238            **tokens.Tokenizer.SINGLE_TOKENS,
239            "$": TokenType.PARAMETER,
240        }
241
242        KEYWORDS = {
243            **tokens.Tokenizer.KEYWORDS,
244            "ADD ARCHIVE": TokenType.COMMAND,
245            "ADD ARCHIVES": TokenType.COMMAND,
246            "ADD FILE": TokenType.COMMAND,
247            "ADD FILES": TokenType.COMMAND,
248            "ADD JAR": TokenType.COMMAND,
249            "ADD JARS": TokenType.COMMAND,
250            "MSCK REPAIR": TokenType.COMMAND,
251            "REFRESH": TokenType.REFRESH,
252            "TIMESTAMP AS OF": TokenType.TIMESTAMP_SNAPSHOT,
253            "VERSION AS OF": TokenType.VERSION_SNAPSHOT,
254            "WITH SERDEPROPERTIES": TokenType.SERDE_PROPERTIES,
255        }
256
257        NUMERIC_LITERALS = {
258            "L": "BIGINT",
259            "S": "SMALLINT",
260            "Y": "TINYINT",
261            "D": "DOUBLE",
262            "F": "FLOAT",
263            "BD": "DECIMAL",
264        }
QUOTES = ["'", '"']
IDENTIFIERS = ['`']
STRING_ESCAPES = ['\\']
ENCODE = 'utf-8'
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'>}
KEYWORDS = {'{%': <TokenType.BLOCK_START: 'BLOCK_START'>, '{%+': <TokenType.BLOCK_START: 'BLOCK_START'>, '{%-': <TokenType.BLOCK_START: 'BLOCK_START'>, '%}': <TokenType.BLOCK_END: 'BLOCK_END'>, '+%}': <TokenType.BLOCK_END: 'BLOCK_END'>, '-%}': <TokenType.BLOCK_END: 'BLOCK_END'>, '{{+': <TokenType.BLOCK_START: 'BLOCK_START'>, '{{-': <TokenType.BLOCK_START: 'BLOCK_START'>, '+}}': <TokenType.BLOCK_END: 'BLOCK_END'>, '-}}': <TokenType.BLOCK_END: 'BLOCK_END'>, '/*+': <TokenType.HINT: 'HINT'>, '==': <TokenType.EQ: 'EQ'>, '::': <TokenType.DCOLON: 'DCOLON'>, '||': <TokenType.DPIPE: 'DPIPE'>, '>=': <TokenType.GTE: 'GTE'>, '<=': <TokenType.LTE: 'LTE'>, '<>': <TokenType.NEQ: 'NEQ'>, '!=': <TokenType.NEQ: 'NEQ'>, '<=>': <TokenType.NULLSAFE_EQ: 'NULLSAFE_EQ'>, '->': <TokenType.ARROW: 'ARROW'>, '->>': <TokenType.DARROW: 'DARROW'>, '=>': <TokenType.FARROW: 'FARROW'>, '#>': <TokenType.HASH_ARROW: 'HASH_ARROW'>, '#>>': <TokenType.DHASH_ARROW: 'DHASH_ARROW'>, '<->': <TokenType.LR_ARROW: 'LR_ARROW'>, '&&': <TokenType.DAMP: 'DAMP'>, '??': <TokenType.DQMARK: 'DQMARK'>, 'ALL': <TokenType.ALL: 'ALL'>, 'ALWAYS': <TokenType.ALWAYS: 'ALWAYS'>, 'AND': <TokenType.AND: 'AND'>, 'ANTI': <TokenType.ANTI: 'ANTI'>, 'ANY': <TokenType.ANY: 'ANY'>, 'ASC': <TokenType.ASC: 'ASC'>, 'AS': <TokenType.ALIAS: 'ALIAS'>, 'ASOF': <TokenType.ASOF: 'ASOF'>, 'AUTOINCREMENT': <TokenType.AUTO_INCREMENT: 'AUTO_INCREMENT'>, 'AUTO_INCREMENT': <TokenType.AUTO_INCREMENT: 'AUTO_INCREMENT'>, 'BEGIN': <TokenType.BEGIN: 'BEGIN'>, 'BETWEEN': <TokenType.BETWEEN: 'BETWEEN'>, 'CACHE': <TokenType.CACHE: 'CACHE'>, 'UNCACHE': <TokenType.UNCACHE: 'UNCACHE'>, 'CASE': <TokenType.CASE: 'CASE'>, 'CHARACTER SET': <TokenType.CHARACTER_SET: 'CHARACTER_SET'>, 'CLUSTER BY': <TokenType.CLUSTER_BY: 'CLUSTER_BY'>, 'COLLATE': <TokenType.COLLATE: 'COLLATE'>, 'COLUMN': <TokenType.COLUMN: 'COLUMN'>, 'COMMIT': <TokenType.COMMIT: 'COMMIT'>, 'CONNECT BY': <TokenType.CONNECT_BY: 'CONNECT_BY'>, 'CONSTRAINT': <TokenType.CONSTRAINT: 'CONSTRAINT'>, 'CREATE': <TokenType.CREATE: 'CREATE'>, 'CROSS': <TokenType.CROSS: 'CROSS'>, 'CUBE': <TokenType.CUBE: 'CUBE'>, 'CURRENT_DATE': <TokenType.CURRENT_DATE: 'CURRENT_DATE'>, 'CURRENT_TIME': <TokenType.CURRENT_TIME: 'CURRENT_TIME'>, 'CURRENT_TIMESTAMP': <TokenType.CURRENT_TIMESTAMP: 'CURRENT_TIMESTAMP'>, 'CURRENT_USER': <TokenType.CURRENT_USER: 'CURRENT_USER'>, 'DATABASE': <TokenType.DATABASE: 'DATABASE'>, 'DEFAULT': <TokenType.DEFAULT: 'DEFAULT'>, 'DELETE': <TokenType.DELETE: 'DELETE'>, 'DESC': <TokenType.DESC: 'DESC'>, 'DESCRIBE': <TokenType.DESCRIBE: 'DESCRIBE'>, 'DISTINCT': <TokenType.DISTINCT: 'DISTINCT'>, 'DISTRIBUTE BY': <TokenType.DISTRIBUTE_BY: 'DISTRIBUTE_BY'>, 'DIV': <TokenType.DIV: 'DIV'>, 'DROP': <TokenType.DROP: 'DROP'>, 'ELSE': <TokenType.ELSE: 'ELSE'>, 'END': <TokenType.END: 'END'>, 'ESCAPE': <TokenType.ESCAPE: 'ESCAPE'>, 'EXCEPT': <TokenType.EXCEPT: 'EXCEPT'>, 'EXECUTE': <TokenType.EXECUTE: 'EXECUTE'>, 'EXISTS': <TokenType.EXISTS: 'EXISTS'>, 'FALSE': <TokenType.FALSE: 'FALSE'>, 'FETCH': <TokenType.FETCH: 'FETCH'>, 'FILTER': <TokenType.FILTER: 'FILTER'>, 'FIRST': <TokenType.FIRST: 'FIRST'>, 'FULL': <TokenType.FULL: 'FULL'>, 'FUNCTION': <TokenType.FUNCTION: 'FUNCTION'>, 'FOR': <TokenType.FOR: 'FOR'>, 'FOREIGN KEY': <TokenType.FOREIGN_KEY: 'FOREIGN_KEY'>, 'FORMAT': <TokenType.FORMAT: 'FORMAT'>, 'FROM': <TokenType.FROM: 'FROM'>, 'GEOGRAPHY': <TokenType.GEOGRAPHY: 'GEOGRAPHY'>, 'GEOMETRY': <TokenType.GEOMETRY: 'GEOMETRY'>, 'GLOB': <TokenType.GLOB: 'GLOB'>, 'GROUP BY': <TokenType.GROUP_BY: 'GROUP_BY'>, 'GROUPING SETS': <TokenType.GROUPING_SETS: 'GROUPING_SETS'>, 'HAVING': <TokenType.HAVING: 'HAVING'>, 'ILIKE': <TokenType.ILIKE: 'ILIKE'>, 'IN': <TokenType.IN: 'IN'>, 'INDEX': <TokenType.INDEX: 'INDEX'>, 'INET': <TokenType.INET: 'INET'>, 'INNER': <TokenType.INNER: 'INNER'>, 'INSERT': <TokenType.INSERT: 'INSERT'>, 'INTERVAL': <TokenType.INTERVAL: 'INTERVAL'>, 'INTERSECT': <TokenType.INTERSECT: 'INTERSECT'>, 'INTO': <TokenType.INTO: 'INTO'>, 'IS': <TokenType.IS: 'IS'>, 'ISNULL': <TokenType.ISNULL: 'ISNULL'>, 'JOIN': <TokenType.JOIN: 'JOIN'>, 'KEEP': <TokenType.KEEP: 'KEEP'>, 'KILL': <TokenType.KILL: 'KILL'>, 'LATERAL': <TokenType.LATERAL: 'LATERAL'>, 'LEFT': <TokenType.LEFT: 'LEFT'>, 'LIKE': <TokenType.LIKE: 'LIKE'>, 'LIMIT': <TokenType.LIMIT: 'LIMIT'>, 'LOAD': <TokenType.LOAD: 'LOAD'>, 'LOCK': <TokenType.LOCK: 'LOCK'>, 'MERGE': <TokenType.MERGE: 'MERGE'>, 'NATURAL': <TokenType.NATURAL: 'NATURAL'>, 'NEXT': <TokenType.NEXT: 'NEXT'>, 'NOT': <TokenType.NOT: 'NOT'>, 'NOTNULL': <TokenType.NOTNULL: 'NOTNULL'>, 'NULL': <TokenType.NULL: 'NULL'>, 'OBJECT': <TokenType.OBJECT: 'OBJECT'>, 'OFFSET': <TokenType.OFFSET: 'OFFSET'>, 'ON': <TokenType.ON: 'ON'>, 'OR': <TokenType.OR: 'OR'>, 'XOR': <TokenType.XOR: 'XOR'>, 'ORDER BY': <TokenType.ORDER_BY: 'ORDER_BY'>, 'ORDINALITY': <TokenType.ORDINALITY: 'ORDINALITY'>, 'OUTER': <TokenType.OUTER: 'OUTER'>, 'OVER': <TokenType.OVER: 'OVER'>, 'OVERLAPS': <TokenType.OVERLAPS: 'OVERLAPS'>, 'OVERWRITE': <TokenType.OVERWRITE: 'OVERWRITE'>, 'PARTITION': <TokenType.PARTITION: 'PARTITION'>, 'PARTITION BY': <TokenType.PARTITION_BY: 'PARTITION_BY'>, 'PARTITIONED BY': <TokenType.PARTITION_BY: 'PARTITION_BY'>, 'PARTITIONED_BY': <TokenType.PARTITION_BY: 'PARTITION_BY'>, 'PERCENT': <TokenType.PERCENT: 'PERCENT'>, 'PIVOT': <TokenType.PIVOT: 'PIVOT'>, 'PRAGMA': <TokenType.PRAGMA: 'PRAGMA'>, 'PRIMARY KEY': <TokenType.PRIMARY_KEY: 'PRIMARY_KEY'>, 'PROCEDURE': <TokenType.PROCEDURE: 'PROCEDURE'>, 'QUALIFY': <TokenType.QUALIFY: 'QUALIFY'>, 'RANGE': <TokenType.RANGE: 'RANGE'>, 'RECURSIVE': <TokenType.RECURSIVE: 'RECURSIVE'>, 'REGEXP': <TokenType.RLIKE: 'RLIKE'>, 'REPLACE': <TokenType.REPLACE: 'REPLACE'>, 'RETURNING': <TokenType.RETURNING: 'RETURNING'>, 'REFERENCES': <TokenType.REFERENCES: 'REFERENCES'>, 'RIGHT': <TokenType.RIGHT: 'RIGHT'>, 'RLIKE': <TokenType.RLIKE: 'RLIKE'>, 'ROLLBACK': <TokenType.ROLLBACK: 'ROLLBACK'>, 'ROLLUP': <TokenType.ROLLUP: 'ROLLUP'>, 'ROW': <TokenType.ROW: 'ROW'>, 'ROWS': <TokenType.ROWS: 'ROWS'>, 'SCHEMA': <TokenType.SCHEMA: 'SCHEMA'>, 'SELECT': <TokenType.SELECT: 'SELECT'>, 'SEMI': <TokenType.SEMI: 'SEMI'>, 'SET': <TokenType.SET: 'SET'>, 'SETTINGS': <TokenType.SETTINGS: 'SETTINGS'>, 'SHOW': <TokenType.SHOW: 'SHOW'>, 'SIMILAR TO': <TokenType.SIMILAR_TO: 'SIMILAR_TO'>, 'SOME': <TokenType.SOME: 'SOME'>, 'SORT BY': <TokenType.SORT_BY: 'SORT_BY'>, 'START WITH': <TokenType.START_WITH: 'START_WITH'>, 'TABLE': <TokenType.TABLE: 'TABLE'>, 'TABLESAMPLE': <TokenType.TABLE_SAMPLE: 'TABLE_SAMPLE'>, 'TEMP': <TokenType.TEMPORARY: 'TEMPORARY'>, 'TEMPORARY': <TokenType.TEMPORARY: 'TEMPORARY'>, 'THEN': <TokenType.THEN: 'THEN'>, 'TRUE': <TokenType.TRUE: 'TRUE'>, 'UNION': <TokenType.UNION: 'UNION'>, 'UNKNOWN': <TokenType.UNKNOWN: 'UNKNOWN'>, 'UNNEST': <TokenType.UNNEST: 'UNNEST'>, 'UNPIVOT': <TokenType.UNPIVOT: 'UNPIVOT'>, 'UPDATE': <TokenType.UPDATE: 'UPDATE'>, 'USE': <TokenType.USE: 'USE'>, 'USING': <TokenType.USING: 'USING'>, 'UUID': <TokenType.UUID: 'UUID'>, 'VALUES': <TokenType.VALUES: 'VALUES'>, 'VIEW': <TokenType.VIEW: 'VIEW'>, 'VOLATILE': <TokenType.VOLATILE: 'VOLATILE'>, 'WHEN': <TokenType.WHEN: 'WHEN'>, 'WHERE': <TokenType.WHERE: 'WHERE'>, 'WINDOW': <TokenType.WINDOW: 'WINDOW'>, 'WITH': <TokenType.WITH: 'WITH'>, 'APPLY': <TokenType.APPLY: 'APPLY'>, 'ARRAY': <TokenType.ARRAY: 'ARRAY'>, 'BIT': <TokenType.BIT: 'BIT'>, 'BOOL': <TokenType.BOOLEAN: 'BOOLEAN'>, 'BOOLEAN': <TokenType.BOOLEAN: 'BOOLEAN'>, 'BYTE': <TokenType.TINYINT: 'TINYINT'>, 'MEDIUMINT': <TokenType.MEDIUMINT: 'MEDIUMINT'>, '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.CHAR: 'CHAR'>, 'CHARACTER': <TokenType.CHAR: 'CHAR'>, 'NCHAR': <TokenType.NCHAR: 'NCHAR'>, 'VARCHAR': <TokenType.VARCHAR: 'VARCHAR'>, 'VARCHAR2': <TokenType.VARCHAR: 'VARCHAR'>, 'NVARCHAR': <TokenType.NVARCHAR: 'NVARCHAR'>, 'NVARCHAR2': <TokenType.NVARCHAR: 'NVARCHAR'>, 'STR': <TokenType.TEXT: 'TEXT'>, 'STRING': <TokenType.TEXT: 'TEXT'>, 'TEXT': <TokenType.TEXT: 'TEXT'>, 'LONGTEXT': <TokenType.LONGTEXT: 'LONGTEXT'>, 'MEDIUMTEXT': <TokenType.MEDIUMTEXT: 'MEDIUMTEXT'>, 'TINYTEXT': <TokenType.TINYTEXT: 'TINYTEXT'>, 'CLOB': <TokenType.TEXT: 'TEXT'>, 'LONGVARCHAR': <TokenType.TEXT: 'TEXT'>, 'BINARY': <TokenType.BINARY: 'BINARY'>, 'BLOB': <TokenType.VARBINARY: 'VARBINARY'>, 'LONGBLOB': <TokenType.LONGBLOB: 'LONGBLOB'>, 'MEDIUMBLOB': <TokenType.MEDIUMBLOB: 'MEDIUMBLOB'>, 'TINYBLOB': <TokenType.TINYBLOB: 'TINYBLOB'>, 'BYTEA': <TokenType.VARBINARY: 'VARBINARY'>, 'VARBINARY': <TokenType.VARBINARY: 'VARBINARY'>, 'TIME': <TokenType.TIME: 'TIME'>, 'TIMETZ': <TokenType.TIMETZ: 'TIMETZ'>, 'TIMESTAMP': <TokenType.TIMESTAMP: 'TIMESTAMP'>, 'TIMESTAMPTZ': <TokenType.TIMESTAMPTZ: 'TIMESTAMPTZ'>, 'TIMESTAMPLTZ': <TokenType.TIMESTAMPLTZ: 'TIMESTAMPLTZ'>, 'DATE': <TokenType.DATE: 'DATE'>, 'DATETIME': <TokenType.DATETIME: 'DATETIME'>, 'INT4RANGE': <TokenType.INT4RANGE: 'INT4RANGE'>, 'INT4MULTIRANGE': <TokenType.INT4MULTIRANGE: 'INT4MULTIRANGE'>, 'INT8RANGE': <TokenType.INT8RANGE: 'INT8RANGE'>, 'INT8MULTIRANGE': <TokenType.INT8MULTIRANGE: 'INT8MULTIRANGE'>, 'NUMRANGE': <TokenType.NUMRANGE: 'NUMRANGE'>, 'NUMMULTIRANGE': <TokenType.NUMMULTIRANGE: 'NUMMULTIRANGE'>, 'TSRANGE': <TokenType.TSRANGE: 'TSRANGE'>, 'TSMULTIRANGE': <TokenType.TSMULTIRANGE: 'TSMULTIRANGE'>, 'TSTZRANGE': <TokenType.TSTZRANGE: 'TSTZRANGE'>, 'TSTZMULTIRANGE': <TokenType.TSTZMULTIRANGE: 'TSTZMULTIRANGE'>, 'DATERANGE': <TokenType.DATERANGE: 'DATERANGE'>, 'DATEMULTIRANGE': <TokenType.DATEMULTIRANGE: 'DATEMULTIRANGE'>, 'UNIQUE': <TokenType.UNIQUE: 'UNIQUE'>, 'STRUCT': <TokenType.STRUCT: 'STRUCT'>, 'VARIANT': <TokenType.VARIANT: 'VARIANT'>, 'ALTER': <TokenType.ALTER: 'ALTER'>, 'ANALYZE': <TokenType.COMMAND: 'COMMAND'>, 'CALL': <TokenType.COMMAND: 'COMMAND'>, 'COMMENT': <TokenType.COMMENT: 'COMMENT'>, 'COPY': <TokenType.COMMAND: 'COMMAND'>, 'EXPLAIN': <TokenType.COMMAND: 'COMMAND'>, 'GRANT': <TokenType.COMMAND: 'COMMAND'>, 'OPTIMIZE': <TokenType.COMMAND: 'COMMAND'>, 'PREPARE': <TokenType.COMMAND: 'COMMAND'>, 'TRUNCATE': <TokenType.COMMAND: 'COMMAND'>, 'VACUUM': <TokenType.COMMAND: 'COMMAND'>, 'USER-DEFINED': <TokenType.USERDEFINED: 'USERDEFINED'>, 'FOR VERSION': <TokenType.VERSION_SNAPSHOT: 'VERSION_SNAPSHOT'>, 'FOR TIMESTAMP': <TokenType.TIMESTAMP_SNAPSHOT: 'TIMESTAMP_SNAPSHOT'>, 'ADD ARCHIVE': <TokenType.COMMAND: 'COMMAND'>, 'ADD ARCHIVES': <TokenType.COMMAND: 'COMMAND'>, 'ADD FILE': <TokenType.COMMAND: 'COMMAND'>, 'ADD FILES': <TokenType.COMMAND: 'COMMAND'>, 'ADD JAR': <TokenType.COMMAND: 'COMMAND'>, 'ADD JARS': <TokenType.COMMAND: 'COMMAND'>, 'MSCK REPAIR': <TokenType.COMMAND: 'COMMAND'>, 'REFRESH': <TokenType.REFRESH: 'REFRESH'>, 'TIMESTAMP AS OF': <TokenType.TIMESTAMP_SNAPSHOT: 'TIMESTAMP_SNAPSHOT'>, 'VERSION AS OF': <TokenType.VERSION_SNAPSHOT: 'VERSION_SNAPSHOT'>, 'WITH SERDEPROPERTIES': <TokenType.SERDE_PROPERTIES: 'SERDE_PROPERTIES'>}
NUMERIC_LITERALS = {'L': 'BIGINT', 'S': 'SMALLINT', 'Y': 'TINYINT', 'D': 'DOUBLE', 'F': 'FLOAT', 'BD': 'DECIMAL'}
IDENTIFIERS_CAN_START_WITH_DIGIT: bool = True
class Hive.Parser(sqlglot.parser.Parser):
266    class Parser(parser.Parser):
267        LOG_DEFAULTS_TO_LN = True
268        STRICT_CAST = False
269
270        FUNCTIONS = {
271            **parser.Parser.FUNCTIONS,
272            "BASE64": exp.ToBase64.from_arg_list,
273            "COLLECT_LIST": exp.ArrayAgg.from_arg_list,
274            "COLLECT_SET": exp.SetAgg.from_arg_list,
275            "DATE_ADD": lambda args: exp.TsOrDsAdd(
276                this=seq_get(args, 0), expression=seq_get(args, 1), unit=exp.Literal.string("DAY")
277            ),
278            "DATE_FORMAT": lambda args: format_time_lambda(exp.TimeToStr, "hive")(
279                [
280                    exp.TimeStrToTime(this=seq_get(args, 0)),
281                    seq_get(args, 1),
282                ]
283            ),
284            "DATE_SUB": lambda args: exp.TsOrDsAdd(
285                this=seq_get(args, 0),
286                expression=exp.Mul(this=seq_get(args, 1), expression=exp.Literal.number(-1)),
287                unit=exp.Literal.string("DAY"),
288            ),
289            "DATEDIFF": lambda args: exp.DateDiff(
290                this=exp.TsOrDsToDate(this=seq_get(args, 0)),
291                expression=exp.TsOrDsToDate(this=seq_get(args, 1)),
292            ),
293            "DAY": lambda args: exp.Day(this=exp.TsOrDsToDate(this=seq_get(args, 0))),
294            "FROM_UNIXTIME": format_time_lambda(exp.UnixToStr, "hive", True),
295            "GET_JSON_OBJECT": exp.JSONExtractScalar.from_arg_list,
296            "LOCATE": locate_to_strposition,
297            "MAP": parse_var_map,
298            "MONTH": lambda args: exp.Month(this=exp.TsOrDsToDate.from_arg_list(args)),
299            "PERCENTILE": exp.Quantile.from_arg_list,
300            "PERCENTILE_APPROX": exp.ApproxQuantile.from_arg_list,
301            "REGEXP_EXTRACT": lambda args: exp.RegexpExtract(
302                this=seq_get(args, 0), expression=seq_get(args, 1), group=seq_get(args, 2)
303            ),
304            "SIZE": exp.ArraySize.from_arg_list,
305            "SPLIT": exp.RegexpSplit.from_arg_list,
306            "STR_TO_MAP": lambda args: exp.StrToMap(
307                this=seq_get(args, 0),
308                pair_delim=seq_get(args, 1) or exp.Literal.string(","),
309                key_value_delim=seq_get(args, 2) or exp.Literal.string(":"),
310            ),
311            "TO_DATE": format_time_lambda(exp.TsOrDsToDate, "hive"),
312            "TO_JSON": exp.JSONFormat.from_arg_list,
313            "UNBASE64": exp.FromBase64.from_arg_list,
314            "UNIX_TIMESTAMP": format_time_lambda(exp.StrToUnix, "hive", True),
315            "YEAR": lambda args: exp.Year(this=exp.TsOrDsToDate.from_arg_list(args)),
316        }
317
318        NO_PAREN_FUNCTION_PARSERS = {
319            **parser.Parser.NO_PAREN_FUNCTION_PARSERS,
320            "TRANSFORM": lambda self: self._parse_transform(),
321        }
322
323        PROPERTY_PARSERS = {
324            **parser.Parser.PROPERTY_PARSERS,
325            "WITH SERDEPROPERTIES": lambda self: exp.SerdeProperties(
326                expressions=self._parse_wrapped_csv(self._parse_property)
327            ),
328        }
329
330        def _parse_transform(self) -> t.Optional[exp.Transform | exp.QueryTransform]:
331            if not self._match(TokenType.L_PAREN, advance=False):
332                self._retreat(self._index - 1)
333                return None
334
335            args = self._parse_wrapped_csv(self._parse_lambda)
336            row_format_before = self._parse_row_format(match_row=True)
337
338            record_writer = None
339            if self._match_text_seq("RECORDWRITER"):
340                record_writer = self._parse_string()
341
342            if not self._match(TokenType.USING):
343                return exp.Transform.from_arg_list(args)
344
345            command_script = self._parse_string()
346
347            self._match(TokenType.ALIAS)
348            schema = self._parse_schema()
349
350            row_format_after = self._parse_row_format(match_row=True)
351            record_reader = None
352            if self._match_text_seq("RECORDREADER"):
353                record_reader = self._parse_string()
354
355            return self.expression(
356                exp.QueryTransform,
357                expressions=args,
358                command_script=command_script,
359                schema=schema,
360                row_format_before=row_format_before,
361                record_writer=record_writer,
362                row_format_after=row_format_after,
363                record_reader=record_reader,
364            )
365
366        def _parse_types(
367            self, check_func: bool = False, schema: bool = False, allow_identifiers: bool = True
368        ) -> t.Optional[exp.Expression]:
369            """
370            Spark (and most likely Hive) treats casts to CHAR(length) and VARCHAR(length) as casts to
371            STRING in all contexts except for schema definitions. For example, this is in Spark v3.4.0:
372
373                spark-sql (default)> select cast(1234 as varchar(2));
374                23/06/06 15:51:18 WARN CharVarcharUtils: The Spark cast operator does not support
375                char/varchar type and simply treats them as string type. Please use string type
376                directly to avoid confusion. Otherwise, you can set spark.sql.legacy.charVarcharAsString
377                to true, so that Spark treat them as string type as same as Spark 3.0 and earlier
378
379                1234
380                Time taken: 4.265 seconds, Fetched 1 row(s)
381
382            This shows that Spark doesn't truncate the value into '12', which is inconsistent with
383            what other dialects (e.g. postgres) do, so we need to drop the length to transpile correctly.
384
385            Reference: https://spark.apache.org/docs/latest/sql-ref-datatypes.html
386            """
387            this = super()._parse_types(
388                check_func=check_func, schema=schema, allow_identifiers=allow_identifiers
389            )
390
391            if this and not schema:
392                return this.transform(
393                    lambda node: node.replace(exp.DataType.build("text"))
394                    if isinstance(node, exp.DataType) and node.is_type("char", "varchar")
395                    else node,
396                    copy=False,
397                )
398
399            return this
400
401        def _parse_partition_and_order(
402            self,
403        ) -> t.Tuple[t.List[exp.Expression], t.Optional[exp.Expression]]:
404            return (
405                self._parse_csv(self._parse_conjunction)
406                if self._match_set({TokenType.PARTITION_BY, TokenType.DISTRIBUTE_BY})
407                else [],
408                super()._parse_order(skip_order_token=self._match(TokenType.SORT_BY)),
409            )

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
LOG_DEFAULTS_TO_LN = True
STRICT_CAST = False
FUNCTIONS = {'ABS': <bound method Func.from_arg_list of <class 'sqlglot.expressions.Abs'>>, 'ANY_VALUE': <bound method Func.from_arg_list of <class 'sqlglot.expressions.AnyValue'>>, 'APPROX_DISTINCT': <bound method Func.from_arg_list of <class 'sqlglot.expressions.ApproxDistinct'>>, 'APPROX_COUNT_DISTINCT': <bound method Func.from_arg_list of <class 'sqlglot.expressions.ApproxDistinct'>>, 'APPROX_QUANTILE': <bound method Func.from_arg_list of <class 'sqlglot.expressions.ApproxQuantile'>>, 'APPROX_TOP_K': <bound method Func.from_arg_list of <class 'sqlglot.expressions.ApproxTopK'>>, 'ARG_MAX': <bound method Func.from_arg_list of <class 'sqlglot.expressions.ArgMax'>>, 'ARGMAX': <bound method Func.from_arg_list of <class 'sqlglot.expressions.ArgMax'>>, 'MAX_BY': <bound method Func.from_arg_list of <class 'sqlglot.expressions.ArgMax'>>, 'ARG_MIN': <bound method Func.from_arg_list of <class 'sqlglot.expressions.ArgMin'>>, 'ARGMIN': <bound method Func.from_arg_list of <class 'sqlglot.expressions.ArgMin'>>, 'MIN_BY': <bound method Func.from_arg_list of <class 'sqlglot.expressions.ArgMin'>>, 'ARRAY': <bound method Func.from_arg_list of <class 'sqlglot.expressions.Array'>>, 'ARRAY_AGG': <bound method Func.from_arg_list of <class 'sqlglot.expressions.ArrayAgg'>>, 'ARRAY_ALL': <bound method Func.from_arg_list of <class 'sqlglot.expressions.ArrayAll'>>, 'ARRAY_ANY': <bound method Func.from_arg_list of <class 'sqlglot.expressions.ArrayAny'>>, 'ARRAY_CONCAT': <bound method Func.from_arg_list of <class 'sqlglot.expressions.ArrayConcat'>>, 'ARRAY_CAT': <bound method Func.from_arg_list of <class 'sqlglot.expressions.ArrayConcat'>>, 'ARRAY_CONTAINS': <bound method Func.from_arg_list of <class 'sqlglot.expressions.ArrayContains'>>, 'FILTER': <bound method Func.from_arg_list of <class 'sqlglot.expressions.ArrayFilter'>>, 'ARRAY_FILTER': <bound method Func.from_arg_list of <class 'sqlglot.expressions.ArrayFilter'>>, 'ARRAY_JOIN': <bound method Func.from_arg_list of <class 'sqlglot.expressions.ArrayJoin'>>, 'ARRAY_SIZE': <bound method Func.from_arg_list of <class 'sqlglot.expressions.ArraySize'>>, 'ARRAY_SORT': <bound method Func.from_arg_list of <class 'sqlglot.expressions.ArraySort'>>, 'ARRAY_SUM': <bound method Func.from_arg_list of <class 'sqlglot.expressions.ArraySum'>>, 'ARRAY_UNION_AGG': <bound method Func.from_arg_list of <class 'sqlglot.expressions.ArrayUnionAgg'>>, 'AVG': <bound method Func.from_arg_list of <class 'sqlglot.expressions.Avg'>>, 'CASE': <bound method Func.from_arg_list of <class 'sqlglot.expressions.Case'>>, 'CAST': <bound method Func.from_arg_list of <class 'sqlglot.expressions.Cast'>>, 'CAST_TO_STR_TYPE': <bound method Func.from_arg_list of <class 'sqlglot.expressions.CastToStrType'>>, 'CEIL': <bound method Func.from_arg_list of <class 'sqlglot.expressions.Ceil'>>, 'CEILING': <bound method Func.from_arg_list of <class 'sqlglot.expressions.Ceil'>>, 'CHR': <bound method Func.from_arg_list of <class 'sqlglot.expressions.Chr'>>, 'CHAR': <bound method Func.from_arg_list of <class 'sqlglot.expressions.Chr'>>, 'COALESCE': <bound method Func.from_arg_list of <class 'sqlglot.expressions.Coalesce'>>, 'IFNULL': <bound method Func.from_arg_list of <class 'sqlglot.expressions.Coalesce'>>, 'NVL': <bound method Func.from_arg_list of <class 'sqlglot.expressions.Coalesce'>>, 'COLLATE': <bound method Func.from_arg_list of <class 'sqlglot.expressions.Collate'>>, 'CONCAT': <bound method Func.from_arg_list of <class 'sqlglot.expressions.Concat'>>, 'CONCAT_WS': <bound method Func.from_arg_list of <class 'sqlglot.expressions.ConcatWs'>>, 'COUNT': <bound method Func.from_arg_list of <class 'sqlglot.expressions.Count'>>, 'COUNT_IF': <bound method Func.from_arg_list of <class 'sqlglot.expressions.CountIf'>>, 'CURRENT_DATE': <bound method Func.from_arg_list of <class 'sqlglot.expressions.CurrentDate'>>, 'CURRENT_DATETIME': <bound method Func.from_arg_list of <class 'sqlglot.expressions.CurrentDatetime'>>, 'CURRENT_TIME': <bound method Func.from_arg_list of <class 'sqlglot.expressions.CurrentTime'>>, 'CURRENT_TIMESTAMP': <bound method Func.from_arg_list of <class 'sqlglot.expressions.CurrentTimestamp'>>, 'CURRENT_USER': <bound method Func.from_arg_list of <class 'sqlglot.expressions.CurrentUser'>>, 'DATE': <bound method Func.from_arg_list of <class 'sqlglot.expressions.Date'>>, 'DATE_ADD': <function Hive.Parser.<lambda>>, 'DATEDIFF': <function Hive.Parser.<lambda>>, 'DATE_DIFF': <bound method Func.from_arg_list of <class 'sqlglot.expressions.DateDiff'>>, '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': <function Hive.Parser.<lambda>>, 'DATE_TO_DATE_STR': <function Parser.<lambda>>, 'DATE_TO_DI': <bound method Func.from_arg_list of <class 'sqlglot.expressions.DateToDi'>>, 'DATE_TRUNC': <bound method Func.from_arg_list of <class 'sqlglot.expressions.DateTrunc'>>, 'DATETIME_ADD': <bound method Func.from_arg_list of <class 'sqlglot.expressions.DatetimeAdd'>>, 'DATETIME_DIFF': <bound method Func.from_arg_list of <class 'sqlglot.expressions.DatetimeDiff'>>, 'DATETIME_SUB': <bound method Func.from_arg_list of <class 'sqlglot.expressions.DatetimeSub'>>, 'DATETIME_TRUNC': <bound method Func.from_arg_list of <class 'sqlglot.expressions.DatetimeTrunc'>>, 'DAY': <function Hive.Parser.<lambda>>, 'DAY_OF_MONTH': <bound method Func.from_arg_list of <class 'sqlglot.expressions.DayOfMonth'>>, 'DAYOFMONTH': <bound method Func.from_arg_list of <class 'sqlglot.expressions.DayOfMonth'>>, 'DAY_OF_WEEK': <bound method Func.from_arg_list of <class 'sqlglot.expressions.DayOfWeek'>>, 'DAYOFWEEK': <bound method Func.from_arg_list of <class 'sqlglot.expressions.DayOfWeek'>>, 'DAY_OF_YEAR': <bound method Func.from_arg_list of <class 'sqlglot.expressions.DayOfYear'>>, 'DAYOFYEAR': <bound method Func.from_arg_list of <class 'sqlglot.expressions.DayOfYear'>>, 'DECODE': <bound method Func.from_arg_list of <class 'sqlglot.expressions.Decode'>>, 'DI_TO_DATE': <bound method Func.from_arg_list of <class 'sqlglot.expressions.DiToDate'>>, 'ENCODE': <bound method Func.from_arg_list of <class 'sqlglot.expressions.Encode'>>, 'EXP': <bound method Func.from_arg_list of <class 'sqlglot.expressions.Exp'>>, 'EXPLODE': <bound method Func.from_arg_list of <class 'sqlglot.expressions.Explode'>>, 'EXPLODE_OUTER': <bound method Func.from_arg_list of <class 'sqlglot.expressions.ExplodeOuter'>>, 'EXTRACT': <bound method Func.from_arg_list of <class 'sqlglot.expressions.Extract'>>, 'FIRST': <bound method Func.from_arg_list of <class 'sqlglot.expressions.First'>>, 'FLATTEN': <bound method Func.from_arg_list of <class 'sqlglot.expressions.Flatten'>>, 'FLOOR': <bound method Func.from_arg_list of <class 'sqlglot.expressions.Floor'>>, 'FROM_BASE': <bound method Func.from_arg_list of <class 'sqlglot.expressions.FromBase'>>, 'FROM_BASE64': <bound method Func.from_arg_list of <class 'sqlglot.expressions.FromBase64'>>, 'GENERATE_SERIES': <bound method Func.from_arg_list of <class 'sqlglot.expressions.GenerateSeries'>>, 'GREATEST': <bound method Func.from_arg_list of <class 'sqlglot.expressions.Greatest'>>, 'GROUP_CONCAT': <bound method Func.from_arg_list of <class 'sqlglot.expressions.GroupConcat'>>, 'HEX': <bound method Func.from_arg_list of <class 'sqlglot.expressions.Hex'>>, 'HLL': <bound method Func.from_arg_list of <class 'sqlglot.expressions.Hll'>>, 'IF': <bound method Func.from_arg_list of <class 'sqlglot.expressions.If'>>, 'INITCAP': <bound method Func.from_arg_list of <class 'sqlglot.expressions.Initcap'>>, 'IS_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_TABLE': <bound method Func.from_arg_list of <class 'sqlglot.expressions.JSONTable'>>, 'LAST': <bound method Func.from_arg_list of <class 'sqlglot.expressions.Last'>>, 'LAST_DATE_OF_MONTH': <bound method Func.from_arg_list of <class 'sqlglot.expressions.LastDateOfMonth'>>, 'LEAST': <bound method Func.from_arg_list of <class 'sqlglot.expressions.Least'>>, 'LEFT': <bound method Func.from_arg_list of <class 'sqlglot.expressions.Left'>>, 'LENGTH': <bound method Func.from_arg_list of <class 'sqlglot.expressions.Length'>>, 'LEN': <bound method Func.from_arg_list of <class 'sqlglot.expressions.Length'>>, 'LEVENSHTEIN': <bound method Func.from_arg_list of <class 'sqlglot.expressions.Levenshtein'>>, 'LN': <bound method Func.from_arg_list of <class 'sqlglot.expressions.Ln'>>, 'LOG': <bound method Func.from_arg_list of <class 'sqlglot.expressions.Log'>>, 'LOG10': <bound method Func.from_arg_list of <class 'sqlglot.expressions.Log10'>>, 'LOG2': <bound method Func.from_arg_list of <class 'sqlglot.expressions.Log2'>>, 'LOGICAL_AND': <bound method Func.from_arg_list of <class 'sqlglot.expressions.LogicalAnd'>>, 'BOOL_AND': <bound method Func.from_arg_list of <class 'sqlglot.expressions.LogicalAnd'>>, 'BOOLAND_AGG': <bound method Func.from_arg_list of <class 'sqlglot.expressions.LogicalAnd'>>, 'LOGICAL_OR': <bound method Func.from_arg_list of <class 'sqlglot.expressions.LogicalOr'>>, 'BOOL_OR': <bound method Func.from_arg_list of <class 'sqlglot.expressions.LogicalOr'>>, 'BOOLOR_AGG': <bound method Func.from_arg_list of <class 'sqlglot.expressions.LogicalOr'>>, 'LOWER': <bound method Func.from_arg_list of <class 'sqlglot.expressions.Lower'>>, 'LCASE': <bound method Func.from_arg_list of <class 'sqlglot.expressions.Lower'>>, 'MD5': <bound method Func.from_arg_list of <class 'sqlglot.expressions.MD5'>>, 'MD5_DIGEST': <bound method Func.from_arg_list of <class 'sqlglot.expressions.MD5Digest'>>, 'MAP': <function parse_var_map>, 'MAP_FROM_ENTRIES': <bound method Func.from_arg_list of <class 'sqlglot.expressions.MapFromEntries'>>, 'MATCH_AGAINST': <bound method Func.from_arg_list of <class 'sqlglot.expressions.MatchAgainst'>>, 'MAX': <bound method Func.from_arg_list of <class 'sqlglot.expressions.Max'>>, 'MIN': <bound method Func.from_arg_list of <class 'sqlglot.expressions.Min'>>, 'MONTH': <function Hive.Parser.<lambda>>, 'MONTHS_BETWEEN': <bound method Func.from_arg_list of <class 'sqlglot.expressions.MonthsBetween'>>, 'NEXT_VALUE_FOR': <bound method Func.from_arg_list of <class 'sqlglot.expressions.NextValueFor'>>, 'NULLIF': <bound method Func.from_arg_list of <class 'sqlglot.expressions.Nullif'>>, 'NUMBER_TO_STR': <bound method Func.from_arg_list of <class 'sqlglot.expressions.NumberToStr'>>, 'NVL2': <bound method Func.from_arg_list of <class 'sqlglot.expressions.Nvl2'>>, 'OPEN_J_S_O_N': <bound method Func.from_arg_list of <class 'sqlglot.expressions.OpenJSON'>>, 'PARAMETERIZED_AGG': <bound method Func.from_arg_list of <class 'sqlglot.expressions.ParameterizedAgg'>>, 'PARSE_JSON': <bound method Func.from_arg_list of <class 'sqlglot.expressions.ParseJSON'>>, 'JSON_PARSE': <bound method Func.from_arg_list of <class 'sqlglot.expressions.ParseJSON'>>, 'PERCENTILE_CONT': <bound method Func.from_arg_list of <class 'sqlglot.expressions.PercentileCont'>>, 'PERCENTILE_DISC': <bound method Func.from_arg_list of <class 'sqlglot.expressions.PercentileDisc'>>, 'POSEXPLODE': <bound method Func.from_arg_list of <class 'sqlglot.expressions.Posexplode'>>, 'POSEXPLODE_OUTER': <bound method Func.from_arg_list of <class 'sqlglot.expressions.PosexplodeOuter'>>, 'POWER': <bound method Func.from_arg_list of <class 'sqlglot.expressions.Pow'>>, 'POW': <bound method Func.from_arg_list of <class 'sqlglot.expressions.Pow'>>, 'PREDICT': <bound method Func.from_arg_list of <class 'sqlglot.expressions.Predict'>>, 'QUANTILE': <bound method Func.from_arg_list of <class 'sqlglot.expressions.Quantile'>>, 'RANGE_N': <bound method Func.from_arg_list of <class 'sqlglot.expressions.RangeN'>>, 'READ_CSV': <bound method Func.from_arg_list of <class 'sqlglot.expressions.ReadCSV'>>, 'REDUCE': <bound method Func.from_arg_list of <class 'sqlglot.expressions.Reduce'>>, 'REGEXP_EXTRACT': <function Hive.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': <bound method Func.from_arg_list of <class 'sqlglot.expressions.RegexpReplace'>>, 'REGEXP_SPLIT': <bound method Func.from_arg_list of <class 'sqlglot.expressions.RegexpSplit'>>, 'REPEAT': <bound method Func.from_arg_list of <class 'sqlglot.expressions.Repeat'>>, 'RIGHT': <bound method Func.from_arg_list of <class 'sqlglot.expressions.Right'>>, 'ROUND': <bound method Func.from_arg_list of <class 'sqlglot.expressions.Round'>>, 'ROW_NUMBER': <bound method Func.from_arg_list of <class 'sqlglot.expressions.RowNumber'>>, 'SHA': <bound method Func.from_arg_list of <class 'sqlglot.expressions.SHA'>>, 'SHA1': <bound method Func.from_arg_list of <class 'sqlglot.expressions.SHA'>>, 'SHA2': <bound method Func.from_arg_list of <class 'sqlglot.expressions.SHA2'>>, 'SAFE_CONCAT': <bound method Func.from_arg_list of <class 'sqlglot.expressions.SafeConcat'>>, 'SAFE_DIVIDE': <bound method Func.from_arg_list of <class 'sqlglot.expressions.SafeDivide'>>, 'SET_AGG': <bound method Func.from_arg_list of <class 'sqlglot.expressions.SetAgg'>>, 'SORT_ARRAY': <bound method Func.from_arg_list of <class 'sqlglot.expressions.SortArray'>>, 'SPLIT': <bound method Func.from_arg_list of <class 'sqlglot.expressions.RegexpSplit'>>, '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': <function Hive.Parser.<lambda>>, 'STR_TO_TIME': <bound method Func.from_arg_list of <class 'sqlglot.expressions.StrToTime'>>, 'STR_TO_UNIX': <bound method Func.from_arg_list of <class 'sqlglot.expressions.StrToUnix'>>, 'STRUCT': <bound method Func.from_arg_list of <class 'sqlglot.expressions.Struct'>>, 'STRUCT_EXTRACT': <bound method Func.from_arg_list of <class 'sqlglot.expressions.StructExtract'>>, 'STUFF': <bound method Func.from_arg_list of <class 'sqlglot.expressions.Stuff'>>, 'INSERT': <bound method Func.from_arg_list of <class 'sqlglot.expressions.Stuff'>>, 'SUBSTRING': <bound method Func.from_arg_list of <class 'sqlglot.expressions.Substring'>>, 'SUM': <bound method Func.from_arg_list of <class 'sqlglot.expressions.Sum'>>, 'TIME_ADD': <bound method Func.from_arg_list of <class 'sqlglot.expressions.TimeAdd'>>, 'TIME_DIFF': <bound method Func.from_arg_list of <class 'sqlglot.expressions.TimeDiff'>>, 'TIME_STR_TO_DATE': <bound method Func.from_arg_list of <class 'sqlglot.expressions.TimeStrToDate'>>, 'TIME_STR_TO_TIME': <bound method Func.from_arg_list of <class 'sqlglot.expressions.TimeStrToTime'>>, 'TIME_STR_TO_UNIX': <bound method Func.from_arg_list of <class 'sqlglot.expressions.TimeStrToUnix'>>, 'TIME_SUB': <bound method Func.from_arg_list of <class 'sqlglot.expressions.TimeSub'>>, 'TIME_TO_STR': <bound method Func.from_arg_list of <class 'sqlglot.expressions.TimeToStr'>>, 'TIME_TO_TIME_STR': <function Parser.<lambda>>, 'TIME_TO_UNIX': <bound method Func.from_arg_list of <class 'sqlglot.expressions.TimeToUnix'>>, 'TIME_TRUNC': <bound method Func.from_arg_list of <class 'sqlglot.expressions.TimeTrunc'>>, 'TIMESTAMP': <bound method Func.from_arg_list of <class 'sqlglot.expressions.Timestamp'>>, 'TIMESTAMP_ADD': <bound method Func.from_arg_list of <class 'sqlglot.expressions.TimestampAdd'>>, 'TIMESTAMP_DIFF': <bound method Func.from_arg_list of <class 'sqlglot.expressions.TimestampDiff'>>, 'TIMESTAMP_SUB': <bound method Func.from_arg_list of <class 'sqlglot.expressions.TimestampSub'>>, 'TIMESTAMP_TRUNC': <bound method Func.from_arg_list of <class 'sqlglot.expressions.TimestampTrunc'>>, 'TO_BASE64': <bound method Func.from_arg_list of <class 'sqlglot.expressions.ToBase64'>>, 'TO_CHAR': <bound method Func.from_arg_list of <class 'sqlglot.expressions.ToChar'>>, 'TO_DAYS': <bound method Func.from_arg_list of <class 'sqlglot.expressions.ToDays'>>, 'TRANSFORM': <bound method Func.from_arg_list of <class 'sqlglot.expressions.Transform'>>, 'TRIM': <bound method Func.from_arg_list of <class 'sqlglot.expressions.Trim'>>, 'TRY_CAST': <bound method Func.from_arg_list of <class 'sqlglot.expressions.TryCast'>>, 'TS_OR_DI_TO_DI': <bound method Func.from_arg_list of <class 'sqlglot.expressions.TsOrDiToDi'>>, 'TS_OR_DS_ADD': <bound method Func.from_arg_list of <class 'sqlglot.expressions.TsOrDsAdd'>>, 'TS_OR_DS_TO_DATE': <bound method Func.from_arg_list of <class 'sqlglot.expressions.TsOrDsToDate'>>, 'TS_OR_DS_TO_DATE_STR': <function Parser.<lambda>>, 'UNHEX': <bound method Func.from_arg_list of <class 'sqlglot.expressions.Unhex'>>, 'UNIX_TO_STR': <bound method Func.from_arg_list of <class 'sqlglot.expressions.UnixToStr'>>, 'UNIX_TO_TIME': <bound method Func.from_arg_list of <class 'sqlglot.expressions.UnixToTime'>>, 'UNIX_TO_TIME_STR': <bound method Func.from_arg_list of <class 'sqlglot.expressions.UnixToTimeStr'>>, 'UPPER': <bound method Func.from_arg_list of <class 'sqlglot.expressions.Upper'>>, 'UCASE': <bound method Func.from_arg_list of <class 'sqlglot.expressions.Upper'>>, 'VAR_MAP': <function parse_var_map>, 'VARIANCE': <bound method Func.from_arg_list of <class 'sqlglot.expressions.Variance'>>, 'VARIANCE_SAMP': <bound method Func.from_arg_list of <class 'sqlglot.expressions.Variance'>>, 'VAR_SAMP': <bound method Func.from_arg_list of <class 'sqlglot.expressions.Variance'>>, 'VARIANCE_POP': <bound method Func.from_arg_list of <class 'sqlglot.expressions.VariancePop'>>, 'VAR_POP': <bound method Func.from_arg_list of <class 'sqlglot.expressions.VariancePop'>>, 'WEEK': <bound method Func.from_arg_list of <class 'sqlglot.expressions.Week'>>, 'WEEK_OF_YEAR': <bound method Func.from_arg_list of <class 'sqlglot.expressions.WeekOfYear'>>, 'WEEKOFYEAR': <bound method Func.from_arg_list of <class 'sqlglot.expressions.WeekOfYear'>>, 'WHEN': <bound method Func.from_arg_list of <class 'sqlglot.expressions.When'>>, 'X_M_L_TABLE': <bound method Func.from_arg_list of <class 'sqlglot.expressions.XMLTable'>>, 'XOR': <bound method Func.from_arg_list of <class 'sqlglot.expressions.Xor'>>, 'YEAR': <function Hive.Parser.<lambda>>, 'GLOB': <function Parser.<lambda>>, 'LIKE': <function parse_like>, 'BASE64': <bound method Func.from_arg_list of <class 'sqlglot.expressions.ToBase64'>>, 'COLLECT_LIST': <bound method Func.from_arg_list of <class 'sqlglot.expressions.ArrayAgg'>>, 'COLLECT_SET': <bound method Func.from_arg_list of <class 'sqlglot.expressions.SetAgg'>>, 'DATE_FORMAT': <function Hive.Parser.<lambda>>, 'FROM_UNIXTIME': <function format_time_lambda.<locals>._format_time>, 'GET_JSON_OBJECT': <bound method Func.from_arg_list of <class 'sqlglot.expressions.JSONExtractScalar'>>, 'LOCATE': <function locate_to_strposition>, 'PERCENTILE': <bound method Func.from_arg_list of <class 'sqlglot.expressions.Quantile'>>, 'PERCENTILE_APPROX': <bound method Func.from_arg_list of <class 'sqlglot.expressions.ApproxQuantile'>>, 'SIZE': <bound method Func.from_arg_list of <class 'sqlglot.expressions.ArraySize'>>, 'TO_DATE': <function format_time_lambda.<locals>._format_time>, 'TO_JSON': <bound method Func.from_arg_list of <class 'sqlglot.expressions.JSONFormat'>>, 'UNBASE64': <bound method Func.from_arg_list of <class 'sqlglot.expressions.FromBase64'>>, 'UNIX_TIMESTAMP': <function format_time_lambda.<locals>._format_time>}
NO_PAREN_FUNCTION_PARSERS = {'ANY': <function Parser.<lambda>>, 'CASE': <function Parser.<lambda>>, 'IF': <function Parser.<lambda>>, 'NEXT': <function Parser.<lambda>>, 'TRANSFORM': <function Hive.Parser.<lambda>>}
PROPERTY_PARSERS = {'ALGORITHM': <function Parser.<lambda>>, 'AUTO_INCREMENT': <function Parser.<lambda>>, 'BLOCKCOMPRESSION': <function Parser.<lambda>>, 'CHARSET': <function Parser.<lambda>>, 'CHARACTER SET': <function Parser.<lambda>>, 'CHECKSUM': <function Parser.<lambda>>, 'CLUSTER BY': <function Parser.<lambda>>, 'CLUSTERED': <function Parser.<lambda>>, 'COLLATE': <function Parser.<lambda>>, 'COMMENT': <function Parser.<lambda>>, 'COPY': <function Parser.<lambda>>, 'DATABLOCKSIZE': <function Parser.<lambda>>, 'DEFINER': <function Parser.<lambda>>, 'DETERMINISTIC': <function Parser.<lambda>>, 'DISTKEY': <function Parser.<lambda>>, 'DISTSTYLE': <function Parser.<lambda>>, 'ENGINE': <function Parser.<lambda>>, 'EXECUTE': <function Parser.<lambda>>, 'EXTERNAL': <function Parser.<lambda>>, 'FALLBACK': <function Parser.<lambda>>, 'FORMAT': <function Parser.<lambda>>, 'FREESPACE': <function Parser.<lambda>>, 'HEAP': <function Parser.<lambda>>, 'IMMUTABLE': <function Parser.<lambda>>, 'INPUT': <function Parser.<lambda>>, 'JOURNAL': <function Parser.<lambda>>, 'LANGUAGE': <function Parser.<lambda>>, 'LAYOUT': <function Parser.<lambda>>, 'LIFETIME': <function Parser.<lambda>>, 'LIKE': <function Parser.<lambda>>, 'LOCATION': <function Parser.<lambda>>, 'LOCK': <function Parser.<lambda>>, 'LOCKING': <function Parser.<lambda>>, 'LOG': <function Parser.<lambda>>, 'MATERIALIZED': <function Parser.<lambda>>, 'MERGEBLOCKRATIO': <function Parser.<lambda>>, 'MULTISET': <function Parser.<lambda>>, 'NO': <function Parser.<lambda>>, 'ON': <function Parser.<lambda>>, 'ORDER BY': <function Parser.<lambda>>, 'OUTPUT': <function Parser.<lambda>>, 'PARTITION': <function Parser.<lambda>>, 'PARTITION BY': <function Parser.<lambda>>, 'PARTITIONED BY': <function Parser.<lambda>>, 'PARTITIONED_BY': <function Parser.<lambda>>, 'PRIMARY KEY': <function Parser.<lambda>>, 'RANGE': <function Parser.<lambda>>, 'REMOTE': <function Parser.<lambda>>, 'RETURNS': <function Parser.<lambda>>, 'ROW': <function Parser.<lambda>>, 'ROW_FORMAT': <function Parser.<lambda>>, 'SAMPLE': <function Parser.<lambda>>, 'SET': <function Parser.<lambda>>, 'SETTINGS': <function Parser.<lambda>>, 'SORTKEY': <function Parser.<lambda>>, 'SOURCE': <function Parser.<lambda>>, 'STABLE': <function Parser.<lambda>>, 'STORED': <function Parser.<lambda>>, 'SYSTEM_VERSIONING': <function Parser.<lambda>>, 'TBLPROPERTIES': <function Parser.<lambda>>, 'TEMP': <function Parser.<lambda>>, 'TEMPORARY': <function Parser.<lambda>>, 'TO': <function Parser.<lambda>>, 'TRANSIENT': <function Parser.<lambda>>, 'TRANSFORM': <function Parser.<lambda>>, 'TTL': <function Parser.<lambda>>, 'USING': <function Parser.<lambda>>, 'VOLATILE': <function Parser.<lambda>>, 'WITH': <function Parser.<lambda>>, 'WITH SERDEPROPERTIES': <function Hive.Parser.<lambda>>}
SAFE_DIVISION = True
TOKENIZER_CLASS: Type[sqlglot.tokens.Tokenizer] = <class 'Hive.Tokenizer'>
ALIAS_POST_TABLESAMPLE: bool = True
SUPPORTS_USER_DEFINED_TYPES = False
SHOW_TRIE: Dict = {}
SET_TRIE: Dict = {'GLOBAL': {0: True}, 'LOCAL': {0: True}, 'SESSION': {0: True}, 'TRANSACTION': {0: True}}
FORMAT_TRIE: Dict = {'y': {0: True, 'y': {'y': {'y': {0: True}}, 0: True}}, 'Y': {0: True, 'Y': {'Y': {'Y': {0: True}}, 0: True}}, 'M': {'M': {'M': {'M': {0: True}, 0: True}, 0: True}, 0: True}, 'd': {'d': {0: True}, 0: True}, 'H': {'H': {0: True}, 0: True}, 'h': {'h': {0: True}, 0: True}, 'm': {'m': {0: True}, 0: True}, 's': {'s': {0: True}, 0: True}, 'S': {'S': {'S': {'S': {'S': {'S': {0: True}}}}}}, 'a': {0: True}, 'D': {'D': {0: True}, 0: True}, 'E': {0: True, 'E': {0: True, 'E': {0: True, 'E': {0: True}}}}}
TIME_MAPPING: Dict[str, str] = {'y': '%Y', 'Y': '%Y', 'YYYY': '%Y', 'yyyy': '%Y', 'YY': '%y', 'yy': '%y', 'MMMM': '%B', 'MMM': '%b', 'MM': '%m', 'M': '%-m', 'dd': '%d', 'd': '%-d', 'HH': '%H', 'H': '%-H', 'hh': '%I', 'h': '%-I', 'mm': '%M', 'm': '%-M', 'ss': '%S', 's': '%-S', 'SSSSSS': '%f', 'a': '%p', 'DD': '%j', 'D': '%-j', 'E': '%a', 'EE': '%a', 'EEE': '%a', 'EEEE': '%A'}
TIME_TRIE: Dict = {'y': {0: True, 'y': {'y': {'y': {0: True}}, 0: True}}, 'Y': {0: True, 'Y': {'Y': {'Y': {0: True}}, 0: True}}, 'M': {'M': {'M': {'M': {0: True}, 0: True}, 0: True}, 0: True}, 'd': {'d': {0: True}, 0: True}, 'H': {'H': {0: True}, 0: True}, 'h': {'h': {0: True}, 0: True}, 'm': {'m': {0: True}, 0: True}, 's': {'s': {0: True}, 0: True}, 'S': {'S': {'S': {'S': {'S': {'S': {0: True}}}}}}, 'a': {0: True}, 'D': {'D': {0: True}, 0: True}, 'E': {0: True, 'E': {0: True, 'E': {0: True, 'E': {0: True}}}}}
Inherited Members
sqlglot.parser.Parser
Parser
NO_PAREN_FUNCTIONS
STRUCT_TYPE_TOKENS
NESTED_TYPE_TOKENS
ENUM_TYPE_TOKENS
TYPE_TOKENS
SIGNED_TO_UNSIGNED_TYPE_TOKEN
SUBQUERY_PREDICATES
RESERVED_KEYWORDS
DB_CREATABLES
CREATABLES
ID_VAR_TOKENS
INTERVAL_VARS
TABLE_ALIAS_TOKENS
COMMENT_TABLE_ALIAS_TOKENS
UPDATE_ALIAS_TOKENS
TRIM_TYPES
FUNC_TOKENS
CONJUNCTION
EQUALITY
COMPARISON
BITWISE
TERM
FACTOR
EXPONENT
TIMES
TIMESTAMPS
SET_OPERATIONS
JOIN_METHODS
JOIN_SIDES
JOIN_KINDS
JOIN_HINTS
LAMBDAS
COLUMN_OPERATORS
EXPRESSION_PARSERS
STATEMENT_PARSERS
UNARY_PARSERS
PRIMARY_PARSERS
PLACEHOLDER_PARSERS
RANGE_PARSERS
CONSTRAINT_PARSERS
ALTER_PARSERS
SCHEMA_UNNAMED_CONSTRAINTS
INVALID_FUNC_NAME_TOKENS
FUNCTIONS_WITH_ALIASED_ARGS
FUNCTION_PARSERS
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
CLONE_KINDS
OPCLASS_FOLLOW_KEYWORDS
OPTYPE_FOLLOW_TOKENS
TABLE_INDEX_HINT_TOKENS
WINDOW_ALIAS_TOKENS
WINDOW_BEFORE_PAREN_TOKENS
WINDOW_SIDES
FETCH_TOKENS
ADD_CONSTRAINT_TOKENS
DISTINCT_TOKENS
NULL_TOKENS
UNNEST_OFFSET_ALIAS_TOKENS
CONCAT_NULL_OUTPUTS_STRING
PREFIXED_PIVOT_COLUMNS
IDENTIFY_PIVOT_STRINGS
LOG_BASE_FIRST
ALTER_TABLE_ADD_COLUMN_KEYWORD
TABLESAMPLE_CSV
SET_REQUIRES_ASSIGNMENT_DELIMITER
TRIM_PATTERN_FIRST
TYPED_DIVISION
INDEX_OFFSET
UNNEST_COLUMN_ONLY
STRICT_STRING_CONCAT
NORMALIZE_FUNCTIONS
NULL_ORDERING
FORMAT_MAPPING
error_level
error_message_context
max_errors
reset
parse
parse_into
check_errors
raise_error
expression
validate_expression
errors
sql
class Hive.Generator(sqlglot.generator.Generator):
411    class Generator(generator.Generator):
412        LIMIT_FETCH = "LIMIT"
413        TABLESAMPLE_WITH_METHOD = False
414        TABLESAMPLE_SIZE_IS_PERCENT = True
415        JOIN_HINTS = False
416        TABLE_HINTS = False
417        QUERY_HINTS = False
418        INDEX_ON = "ON TABLE"
419        EXTRACT_ALLOWS_QUOTES = False
420        NVL2_SUPPORTED = False
421
422        EXPRESSIONS_WITHOUT_NESTED_CTES = {
423            exp.Insert,
424            exp.Select,
425            exp.Subquery,
426            exp.Union,
427        }
428
429        TYPE_MAPPING = {
430            **generator.Generator.TYPE_MAPPING,
431            exp.DataType.Type.BIT: "BOOLEAN",
432            exp.DataType.Type.DATETIME: "TIMESTAMP",
433            exp.DataType.Type.TEXT: "STRING",
434            exp.DataType.Type.TIME: "TIMESTAMP",
435            exp.DataType.Type.TIMESTAMPTZ: "TIMESTAMP",
436            exp.DataType.Type.VARBINARY: "BINARY",
437        }
438
439        TRANSFORMS = {
440            **generator.Generator.TRANSFORMS,
441            exp.Group: transforms.preprocess([transforms.unalias_group]),
442            exp.Select: transforms.preprocess(
443                [
444                    transforms.eliminate_qualify,
445                    transforms.eliminate_distinct_on,
446                    transforms.unnest_to_explode,
447                ]
448            ),
449            exp.Property: _property_sql,
450            exp.AnyValue: rename_func("FIRST"),
451            exp.ApproxDistinct: approx_count_distinct_sql,
452            exp.ArgMax: arg_max_or_min_no_count("MAX_BY"),
453            exp.ArgMin: arg_max_or_min_no_count("MIN_BY"),
454            exp.ArrayConcat: rename_func("CONCAT"),
455            exp.ArrayJoin: lambda self, e: self.func("CONCAT_WS", e.expression, e.this),
456            exp.ArraySize: rename_func("SIZE"),
457            exp.ArraySort: _array_sort_sql,
458            exp.With: no_recursive_cte_sql,
459            exp.DateAdd: _add_date_sql,
460            exp.DateDiff: _date_diff_sql,
461            exp.DateStrToDate: rename_func("TO_DATE"),
462            exp.DateSub: _add_date_sql,
463            exp.DateToDi: lambda self, e: f"CAST(DATE_FORMAT({self.sql(e, 'this')}, {Hive.DATEINT_FORMAT}) AS INT)",
464            exp.DiToDate: lambda self, e: f"TO_DATE(CAST({self.sql(e, 'this')} AS STRING), {Hive.DATEINT_FORMAT})",
465            exp.FileFormatProperty: lambda self, e: f"STORED AS {self.sql(e, 'this') if isinstance(e.this, exp.InputOutputFormat) else e.name.upper()}",
466            exp.FromBase64: rename_func("UNBASE64"),
467            exp.If: if_sql(),
468            exp.ILike: no_ilike_sql,
469            exp.IsNan: rename_func("ISNAN"),
470            exp.JSONExtract: rename_func("GET_JSON_OBJECT"),
471            exp.JSONExtractScalar: rename_func("GET_JSON_OBJECT"),
472            exp.JSONFormat: _json_format_sql,
473            exp.Left: left_to_substring_sql,
474            exp.Map: var_map_sql,
475            exp.Max: max_or_greatest,
476            exp.MD5Digest: lambda self, e: self.func("UNHEX", self.func("MD5", e.this)),
477            exp.Min: min_or_least,
478            exp.MonthsBetween: lambda self, e: self.func("MONTHS_BETWEEN", e.this, e.expression),
479            exp.NotNullColumnConstraint: lambda self, e: ""
480            if e.args.get("allow_null")
481            else "NOT NULL",
482            exp.VarMap: var_map_sql,
483            exp.Create: _create_sql,
484            exp.Quantile: rename_func("PERCENTILE"),
485            exp.ApproxQuantile: rename_func("PERCENTILE_APPROX"),
486            exp.RegexpExtract: regexp_extract_sql,
487            exp.RegexpReplace: regexp_replace_sql,
488            exp.RegexpLike: lambda self, e: self.binary(e, "RLIKE"),
489            exp.RegexpSplit: rename_func("SPLIT"),
490            exp.Right: right_to_substring_sql,
491            exp.SafeDivide: no_safe_divide_sql,
492            exp.SchemaCommentProperty: lambda self, e: self.naked_property(e),
493            exp.SetAgg: rename_func("COLLECT_SET"),
494            exp.Split: lambda self, e: f"SPLIT({self.sql(e, 'this')}, CONCAT('\\\\Q', {self.sql(e, 'expression')}))",
495            exp.StrPosition: strposition_to_locate_sql,
496            exp.StrToDate: _str_to_date_sql,
497            exp.StrToTime: _str_to_time_sql,
498            exp.StrToUnix: _str_to_unix_sql,
499            exp.StructExtract: struct_extract_sql,
500            exp.TimeStrToDate: rename_func("TO_DATE"),
501            exp.TimeStrToTime: timestrtotime_sql,
502            exp.TimeStrToUnix: rename_func("UNIX_TIMESTAMP"),
503            exp.TimeToStr: _time_to_str,
504            exp.TimeToUnix: rename_func("UNIX_TIMESTAMP"),
505            exp.ToBase64: rename_func("BASE64"),
506            exp.TsOrDiToDi: lambda self, e: f"CAST(SUBSTR(REPLACE(CAST({self.sql(e, 'this')} AS STRING), '-', ''), 1, 8) AS INT)",
507            exp.TsOrDsAdd: lambda self, e: f"DATE_ADD({self.sql(e, 'this')}, {self.sql(e, 'expression')})",
508            exp.TsOrDsToDate: _to_date_sql,
509            exp.TryCast: no_trycast_sql,
510            exp.UnixToStr: lambda self, e: self.func(
511                "FROM_UNIXTIME", e.this, time_format("hive")(self, e)
512            ),
513            exp.UnixToTime: rename_func("FROM_UNIXTIME"),
514            exp.UnixToTimeStr: rename_func("FROM_UNIXTIME"),
515            exp.PartitionedByProperty: lambda self, e: f"PARTITIONED BY {self.sql(e, 'this')}",
516            exp.SerdeProperties: lambda self, e: self.properties(e, prefix="WITH SERDEPROPERTIES"),
517            exp.NumberToStr: rename_func("FORMAT_NUMBER"),
518            exp.LastDateOfMonth: rename_func("LAST_DAY"),
519            exp.National: lambda self, e: self.national_sql(e, prefix=""),
520            exp.ClusteredColumnConstraint: lambda self, e: f"({self.expressions(e, 'this', indent=False)})",
521            exp.NonClusteredColumnConstraint: lambda self, e: f"({self.expressions(e, 'this', indent=False)})",
522            exp.NotForReplicationColumnConstraint: lambda self, e: "",
523            exp.OnProperty: lambda self, e: "",
524            exp.PrimaryKeyColumnConstraint: lambda self, e: "PRIMARY KEY",
525        }
526
527        PROPERTIES_LOCATION = {
528            **generator.Generator.PROPERTIES_LOCATION,
529            exp.FileFormatProperty: exp.Properties.Location.POST_SCHEMA,
530            exp.PartitionedByProperty: exp.Properties.Location.POST_SCHEMA,
531            exp.VolatileProperty: exp.Properties.Location.UNSUPPORTED,
532            exp.WithDataProperty: exp.Properties.Location.UNSUPPORTED,
533        }
534
535        def temporary_storage_provider(self, expression: exp.Create) -> exp.Create:
536            # Hive has no temporary storage provider (there are hive settings though)
537            return expression
538
539        def parameter_sql(self, expression: exp.Parameter) -> str:
540            this = self.sql(expression, "this")
541            expression_sql = self.sql(expression, "expression")
542
543            parent = expression.parent
544            this = f"{this}:{expression_sql}" if expression_sql else this
545
546            if isinstance(parent, exp.EQ) and isinstance(parent.parent, exp.SetItem):
547                # We need to produce SET key = value instead of SET ${key} = value
548                return this
549
550            return f"${{{this}}}"
551
552        def schema_sql(self, expression: exp.Schema) -> str:
553            for ordered in expression.find_all(exp.Ordered):
554                if ordered.args.get("desc") is False:
555                    ordered.set("desc", None)
556
557            return super().schema_sql(expression)
558
559        def constraint_sql(self, expression: exp.Constraint) -> str:
560            for prop in list(expression.find_all(exp.Properties)):
561                prop.pop()
562
563            this = self.sql(expression, "this")
564            expressions = self.expressions(expression, sep=" ", flat=True)
565            return f"CONSTRAINT {this} {expressions}"
566
567        def rowformatserdeproperty_sql(self, expression: exp.RowFormatSerdeProperty) -> str:
568            serde_props = self.sql(expression, "serde_properties")
569            serde_props = f" {serde_props}" if serde_props else ""
570            return f"ROW FORMAT SERDE {self.sql(expression, 'this')}{serde_props}"
571
572        def arrayagg_sql(self, expression: exp.ArrayAgg) -> str:
573            return self.func(
574                "COLLECT_LIST",
575                expression.this.this if isinstance(expression.this, exp.Order) else expression.this,
576            )
577
578        def with_properties(self, properties: exp.Properties) -> str:
579            return self.properties(properties, prefix=self.seg("TBLPROPERTIES"))
580
581        def datatype_sql(self, expression: exp.DataType) -> str:
582            if (
583                expression.this in (exp.DataType.Type.VARCHAR, exp.DataType.Type.NVARCHAR)
584                and not expression.expressions
585            ):
586                expression = exp.DataType.build("text")
587            elif expression.this in exp.DataType.TEMPORAL_TYPES:
588                expression = exp.DataType.build(expression.this)
589            elif expression.is_type("float"):
590                size_expression = expression.find(exp.DataTypeParam)
591                if size_expression:
592                    size = int(size_expression.name)
593                    expression = (
594                        exp.DataType.build("float") if size <= 32 else exp.DataType.build("double")
595                    )
596
597            return super().datatype_sql(expression)
598
599        def version_sql(self, expression: exp.Version) -> str:
600            sql = super().version_sql(expression)
601            return sql.replace("FOR ", "", 1)

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
LIMIT_FETCH = 'LIMIT'
TABLESAMPLE_WITH_METHOD = False
TABLESAMPLE_SIZE_IS_PERCENT = True
JOIN_HINTS = False
TABLE_HINTS = False
QUERY_HINTS = False
INDEX_ON = 'ON TABLE'
EXTRACT_ALLOWS_QUOTES = False
NVL2_SUPPORTED = False
EXPRESSIONS_WITHOUT_NESTED_CTES = {<class 'sqlglot.expressions.Select'>, <class 'sqlglot.expressions.Insert'>, <class 'sqlglot.expressions.Subquery'>, <class 'sqlglot.expressions.Union'>}
TYPE_MAPPING = {<Type.NCHAR: 'NCHAR'>: 'CHAR', <Type.NVARCHAR: 'NVARCHAR'>: 'VARCHAR', <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.BIT: 'BIT'>: 'BOOLEAN', <Type.DATETIME: 'DATETIME'>: 'TIMESTAMP', <Type.TEXT: 'TEXT'>: 'STRING', <Type.TIME: 'TIME'>: 'TIMESTAMP', <Type.TIMESTAMPTZ: 'TIMESTAMPTZ'>: 'TIMESTAMP', <Type.VARBINARY: 'VARBINARY'>: 'BINARY'}
TRANSFORMS = {<class 'sqlglot.expressions.DateAdd'>: <function _add_date_sql>, <class 'sqlglot.expressions.TsOrDsAdd'>: <function Hive.Generator.<lambda>>, <class 'sqlglot.expressions.CaseSpecificColumnConstraint'>: <function Generator.<lambda>>, <class 'sqlglot.expressions.CharacterSetColumnConstraint'>: <function Generator.<lambda>>, <class 'sqlglot.expressions.CharacterSetProperty'>: <function Generator.<lambda>>, <class 'sqlglot.expressions.CheckColumnConstraint'>: <function Generator.<lambda>>, <class 'sqlglot.expressions.ClusteredColumnConstraint'>: <function Hive.Generator.<lambda>>, <class 'sqlglot.expressions.CollateColumnConstraint'>: <function Generator.<lambda>>, <class 'sqlglot.expressions.CopyGrantsProperty'>: <function Generator.<lambda>>, <class 'sqlglot.expressions.CommentColumnConstraint'>: <function Generator.<lambda>>, <class 'sqlglot.expressions.DateFormatColumnConstraint'>: <function Generator.<lambda>>, <class 'sqlglot.expressions.DefaultColumnConstraint'>: <function Generator.<lambda>>, <class 'sqlglot.expressions.EncodeColumnConstraint'>: <function Generator.<lambda>>, <class 'sqlglot.expressions.ExecuteAsProperty'>: <function Generator.<lambda>>, <class 'sqlglot.expressions.ExternalProperty'>: <function Generator.<lambda>>, <class 'sqlglot.expressions.HeapProperty'>: <function Generator.<lambda>>, <class 'sqlglot.expressions.InlineLengthColumnConstraint'>: <function Generator.<lambda>>, <class 'sqlglot.expressions.InputModelProperty'>: <function Generator.<lambda>>, <class 'sqlglot.expressions.IntervalSpan'>: <function Generator.<lambda>>, <class 'sqlglot.expressions.LanguageProperty'>: <function Generator.<lambda>>, <class 'sqlglot.expressions.LocationProperty'>: <function Generator.<lambda>>, <class 'sqlglot.expressions.LogProperty'>: <function Generator.<lambda>>, <class 'sqlglot.expressions.MaterializedProperty'>: <function Generator.<lambda>>, <class 'sqlglot.expressions.NoPrimaryIndexProperty'>: <function Generator.<lambda>>, <class 'sqlglot.expressions.NonClusteredColumnConstraint'>: <function Hive.Generator.<lambda>>, <class 'sqlglot.expressions.NotForReplicationColumnConstraint'>: <function Hive.Generator.<lambda>>, <class 'sqlglot.expressions.OnCommitProperty'>: <function Generator.<lambda>>, <class 'sqlglot.expressions.OnProperty'>: <function Hive.Generator.<lambda>>, <class 'sqlglot.expressions.OnUpdateColumnConstraint'>: <function Generator.<lambda>>, <class 'sqlglot.expressions.OutputModelProperty'>: <function Generator.<lambda>>, <class 'sqlglot.expressions.PathColumnConstraint'>: <function Generator.<lambda>>, <class 'sqlglot.expressions.RemoteWithConnectionModelProperty'>: <function Generator.<lambda>>, <class 'sqlglot.expressions.ReturnsProperty'>: <function Generator.<lambda>>, <class 'sqlglot.expressions.SampleProperty'>: <function Generator.<lambda>>, <class 'sqlglot.expressions.SetProperty'>: <function Generator.<lambda>>, <class 'sqlglot.expressions.SettingsProperty'>: <function Generator.<lambda>>, <class 'sqlglot.expressions.SqlSecurityProperty'>: <function Generator.<lambda>>, <class 'sqlglot.expressions.StabilityProperty'>: <function Generator.<lambda>>, <class 'sqlglot.expressions.TemporaryProperty'>: <function Generator.<lambda>>, <class 'sqlglot.expressions.ToTableProperty'>: <function Generator.<lambda>>, <class 'sqlglot.expressions.TransientProperty'>: <function Generator.<lambda>>, <class 'sqlglot.expressions.TransformModelProperty'>: <function Generator.<lambda>>, <class 'sqlglot.expressions.TitleColumnConstraint'>: <function Generator.<lambda>>, <class 'sqlglot.expressions.UppercaseColumnConstraint'>: <function Generator.<lambda>>, <class 'sqlglot.expressions.VarMap'>: <function var_map_sql>, <class 'sqlglot.expressions.VolatileProperty'>: <function Generator.<lambda>>, <class 'sqlglot.expressions.WithJournalTableProperty'>: <function Generator.<lambda>>, <class 'sqlglot.expressions.Group'>: <function preprocess.<locals>._to_sql>, <class 'sqlglot.expressions.Select'>: <function preprocess.<locals>._to_sql>, <class 'sqlglot.expressions.Property'>: <function _property_sql>, <class 'sqlglot.expressions.AnyValue'>: <function rename_func.<locals>.<lambda>>, <class 'sqlglot.expressions.ApproxDistinct'>: <function approx_count_distinct_sql>, <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.ArrayConcat'>: <function rename_func.<locals>.<lambda>>, <class 'sqlglot.expressions.ArrayJoin'>: <function Hive.Generator.<lambda>>, <class 'sqlglot.expressions.ArraySize'>: <function rename_func.<locals>.<lambda>>, <class 'sqlglot.expressions.ArraySort'>: <function _array_sort_sql>, <class 'sqlglot.expressions.With'>: <function no_recursive_cte_sql>, <class 'sqlglot.expressions.DateDiff'>: <function _date_diff_sql>, <class 'sqlglot.expressions.DateStrToDate'>: <function rename_func.<locals>.<lambda>>, <class 'sqlglot.expressions.DateSub'>: <function _add_date_sql>, <class 'sqlglot.expressions.DateToDi'>: <function Hive.Generator.<lambda>>, <class 'sqlglot.expressions.DiToDate'>: <function Hive.Generator.<lambda>>, <class 'sqlglot.expressions.FileFormatProperty'>: <function Hive.Generator.<lambda>>, <class 'sqlglot.expressions.FromBase64'>: <function rename_func.<locals>.<lambda>>, <class 'sqlglot.expressions.If'>: <function if_sql.<locals>._if_sql>, <class 'sqlglot.expressions.ILike'>: <function no_ilike_sql>, <class 'sqlglot.expressions.IsNan'>: <function rename_func.<locals>.<lambda>>, <class 'sqlglot.expressions.JSONExtract'>: <function rename_func.<locals>.<lambda>>, <class 'sqlglot.expressions.JSONExtractScalar'>: <function rename_func.<locals>.<lambda>>, <class 'sqlglot.expressions.JSONFormat'>: <function _json_format_sql>, <class 'sqlglot.expressions.Left'>: <function left_to_substring_sql>, <class 'sqlglot.expressions.Map'>: <function var_map_sql>, <class 'sqlglot.expressions.Max'>: <function max_or_greatest>, <class 'sqlglot.expressions.MD5Digest'>: <function Hive.Generator.<lambda>>, <class 'sqlglot.expressions.Min'>: <function min_or_least>, <class 'sqlglot.expressions.MonthsBetween'>: <function Hive.Generator.<lambda>>, <class 'sqlglot.expressions.NotNullColumnConstraint'>: <function Hive.Generator.<lambda>>, <class 'sqlglot.expressions.Create'>: <function _create_sql>, <class 'sqlglot.expressions.Quantile'>: <function rename_func.<locals>.<lambda>>, <class 'sqlglot.expressions.ApproxQuantile'>: <function rename_func.<locals>.<lambda>>, <class 'sqlglot.expressions.RegexpExtract'>: <function regexp_extract_sql>, <class 'sqlglot.expressions.RegexpReplace'>: <function regexp_replace_sql>, <class 'sqlglot.expressions.RegexpLike'>: <function Hive.Generator.<lambda>>, <class 'sqlglot.expressions.RegexpSplit'>: <function rename_func.<locals>.<lambda>>, <class 'sqlglot.expressions.Right'>: <function right_to_substring_sql>, <class 'sqlglot.expressions.SafeDivide'>: <function no_safe_divide_sql>, <class 'sqlglot.expressions.SchemaCommentProperty'>: <function Hive.Generator.<lambda>>, <class 'sqlglot.expressions.SetAgg'>: <function rename_func.<locals>.<lambda>>, <class 'sqlglot.expressions.Split'>: <function Hive.Generator.<lambda>>, <class 'sqlglot.expressions.StrPosition'>: <function strposition_to_locate_sql>, <class 'sqlglot.expressions.StrToDate'>: <function _str_to_date_sql>, <class 'sqlglot.expressions.StrToTime'>: <function _str_to_time_sql>, <class 'sqlglot.expressions.StrToUnix'>: <function _str_to_unix_sql>, <class 'sqlglot.expressions.StructExtract'>: <function struct_extract_sql>, <class 'sqlglot.expressions.TimeStrToDate'>: <function rename_func.<locals>.<lambda>>, <class 'sqlglot.expressions.TimeStrToTime'>: <function timestrtotime_sql>, <class 'sqlglot.expressions.TimeStrToUnix'>: <function rename_func.<locals>.<lambda>>, <class 'sqlglot.expressions.TimeToStr'>: <function _time_to_str>, <class 'sqlglot.expressions.TimeToUnix'>: <function rename_func.<locals>.<lambda>>, <class 'sqlglot.expressions.ToBase64'>: <function rename_func.<locals>.<lambda>>, <class 'sqlglot.expressions.TsOrDiToDi'>: <function Hive.Generator.<lambda>>, <class 'sqlglot.expressions.TsOrDsToDate'>: <function _to_date_sql>, <class 'sqlglot.expressions.TryCast'>: <function no_trycast_sql>, <class 'sqlglot.expressions.UnixToStr'>: <function Hive.Generator.<lambda>>, <class 'sqlglot.expressions.UnixToTime'>: <function rename_func.<locals>.<lambda>>, <class 'sqlglot.expressions.UnixToTimeStr'>: <function rename_func.<locals>.<lambda>>, <class 'sqlglot.expressions.PartitionedByProperty'>: <function Hive.Generator.<lambda>>, <class 'sqlglot.expressions.SerdeProperties'>: <function Hive.Generator.<lambda>>, <class 'sqlglot.expressions.NumberToStr'>: <function rename_func.<locals>.<lambda>>, <class 'sqlglot.expressions.LastDateOfMonth'>: <function rename_func.<locals>.<lambda>>, <class 'sqlglot.expressions.National'>: <function Hive.Generator.<lambda>>, <class 'sqlglot.expressions.PrimaryKeyColumnConstraint'>: <function Hive.Generator.<lambda>>}
PROPERTIES_LOCATION = {<class 'sqlglot.expressions.AlgorithmProperty'>: <Location.POST_CREATE: 'POST_CREATE'>, <class 'sqlglot.expressions.AutoIncrementProperty'>: <Location.POST_SCHEMA: 'POST_SCHEMA'>, <class 'sqlglot.expressions.BlockCompressionProperty'>: <Location.POST_NAME: 'POST_NAME'>, <class 'sqlglot.expressions.CharacterSetProperty'>: <Location.POST_SCHEMA: 'POST_SCHEMA'>, <class 'sqlglot.expressions.ChecksumProperty'>: <Location.POST_NAME: 'POST_NAME'>, <class 'sqlglot.expressions.CollateProperty'>: <Location.POST_SCHEMA: 'POST_SCHEMA'>, <class 'sqlglot.expressions.CopyGrantsProperty'>: <Location.POST_SCHEMA: 'POST_SCHEMA'>, <class 'sqlglot.expressions.Cluster'>: <Location.POST_SCHEMA: 'POST_SCHEMA'>, <class 'sqlglot.expressions.ClusteredByProperty'>: <Location.POST_SCHEMA: 'POST_SCHEMA'>, <class 'sqlglot.expressions.DataBlocksizeProperty'>: <Location.POST_NAME: 'POST_NAME'>, <class 'sqlglot.expressions.DefinerProperty'>: <Location.POST_CREATE: 'POST_CREATE'>, <class 'sqlglot.expressions.DictRange'>: <Location.POST_SCHEMA: 'POST_SCHEMA'>, <class 'sqlglot.expressions.DictProperty'>: <Location.POST_SCHEMA: 'POST_SCHEMA'>, <class 'sqlglot.expressions.DistKeyProperty'>: <Location.POST_SCHEMA: 'POST_SCHEMA'>, <class 'sqlglot.expressions.DistStyleProperty'>: <Location.POST_SCHEMA: 'POST_SCHEMA'>, <class 'sqlglot.expressions.EngineProperty'>: <Location.POST_SCHEMA: 'POST_SCHEMA'>, <class 'sqlglot.expressions.ExecuteAsProperty'>: <Location.POST_SCHEMA: 'POST_SCHEMA'>, <class 'sqlglot.expressions.ExternalProperty'>: <Location.POST_CREATE: 'POST_CREATE'>, <class 'sqlglot.expressions.FallbackProperty'>: <Location.POST_NAME: 'POST_NAME'>, <class 'sqlglot.expressions.FileFormatProperty'>: <Location.POST_SCHEMA: 'POST_SCHEMA'>, <class 'sqlglot.expressions.FreespaceProperty'>: <Location.POST_NAME: 'POST_NAME'>, <class 'sqlglot.expressions.HeapProperty'>: <Location.POST_WITH: 'POST_WITH'>, <class 'sqlglot.expressions.InputModelProperty'>: <Location.POST_SCHEMA: 'POST_SCHEMA'>, <class 'sqlglot.expressions.IsolatedLoadingProperty'>: <Location.POST_NAME: 'POST_NAME'>, <class 'sqlglot.expressions.JournalProperty'>: <Location.POST_NAME: 'POST_NAME'>, <class 'sqlglot.expressions.LanguageProperty'>: <Location.POST_SCHEMA: 'POST_SCHEMA'>, <class 'sqlglot.expressions.LikeProperty'>: <Location.POST_SCHEMA: 'POST_SCHEMA'>, <class 'sqlglot.expressions.LocationProperty'>: <Location.POST_SCHEMA: 'POST_SCHEMA'>, <class 'sqlglot.expressions.LockingProperty'>: <Location.POST_ALIAS: 'POST_ALIAS'>, <class 'sqlglot.expressions.LogProperty'>: <Location.POST_NAME: 'POST_NAME'>, <class 'sqlglot.expressions.MaterializedProperty'>: <Location.POST_CREATE: 'POST_CREATE'>, <class 'sqlglot.expressions.MergeBlockRatioProperty'>: <Location.POST_NAME: 'POST_NAME'>, <class 'sqlglot.expressions.NoPrimaryIndexProperty'>: <Location.POST_EXPRESSION: 'POST_EXPRESSION'>, <class 'sqlglot.expressions.OnProperty'>: <Location.POST_SCHEMA: 'POST_SCHEMA'>, <class 'sqlglot.expressions.OnCommitProperty'>: <Location.POST_EXPRESSION: 'POST_EXPRESSION'>, <class 'sqlglot.expressions.Order'>: <Location.POST_SCHEMA: 'POST_SCHEMA'>, <class 'sqlglot.expressions.OutputModelProperty'>: <Location.POST_SCHEMA: 'POST_SCHEMA'>, <class 'sqlglot.expressions.PartitionedByProperty'>: <Location.POST_SCHEMA: 'POST_SCHEMA'>, <class 'sqlglot.expressions.PartitionedOfProperty'>: <Location.POST_SCHEMA: 'POST_SCHEMA'>, <class 'sqlglot.expressions.PrimaryKey'>: <Location.POST_SCHEMA: 'POST_SCHEMA'>, <class 'sqlglot.expressions.Property'>: <Location.POST_WITH: 'POST_WITH'>, <class 'sqlglot.expressions.RemoteWithConnectionModelProperty'>: <Location.POST_SCHEMA: 'POST_SCHEMA'>, <class 'sqlglot.expressions.ReturnsProperty'>: <Location.POST_SCHEMA: 'POST_SCHEMA'>, <class 'sqlglot.expressions.RowFormatProperty'>: <Location.POST_SCHEMA: 'POST_SCHEMA'>, <class 'sqlglot.expressions.RowFormatDelimitedProperty'>: <Location.POST_SCHEMA: 'POST_SCHEMA'>, <class 'sqlglot.expressions.RowFormatSerdeProperty'>: <Location.POST_SCHEMA: 'POST_SCHEMA'>, <class 'sqlglot.expressions.SampleProperty'>: <Location.POST_SCHEMA: 'POST_SCHEMA'>, <class 'sqlglot.expressions.SchemaCommentProperty'>: <Location.POST_SCHEMA: 'POST_SCHEMA'>, <class 'sqlglot.expressions.SerdeProperties'>: <Location.POST_SCHEMA: 'POST_SCHEMA'>, <class 'sqlglot.expressions.Set'>: <Location.POST_SCHEMA: 'POST_SCHEMA'>, <class 'sqlglot.expressions.SettingsProperty'>: <Location.POST_SCHEMA: 'POST_SCHEMA'>, <class 'sqlglot.expressions.SetProperty'>: <Location.POST_CREATE: 'POST_CREATE'>, <class 'sqlglot.expressions.SortKeyProperty'>: <Location.POST_SCHEMA: 'POST_SCHEMA'>, <class 'sqlglot.expressions.SqlSecurityProperty'>: <Location.POST_CREATE: 'POST_CREATE'>, <class 'sqlglot.expressions.StabilityProperty'>: <Location.POST_SCHEMA: 'POST_SCHEMA'>, <class 'sqlglot.expressions.TemporaryProperty'>: <Location.POST_CREATE: 'POST_CREATE'>, <class 'sqlglot.expressions.ToTableProperty'>: <Location.POST_SCHEMA: 'POST_SCHEMA'>, <class 'sqlglot.expressions.TransientProperty'>: <Location.POST_CREATE: 'POST_CREATE'>, <class 'sqlglot.expressions.TransformModelProperty'>: <Location.POST_SCHEMA: 'POST_SCHEMA'>, <class 'sqlglot.expressions.MergeTreeTTL'>: <Location.POST_SCHEMA: 'POST_SCHEMA'>, <class 'sqlglot.expressions.VolatileProperty'>: <Location.UNSUPPORTED: 'UNSUPPORTED'>, <class 'sqlglot.expressions.WithDataProperty'>: <Location.UNSUPPORTED: 'UNSUPPORTED'>, <class 'sqlglot.expressions.WithJournalTableProperty'>: <Location.POST_NAME: 'POST_NAME'>, <class 'sqlglot.expressions.WithSystemVersioningProperty'>: <Location.POST_SCHEMA: 'POST_SCHEMA'>}
def temporary_storage_provider( self, expression: sqlglot.expressions.Create) -> sqlglot.expressions.Create:
535        def temporary_storage_provider(self, expression: exp.Create) -> exp.Create:
536            # Hive has no temporary storage provider (there are hive settings though)
537            return expression
def parameter_sql(self, expression: sqlglot.expressions.Parameter) -> str:
539        def parameter_sql(self, expression: exp.Parameter) -> str:
540            this = self.sql(expression, "this")
541            expression_sql = self.sql(expression, "expression")
542
543            parent = expression.parent
544            this = f"{this}:{expression_sql}" if expression_sql else this
545
546            if isinstance(parent, exp.EQ) and isinstance(parent.parent, exp.SetItem):
547                # We need to produce SET key = value instead of SET ${key} = value
548                return this
549
550            return f"${{{this}}}"
def schema_sql(self, expression: sqlglot.expressions.Schema) -> str:
552        def schema_sql(self, expression: exp.Schema) -> str:
553            for ordered in expression.find_all(exp.Ordered):
554                if ordered.args.get("desc") is False:
555                    ordered.set("desc", None)
556
557            return super().schema_sql(expression)
def constraint_sql(self, expression: sqlglot.expressions.Constraint) -> str:
559        def constraint_sql(self, expression: exp.Constraint) -> str:
560            for prop in list(expression.find_all(exp.Properties)):
561                prop.pop()
562
563            this = self.sql(expression, "this")
564            expressions = self.expressions(expression, sep=" ", flat=True)
565            return f"CONSTRAINT {this} {expressions}"
def rowformatserdeproperty_sql(self, expression: sqlglot.expressions.RowFormatSerdeProperty) -> str:
567        def rowformatserdeproperty_sql(self, expression: exp.RowFormatSerdeProperty) -> str:
568            serde_props = self.sql(expression, "serde_properties")
569            serde_props = f" {serde_props}" if serde_props else ""
570            return f"ROW FORMAT SERDE {self.sql(expression, 'this')}{serde_props}"
def arrayagg_sql(self, expression: sqlglot.expressions.ArrayAgg) -> str:
572        def arrayagg_sql(self, expression: exp.ArrayAgg) -> str:
573            return self.func(
574                "COLLECT_LIST",
575                expression.this.this if isinstance(expression.this, exp.Order) else expression.this,
576            )
def with_properties(self, properties: sqlglot.expressions.Properties) -> str:
578        def with_properties(self, properties: exp.Properties) -> str:
579            return self.properties(properties, prefix=self.seg("TBLPROPERTIES"))
def datatype_sql(self, expression: sqlglot.expressions.DataType) -> str:
581        def datatype_sql(self, expression: exp.DataType) -> str:
582            if (
583                expression.this in (exp.DataType.Type.VARCHAR, exp.DataType.Type.NVARCHAR)
584                and not expression.expressions
585            ):
586                expression = exp.DataType.build("text")
587            elif expression.this in exp.DataType.TEMPORAL_TYPES:
588                expression = exp.DataType.build(expression.this)
589            elif expression.is_type("float"):
590                size_expression = expression.find(exp.DataTypeParam)
591                if size_expression:
592                    size = int(size_expression.name)
593                    expression = (
594                        exp.DataType.build("float") if size <= 32 else exp.DataType.build("double")
595                    )
596
597            return super().datatype_sql(expression)
def version_sql(self, expression: sqlglot.expressions.Version) -> str:
599        def version_sql(self, expression: exp.Version) -> str:
600            sql = super().version_sql(expression)
601            return sql.replace("FOR ", "", 1)
SELECT_KINDS: Tuple[str, ...] = ()
SAFE_DIVISION = True
INVERSE_TIME_MAPPING: Dict[str, str] = {'%Y': 'yyyy', '%y': 'yy', '%B': 'MMMM', '%b': 'MMM', '%m': 'MM', '%-m': 'M', '%d': 'dd', '%-d': 'd', '%H': 'HH', '%-H': 'H', '%I': 'hh', '%-I': 'h', '%M': 'mm', '%-M': 'm', '%S': 'ss', '%-S': 's', '%f': 'SSSSSS', '%p': 'a', '%j': 'DD', '%-j': 'D', '%a': 'EEE', '%A': 'EEEE'}
INVERSE_TIME_TRIE: Dict = {'%': {'Y': {0: True}, 'y': {0: True}, 'B': {0: True}, 'b': {0: True}, 'm': {0: True}, '-': {'m': {0: True}, 'd': {0: True}, 'H': {0: True}, 'I': {0: True}, 'M': {0: True}, 'S': {0: True}, 'j': {0: True}}, 'd': {0: True}, 'H': {0: True}, 'I': {0: True}, 'M': {0: True}, 'S': {0: True}, 'f': {0: True}, 'p': {0: True}, 'j': {0: True}, 'a': {0: True}, 'A': {0: True}}}
INVERSE_ESCAPE_SEQUENCES: Dict[str, str] = {}
ALIAS_POST_TABLESAMPLE = True
IDENTIFIERS_CAN_START_WITH_DIGIT = True
@classmethod
def can_identify(text: str, identify: str | bool = 'safe') -> bool:
288    @classmethod
289    def can_identify(cls, text: str, identify: str | bool = "safe") -> bool:
290        """Checks if text can be identified given an identify option.
291
292        Args:
293            text: The text to check.
294            identify:
295                "always" or `True`: Always returns true.
296                "safe": True if the identifier is case-insensitive.
297
298        Returns:
299            Whether or not the given text can be identified.
300        """
301        if identify is True or identify == "always":
302            return True
303
304        if identify == "safe":
305            return not cls.case_sensitive(text)
306
307        return False

Checks if text can be identified given an identify option.

Arguments:
  • text: The text to check.
  • identify: "always" or True: Always returns true. "safe": True if the identifier is case-insensitive.
Returns:

Whether or not the given text can be identified.

QUOTE_START = "'"
QUOTE_END = "'"
IDENTIFIER_START = '`'
IDENTIFIER_END = '`'
TOKENIZER_CLASS = <class 'Hive.Tokenizer'>
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
Inherited Members
sqlglot.generator.Generator
Generator
LOG_BASE_FIRST
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
RENAME_TABLE_WITH_DB
GROUPINGS_SEP
QUERY_HINT_SEP
IS_BOOL_ALLOWED
DUPLICATE_KEY_UPDATE_WITH_SET
LIMIT_IS_TOP
RETURNING_END
COLUMN_JOIN_MARKS_SUPPORTED
TZ_TO_WITH_TIME_ZONE
VALUES_AS_TABLE
ALTER_TABLE_ADD_COLUMN_KEYWORD
UNNEST_WITH_ORDINALITY
AGGREGATE_FILTER_SUPPORTED
SEMI_ANTI_JOIN_WITH_SIDE
SUPPORTS_PARAMETERS
COMPUTED_COLUMN_WITH_TYPE
SUPPORTS_TABLE_COPY
TABLESAMPLE_REQUIRES_PARENS
COLLATE_IS_FUNC
DATA_TYPE_SPECIFIERS_ALLOWED
ENSURE_BOOLS
CTE_RECURSIVE_KEYWORD_REQUIRED
TYPED_DIVISION
STAR_MAPPING
TIME_PART_SINGULARS
TOKEN_MAPPING
STRUCT_DELIMITER
PARAMETER_TOKEN
RESERVED_KEYWORDS
WITH_SEPARATED_COMMENTS
EXCLUDE_COMMENTS
UNWRAPPED_INTERVAL_VALUES
KEY_VALUE_DEFINITONS
SENTINEL_LINE_BREAK
INDEX_OFFSET
UNNEST_COLUMN_ONLY
STRICT_STRING_CONCAT
NORMALIZE_FUNCTIONS
NULL_ORDERING
pretty
identify
normalize
pad
unsupported_level
max_unsupported
leading_comma
max_text_width
comments
normalize_functions
unsupported_messages
generate
unsupported
sep
seg
pad_comment
maybe_comment
wrap
no_identify
normalize_func
indent
sql
uncache_sql
cache_sql
characterset_sql
column_sql
columnposition_sql
columndef_sql
columnconstraint_sql
computedcolumnconstraint_sql
autoincrementcolumnconstraint_sql
compresscolumnconstraint_sql
generatedasidentitycolumnconstraint_sql
generatedasrowcolumnconstraint_sql
periodforsystemtimeconstraint_sql
notnullcolumnconstraint_sql
primarykeycolumnconstraint_sql
uniquecolumnconstraint_sql
createable_sql
create_sql
clone_sql
describe_sql
prepend_ctes
with_sql
cte_sql
tablealias_sql
bitstring_sql
hexstring_sql
bytestring_sql
rawstring_sql
datatypeparam_sql
directory_sql
delete_sql
drop_sql
except_sql
except_op
fetch_sql
filter_sql
hint_sql
index_sql
identifier_sql
inputoutputformat_sql
national_sql
partition_sql
properties_sql
root_properties
properties
locate_properties
property_name
property_sql
likeproperty_sql
fallbackproperty_sql
journalproperty_sql
freespaceproperty_sql
checksumproperty_sql
mergeblockratioproperty_sql
datablocksizeproperty_sql
blockcompressionproperty_sql
isolatedloadingproperty_sql
partitionboundspec_sql
partitionedofproperty_sql
lockingproperty_sql
withdataproperty_sql
withsystemversioningproperty_sql
insert_sql
intersect_sql
intersect_op
introducer_sql
kill_sql
pseudotype_sql
objectidentifier_sql
onconflict_sql
returning_sql
rowformatdelimitedproperty_sql
withtablehint_sql
indextablehint_sql
table_sql
tablesample_sql
pivot_sql
tuple_sql
update_sql
values_sql
var_sql
into_sql
from_sql
group_sql
having_sql
connect_sql
prior_sql
join_sql
lambda_sql
lateral_sql
limit_sql
offset_sql
setitem_sql
set_sql
pragma_sql
lock_sql
literal_sql
escape_str
loaddata_sql
null_sql
boolean_sql
order_sql
cluster_sql
distribute_sql
sort_sql
ordered_sql
matchrecognize_sql
query_modifiers
offset_limit_modifiers
after_having_modifiers
after_limit_modifiers
select_sql
schema_columns_sql
star_sql
sessionparameter_sql
placeholder_sql
subquery_sql
qualify_sql
union_sql
union_op
unnest_sql
where_sql
window_sql
partition_by_sql
windowspec_sql
withingroup_sql
between_sql
bracket_sql
safebracket_sql
all_sql
any_sql
exists_sql
case_sql
nextvaluefor_sql
extract_sql
trim_sql
safeconcat_sql
check_sql
foreignkey_sql
primarykey_sql
if_sql
matchagainst_sql
jsonkeyvalue_sql
formatjson_sql
jsonobject_sql
jsonarray_sql
jsonarrayagg_sql
jsoncolumndef_sql
jsonschema_sql
jsontable_sql
openjsoncolumndef_sql
openjson_sql
in_sql
in_unnest_op
interval_sql
return_sql
reference_sql
anonymous_sql
paren_sql
neg_sql
not_sql
alias_sql
aliases_sql
attimezone_sql
add_sql
and_sql
xor_sql
connector_sql
bitwiseand_sql
bitwiseleftshift_sql
bitwisenot_sql
bitwiseor_sql
bitwiserightshift_sql
bitwisexor_sql
cast_sql
currentdate_sql
collate_sql
command_sql
comment_sql
mergetreettlaction_sql
mergetreettl_sql
transaction_sql
commit_sql
rollback_sql
altercolumn_sql
renametable_sql
altertable_sql
add_column_sql
droppartition_sql
addconstraint_sql
distinct_sql
ignorenulls_sql
respectnulls_sql
intdiv_sql
dpipe_sql
safedpipe_sql
div_sql
overlaps_sql
distance_sql
dot_sql
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