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    expression = expression.copy()
 57
 58    # remove UNIQUE column constraints
 59    for constraint in expression.find_all(exp.UniqueColumnConstraint):
 60        if constraint.parent:
 61            constraint.parent.pop()
 62
 63    properties = expression.args.get("properties")
 64    temporary = any(
 65        isinstance(prop, exp.TemporaryProperty)
 66        for prop in (properties.expressions if properties else [])
 67    )
 68
 69    # CTAS with temp tables map to CREATE TEMPORARY VIEW
 70    kind = expression.args["kind"]
 71    if kind.upper() == "TABLE" and temporary:
 72        if expression.expression:
 73            return f"CREATE TEMPORARY VIEW {self.sql(expression, 'this')} AS {self.sql(expression, 'expression')}"
 74        else:
 75            # CREATE TEMPORARY TABLE may require storage provider
 76            expression = self.temporary_storage_provider(expression)
 77
 78    return create_with_partitions_sql(self, expression)
 79
 80
 81def _add_date_sql(self: Hive.Generator, expression: exp.DateAdd | exp.DateSub) -> str:
 82    unit = expression.text("unit").upper()
 83    func, multiplier = DATE_DELTA_INTERVAL.get(unit, ("DATE_ADD", 1))
 84
 85    if isinstance(expression, exp.DateSub):
 86        multiplier *= -1
 87
 88    if expression.expression.is_number:
 89        modified_increment = exp.Literal.number(int(expression.text("expression")) * multiplier)
 90    else:
 91        modified_increment = expression.expression.copy()
 92        if multiplier != 1:
 93            modified_increment = exp.Mul(  # type: ignore
 94                this=modified_increment, expression=exp.Literal.number(multiplier)
 95            )
 96
 97    return self.func(func, expression.this, modified_increment)
 98
 99
100def _date_diff_sql(self: Hive.Generator, expression: exp.DateDiff) -> str:
101    unit = expression.text("unit").upper()
102
103    factor = TIME_DIFF_FACTOR.get(unit)
104    if factor is not None:
105        left = self.sql(expression, "this")
106        right = self.sql(expression, "expression")
107        sec_diff = f"UNIX_TIMESTAMP({left}) - UNIX_TIMESTAMP({right})"
108        return f"({sec_diff}){factor}" if factor else sec_diff
109
110    months_between = unit in DIFF_MONTH_SWITCH
111    sql_func = "MONTHS_BETWEEN" if months_between else "DATEDIFF"
112    _, multiplier = DATE_DELTA_INTERVAL.get(unit, ("", 1))
113    multiplier_sql = f" / {multiplier}" if multiplier > 1 else ""
114    diff_sql = f"{sql_func}({self.format_args(expression.this, expression.expression)})"
115
116    if months_between:
117        # MONTHS_BETWEEN returns a float, so we need to truncate the fractional part
118        diff_sql = f"CAST({diff_sql} AS INT)"
119
120    return f"{diff_sql}{multiplier_sql}"
121
122
123def _json_format_sql(self: Hive.Generator, expression: exp.JSONFormat) -> str:
124    this = expression.this
125    if is_parse_json(this) and 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("FROM_JSON", wrapped_json, self.func("SCHEMA_OF_JSON", wrapped_json))
131        to_json = self.func("TO_JSON", from_json)
132
133        # This strips the [, ] delimiters of the dummy array printed by TO_JSON
134        return self.func("REGEXP_EXTRACT", to_json, "'^.(.*).$'", "1")
135
136    return self.func("TO_JSON", this, expression.args.get("options"))
137
138
139def _array_sort_sql(self: Hive.Generator, expression: exp.ArraySort) -> str:
140    if expression.expression:
141        self.unsupported("Hive SORT_ARRAY does not support a comparator")
142    return f"SORT_ARRAY({self.sql(expression, 'this')})"
143
144
145def _property_sql(self: Hive.Generator, expression: exp.Property) -> str:
146    return f"{self.property_name(expression, string_key=True)}={self.sql(expression, 'value')}"
147
148
149def _str_to_unix_sql(self: Hive.Generator, expression: exp.StrToUnix) -> str:
150    return self.func("UNIX_TIMESTAMP", expression.this, time_format("hive")(self, expression))
151
152
153def _str_to_date_sql(self: Hive.Generator, expression: exp.StrToDate) -> str:
154    this = self.sql(expression, "this")
155    time_format = self.format_time(expression)
156    if time_format not in (Hive.TIME_FORMAT, Hive.DATE_FORMAT):
157        this = f"FROM_UNIXTIME(UNIX_TIMESTAMP({this}, {time_format}))"
158    return f"CAST({this} AS DATE)"
159
160
161def _str_to_time_sql(self: Hive.Generator, expression: exp.StrToTime) -> str:
162    this = self.sql(expression, "this")
163    time_format = self.format_time(expression)
164    if time_format not in (Hive.TIME_FORMAT, Hive.DATE_FORMAT):
165        this = f"FROM_UNIXTIME(UNIX_TIMESTAMP({this}, {time_format}))"
166    return f"CAST({this} AS TIMESTAMP)"
167
168
169def _time_to_str(self: Hive.Generator, expression: exp.TimeToStr) -> str:
170    this = self.sql(expression, "this")
171    time_format = self.format_time(expression)
172    return f"DATE_FORMAT({this}, {time_format})"
173
174
175def _to_date_sql(self: Hive.Generator, expression: exp.TsOrDsToDate) -> str:
176    this = self.sql(expression, "this")
177    time_format = self.format_time(expression)
178    if time_format and time_format not in (Hive.TIME_FORMAT, Hive.DATE_FORMAT):
179        return f"TO_DATE({this}, {time_format})"
180    return f"TO_DATE({this})"
181
182
183class Hive(Dialect):
184    ALIAS_POST_TABLESAMPLE = True
185    IDENTIFIERS_CAN_START_WITH_DIGIT = True
186    SUPPORTS_USER_DEFINED_TYPES = False
187
188    # https://spark.apache.org/docs/latest/sql-ref-identifier.html#description
189    RESOLVES_IDENTIFIERS_AS_UPPERCASE = None
190
191    TIME_MAPPING = {
192        "y": "%Y",
193        "Y": "%Y",
194        "YYYY": "%Y",
195        "yyyy": "%Y",
196        "YY": "%y",
197        "yy": "%y",
198        "MMMM": "%B",
199        "MMM": "%b",
200        "MM": "%m",
201        "M": "%-m",
202        "dd": "%d",
203        "d": "%-d",
204        "HH": "%H",
205        "H": "%-H",
206        "hh": "%I",
207        "h": "%-I",
208        "mm": "%M",
209        "m": "%-M",
210        "ss": "%S",
211        "s": "%-S",
212        "SSSSSS": "%f",
213        "a": "%p",
214        "DD": "%j",
215        "D": "%-j",
216        "E": "%a",
217        "EE": "%a",
218        "EEE": "%a",
219        "EEEE": "%A",
220    }
221
222    DATE_FORMAT = "'yyyy-MM-dd'"
223    DATEINT_FORMAT = "'yyyyMMdd'"
224    TIME_FORMAT = "'yyyy-MM-dd HH:mm:ss'"
225
226    class Tokenizer(tokens.Tokenizer):
227        QUOTES = ["'", '"']
228        IDENTIFIERS = ["`"]
229        STRING_ESCAPES = ["\\"]
230        ENCODE = "utf-8"
231
232        KEYWORDS = {
233            **tokens.Tokenizer.KEYWORDS,
234            "ADD ARCHIVE": TokenType.COMMAND,
235            "ADD ARCHIVES": TokenType.COMMAND,
236            "ADD FILE": TokenType.COMMAND,
237            "ADD FILES": TokenType.COMMAND,
238            "ADD JAR": TokenType.COMMAND,
239            "ADD JARS": TokenType.COMMAND,
240            "MSCK REPAIR": TokenType.COMMAND,
241            "REFRESH": TokenType.COMMAND,
242            "WITH SERDEPROPERTIES": TokenType.SERDE_PROPERTIES,
243            "TIMESTAMP AS OF": TokenType.TIMESTAMP_SNAPSHOT,
244            "VERSION AS OF": TokenType.VERSION_SNAPSHOT,
245        }
246
247        NUMERIC_LITERALS = {
248            "L": "BIGINT",
249            "S": "SMALLINT",
250            "Y": "TINYINT",
251            "D": "DOUBLE",
252            "F": "FLOAT",
253            "BD": "DECIMAL",
254        }
255
256    class Parser(parser.Parser):
257        LOG_DEFAULTS_TO_LN = True
258        STRICT_CAST = False
259
260        FUNCTIONS = {
261            **parser.Parser.FUNCTIONS,
262            "BASE64": exp.ToBase64.from_arg_list,
263            "COLLECT_LIST": exp.ArrayAgg.from_arg_list,
264            "COLLECT_SET": exp.SetAgg.from_arg_list,
265            "DATE_ADD": lambda args: exp.TsOrDsAdd(
266                this=seq_get(args, 0), expression=seq_get(args, 1), unit=exp.Literal.string("DAY")
267            ),
268            "DATE_FORMAT": lambda args: format_time_lambda(exp.TimeToStr, "hive")(
269                [
270                    exp.TimeStrToTime(this=seq_get(args, 0)),
271                    seq_get(args, 1),
272                ]
273            ),
274            "DATE_SUB": lambda args: exp.TsOrDsAdd(
275                this=seq_get(args, 0),
276                expression=exp.Mul(this=seq_get(args, 1), expression=exp.Literal.number(-1)),
277                unit=exp.Literal.string("DAY"),
278            ),
279            "DATEDIFF": lambda args: exp.DateDiff(
280                this=exp.TsOrDsToDate(this=seq_get(args, 0)),
281                expression=exp.TsOrDsToDate(this=seq_get(args, 1)),
282            ),
283            "DAY": lambda args: exp.Day(this=exp.TsOrDsToDate(this=seq_get(args, 0))),
284            "FROM_UNIXTIME": format_time_lambda(exp.UnixToStr, "hive", True),
285            "GET_JSON_OBJECT": exp.JSONExtractScalar.from_arg_list,
286            "LOCATE": locate_to_strposition,
287            "MAP": parse_var_map,
288            "MONTH": lambda args: exp.Month(this=exp.TsOrDsToDate.from_arg_list(args)),
289            "PERCENTILE": exp.Quantile.from_arg_list,
290            "PERCENTILE_APPROX": exp.ApproxQuantile.from_arg_list,
291            "REGEXP_EXTRACT": lambda args: exp.RegexpExtract(
292                this=seq_get(args, 0), expression=seq_get(args, 1), group=seq_get(args, 2)
293            ),
294            "SIZE": exp.ArraySize.from_arg_list,
295            "SPLIT": exp.RegexpSplit.from_arg_list,
296            "STR_TO_MAP": lambda args: exp.StrToMap(
297                this=seq_get(args, 0),
298                pair_delim=seq_get(args, 1) or exp.Literal.string(","),
299                key_value_delim=seq_get(args, 2) or exp.Literal.string(":"),
300            ),
301            "TO_DATE": format_time_lambda(exp.TsOrDsToDate, "hive"),
302            "TO_JSON": exp.JSONFormat.from_arg_list,
303            "UNBASE64": exp.FromBase64.from_arg_list,
304            "UNIX_TIMESTAMP": format_time_lambda(exp.StrToUnix, "hive", True),
305            "YEAR": lambda args: exp.Year(this=exp.TsOrDsToDate.from_arg_list(args)),
306        }
307
308        NO_PAREN_FUNCTION_PARSERS = {
309            **parser.Parser.NO_PAREN_FUNCTION_PARSERS,
310            "TRANSFORM": lambda self: self._parse_transform(),
311        }
312
313        PROPERTY_PARSERS = {
314            **parser.Parser.PROPERTY_PARSERS,
315            "WITH SERDEPROPERTIES": lambda self: exp.SerdeProperties(
316                expressions=self._parse_wrapped_csv(self._parse_property)
317            ),
318        }
319
320        def _parse_transform(self) -> t.Optional[exp.Transform | exp.QueryTransform]:
321            if not self._match(TokenType.L_PAREN, advance=False):
322                self._retreat(self._index - 1)
323                return None
324
325            args = self._parse_wrapped_csv(self._parse_lambda)
326            row_format_before = self._parse_row_format(match_row=True)
327
328            record_writer = None
329            if self._match_text_seq("RECORDWRITER"):
330                record_writer = self._parse_string()
331
332            if not self._match(TokenType.USING):
333                return exp.Transform.from_arg_list(args)
334
335            command_script = self._parse_string()
336
337            self._match(TokenType.ALIAS)
338            schema = self._parse_schema()
339
340            row_format_after = self._parse_row_format(match_row=True)
341            record_reader = None
342            if self._match_text_seq("RECORDREADER"):
343                record_reader = self._parse_string()
344
345            return self.expression(
346                exp.QueryTransform,
347                expressions=args,
348                command_script=command_script,
349                schema=schema,
350                row_format_before=row_format_before,
351                record_writer=record_writer,
352                row_format_after=row_format_after,
353                record_reader=record_reader,
354            )
355
356        def _parse_types(
357            self, check_func: bool = False, schema: bool = False, allow_identifiers: bool = True
358        ) -> t.Optional[exp.Expression]:
359            """
360            Spark (and most likely Hive) treats casts to CHAR(length) and VARCHAR(length) as casts to
361            STRING in all contexts except for schema definitions. For example, this is in Spark v3.4.0:
362
363                spark-sql (default)> select cast(1234 as varchar(2));
364                23/06/06 15:51:18 WARN CharVarcharUtils: The Spark cast operator does not support
365                char/varchar type and simply treats them as string type. Please use string type
366                directly to avoid confusion. Otherwise, you can set spark.sql.legacy.charVarcharAsString
367                to true, so that Spark treat them as string type as same as Spark 3.0 and earlier
368
369                1234
370                Time taken: 4.265 seconds, Fetched 1 row(s)
371
372            This shows that Spark doesn't truncate the value into '12', which is inconsistent with
373            what other dialects (e.g. postgres) do, so we need to drop the length to transpile correctly.
374
375            Reference: https://spark.apache.org/docs/latest/sql-ref-datatypes.html
376            """
377            this = super()._parse_types(
378                check_func=check_func, schema=schema, allow_identifiers=allow_identifiers
379            )
380
381            if this and not schema:
382                return this.transform(
383                    lambda node: node.replace(exp.DataType.build("text"))
384                    if isinstance(node, exp.DataType) and node.is_type("char", "varchar")
385                    else node,
386                    copy=False,
387                )
388
389            return this
390
391        def _parse_partition_and_order(
392            self,
393        ) -> t.Tuple[t.List[exp.Expression], t.Optional[exp.Expression]]:
394            return (
395                self._parse_csv(self._parse_conjunction)
396                if self._match_set({TokenType.PARTITION_BY, TokenType.DISTRIBUTE_BY})
397                else [],
398                super()._parse_order(skip_order_token=self._match(TokenType.SORT_BY)),
399            )
400
401    class Generator(generator.Generator):
402        LIMIT_FETCH = "LIMIT"
403        TABLESAMPLE_WITH_METHOD = False
404        TABLESAMPLE_SIZE_IS_PERCENT = True
405        JOIN_HINTS = False
406        TABLE_HINTS = False
407        QUERY_HINTS = False
408        INDEX_ON = "ON TABLE"
409        EXTRACT_ALLOWS_QUOTES = False
410        NVL2_SUPPORTED = False
411
412        TYPE_MAPPING = {
413            **generator.Generator.TYPE_MAPPING,
414            exp.DataType.Type.BIT: "BOOLEAN",
415            exp.DataType.Type.DATETIME: "TIMESTAMP",
416            exp.DataType.Type.TEXT: "STRING",
417            exp.DataType.Type.TIME: "TIMESTAMP",
418            exp.DataType.Type.TIMESTAMPTZ: "TIMESTAMP",
419            exp.DataType.Type.VARBINARY: "BINARY",
420        }
421
422        TRANSFORMS = {
423            **generator.Generator.TRANSFORMS,
424            exp.Group: transforms.preprocess([transforms.unalias_group]),
425            exp.Select: transforms.preprocess(
426                [
427                    transforms.eliminate_qualify,
428                    transforms.eliminate_distinct_on,
429                    transforms.unnest_to_explode,
430                ]
431            ),
432            exp.Property: _property_sql,
433            exp.AnyValue: rename_func("FIRST"),
434            exp.ApproxDistinct: approx_count_distinct_sql,
435            exp.ArgMax: arg_max_or_min_no_count("MAX_BY"),
436            exp.ArgMin: arg_max_or_min_no_count("MIN_BY"),
437            exp.ArrayConcat: rename_func("CONCAT"),
438            exp.ArrayJoin: lambda self, e: self.func("CONCAT_WS", e.expression, e.this),
439            exp.ArraySize: rename_func("SIZE"),
440            exp.ArraySort: _array_sort_sql,
441            exp.With: no_recursive_cte_sql,
442            exp.DateAdd: _add_date_sql,
443            exp.DateDiff: _date_diff_sql,
444            exp.DateStrToDate: rename_func("TO_DATE"),
445            exp.DateSub: _add_date_sql,
446            exp.DateToDi: lambda self, e: f"CAST(DATE_FORMAT({self.sql(e, 'this')}, {Hive.DATEINT_FORMAT}) AS INT)",
447            exp.DiToDate: lambda self, e: f"TO_DATE(CAST({self.sql(e, 'this')} AS STRING), {Hive.DATEINT_FORMAT})",
448            exp.FileFormatProperty: lambda self, e: f"STORED AS {self.sql(e, 'this') if isinstance(e.this, exp.InputOutputFormat) else e.name.upper()}",
449            exp.FromBase64: rename_func("UNBASE64"),
450            exp.If: if_sql(),
451            exp.ILike: no_ilike_sql,
452            exp.IsNan: rename_func("ISNAN"),
453            exp.JSONExtract: rename_func("GET_JSON_OBJECT"),
454            exp.JSONExtractScalar: rename_func("GET_JSON_OBJECT"),
455            exp.JSONFormat: _json_format_sql,
456            exp.Left: left_to_substring_sql,
457            exp.Map: var_map_sql,
458            exp.Max: max_or_greatest,
459            exp.MD5Digest: lambda self, e: self.func("UNHEX", self.func("MD5", e.this)),
460            exp.Min: min_or_least,
461            exp.MonthsBetween: lambda self, e: self.func("MONTHS_BETWEEN", e.this, e.expression),
462            exp.NotNullColumnConstraint: lambda self, e: ""
463            if e.args.get("allow_null")
464            else "NOT NULL",
465            exp.VarMap: var_map_sql,
466            exp.Create: _create_sql,
467            exp.Quantile: rename_func("PERCENTILE"),
468            exp.ApproxQuantile: rename_func("PERCENTILE_APPROX"),
469            exp.RegexpExtract: regexp_extract_sql,
470            exp.RegexpReplace: regexp_replace_sql,
471            exp.RegexpLike: lambda self, e: self.binary(e, "RLIKE"),
472            exp.RegexpSplit: rename_func("SPLIT"),
473            exp.Right: right_to_substring_sql,
474            exp.SafeDivide: no_safe_divide_sql,
475            exp.SchemaCommentProperty: lambda self, e: self.naked_property(e),
476            exp.SetAgg: rename_func("COLLECT_SET"),
477            exp.Split: lambda self, e: f"SPLIT({self.sql(e, 'this')}, CONCAT('\\\\Q', {self.sql(e, 'expression')}))",
478            exp.StrPosition: strposition_to_locate_sql,
479            exp.StrToDate: _str_to_date_sql,
480            exp.StrToTime: _str_to_time_sql,
481            exp.StrToUnix: _str_to_unix_sql,
482            exp.StructExtract: struct_extract_sql,
483            exp.TimeStrToDate: rename_func("TO_DATE"),
484            exp.TimeStrToTime: timestrtotime_sql,
485            exp.TimeStrToUnix: rename_func("UNIX_TIMESTAMP"),
486            exp.TimeToStr: _time_to_str,
487            exp.TimeToUnix: rename_func("UNIX_TIMESTAMP"),
488            exp.ToBase64: rename_func("BASE64"),
489            exp.TsOrDiToDi: lambda self, e: f"CAST(SUBSTR(REPLACE(CAST({self.sql(e, 'this')} AS STRING), '-', ''), 1, 8) AS INT)",
490            exp.TsOrDsAdd: lambda self, e: f"DATE_ADD({self.sql(e, 'this')}, {self.sql(e, 'expression')})",
491            exp.TsOrDsToDate: _to_date_sql,
492            exp.TryCast: no_trycast_sql,
493            exp.UnixToStr: lambda self, e: self.func(
494                "FROM_UNIXTIME", e.this, time_format("hive")(self, e)
495            ),
496            exp.UnixToTime: rename_func("FROM_UNIXTIME"),
497            exp.UnixToTimeStr: rename_func("FROM_UNIXTIME"),
498            exp.PartitionedByProperty: lambda self, e: f"PARTITIONED BY {self.sql(e, 'this')}",
499            exp.SerdeProperties: lambda self, e: self.properties(e, prefix="WITH SERDEPROPERTIES"),
500            exp.NumberToStr: rename_func("FORMAT_NUMBER"),
501            exp.LastDateOfMonth: rename_func("LAST_DAY"),
502            exp.National: lambda self, e: self.national_sql(e, prefix=""),
503            exp.ClusteredColumnConstraint: lambda self, e: f"({self.expressions(e, 'this', indent=False)})",
504            exp.NonClusteredColumnConstraint: lambda self, e: f"({self.expressions(e, 'this', indent=False)})",
505            exp.NotForReplicationColumnConstraint: lambda self, e: "",
506            exp.OnProperty: lambda self, e: "",
507            exp.PrimaryKeyColumnConstraint: lambda self, e: "PRIMARY KEY",
508        }
509
510        PROPERTIES_LOCATION = {
511            **generator.Generator.PROPERTIES_LOCATION,
512            exp.FileFormatProperty: exp.Properties.Location.POST_SCHEMA,
513            exp.PartitionedByProperty: exp.Properties.Location.POST_SCHEMA,
514            exp.VolatileProperty: exp.Properties.Location.UNSUPPORTED,
515            exp.WithDataProperty: exp.Properties.Location.UNSUPPORTED,
516        }
517
518        def temporary_storage_provider(self, expression: exp.Create) -> exp.Create:
519            # Hive has no temporary storage provider (there are hive settings though)
520            return expression
521
522        def parameter_sql(self, expression: exp.Parameter) -> str:
523            this = self.sql(expression, "this")
524            parent = expression.parent
525
526            if isinstance(parent, exp.EQ) and isinstance(parent.parent, exp.SetItem):
527                # We need to produce SET key = value instead of SET ${key} = value
528                return this
529
530            return f"${{{this}}}"
531
532        def schema_sql(self, expression: exp.Schema) -> str:
533            expression = expression.copy()
534
535            for ordered in expression.find_all(exp.Ordered):
536                if ordered.args.get("desc") is False:
537                    ordered.set("desc", None)
538
539            return super().schema_sql(expression)
540
541        def constraint_sql(self, expression: exp.Constraint) -> str:
542            expression = expression.copy()
543
544            for prop in list(expression.find_all(exp.Properties)):
545                prop.pop()
546
547            this = self.sql(expression, "this")
548            expressions = self.expressions(expression, sep=" ", flat=True)
549            return f"CONSTRAINT {this} {expressions}"
550
551        def rowformatserdeproperty_sql(self, expression: exp.RowFormatSerdeProperty) -> str:
552            serde_props = self.sql(expression, "serde_properties")
553            serde_props = f" {serde_props}" if serde_props else ""
554            return f"ROW FORMAT SERDE {self.sql(expression, 'this')}{serde_props}"
555
556        def arrayagg_sql(self, expression: exp.ArrayAgg) -> str:
557            return self.func(
558                "COLLECT_LIST",
559                expression.this.this if isinstance(expression.this, exp.Order) else expression.this,
560            )
561
562        def with_properties(self, properties: exp.Properties) -> str:
563            return self.properties(properties, prefix=self.seg("TBLPROPERTIES"))
564
565        def datatype_sql(self, expression: exp.DataType) -> str:
566            if (
567                expression.this in (exp.DataType.Type.VARCHAR, exp.DataType.Type.NVARCHAR)
568                and not expression.expressions
569            ):
570                expression = exp.DataType.build("text")
571            elif expression.this in exp.DataType.TEMPORAL_TYPES:
572                expression = exp.DataType.build(expression.this)
573            elif expression.is_type("float"):
574                size_expression = expression.find(exp.DataTypeParam)
575                if size_expression:
576                    size = int(size_expression.name)
577                    expression = (
578                        exp.DataType.build("float") if size <= 32 else exp.DataType.build("double")
579                    )
580
581            return super().datatype_sql(expression)
582
583        def version_sql(self, expression: exp.Version) -> str:
584            sql = super().version_sql(expression)
585            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):
184class Hive(Dialect):
185    ALIAS_POST_TABLESAMPLE = True
186    IDENTIFIERS_CAN_START_WITH_DIGIT = True
187    SUPPORTS_USER_DEFINED_TYPES = False
188
189    # https://spark.apache.org/docs/latest/sql-ref-identifier.html#description
190    RESOLVES_IDENTIFIERS_AS_UPPERCASE = None
191
192    TIME_MAPPING = {
193        "y": "%Y",
194        "Y": "%Y",
195        "YYYY": "%Y",
196        "yyyy": "%Y",
197        "YY": "%y",
198        "yy": "%y",
199        "MMMM": "%B",
200        "MMM": "%b",
201        "MM": "%m",
202        "M": "%-m",
203        "dd": "%d",
204        "d": "%-d",
205        "HH": "%H",
206        "H": "%-H",
207        "hh": "%I",
208        "h": "%-I",
209        "mm": "%M",
210        "m": "%-M",
211        "ss": "%S",
212        "s": "%-S",
213        "SSSSSS": "%f",
214        "a": "%p",
215        "DD": "%j",
216        "D": "%-j",
217        "E": "%a",
218        "EE": "%a",
219        "EEE": "%a",
220        "EEEE": "%A",
221    }
222
223    DATE_FORMAT = "'yyyy-MM-dd'"
224    DATEINT_FORMAT = "'yyyyMMdd'"
225    TIME_FORMAT = "'yyyy-MM-dd HH:mm:ss'"
226
227    class Tokenizer(tokens.Tokenizer):
228        QUOTES = ["'", '"']
229        IDENTIFIERS = ["`"]
230        STRING_ESCAPES = ["\\"]
231        ENCODE = "utf-8"
232
233        KEYWORDS = {
234            **tokens.Tokenizer.KEYWORDS,
235            "ADD ARCHIVE": TokenType.COMMAND,
236            "ADD ARCHIVES": TokenType.COMMAND,
237            "ADD FILE": TokenType.COMMAND,
238            "ADD FILES": TokenType.COMMAND,
239            "ADD JAR": TokenType.COMMAND,
240            "ADD JARS": TokenType.COMMAND,
241            "MSCK REPAIR": TokenType.COMMAND,
242            "REFRESH": TokenType.COMMAND,
243            "WITH SERDEPROPERTIES": TokenType.SERDE_PROPERTIES,
244            "TIMESTAMP AS OF": TokenType.TIMESTAMP_SNAPSHOT,
245            "VERSION AS OF": TokenType.VERSION_SNAPSHOT,
246        }
247
248        NUMERIC_LITERALS = {
249            "L": "BIGINT",
250            "S": "SMALLINT",
251            "Y": "TINYINT",
252            "D": "DOUBLE",
253            "F": "FLOAT",
254            "BD": "DECIMAL",
255        }
256
257    class Parser(parser.Parser):
258        LOG_DEFAULTS_TO_LN = True
259        STRICT_CAST = False
260
261        FUNCTIONS = {
262            **parser.Parser.FUNCTIONS,
263            "BASE64": exp.ToBase64.from_arg_list,
264            "COLLECT_LIST": exp.ArrayAgg.from_arg_list,
265            "COLLECT_SET": exp.SetAgg.from_arg_list,
266            "DATE_ADD": lambda args: exp.TsOrDsAdd(
267                this=seq_get(args, 0), expression=seq_get(args, 1), unit=exp.Literal.string("DAY")
268            ),
269            "DATE_FORMAT": lambda args: format_time_lambda(exp.TimeToStr, "hive")(
270                [
271                    exp.TimeStrToTime(this=seq_get(args, 0)),
272                    seq_get(args, 1),
273                ]
274            ),
275            "DATE_SUB": lambda args: exp.TsOrDsAdd(
276                this=seq_get(args, 0),
277                expression=exp.Mul(this=seq_get(args, 1), expression=exp.Literal.number(-1)),
278                unit=exp.Literal.string("DAY"),
279            ),
280            "DATEDIFF": lambda args: exp.DateDiff(
281                this=exp.TsOrDsToDate(this=seq_get(args, 0)),
282                expression=exp.TsOrDsToDate(this=seq_get(args, 1)),
283            ),
284            "DAY": lambda args: exp.Day(this=exp.TsOrDsToDate(this=seq_get(args, 0))),
285            "FROM_UNIXTIME": format_time_lambda(exp.UnixToStr, "hive", True),
286            "GET_JSON_OBJECT": exp.JSONExtractScalar.from_arg_list,
287            "LOCATE": locate_to_strposition,
288            "MAP": parse_var_map,
289            "MONTH": lambda args: exp.Month(this=exp.TsOrDsToDate.from_arg_list(args)),
290            "PERCENTILE": exp.Quantile.from_arg_list,
291            "PERCENTILE_APPROX": exp.ApproxQuantile.from_arg_list,
292            "REGEXP_EXTRACT": lambda args: exp.RegexpExtract(
293                this=seq_get(args, 0), expression=seq_get(args, 1), group=seq_get(args, 2)
294            ),
295            "SIZE": exp.ArraySize.from_arg_list,
296            "SPLIT": exp.RegexpSplit.from_arg_list,
297            "STR_TO_MAP": lambda args: exp.StrToMap(
298                this=seq_get(args, 0),
299                pair_delim=seq_get(args, 1) or exp.Literal.string(","),
300                key_value_delim=seq_get(args, 2) or exp.Literal.string(":"),
301            ),
302            "TO_DATE": format_time_lambda(exp.TsOrDsToDate, "hive"),
303            "TO_JSON": exp.JSONFormat.from_arg_list,
304            "UNBASE64": exp.FromBase64.from_arg_list,
305            "UNIX_TIMESTAMP": format_time_lambda(exp.StrToUnix, "hive", True),
306            "YEAR": lambda args: exp.Year(this=exp.TsOrDsToDate.from_arg_list(args)),
307        }
308
309        NO_PAREN_FUNCTION_PARSERS = {
310            **parser.Parser.NO_PAREN_FUNCTION_PARSERS,
311            "TRANSFORM": lambda self: self._parse_transform(),
312        }
313
314        PROPERTY_PARSERS = {
315            **parser.Parser.PROPERTY_PARSERS,
316            "WITH SERDEPROPERTIES": lambda self: exp.SerdeProperties(
317                expressions=self._parse_wrapped_csv(self._parse_property)
318            ),
319        }
320
321        def _parse_transform(self) -> t.Optional[exp.Transform | exp.QueryTransform]:
322            if not self._match(TokenType.L_PAREN, advance=False):
323                self._retreat(self._index - 1)
324                return None
325
326            args = self._parse_wrapped_csv(self._parse_lambda)
327            row_format_before = self._parse_row_format(match_row=True)
328
329            record_writer = None
330            if self._match_text_seq("RECORDWRITER"):
331                record_writer = self._parse_string()
332
333            if not self._match(TokenType.USING):
334                return exp.Transform.from_arg_list(args)
335
336            command_script = self._parse_string()
337
338            self._match(TokenType.ALIAS)
339            schema = self._parse_schema()
340
341            row_format_after = self._parse_row_format(match_row=True)
342            record_reader = None
343            if self._match_text_seq("RECORDREADER"):
344                record_reader = self._parse_string()
345
346            return self.expression(
347                exp.QueryTransform,
348                expressions=args,
349                command_script=command_script,
350                schema=schema,
351                row_format_before=row_format_before,
352                record_writer=record_writer,
353                row_format_after=row_format_after,
354                record_reader=record_reader,
355            )
356
357        def _parse_types(
358            self, check_func: bool = False, schema: bool = False, allow_identifiers: bool = True
359        ) -> t.Optional[exp.Expression]:
360            """
361            Spark (and most likely Hive) treats casts to CHAR(length) and VARCHAR(length) as casts to
362            STRING in all contexts except for schema definitions. For example, this is in Spark v3.4.0:
363
364                spark-sql (default)> select cast(1234 as varchar(2));
365                23/06/06 15:51:18 WARN CharVarcharUtils: The Spark cast operator does not support
366                char/varchar type and simply treats them as string type. Please use string type
367                directly to avoid confusion. Otherwise, you can set spark.sql.legacy.charVarcharAsString
368                to true, so that Spark treat them as string type as same as Spark 3.0 and earlier
369
370                1234
371                Time taken: 4.265 seconds, Fetched 1 row(s)
372
373            This shows that Spark doesn't truncate the value into '12', which is inconsistent with
374            what other dialects (e.g. postgres) do, so we need to drop the length to transpile correctly.
375
376            Reference: https://spark.apache.org/docs/latest/sql-ref-datatypes.html
377            """
378            this = super()._parse_types(
379                check_func=check_func, schema=schema, allow_identifiers=allow_identifiers
380            )
381
382            if this and not schema:
383                return this.transform(
384                    lambda node: node.replace(exp.DataType.build("text"))
385                    if isinstance(node, exp.DataType) and node.is_type("char", "varchar")
386                    else node,
387                    copy=False,
388                )
389
390            return this
391
392        def _parse_partition_and_order(
393            self,
394        ) -> t.Tuple[t.List[exp.Expression], t.Optional[exp.Expression]]:
395            return (
396                self._parse_csv(self._parse_conjunction)
397                if self._match_set({TokenType.PARTITION_BY, TokenType.DISTRIBUTE_BY})
398                else [],
399                super()._parse_order(skip_order_token=self._match(TokenType.SORT_BY)),
400            )
401
402    class Generator(generator.Generator):
403        LIMIT_FETCH = "LIMIT"
404        TABLESAMPLE_WITH_METHOD = False
405        TABLESAMPLE_SIZE_IS_PERCENT = True
406        JOIN_HINTS = False
407        TABLE_HINTS = False
408        QUERY_HINTS = False
409        INDEX_ON = "ON TABLE"
410        EXTRACT_ALLOWS_QUOTES = False
411        NVL2_SUPPORTED = False
412
413        TYPE_MAPPING = {
414            **generator.Generator.TYPE_MAPPING,
415            exp.DataType.Type.BIT: "BOOLEAN",
416            exp.DataType.Type.DATETIME: "TIMESTAMP",
417            exp.DataType.Type.TEXT: "STRING",
418            exp.DataType.Type.TIME: "TIMESTAMP",
419            exp.DataType.Type.TIMESTAMPTZ: "TIMESTAMP",
420            exp.DataType.Type.VARBINARY: "BINARY",
421        }
422
423        TRANSFORMS = {
424            **generator.Generator.TRANSFORMS,
425            exp.Group: transforms.preprocess([transforms.unalias_group]),
426            exp.Select: transforms.preprocess(
427                [
428                    transforms.eliminate_qualify,
429                    transforms.eliminate_distinct_on,
430                    transforms.unnest_to_explode,
431                ]
432            ),
433            exp.Property: _property_sql,
434            exp.AnyValue: rename_func("FIRST"),
435            exp.ApproxDistinct: approx_count_distinct_sql,
436            exp.ArgMax: arg_max_or_min_no_count("MAX_BY"),
437            exp.ArgMin: arg_max_or_min_no_count("MIN_BY"),
438            exp.ArrayConcat: rename_func("CONCAT"),
439            exp.ArrayJoin: lambda self, e: self.func("CONCAT_WS", e.expression, e.this),
440            exp.ArraySize: rename_func("SIZE"),
441            exp.ArraySort: _array_sort_sql,
442            exp.With: no_recursive_cte_sql,
443            exp.DateAdd: _add_date_sql,
444            exp.DateDiff: _date_diff_sql,
445            exp.DateStrToDate: rename_func("TO_DATE"),
446            exp.DateSub: _add_date_sql,
447            exp.DateToDi: lambda self, e: f"CAST(DATE_FORMAT({self.sql(e, 'this')}, {Hive.DATEINT_FORMAT}) AS INT)",
448            exp.DiToDate: lambda self, e: f"TO_DATE(CAST({self.sql(e, 'this')} AS STRING), {Hive.DATEINT_FORMAT})",
449            exp.FileFormatProperty: lambda self, e: f"STORED AS {self.sql(e, 'this') if isinstance(e.this, exp.InputOutputFormat) else e.name.upper()}",
450            exp.FromBase64: rename_func("UNBASE64"),
451            exp.If: if_sql(),
452            exp.ILike: no_ilike_sql,
453            exp.IsNan: rename_func("ISNAN"),
454            exp.JSONExtract: rename_func("GET_JSON_OBJECT"),
455            exp.JSONExtractScalar: rename_func("GET_JSON_OBJECT"),
456            exp.JSONFormat: _json_format_sql,
457            exp.Left: left_to_substring_sql,
458            exp.Map: var_map_sql,
459            exp.Max: max_or_greatest,
460            exp.MD5Digest: lambda self, e: self.func("UNHEX", self.func("MD5", e.this)),
461            exp.Min: min_or_least,
462            exp.MonthsBetween: lambda self, e: self.func("MONTHS_BETWEEN", e.this, e.expression),
463            exp.NotNullColumnConstraint: lambda self, e: ""
464            if e.args.get("allow_null")
465            else "NOT NULL",
466            exp.VarMap: var_map_sql,
467            exp.Create: _create_sql,
468            exp.Quantile: rename_func("PERCENTILE"),
469            exp.ApproxQuantile: rename_func("PERCENTILE_APPROX"),
470            exp.RegexpExtract: regexp_extract_sql,
471            exp.RegexpReplace: regexp_replace_sql,
472            exp.RegexpLike: lambda self, e: self.binary(e, "RLIKE"),
473            exp.RegexpSplit: rename_func("SPLIT"),
474            exp.Right: right_to_substring_sql,
475            exp.SafeDivide: no_safe_divide_sql,
476            exp.SchemaCommentProperty: lambda self, e: self.naked_property(e),
477            exp.SetAgg: rename_func("COLLECT_SET"),
478            exp.Split: lambda self, e: f"SPLIT({self.sql(e, 'this')}, CONCAT('\\\\Q', {self.sql(e, 'expression')}))",
479            exp.StrPosition: strposition_to_locate_sql,
480            exp.StrToDate: _str_to_date_sql,
481            exp.StrToTime: _str_to_time_sql,
482            exp.StrToUnix: _str_to_unix_sql,
483            exp.StructExtract: struct_extract_sql,
484            exp.TimeStrToDate: rename_func("TO_DATE"),
485            exp.TimeStrToTime: timestrtotime_sql,
486            exp.TimeStrToUnix: rename_func("UNIX_TIMESTAMP"),
487            exp.TimeToStr: _time_to_str,
488            exp.TimeToUnix: rename_func("UNIX_TIMESTAMP"),
489            exp.ToBase64: rename_func("BASE64"),
490            exp.TsOrDiToDi: lambda self, e: f"CAST(SUBSTR(REPLACE(CAST({self.sql(e, 'this')} AS STRING), '-', ''), 1, 8) AS INT)",
491            exp.TsOrDsAdd: lambda self, e: f"DATE_ADD({self.sql(e, 'this')}, {self.sql(e, 'expression')})",
492            exp.TsOrDsToDate: _to_date_sql,
493            exp.TryCast: no_trycast_sql,
494            exp.UnixToStr: lambda self, e: self.func(
495                "FROM_UNIXTIME", e.this, time_format("hive")(self, e)
496            ),
497            exp.UnixToTime: rename_func("FROM_UNIXTIME"),
498            exp.UnixToTimeStr: rename_func("FROM_UNIXTIME"),
499            exp.PartitionedByProperty: lambda self, e: f"PARTITIONED BY {self.sql(e, 'this')}",
500            exp.SerdeProperties: lambda self, e: self.properties(e, prefix="WITH SERDEPROPERTIES"),
501            exp.NumberToStr: rename_func("FORMAT_NUMBER"),
502            exp.LastDateOfMonth: rename_func("LAST_DAY"),
503            exp.National: lambda self, e: self.national_sql(e, prefix=""),
504            exp.ClusteredColumnConstraint: lambda self, e: f"({self.expressions(e, 'this', indent=False)})",
505            exp.NonClusteredColumnConstraint: lambda self, e: f"({self.expressions(e, 'this', indent=False)})",
506            exp.NotForReplicationColumnConstraint: lambda self, e: "",
507            exp.OnProperty: lambda self, e: "",
508            exp.PrimaryKeyColumnConstraint: lambda self, e: "PRIMARY KEY",
509        }
510
511        PROPERTIES_LOCATION = {
512            **generator.Generator.PROPERTIES_LOCATION,
513            exp.FileFormatProperty: exp.Properties.Location.POST_SCHEMA,
514            exp.PartitionedByProperty: exp.Properties.Location.POST_SCHEMA,
515            exp.VolatileProperty: exp.Properties.Location.UNSUPPORTED,
516            exp.WithDataProperty: exp.Properties.Location.UNSUPPORTED,
517        }
518
519        def temporary_storage_provider(self, expression: exp.Create) -> exp.Create:
520            # Hive has no temporary storage provider (there are hive settings though)
521            return expression
522
523        def parameter_sql(self, expression: exp.Parameter) -> str:
524            this = self.sql(expression, "this")
525            parent = expression.parent
526
527            if isinstance(parent, exp.EQ) and isinstance(parent.parent, exp.SetItem):
528                # We need to produce SET key = value instead of SET ${key} = value
529                return this
530
531            return f"${{{this}}}"
532
533        def schema_sql(self, expression: exp.Schema) -> str:
534            expression = expression.copy()
535
536            for ordered in expression.find_all(exp.Ordered):
537                if ordered.args.get("desc") is False:
538                    ordered.set("desc", None)
539
540            return super().schema_sql(expression)
541
542        def constraint_sql(self, expression: exp.Constraint) -> str:
543            expression = expression.copy()
544
545            for prop in list(expression.find_all(exp.Properties)):
546                prop.pop()
547
548            this = self.sql(expression, "this")
549            expressions = self.expressions(expression, sep=" ", flat=True)
550            return f"CONSTRAINT {this} {expressions}"
551
552        def rowformatserdeproperty_sql(self, expression: exp.RowFormatSerdeProperty) -> str:
553            serde_props = self.sql(expression, "serde_properties")
554            serde_props = f" {serde_props}" if serde_props else ""
555            return f"ROW FORMAT SERDE {self.sql(expression, 'this')}{serde_props}"
556
557        def arrayagg_sql(self, expression: exp.ArrayAgg) -> str:
558            return self.func(
559                "COLLECT_LIST",
560                expression.this.this if isinstance(expression.this, exp.Order) else expression.this,
561            )
562
563        def with_properties(self, properties: exp.Properties) -> str:
564            return self.properties(properties, prefix=self.seg("TBLPROPERTIES"))
565
566        def datatype_sql(self, expression: exp.DataType) -> str:
567            if (
568                expression.this in (exp.DataType.Type.VARCHAR, exp.DataType.Type.NVARCHAR)
569                and not expression.expressions
570            ):
571                expression = exp.DataType.build("text")
572            elif expression.this in exp.DataType.TEMPORAL_TYPES:
573                expression = exp.DataType.build(expression.this)
574            elif expression.is_type("float"):
575                size_expression = expression.find(exp.DataTypeParam)
576                if size_expression:
577                    size = int(size_expression.name)
578                    expression = (
579                        exp.DataType.build("float") if size <= 32 else exp.DataType.build("double")
580                    )
581
582            return super().datatype_sql(expression)
583
584        def version_sql(self, expression: exp.Version) -> str:
585            sql = super().version_sql(expression)
586            return sql.replace("FOR ", "", 1)
ALIAS_POST_TABLESAMPLE = True
IDENTIFIERS_CAN_START_WITH_DIGIT = True
SUPPORTS_USER_DEFINED_TYPES = False
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):
227    class Tokenizer(tokens.Tokenizer):
228        QUOTES = ["'", '"']
229        IDENTIFIERS = ["`"]
230        STRING_ESCAPES = ["\\"]
231        ENCODE = "utf-8"
232
233        KEYWORDS = {
234            **tokens.Tokenizer.KEYWORDS,
235            "ADD ARCHIVE": TokenType.COMMAND,
236            "ADD ARCHIVES": TokenType.COMMAND,
237            "ADD FILE": TokenType.COMMAND,
238            "ADD FILES": TokenType.COMMAND,
239            "ADD JAR": TokenType.COMMAND,
240            "ADD JARS": TokenType.COMMAND,
241            "MSCK REPAIR": TokenType.COMMAND,
242            "REFRESH": TokenType.COMMAND,
243            "WITH SERDEPROPERTIES": TokenType.SERDE_PROPERTIES,
244            "TIMESTAMP AS OF": TokenType.TIMESTAMP_SNAPSHOT,
245            "VERSION AS OF": TokenType.VERSION_SNAPSHOT,
246        }
247
248        NUMERIC_LITERALS = {
249            "L": "BIGINT",
250            "S": "SMALLINT",
251            "Y": "TINYINT",
252            "D": "DOUBLE",
253            "F": "FLOAT",
254            "BD": "DECIMAL",
255        }
QUOTES = ["'", '"']
IDENTIFIERS = ['`']
STRING_ESCAPES = ['\\']
ENCODE = 'utf-8'
KEYWORDS = {'{%': <TokenType.BLOCK_START: 'BLOCK_START'>, '{%+': <TokenType.BLOCK_START: 'BLOCK_START'>, '{%-': <TokenType.BLOCK_START: 'BLOCK_START'>, '%}': <TokenType.BLOCK_END: 'BLOCK_END'>, '+%}': <TokenType.BLOCK_END: 'BLOCK_END'>, '-%}': <TokenType.BLOCK_END: 'BLOCK_END'>, '{{+': <TokenType.BLOCK_START: 'BLOCK_START'>, '{{-': <TokenType.BLOCK_START: 'BLOCK_START'>, '+}}': <TokenType.BLOCK_END: 'BLOCK_END'>, '-}}': <TokenType.BLOCK_END: 'BLOCK_END'>, '/*+': <TokenType.HINT: 'HINT'>, '==': <TokenType.EQ: 'EQ'>, '::': <TokenType.DCOLON: 'DCOLON'>, '||': <TokenType.DPIPE: 'DPIPE'>, '>=': <TokenType.GTE: 'GTE'>, '<=': <TokenType.LTE: 'LTE'>, '<>': <TokenType.NEQ: 'NEQ'>, '!=': <TokenType.NEQ: 'NEQ'>, '<=>': <TokenType.NULLSAFE_EQ: 'NULLSAFE_EQ'>, '->': <TokenType.ARROW: 'ARROW'>, '->>': <TokenType.DARROW: 'DARROW'>, '=>': <TokenType.FARROW: 'FARROW'>, '#>': <TokenType.HASH_ARROW: 'HASH_ARROW'>, '#>>': <TokenType.DHASH_ARROW: 'DHASH_ARROW'>, '<->': <TokenType.LR_ARROW: 'LR_ARROW'>, '&&': <TokenType.DAMP: 'DAMP'>, '??': <TokenType.DQMARK: 'DQMARK'>, 'ALL': <TokenType.ALL: 'ALL'>, 'ALWAYS': <TokenType.ALWAYS: 'ALWAYS'>, 'AND': <TokenType.AND: 'AND'>, 'ANTI': <TokenType.ANTI: 'ANTI'>, 'ANY': <TokenType.ANY: 'ANY'>, 'ASC': <TokenType.ASC: 'ASC'>, 'AS': <TokenType.ALIAS: 'ALIAS'>, 'ASOF': <TokenType.ASOF: 'ASOF'>, 'AUTOINCREMENT': <TokenType.AUTO_INCREMENT: 'AUTO_INCREMENT'>, 'AUTO_INCREMENT': <TokenType.AUTO_INCREMENT: 'AUTO_INCREMENT'>, 'BEGIN': <TokenType.BEGIN: 'BEGIN'>, 'BETWEEN': <TokenType.BETWEEN: 'BETWEEN'>, 'CACHE': <TokenType.CACHE: 'CACHE'>, 'UNCACHE': <TokenType.UNCACHE: 'UNCACHE'>, 'CASE': <TokenType.CASE: 'CASE'>, 'CHARACTER SET': <TokenType.CHARACTER_SET: 'CHARACTER_SET'>, 'CLUSTER BY': <TokenType.CLUSTER_BY: 'CLUSTER_BY'>, 'COLLATE': <TokenType.COLLATE: 'COLLATE'>, 'COLUMN': <TokenType.COLUMN: 'COLUMN'>, 'COMMIT': <TokenType.COMMIT: 'COMMIT'>, 'CONNECT BY': <TokenType.CONNECT_BY: 'CONNECT_BY'>, 'CONSTRAINT': <TokenType.CONSTRAINT: 'CONSTRAINT'>, 'CREATE': <TokenType.CREATE: 'CREATE'>, 'CROSS': <TokenType.CROSS: 'CROSS'>, 'CUBE': <TokenType.CUBE: 'CUBE'>, 'CURRENT_DATE': <TokenType.CURRENT_DATE: 'CURRENT_DATE'>, 'CURRENT_TIME': <TokenType.CURRENT_TIME: 'CURRENT_TIME'>, 'CURRENT_TIMESTAMP': <TokenType.CURRENT_TIMESTAMP: 'CURRENT_TIMESTAMP'>, 'CURRENT_USER': <TokenType.CURRENT_USER: 'CURRENT_USER'>, 'DATABASE': <TokenType.DATABASE: 'DATABASE'>, 'DEFAULT': <TokenType.DEFAULT: 'DEFAULT'>, 'DELETE': <TokenType.DELETE: 'DELETE'>, 'DESC': <TokenType.DESC: 'DESC'>, 'DESCRIBE': <TokenType.DESCRIBE: 'DESCRIBE'>, 'DISTINCT': <TokenType.DISTINCT: 'DISTINCT'>, 'DISTRIBUTE BY': <TokenType.DISTRIBUTE_BY: 'DISTRIBUTE_BY'>, 'DIV': <TokenType.DIV: 'DIV'>, 'DROP': <TokenType.DROP: 'DROP'>, 'ELSE': <TokenType.ELSE: 'ELSE'>, 'END': <TokenType.END: 'END'>, 'ESCAPE': <TokenType.ESCAPE: 'ESCAPE'>, 'EXCEPT': <TokenType.EXCEPT: 'EXCEPT'>, 'EXECUTE': <TokenType.EXECUTE: 'EXECUTE'>, 'EXISTS': <TokenType.EXISTS: 'EXISTS'>, 'FALSE': <TokenType.FALSE: 'FALSE'>, 'FETCH': <TokenType.FETCH: 'FETCH'>, 'FILTER': <TokenType.FILTER: 'FILTER'>, 'FIRST': <TokenType.FIRST: 'FIRST'>, 'FULL': <TokenType.FULL: 'FULL'>, 'FUNCTION': <TokenType.FUNCTION: 'FUNCTION'>, 'FOR': <TokenType.FOR: 'FOR'>, 'FOREIGN KEY': <TokenType.FOREIGN_KEY: 'FOREIGN_KEY'>, 'FORMAT': <TokenType.FORMAT: 'FORMAT'>, 'FROM': <TokenType.FROM: 'FROM'>, 'GEOGRAPHY': <TokenType.GEOGRAPHY: 'GEOGRAPHY'>, 'GEOMETRY': <TokenType.GEOMETRY: 'GEOMETRY'>, 'GLOB': <TokenType.GLOB: 'GLOB'>, 'GROUP BY': <TokenType.GROUP_BY: 'GROUP_BY'>, 'GROUPING SETS': <TokenType.GROUPING_SETS: 'GROUPING_SETS'>, 'HAVING': <TokenType.HAVING: 'HAVING'>, 'ILIKE': <TokenType.ILIKE: 'ILIKE'>, 'IN': <TokenType.IN: 'IN'>, 'INDEX': <TokenType.INDEX: 'INDEX'>, 'INET': <TokenType.INET: 'INET'>, 'INNER': <TokenType.INNER: 'INNER'>, 'INSERT': <TokenType.INSERT: 'INSERT'>, 'INTERVAL': <TokenType.INTERVAL: 'INTERVAL'>, 'INTERSECT': <TokenType.INTERSECT: 'INTERSECT'>, 'INTO': <TokenType.INTO: 'INTO'>, 'IS': <TokenType.IS: 'IS'>, 'ISNULL': <TokenType.ISNULL: 'ISNULL'>, 'JOIN': <TokenType.JOIN: 'JOIN'>, 'KEEP': <TokenType.KEEP: 'KEEP'>, 'KILL': <TokenType.KILL: 'KILL'>, 'LATERAL': <TokenType.LATERAL: 'LATERAL'>, 'LEFT': <TokenType.LEFT: 'LEFT'>, 'LIKE': <TokenType.LIKE: 'LIKE'>, 'LIMIT': <TokenType.LIMIT: 'LIMIT'>, 'LOAD': <TokenType.LOAD: 'LOAD'>, 'LOCK': <TokenType.LOCK: 'LOCK'>, 'MERGE': <TokenType.MERGE: 'MERGE'>, 'NATURAL': <TokenType.NATURAL: 'NATURAL'>, 'NEXT': <TokenType.NEXT: 'NEXT'>, 'NOT': <TokenType.NOT: 'NOT'>, 'NOTNULL': <TokenType.NOTNULL: 'NOTNULL'>, 'NULL': <TokenType.NULL: 'NULL'>, 'OBJECT': <TokenType.OBJECT: 'OBJECT'>, 'OFFSET': <TokenType.OFFSET: 'OFFSET'>, 'ON': <TokenType.ON: 'ON'>, 'OR': <TokenType.OR: 'OR'>, 'XOR': <TokenType.XOR: 'XOR'>, 'ORDER BY': <TokenType.ORDER_BY: 'ORDER_BY'>, 'ORDINALITY': <TokenType.ORDINALITY: 'ORDINALITY'>, 'OUTER': <TokenType.OUTER: 'OUTER'>, 'OVER': <TokenType.OVER: 'OVER'>, 'OVERLAPS': <TokenType.OVERLAPS: 'OVERLAPS'>, 'OVERWRITE': <TokenType.OVERWRITE: 'OVERWRITE'>, 'PARTITION': <TokenType.PARTITION: 'PARTITION'>, 'PARTITION BY': <TokenType.PARTITION_BY: 'PARTITION_BY'>, 'PARTITIONED BY': <TokenType.PARTITION_BY: 'PARTITION_BY'>, 'PARTITIONED_BY': <TokenType.PARTITION_BY: 'PARTITION_BY'>, 'PERCENT': <TokenType.PERCENT: 'PERCENT'>, 'PIVOT': <TokenType.PIVOT: 'PIVOT'>, 'PRAGMA': <TokenType.PRAGMA: 'PRAGMA'>, 'PRIMARY KEY': <TokenType.PRIMARY_KEY: 'PRIMARY_KEY'>, 'PROCEDURE': <TokenType.PROCEDURE: 'PROCEDURE'>, 'QUALIFY': <TokenType.QUALIFY: 'QUALIFY'>, 'RANGE': <TokenType.RANGE: 'RANGE'>, 'RECURSIVE': <TokenType.RECURSIVE: 'RECURSIVE'>, 'REGEXP': <TokenType.RLIKE: 'RLIKE'>, 'REPLACE': <TokenType.REPLACE: 'REPLACE'>, 'RETURNING': <TokenType.RETURNING: 'RETURNING'>, 'REFERENCES': <TokenType.REFERENCES: 'REFERENCES'>, 'RIGHT': <TokenType.RIGHT: 'RIGHT'>, 'RLIKE': <TokenType.RLIKE: 'RLIKE'>, 'ROLLBACK': <TokenType.ROLLBACK: 'ROLLBACK'>, 'ROLLUP': <TokenType.ROLLUP: 'ROLLUP'>, 'ROW': <TokenType.ROW: 'ROW'>, 'ROWS': <TokenType.ROWS: 'ROWS'>, 'SCHEMA': <TokenType.SCHEMA: 'SCHEMA'>, 'SELECT': <TokenType.SELECT: 'SELECT'>, 'SEMI': <TokenType.SEMI: 'SEMI'>, 'SET': <TokenType.SET: 'SET'>, 'SETTINGS': <TokenType.SETTINGS: 'SETTINGS'>, 'SHOW': <TokenType.SHOW: 'SHOW'>, 'SIMILAR TO': <TokenType.SIMILAR_TO: 'SIMILAR_TO'>, 'SOME': <TokenType.SOME: 'SOME'>, 'SORT BY': <TokenType.SORT_BY: 'SORT_BY'>, 'START WITH': <TokenType.START_WITH: 'START_WITH'>, 'TABLE': <TokenType.TABLE: 'TABLE'>, 'TABLESAMPLE': <TokenType.TABLE_SAMPLE: 'TABLE_SAMPLE'>, 'TEMP': <TokenType.TEMPORARY: 'TEMPORARY'>, 'TEMPORARY': <TokenType.TEMPORARY: 'TEMPORARY'>, 'THEN': <TokenType.THEN: 'THEN'>, 'TRUE': <TokenType.TRUE: 'TRUE'>, 'UNION': <TokenType.UNION: 'UNION'>, 'UNKNOWN': <TokenType.UNKNOWN: 'UNKNOWN'>, 'UNNEST': <TokenType.UNNEST: 'UNNEST'>, 'UNPIVOT': <TokenType.UNPIVOT: 'UNPIVOT'>, 'UPDATE': <TokenType.UPDATE: 'UPDATE'>, 'USE': <TokenType.USE: 'USE'>, 'USING': <TokenType.USING: 'USING'>, 'UUID': <TokenType.UUID: 'UUID'>, 'VALUES': <TokenType.VALUES: 'VALUES'>, 'VIEW': <TokenType.VIEW: 'VIEW'>, 'VOLATILE': <TokenType.VOLATILE: 'VOLATILE'>, 'WHEN': <TokenType.WHEN: 'WHEN'>, 'WHERE': <TokenType.WHERE: 'WHERE'>, 'WINDOW': <TokenType.WINDOW: 'WINDOW'>, 'WITH': <TokenType.WITH: 'WITH'>, 'APPLY': <TokenType.APPLY: 'APPLY'>, 'ARRAY': <TokenType.ARRAY: 'ARRAY'>, 'BIT': <TokenType.BIT: 'BIT'>, 'BOOL': <TokenType.BOOLEAN: 'BOOLEAN'>, 'BOOLEAN': <TokenType.BOOLEAN: 'BOOLEAN'>, 'BYTE': <TokenType.TINYINT: 'TINYINT'>, 'MEDIUMINT': <TokenType.MEDIUMINT: 'MEDIUMINT'>, 'TINYINT': <TokenType.TINYINT: 'TINYINT'>, 'SHORT': <TokenType.SMALLINT: 'SMALLINT'>, 'SMALLINT': <TokenType.SMALLINT: 'SMALLINT'>, 'INT128': <TokenType.INT128: 'INT128'>, 'INT2': <TokenType.SMALLINT: 'SMALLINT'>, 'INTEGER': <TokenType.INT: 'INT'>, 'INT': <TokenType.INT: 'INT'>, 'INT4': <TokenType.INT: 'INT'>, 'LONG': <TokenType.BIGINT: 'BIGINT'>, 'BIGINT': <TokenType.BIGINT: 'BIGINT'>, 'INT8': <TokenType.BIGINT: 'BIGINT'>, '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.COMMAND: 'COMMAND'>, 'WITH SERDEPROPERTIES': <TokenType.SERDE_PROPERTIES: 'SERDE_PROPERTIES'>, 'TIMESTAMP AS OF': <TokenType.TIMESTAMP_SNAPSHOT: 'TIMESTAMP_SNAPSHOT'>, 'VERSION AS OF': <TokenType.VERSION_SNAPSHOT: 'VERSION_SNAPSHOT'>}
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):
257    class Parser(parser.Parser):
258        LOG_DEFAULTS_TO_LN = True
259        STRICT_CAST = False
260
261        FUNCTIONS = {
262            **parser.Parser.FUNCTIONS,
263            "BASE64": exp.ToBase64.from_arg_list,
264            "COLLECT_LIST": exp.ArrayAgg.from_arg_list,
265            "COLLECT_SET": exp.SetAgg.from_arg_list,
266            "DATE_ADD": lambda args: exp.TsOrDsAdd(
267                this=seq_get(args, 0), expression=seq_get(args, 1), unit=exp.Literal.string("DAY")
268            ),
269            "DATE_FORMAT": lambda args: format_time_lambda(exp.TimeToStr, "hive")(
270                [
271                    exp.TimeStrToTime(this=seq_get(args, 0)),
272                    seq_get(args, 1),
273                ]
274            ),
275            "DATE_SUB": lambda args: exp.TsOrDsAdd(
276                this=seq_get(args, 0),
277                expression=exp.Mul(this=seq_get(args, 1), expression=exp.Literal.number(-1)),
278                unit=exp.Literal.string("DAY"),
279            ),
280            "DATEDIFF": lambda args: exp.DateDiff(
281                this=exp.TsOrDsToDate(this=seq_get(args, 0)),
282                expression=exp.TsOrDsToDate(this=seq_get(args, 1)),
283            ),
284            "DAY": lambda args: exp.Day(this=exp.TsOrDsToDate(this=seq_get(args, 0))),
285            "FROM_UNIXTIME": format_time_lambda(exp.UnixToStr, "hive", True),
286            "GET_JSON_OBJECT": exp.JSONExtractScalar.from_arg_list,
287            "LOCATE": locate_to_strposition,
288            "MAP": parse_var_map,
289            "MONTH": lambda args: exp.Month(this=exp.TsOrDsToDate.from_arg_list(args)),
290            "PERCENTILE": exp.Quantile.from_arg_list,
291            "PERCENTILE_APPROX": exp.ApproxQuantile.from_arg_list,
292            "REGEXP_EXTRACT": lambda args: exp.RegexpExtract(
293                this=seq_get(args, 0), expression=seq_get(args, 1), group=seq_get(args, 2)
294            ),
295            "SIZE": exp.ArraySize.from_arg_list,
296            "SPLIT": exp.RegexpSplit.from_arg_list,
297            "STR_TO_MAP": lambda args: exp.StrToMap(
298                this=seq_get(args, 0),
299                pair_delim=seq_get(args, 1) or exp.Literal.string(","),
300                key_value_delim=seq_get(args, 2) or exp.Literal.string(":"),
301            ),
302            "TO_DATE": format_time_lambda(exp.TsOrDsToDate, "hive"),
303            "TO_JSON": exp.JSONFormat.from_arg_list,
304            "UNBASE64": exp.FromBase64.from_arg_list,
305            "UNIX_TIMESTAMP": format_time_lambda(exp.StrToUnix, "hive", True),
306            "YEAR": lambda args: exp.Year(this=exp.TsOrDsToDate.from_arg_list(args)),
307        }
308
309        NO_PAREN_FUNCTION_PARSERS = {
310            **parser.Parser.NO_PAREN_FUNCTION_PARSERS,
311            "TRANSFORM": lambda self: self._parse_transform(),
312        }
313
314        PROPERTY_PARSERS = {
315            **parser.Parser.PROPERTY_PARSERS,
316            "WITH SERDEPROPERTIES": lambda self: exp.SerdeProperties(
317                expressions=self._parse_wrapped_csv(self._parse_property)
318            ),
319        }
320
321        def _parse_transform(self) -> t.Optional[exp.Transform | exp.QueryTransform]:
322            if not self._match(TokenType.L_PAREN, advance=False):
323                self._retreat(self._index - 1)
324                return None
325
326            args = self._parse_wrapped_csv(self._parse_lambda)
327            row_format_before = self._parse_row_format(match_row=True)
328
329            record_writer = None
330            if self._match_text_seq("RECORDWRITER"):
331                record_writer = self._parse_string()
332
333            if not self._match(TokenType.USING):
334                return exp.Transform.from_arg_list(args)
335
336            command_script = self._parse_string()
337
338            self._match(TokenType.ALIAS)
339            schema = self._parse_schema()
340
341            row_format_after = self._parse_row_format(match_row=True)
342            record_reader = None
343            if self._match_text_seq("RECORDREADER"):
344                record_reader = self._parse_string()
345
346            return self.expression(
347                exp.QueryTransform,
348                expressions=args,
349                command_script=command_script,
350                schema=schema,
351                row_format_before=row_format_before,
352                record_writer=record_writer,
353                row_format_after=row_format_after,
354                record_reader=record_reader,
355            )
356
357        def _parse_types(
358            self, check_func: bool = False, schema: bool = False, allow_identifiers: bool = True
359        ) -> t.Optional[exp.Expression]:
360            """
361            Spark (and most likely Hive) treats casts to CHAR(length) and VARCHAR(length) as casts to
362            STRING in all contexts except for schema definitions. For example, this is in Spark v3.4.0:
363
364                spark-sql (default)> select cast(1234 as varchar(2));
365                23/06/06 15:51:18 WARN CharVarcharUtils: The Spark cast operator does not support
366                char/varchar type and simply treats them as string type. Please use string type
367                directly to avoid confusion. Otherwise, you can set spark.sql.legacy.charVarcharAsString
368                to true, so that Spark treat them as string type as same as Spark 3.0 and earlier
369
370                1234
371                Time taken: 4.265 seconds, Fetched 1 row(s)
372
373            This shows that Spark doesn't truncate the value into '12', which is inconsistent with
374            what other dialects (e.g. postgres) do, so we need to drop the length to transpile correctly.
375
376            Reference: https://spark.apache.org/docs/latest/sql-ref-datatypes.html
377            """
378            this = super()._parse_types(
379                check_func=check_func, schema=schema, allow_identifiers=allow_identifiers
380            )
381
382            if this and not schema:
383                return this.transform(
384                    lambda node: node.replace(exp.DataType.build("text"))
385                    if isinstance(node, exp.DataType) and node.is_type("char", "varchar")
386                    else node,
387                    copy=False,
388                )
389
390            return this
391
392        def _parse_partition_and_order(
393            self,
394        ) -> t.Tuple[t.List[exp.Expression], t.Optional[exp.Expression]]:
395            return (
396                self._parse_csv(self._parse_conjunction)
397                if self._match_set({TokenType.PARTITION_BY, TokenType.DISTRIBUTE_BY})
398                else [],
399                super()._parse_order(skip_order_token=self._match(TokenType.SORT_BY)),
400            )

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_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'>>, '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 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>>, '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>>}
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
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
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
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):
402    class Generator(generator.Generator):
403        LIMIT_FETCH = "LIMIT"
404        TABLESAMPLE_WITH_METHOD = False
405        TABLESAMPLE_SIZE_IS_PERCENT = True
406        JOIN_HINTS = False
407        TABLE_HINTS = False
408        QUERY_HINTS = False
409        INDEX_ON = "ON TABLE"
410        EXTRACT_ALLOWS_QUOTES = False
411        NVL2_SUPPORTED = False
412
413        TYPE_MAPPING = {
414            **generator.Generator.TYPE_MAPPING,
415            exp.DataType.Type.BIT: "BOOLEAN",
416            exp.DataType.Type.DATETIME: "TIMESTAMP",
417            exp.DataType.Type.TEXT: "STRING",
418            exp.DataType.Type.TIME: "TIMESTAMP",
419            exp.DataType.Type.TIMESTAMPTZ: "TIMESTAMP",
420            exp.DataType.Type.VARBINARY: "BINARY",
421        }
422
423        TRANSFORMS = {
424            **generator.Generator.TRANSFORMS,
425            exp.Group: transforms.preprocess([transforms.unalias_group]),
426            exp.Select: transforms.preprocess(
427                [
428                    transforms.eliminate_qualify,
429                    transforms.eliminate_distinct_on,
430                    transforms.unnest_to_explode,
431                ]
432            ),
433            exp.Property: _property_sql,
434            exp.AnyValue: rename_func("FIRST"),
435            exp.ApproxDistinct: approx_count_distinct_sql,
436            exp.ArgMax: arg_max_or_min_no_count("MAX_BY"),
437            exp.ArgMin: arg_max_or_min_no_count("MIN_BY"),
438            exp.ArrayConcat: rename_func("CONCAT"),
439            exp.ArrayJoin: lambda self, e: self.func("CONCAT_WS", e.expression, e.this),
440            exp.ArraySize: rename_func("SIZE"),
441            exp.ArraySort: _array_sort_sql,
442            exp.With: no_recursive_cte_sql,
443            exp.DateAdd: _add_date_sql,
444            exp.DateDiff: _date_diff_sql,
445            exp.DateStrToDate: rename_func("TO_DATE"),
446            exp.DateSub: _add_date_sql,
447            exp.DateToDi: lambda self, e: f"CAST(DATE_FORMAT({self.sql(e, 'this')}, {Hive.DATEINT_FORMAT}) AS INT)",
448            exp.DiToDate: lambda self, e: f"TO_DATE(CAST({self.sql(e, 'this')} AS STRING), {Hive.DATEINT_FORMAT})",
449            exp.FileFormatProperty: lambda self, e: f"STORED AS {self.sql(e, 'this') if isinstance(e.this, exp.InputOutputFormat) else e.name.upper()}",
450            exp.FromBase64: rename_func("UNBASE64"),
451            exp.If: if_sql(),
452            exp.ILike: no_ilike_sql,
453            exp.IsNan: rename_func("ISNAN"),
454            exp.JSONExtract: rename_func("GET_JSON_OBJECT"),
455            exp.JSONExtractScalar: rename_func("GET_JSON_OBJECT"),
456            exp.JSONFormat: _json_format_sql,
457            exp.Left: left_to_substring_sql,
458            exp.Map: var_map_sql,
459            exp.Max: max_or_greatest,
460            exp.MD5Digest: lambda self, e: self.func("UNHEX", self.func("MD5", e.this)),
461            exp.Min: min_or_least,
462            exp.MonthsBetween: lambda self, e: self.func("MONTHS_BETWEEN", e.this, e.expression),
463            exp.NotNullColumnConstraint: lambda self, e: ""
464            if e.args.get("allow_null")
465            else "NOT NULL",
466            exp.VarMap: var_map_sql,
467            exp.Create: _create_sql,
468            exp.Quantile: rename_func("PERCENTILE"),
469            exp.ApproxQuantile: rename_func("PERCENTILE_APPROX"),
470            exp.RegexpExtract: regexp_extract_sql,
471            exp.RegexpReplace: regexp_replace_sql,
472            exp.RegexpLike: lambda self, e: self.binary(e, "RLIKE"),
473            exp.RegexpSplit: rename_func("SPLIT"),
474            exp.Right: right_to_substring_sql,
475            exp.SafeDivide: no_safe_divide_sql,
476            exp.SchemaCommentProperty: lambda self, e: self.naked_property(e),
477            exp.SetAgg: rename_func("COLLECT_SET"),
478            exp.Split: lambda self, e: f"SPLIT({self.sql(e, 'this')}, CONCAT('\\\\Q', {self.sql(e, 'expression')}))",
479            exp.StrPosition: strposition_to_locate_sql,
480            exp.StrToDate: _str_to_date_sql,
481            exp.StrToTime: _str_to_time_sql,
482            exp.StrToUnix: _str_to_unix_sql,
483            exp.StructExtract: struct_extract_sql,
484            exp.TimeStrToDate: rename_func("TO_DATE"),
485            exp.TimeStrToTime: timestrtotime_sql,
486            exp.TimeStrToUnix: rename_func("UNIX_TIMESTAMP"),
487            exp.TimeToStr: _time_to_str,
488            exp.TimeToUnix: rename_func("UNIX_TIMESTAMP"),
489            exp.ToBase64: rename_func("BASE64"),
490            exp.TsOrDiToDi: lambda self, e: f"CAST(SUBSTR(REPLACE(CAST({self.sql(e, 'this')} AS STRING), '-', ''), 1, 8) AS INT)",
491            exp.TsOrDsAdd: lambda self, e: f"DATE_ADD({self.sql(e, 'this')}, {self.sql(e, 'expression')})",
492            exp.TsOrDsToDate: _to_date_sql,
493            exp.TryCast: no_trycast_sql,
494            exp.UnixToStr: lambda self, e: self.func(
495                "FROM_UNIXTIME", e.this, time_format("hive")(self, e)
496            ),
497            exp.UnixToTime: rename_func("FROM_UNIXTIME"),
498            exp.UnixToTimeStr: rename_func("FROM_UNIXTIME"),
499            exp.PartitionedByProperty: lambda self, e: f"PARTITIONED BY {self.sql(e, 'this')}",
500            exp.SerdeProperties: lambda self, e: self.properties(e, prefix="WITH SERDEPROPERTIES"),
501            exp.NumberToStr: rename_func("FORMAT_NUMBER"),
502            exp.LastDateOfMonth: rename_func("LAST_DAY"),
503            exp.National: lambda self, e: self.national_sql(e, prefix=""),
504            exp.ClusteredColumnConstraint: lambda self, e: f"({self.expressions(e, 'this', indent=False)})",
505            exp.NonClusteredColumnConstraint: lambda self, e: f"({self.expressions(e, 'this', indent=False)})",
506            exp.NotForReplicationColumnConstraint: lambda self, e: "",
507            exp.OnProperty: lambda self, e: "",
508            exp.PrimaryKeyColumnConstraint: lambda self, e: "PRIMARY KEY",
509        }
510
511        PROPERTIES_LOCATION = {
512            **generator.Generator.PROPERTIES_LOCATION,
513            exp.FileFormatProperty: exp.Properties.Location.POST_SCHEMA,
514            exp.PartitionedByProperty: exp.Properties.Location.POST_SCHEMA,
515            exp.VolatileProperty: exp.Properties.Location.UNSUPPORTED,
516            exp.WithDataProperty: exp.Properties.Location.UNSUPPORTED,
517        }
518
519        def temporary_storage_provider(self, expression: exp.Create) -> exp.Create:
520            # Hive has no temporary storage provider (there are hive settings though)
521            return expression
522
523        def parameter_sql(self, expression: exp.Parameter) -> str:
524            this = self.sql(expression, "this")
525            parent = expression.parent
526
527            if isinstance(parent, exp.EQ) and isinstance(parent.parent, exp.SetItem):
528                # We need to produce SET key = value instead of SET ${key} = value
529                return this
530
531            return f"${{{this}}}"
532
533        def schema_sql(self, expression: exp.Schema) -> str:
534            expression = expression.copy()
535
536            for ordered in expression.find_all(exp.Ordered):
537                if ordered.args.get("desc") is False:
538                    ordered.set("desc", None)
539
540            return super().schema_sql(expression)
541
542        def constraint_sql(self, expression: exp.Constraint) -> str:
543            expression = expression.copy()
544
545            for prop in list(expression.find_all(exp.Properties)):
546                prop.pop()
547
548            this = self.sql(expression, "this")
549            expressions = self.expressions(expression, sep=" ", flat=True)
550            return f"CONSTRAINT {this} {expressions}"
551
552        def rowformatserdeproperty_sql(self, expression: exp.RowFormatSerdeProperty) -> str:
553            serde_props = self.sql(expression, "serde_properties")
554            serde_props = f" {serde_props}" if serde_props else ""
555            return f"ROW FORMAT SERDE {self.sql(expression, 'this')}{serde_props}"
556
557        def arrayagg_sql(self, expression: exp.ArrayAgg) -> str:
558            return self.func(
559                "COLLECT_LIST",
560                expression.this.this if isinstance(expression.this, exp.Order) else expression.this,
561            )
562
563        def with_properties(self, properties: exp.Properties) -> str:
564            return self.properties(properties, prefix=self.seg("TBLPROPERTIES"))
565
566        def datatype_sql(self, expression: exp.DataType) -> str:
567            if (
568                expression.this in (exp.DataType.Type.VARCHAR, exp.DataType.Type.NVARCHAR)
569                and not expression.expressions
570            ):
571                expression = exp.DataType.build("text")
572            elif expression.this in exp.DataType.TEMPORAL_TYPES:
573                expression = exp.DataType.build(expression.this)
574            elif expression.is_type("float"):
575                size_expression = expression.find(exp.DataTypeParam)
576                if size_expression:
577                    size = int(size_expression.name)
578                    expression = (
579                        exp.DataType.build("float") if size <= 32 else exp.DataType.build("double")
580                    )
581
582            return super().datatype_sql(expression)
583
584        def version_sql(self, expression: exp.Version) -> str:
585            sql = super().version_sql(expression)
586            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
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.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'>}
def temporary_storage_provider( self, expression: sqlglot.expressions.Create) -> sqlglot.expressions.Create:
519        def temporary_storage_provider(self, expression: exp.Create) -> exp.Create:
520            # Hive has no temporary storage provider (there are hive settings though)
521            return expression
def parameter_sql(self, expression: sqlglot.expressions.Parameter) -> str:
523        def parameter_sql(self, expression: exp.Parameter) -> str:
524            this = self.sql(expression, "this")
525            parent = expression.parent
526
527            if isinstance(parent, exp.EQ) and isinstance(parent.parent, exp.SetItem):
528                # We need to produce SET key = value instead of SET ${key} = value
529                return this
530
531            return f"${{{this}}}"
def schema_sql(self, expression: sqlglot.expressions.Schema) -> str:
533        def schema_sql(self, expression: exp.Schema) -> str:
534            expression = expression.copy()
535
536            for ordered in expression.find_all(exp.Ordered):
537                if ordered.args.get("desc") is False:
538                    ordered.set("desc", None)
539
540            return super().schema_sql(expression)
def constraint_sql(self, expression: sqlglot.expressions.Constraint) -> str:
542        def constraint_sql(self, expression: exp.Constraint) -> str:
543            expression = expression.copy()
544
545            for prop in list(expression.find_all(exp.Properties)):
546                prop.pop()
547
548            this = self.sql(expression, "this")
549            expressions = self.expressions(expression, sep=" ", flat=True)
550            return f"CONSTRAINT {this} {expressions}"
def rowformatserdeproperty_sql(self, expression: sqlglot.expressions.RowFormatSerdeProperty) -> str:
552        def rowformatserdeproperty_sql(self, expression: exp.RowFormatSerdeProperty) -> str:
553            serde_props = self.sql(expression, "serde_properties")
554            serde_props = f" {serde_props}" if serde_props else ""
555            return f"ROW FORMAT SERDE {self.sql(expression, 'this')}{serde_props}"
def arrayagg_sql(self, expression: sqlglot.expressions.ArrayAgg) -> str:
557        def arrayagg_sql(self, expression: exp.ArrayAgg) -> str:
558            return self.func(
559                "COLLECT_LIST",
560                expression.this.this if isinstance(expression.this, exp.Order) else expression.this,
561            )
def with_properties(self, properties: sqlglot.expressions.Properties) -> str:
563        def with_properties(self, properties: exp.Properties) -> str:
564            return self.properties(properties, prefix=self.seg("TBLPROPERTIES"))
def datatype_sql(self, expression: sqlglot.expressions.DataType) -> str:
566        def datatype_sql(self, expression: exp.DataType) -> str:
567            if (
568                expression.this in (exp.DataType.Type.VARCHAR, exp.DataType.Type.NVARCHAR)
569                and not expression.expressions
570            ):
571                expression = exp.DataType.build("text")
572            elif expression.this in exp.DataType.TEMPORAL_TYPES:
573                expression = exp.DataType.build(expression.this)
574            elif expression.is_type("float"):
575                size_expression = expression.find(exp.DataTypeParam)
576                if size_expression:
577                    size = int(size_expression.name)
578                    expression = (
579                        exp.DataType.build("float") if size <= 32 else exp.DataType.build("double")
580                    )
581
582            return super().datatype_sql(expression)
def version_sql(self, expression: sqlglot.expressions.Version) -> str:
584        def version_sql(self, expression: exp.Version) -> str:
585            sql = super().version_sql(expression)
586            return sql.replace("FOR ", "", 1)
SELECT_KINDS: Tuple[str, ...] = ()
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:
279    @classmethod
280    def can_identify(cls, text: str, identify: str | bool = "safe") -> bool:
281        """Checks if text can be identified given an identify option.
282
283        Args:
284            text: The text to check.
285            identify:
286                "always" or `True`: Always returns true.
287                "safe": True if the identifier is case-insensitive.
288
289        Returns:
290            Whether or not the given text can be identified.
291        """
292        if identify is True or identify == "always":
293            return True
294
295        if identify == "safe":
296            return not cls.case_sensitive(text)
297
298        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
STAR_MAPPING
TIME_PART_SINGULARS
TOKEN_MAPPING
STRUCT_DELIMITER
PARAMETER_TOKEN
RESERVED_KEYWORDS
WITH_SEPARATED_COMMENTS
EXCLUDE_COMMENTS
UNWRAPPED_INTERVAL_VALUES
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
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
lockingproperty_sql
withdataproperty_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
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
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