Edit on GitHub

sqlglot.dialects.spark2

  1from __future__ import annotations
  2
  3import typing as t
  4
  5from sqlglot import exp, parser, transforms
  6from sqlglot.dialects.dialect import (
  7    create_with_partitions_sql,
  8    pivot_column_names,
  9    rename_func,
 10    trim_sql,
 11)
 12from sqlglot.dialects.hive import Hive
 13from sqlglot.helper import seq_get
 14
 15
 16def _create_sql(self: Hive.Generator, e: exp.Create) -> str:
 17    kind = e.args["kind"]
 18    properties = e.args.get("properties")
 19
 20    if kind.upper() == "TABLE" and any(
 21        isinstance(prop, exp.TemporaryProperty)
 22        for prop in (properties.expressions if properties else [])
 23    ):
 24        return f"CREATE TEMPORARY VIEW {self.sql(e, 'this')} AS {self.sql(e, 'expression')}"
 25    return create_with_partitions_sql(self, e)
 26
 27
 28def _map_sql(self: Hive.Generator, expression: exp.Map) -> str:
 29    keys = self.sql(expression.args["keys"])
 30    values = self.sql(expression.args["values"])
 31    return f"MAP_FROM_ARRAYS({keys}, {values})"
 32
 33
 34def _parse_as_cast(to_type: str) -> t.Callable[[t.List], exp.Expression]:
 35    return lambda args: exp.Cast(this=seq_get(args, 0), to=exp.DataType.build(to_type))
 36
 37
 38def _str_to_date(self: Hive.Generator, expression: exp.StrToDate) -> str:
 39    this = self.sql(expression, "this")
 40    time_format = self.format_time(expression)
 41    if time_format == Hive.DATE_FORMAT:
 42        return f"TO_DATE({this})"
 43    return f"TO_DATE({this}, {time_format})"
 44
 45
 46def _unix_to_time_sql(self: Hive.Generator, expression: exp.UnixToTime) -> str:
 47    scale = expression.args.get("scale")
 48    timestamp = self.sql(expression, "this")
 49    if scale is None:
 50        return f"CAST(FROM_UNIXTIME({timestamp}) AS TIMESTAMP)"
 51    if scale == exp.UnixToTime.SECONDS:
 52        return f"TIMESTAMP_SECONDS({timestamp})"
 53    if scale == exp.UnixToTime.MILLIS:
 54        return f"TIMESTAMP_MILLIS({timestamp})"
 55    if scale == exp.UnixToTime.MICROS:
 56        return f"TIMESTAMP_MICROS({timestamp})"
 57
 58    raise ValueError("Improper scale for timestamp")
 59
 60
 61def _unalias_pivot(expression: exp.Expression) -> exp.Expression:
 62    """
 63    Spark doesn't allow PIVOT aliases, so we need to remove them and possibly wrap a
 64    pivoted source in a subquery with the same alias to preserve the query's semantics.
 65
 66    Example:
 67        >>> from sqlglot import parse_one
 68        >>> expr = parse_one("SELECT piv.x FROM tbl PIVOT (SUM(a) FOR b IN ('x')) piv")
 69        >>> print(_unalias_pivot(expr).sql(dialect="spark"))
 70        SELECT piv.x FROM (SELECT * FROM tbl PIVOT(SUM(a) FOR b IN ('x'))) AS piv
 71    """
 72    if isinstance(expression, exp.From) and expression.this.args.get("pivots"):
 73        pivot = expression.this.args["pivots"][0]
 74        if pivot.alias:
 75            alias = pivot.args["alias"].pop()
 76            return exp.From(
 77                this=expression.this.replace(
 78                    exp.select("*").from_(expression.this.copy()).subquery(alias=alias)
 79                )
 80            )
 81
 82    return expression
 83
 84
 85def _unqualify_pivot_columns(expression: exp.Expression) -> exp.Expression:
 86    """
 87    Spark doesn't allow the column referenced in the PIVOT's field to be qualified,
 88    so we need to unqualify it.
 89
 90    Example:
 91        >>> from sqlglot import parse_one
 92        >>> expr = parse_one("SELECT * FROM tbl PIVOT (SUM(tbl.sales) FOR tbl.quarter IN ('Q1', 'Q2'))")
 93        >>> print(_unqualify_pivot_columns(expr).sql(dialect="spark"))
 94        SELECT * FROM tbl PIVOT(SUM(tbl.sales) FOR quarter IN ('Q1', 'Q1'))
 95    """
 96    if isinstance(expression, exp.Pivot):
 97        expression.args["field"].transform(
 98            lambda node: exp.column(node.output_name, quoted=node.this.quoted)
 99            if isinstance(node, exp.Column)
100            else node,
101            copy=False,
102        )
103
104    return expression
105
106
107class Spark2(Hive):
108    class Parser(Hive.Parser):
109        FUNCTIONS = {
110            **Hive.Parser.FUNCTIONS,
111            "MAP_FROM_ARRAYS": exp.Map.from_arg_list,
112            "TO_UNIX_TIMESTAMP": exp.StrToUnix.from_arg_list,
113            "SHIFTLEFT": lambda args: exp.BitwiseLeftShift(
114                this=seq_get(args, 0),
115                expression=seq_get(args, 1),
116            ),
117            "SHIFTRIGHT": lambda args: exp.BitwiseRightShift(
118                this=seq_get(args, 0),
119                expression=seq_get(args, 1),
120            ),
121            "APPROX_PERCENTILE": exp.ApproxQuantile.from_arg_list,
122            "IIF": exp.If.from_arg_list,
123            "AGGREGATE": exp.Reduce.from_arg_list,
124            "DAYOFWEEK": lambda args: exp.DayOfWeek(
125                this=exp.TsOrDsToDate(this=seq_get(args, 0)),
126            ),
127            "DAYOFMONTH": lambda args: exp.DayOfMonth(
128                this=exp.TsOrDsToDate(this=seq_get(args, 0)),
129            ),
130            "DAYOFYEAR": lambda args: exp.DayOfYear(
131                this=exp.TsOrDsToDate(this=seq_get(args, 0)),
132            ),
133            "WEEKOFYEAR": lambda args: exp.WeekOfYear(
134                this=exp.TsOrDsToDate(this=seq_get(args, 0)),
135            ),
136            "DATE_TRUNC": lambda args: exp.TimestampTrunc(
137                this=seq_get(args, 1),
138                unit=exp.var(seq_get(args, 0)),
139            ),
140            "TRUNC": lambda args: exp.DateTrunc(unit=seq_get(args, 1), this=seq_get(args, 0)),
141            "BOOLEAN": _parse_as_cast("boolean"),
142            "DATE": _parse_as_cast("date"),
143            "DOUBLE": _parse_as_cast("double"),
144            "FLOAT": _parse_as_cast("float"),
145            "INT": _parse_as_cast("int"),
146            "STRING": _parse_as_cast("string"),
147            "TIMESTAMP": _parse_as_cast("timestamp"),
148        }
149
150        FUNCTION_PARSERS = {
151            **parser.Parser.FUNCTION_PARSERS,
152            "BROADCAST": lambda self: self._parse_join_hint("BROADCAST"),
153            "BROADCASTJOIN": lambda self: self._parse_join_hint("BROADCASTJOIN"),
154            "MAPJOIN": lambda self: self._parse_join_hint("MAPJOIN"),
155            "MERGE": lambda self: self._parse_join_hint("MERGE"),
156            "SHUFFLEMERGE": lambda self: self._parse_join_hint("SHUFFLEMERGE"),
157            "MERGEJOIN": lambda self: self._parse_join_hint("MERGEJOIN"),
158            "SHUFFLE_HASH": lambda self: self._parse_join_hint("SHUFFLE_HASH"),
159            "SHUFFLE_REPLICATE_NL": lambda self: self._parse_join_hint("SHUFFLE_REPLICATE_NL"),
160        }
161
162        def _parse_add_column(self) -> t.Optional[exp.Expression]:
163            return self._match_text_seq("ADD", "COLUMNS") and self._parse_schema()
164
165        def _parse_drop_column(self) -> t.Optional[exp.Drop | exp.Command]:
166            return self._match_text_seq("DROP", "COLUMNS") and self.expression(
167                exp.Drop, this=self._parse_schema(), kind="COLUMNS"
168            )
169
170        def _pivot_column_names(self, aggregations: t.List[exp.Expression]) -> t.List[str]:
171            if len(aggregations) == 1:
172                return [""]
173            return pivot_column_names(aggregations, dialect="spark")
174
175    class Generator(Hive.Generator):
176        QUERY_HINTS = True
177
178        TYPE_MAPPING = {
179            **Hive.Generator.TYPE_MAPPING,
180        }
181
182        PROPERTIES_LOCATION = {
183            **Hive.Generator.PROPERTIES_LOCATION,
184            exp.EngineProperty: exp.Properties.Location.UNSUPPORTED,
185            exp.AutoIncrementProperty: exp.Properties.Location.UNSUPPORTED,
186            exp.CharacterSetProperty: exp.Properties.Location.UNSUPPORTED,
187            exp.CollateProperty: exp.Properties.Location.UNSUPPORTED,
188        }
189
190        TRANSFORMS = {
191            **Hive.Generator.TRANSFORMS,
192            exp.ApproxDistinct: rename_func("APPROX_COUNT_DISTINCT"),
193            exp.ArraySum: lambda self, e: f"AGGREGATE({self.sql(e, 'this')}, 0, (acc, x) -> acc + x, acc -> acc)",
194            exp.AtTimeZone: lambda self, e: f"FROM_UTC_TIMESTAMP({self.sql(e, 'this')}, {self.sql(e, 'zone')})",
195            exp.BitwiseLeftShift: rename_func("SHIFTLEFT"),
196            exp.BitwiseRightShift: rename_func("SHIFTRIGHT"),
197            exp.Create: _create_sql,
198            exp.DateFromParts: rename_func("MAKE_DATE"),
199            exp.DateTrunc: lambda self, e: self.func("TRUNC", e.this, e.args.get("unit")),
200            exp.DayOfMonth: rename_func("DAYOFMONTH"),
201            exp.DayOfWeek: rename_func("DAYOFWEEK"),
202            exp.DayOfYear: rename_func("DAYOFYEAR"),
203            exp.FileFormatProperty: lambda self, e: f"USING {e.name.upper()}",
204            exp.From: transforms.preprocess([_unalias_pivot]),
205            exp.LogicalAnd: rename_func("BOOL_AND"),
206            exp.LogicalOr: rename_func("BOOL_OR"),
207            exp.Map: _map_sql,
208            exp.Pivot: transforms.preprocess([_unqualify_pivot_columns]),
209            exp.Reduce: rename_func("AGGREGATE"),
210            exp.StrToDate: _str_to_date,
211            exp.StrToTime: lambda self, e: f"TO_TIMESTAMP({self.sql(e, 'this')}, {self.format_time(e)})",
212            exp.TimestampTrunc: lambda self, e: self.func(
213                "DATE_TRUNC", exp.Literal.string(e.text("unit")), e.this
214            ),
215            exp.Trim: trim_sql,
216            exp.UnixToTime: _unix_to_time_sql,
217            exp.VariancePop: rename_func("VAR_POP"),
218            exp.WeekOfYear: rename_func("WEEKOFYEAR"),
219            exp.WithinGroup: transforms.preprocess(
220                [transforms.remove_within_group_for_percentiles]
221            ),
222        }
223        TRANSFORMS.pop(exp.ArrayJoin)
224        TRANSFORMS.pop(exp.ArraySort)
225        TRANSFORMS.pop(exp.ILike)
226        TRANSFORMS.pop(exp.Left)
227        TRANSFORMS.pop(exp.Right)
228
229        WRAP_DERIVED_VALUES = False
230        CREATE_FUNCTION_RETURN_AS = False
231
232        def cast_sql(self, expression: exp.Cast, safe_prefix: t.Optional[str] = None) -> str:
233            if isinstance(expression.this, exp.Cast) and expression.this.is_type("json"):
234                schema = f"'{self.sql(expression, 'to')}'"
235                return self.func("FROM_JSON", expression.this.this, schema)
236            if expression.is_type("json"):
237                return self.func("TO_JSON", expression.this)
238
239            return super(Hive.Generator, self).cast_sql(expression, safe_prefix=safe_prefix)
240
241        def columndef_sql(self, expression: exp.ColumnDef, sep: str = " ") -> str:
242            return super().columndef_sql(
243                expression,
244                sep=": "
245                if isinstance(expression.parent, exp.DataType)
246                and expression.parent.is_type("struct")
247                else sep,
248            )
249
250    class Tokenizer(Hive.Tokenizer):
251        HEX_STRINGS = [("X'", "'")]
class Spark2(sqlglot.dialects.hive.Hive):
108class Spark2(Hive):
109    class Parser(Hive.Parser):
110        FUNCTIONS = {
111            **Hive.Parser.FUNCTIONS,
112            "MAP_FROM_ARRAYS": exp.Map.from_arg_list,
113            "TO_UNIX_TIMESTAMP": exp.StrToUnix.from_arg_list,
114            "SHIFTLEFT": lambda args: exp.BitwiseLeftShift(
115                this=seq_get(args, 0),
116                expression=seq_get(args, 1),
117            ),
118            "SHIFTRIGHT": lambda args: exp.BitwiseRightShift(
119                this=seq_get(args, 0),
120                expression=seq_get(args, 1),
121            ),
122            "APPROX_PERCENTILE": exp.ApproxQuantile.from_arg_list,
123            "IIF": exp.If.from_arg_list,
124            "AGGREGATE": exp.Reduce.from_arg_list,
125            "DAYOFWEEK": lambda args: exp.DayOfWeek(
126                this=exp.TsOrDsToDate(this=seq_get(args, 0)),
127            ),
128            "DAYOFMONTH": lambda args: exp.DayOfMonth(
129                this=exp.TsOrDsToDate(this=seq_get(args, 0)),
130            ),
131            "DAYOFYEAR": lambda args: exp.DayOfYear(
132                this=exp.TsOrDsToDate(this=seq_get(args, 0)),
133            ),
134            "WEEKOFYEAR": lambda args: exp.WeekOfYear(
135                this=exp.TsOrDsToDate(this=seq_get(args, 0)),
136            ),
137            "DATE_TRUNC": lambda args: exp.TimestampTrunc(
138                this=seq_get(args, 1),
139                unit=exp.var(seq_get(args, 0)),
140            ),
141            "TRUNC": lambda args: exp.DateTrunc(unit=seq_get(args, 1), this=seq_get(args, 0)),
142            "BOOLEAN": _parse_as_cast("boolean"),
143            "DATE": _parse_as_cast("date"),
144            "DOUBLE": _parse_as_cast("double"),
145            "FLOAT": _parse_as_cast("float"),
146            "INT": _parse_as_cast("int"),
147            "STRING": _parse_as_cast("string"),
148            "TIMESTAMP": _parse_as_cast("timestamp"),
149        }
150
151        FUNCTION_PARSERS = {
152            **parser.Parser.FUNCTION_PARSERS,
153            "BROADCAST": lambda self: self._parse_join_hint("BROADCAST"),
154            "BROADCASTJOIN": lambda self: self._parse_join_hint("BROADCASTJOIN"),
155            "MAPJOIN": lambda self: self._parse_join_hint("MAPJOIN"),
156            "MERGE": lambda self: self._parse_join_hint("MERGE"),
157            "SHUFFLEMERGE": lambda self: self._parse_join_hint("SHUFFLEMERGE"),
158            "MERGEJOIN": lambda self: self._parse_join_hint("MERGEJOIN"),
159            "SHUFFLE_HASH": lambda self: self._parse_join_hint("SHUFFLE_HASH"),
160            "SHUFFLE_REPLICATE_NL": lambda self: self._parse_join_hint("SHUFFLE_REPLICATE_NL"),
161        }
162
163        def _parse_add_column(self) -> t.Optional[exp.Expression]:
164            return self._match_text_seq("ADD", "COLUMNS") and self._parse_schema()
165
166        def _parse_drop_column(self) -> t.Optional[exp.Drop | exp.Command]:
167            return self._match_text_seq("DROP", "COLUMNS") and self.expression(
168                exp.Drop, this=self._parse_schema(), kind="COLUMNS"
169            )
170
171        def _pivot_column_names(self, aggregations: t.List[exp.Expression]) -> t.List[str]:
172            if len(aggregations) == 1:
173                return [""]
174            return pivot_column_names(aggregations, dialect="spark")
175
176    class Generator(Hive.Generator):
177        QUERY_HINTS = True
178
179        TYPE_MAPPING = {
180            **Hive.Generator.TYPE_MAPPING,
181        }
182
183        PROPERTIES_LOCATION = {
184            **Hive.Generator.PROPERTIES_LOCATION,
185            exp.EngineProperty: exp.Properties.Location.UNSUPPORTED,
186            exp.AutoIncrementProperty: exp.Properties.Location.UNSUPPORTED,
187            exp.CharacterSetProperty: exp.Properties.Location.UNSUPPORTED,
188            exp.CollateProperty: exp.Properties.Location.UNSUPPORTED,
189        }
190
191        TRANSFORMS = {
192            **Hive.Generator.TRANSFORMS,
193            exp.ApproxDistinct: rename_func("APPROX_COUNT_DISTINCT"),
194            exp.ArraySum: lambda self, e: f"AGGREGATE({self.sql(e, 'this')}, 0, (acc, x) -> acc + x, acc -> acc)",
195            exp.AtTimeZone: lambda self, e: f"FROM_UTC_TIMESTAMP({self.sql(e, 'this')}, {self.sql(e, 'zone')})",
196            exp.BitwiseLeftShift: rename_func("SHIFTLEFT"),
197            exp.BitwiseRightShift: rename_func("SHIFTRIGHT"),
198            exp.Create: _create_sql,
199            exp.DateFromParts: rename_func("MAKE_DATE"),
200            exp.DateTrunc: lambda self, e: self.func("TRUNC", e.this, e.args.get("unit")),
201            exp.DayOfMonth: rename_func("DAYOFMONTH"),
202            exp.DayOfWeek: rename_func("DAYOFWEEK"),
203            exp.DayOfYear: rename_func("DAYOFYEAR"),
204            exp.FileFormatProperty: lambda self, e: f"USING {e.name.upper()}",
205            exp.From: transforms.preprocess([_unalias_pivot]),
206            exp.LogicalAnd: rename_func("BOOL_AND"),
207            exp.LogicalOr: rename_func("BOOL_OR"),
208            exp.Map: _map_sql,
209            exp.Pivot: transforms.preprocess([_unqualify_pivot_columns]),
210            exp.Reduce: rename_func("AGGREGATE"),
211            exp.StrToDate: _str_to_date,
212            exp.StrToTime: lambda self, e: f"TO_TIMESTAMP({self.sql(e, 'this')}, {self.format_time(e)})",
213            exp.TimestampTrunc: lambda self, e: self.func(
214                "DATE_TRUNC", exp.Literal.string(e.text("unit")), e.this
215            ),
216            exp.Trim: trim_sql,
217            exp.UnixToTime: _unix_to_time_sql,
218            exp.VariancePop: rename_func("VAR_POP"),
219            exp.WeekOfYear: rename_func("WEEKOFYEAR"),
220            exp.WithinGroup: transforms.preprocess(
221                [transforms.remove_within_group_for_percentiles]
222            ),
223        }
224        TRANSFORMS.pop(exp.ArrayJoin)
225        TRANSFORMS.pop(exp.ArraySort)
226        TRANSFORMS.pop(exp.ILike)
227        TRANSFORMS.pop(exp.Left)
228        TRANSFORMS.pop(exp.Right)
229
230        WRAP_DERIVED_VALUES = False
231        CREATE_FUNCTION_RETURN_AS = False
232
233        def cast_sql(self, expression: exp.Cast, safe_prefix: t.Optional[str] = None) -> str:
234            if isinstance(expression.this, exp.Cast) and expression.this.is_type("json"):
235                schema = f"'{self.sql(expression, 'to')}'"
236                return self.func("FROM_JSON", expression.this.this, schema)
237            if expression.is_type("json"):
238                return self.func("TO_JSON", expression.this)
239
240            return super(Hive.Generator, self).cast_sql(expression, safe_prefix=safe_prefix)
241
242        def columndef_sql(self, expression: exp.ColumnDef, sep: str = " ") -> str:
243            return super().columndef_sql(
244                expression,
245                sep=": "
246                if isinstance(expression.parent, exp.DataType)
247                and expression.parent.is_type("struct")
248                else sep,
249            )
250
251    class Tokenizer(Hive.Tokenizer):
252        HEX_STRINGS = [("X'", "'")]
tokenizer_class = <class 'sqlglot.dialects.spark2.Spark2.Tokenizer'>
parser_class = <class 'sqlglot.dialects.spark2.Spark2.Parser'>
generator_class = <class 'sqlglot.dialects.spark2.Spark2.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}}}
QUOTE_START = "'"
QUOTE_END = "'"
IDENTIFIER_START = '`'
IDENTIFIER_END = '`'
BIT_START = None
BIT_END = None
HEX_START = "X'"
HEX_END = "'"
BYTE_START = None
BYTE_END = None
class Spark2.Parser(sqlglot.dialects.hive.Hive.Parser):
109    class Parser(Hive.Parser):
110        FUNCTIONS = {
111            **Hive.Parser.FUNCTIONS,
112            "MAP_FROM_ARRAYS": exp.Map.from_arg_list,
113            "TO_UNIX_TIMESTAMP": exp.StrToUnix.from_arg_list,
114            "SHIFTLEFT": lambda args: exp.BitwiseLeftShift(
115                this=seq_get(args, 0),
116                expression=seq_get(args, 1),
117            ),
118            "SHIFTRIGHT": lambda args: exp.BitwiseRightShift(
119                this=seq_get(args, 0),
120                expression=seq_get(args, 1),
121            ),
122            "APPROX_PERCENTILE": exp.ApproxQuantile.from_arg_list,
123            "IIF": exp.If.from_arg_list,
124            "AGGREGATE": exp.Reduce.from_arg_list,
125            "DAYOFWEEK": lambda args: exp.DayOfWeek(
126                this=exp.TsOrDsToDate(this=seq_get(args, 0)),
127            ),
128            "DAYOFMONTH": lambda args: exp.DayOfMonth(
129                this=exp.TsOrDsToDate(this=seq_get(args, 0)),
130            ),
131            "DAYOFYEAR": lambda args: exp.DayOfYear(
132                this=exp.TsOrDsToDate(this=seq_get(args, 0)),
133            ),
134            "WEEKOFYEAR": lambda args: exp.WeekOfYear(
135                this=exp.TsOrDsToDate(this=seq_get(args, 0)),
136            ),
137            "DATE_TRUNC": lambda args: exp.TimestampTrunc(
138                this=seq_get(args, 1),
139                unit=exp.var(seq_get(args, 0)),
140            ),
141            "TRUNC": lambda args: exp.DateTrunc(unit=seq_get(args, 1), this=seq_get(args, 0)),
142            "BOOLEAN": _parse_as_cast("boolean"),
143            "DATE": _parse_as_cast("date"),
144            "DOUBLE": _parse_as_cast("double"),
145            "FLOAT": _parse_as_cast("float"),
146            "INT": _parse_as_cast("int"),
147            "STRING": _parse_as_cast("string"),
148            "TIMESTAMP": _parse_as_cast("timestamp"),
149        }
150
151        FUNCTION_PARSERS = {
152            **parser.Parser.FUNCTION_PARSERS,
153            "BROADCAST": lambda self: self._parse_join_hint("BROADCAST"),
154            "BROADCASTJOIN": lambda self: self._parse_join_hint("BROADCASTJOIN"),
155            "MAPJOIN": lambda self: self._parse_join_hint("MAPJOIN"),
156            "MERGE": lambda self: self._parse_join_hint("MERGE"),
157            "SHUFFLEMERGE": lambda self: self._parse_join_hint("SHUFFLEMERGE"),
158            "MERGEJOIN": lambda self: self._parse_join_hint("MERGEJOIN"),
159            "SHUFFLE_HASH": lambda self: self._parse_join_hint("SHUFFLE_HASH"),
160            "SHUFFLE_REPLICATE_NL": lambda self: self._parse_join_hint("SHUFFLE_REPLICATE_NL"),
161        }
162
163        def _parse_add_column(self) -> t.Optional[exp.Expression]:
164            return self._match_text_seq("ADD", "COLUMNS") and self._parse_schema()
165
166        def _parse_drop_column(self) -> t.Optional[exp.Drop | exp.Command]:
167            return self._match_text_seq("DROP", "COLUMNS") and self.expression(
168                exp.Drop, this=self._parse_schema(), kind="COLUMNS"
169            )
170
171        def _pivot_column_names(self, aggregations: t.List[exp.Expression]) -> t.List[str]:
172            if len(aggregations) == 1:
173                return [""]
174            return pivot_column_names(aggregations, dialect="spark")

Parser consumes a list of tokens produced by the Tokenizer and produces a parsed syntax tree.

Arguments:
  • error_level: The desired error level. Default: ErrorLevel.IMMEDIATE
  • error_message_context: Determines the amount of context to capture from a query string when displaying the error message (in number of characters). Default: 100
  • max_errors: Maximum number of error messages to include in a raised ParseError. This is only relevant if error_level is ErrorLevel.RAISE. Default: 3
FUNCTIONS = {'ABS': <bound method Func.from_arg_list of <class 'sqlglot.expressions.Abs'>>, 'ANY_VALUE': <bound method Func.from_arg_list of <class 'sqlglot.expressions.AnyValue'>>, 'APPROX_DISTINCT': <bound method Func.from_arg_list of <class 'sqlglot.expressions.ApproxDistinct'>>, 'APPROX_COUNT_DISTINCT': <bound method Func.from_arg_list of <class 'sqlglot.expressions.ApproxDistinct'>>, 'APPROX_QUANTILE': <bound method Func.from_arg_list of <class 'sqlglot.expressions.ApproxQuantile'>>, '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_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'>>, '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'>>, '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': <function _parse_as_cast.<locals>.<lambda>>, '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': <function Spark2.Parser.<lambda>>, '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': <function Spark2.Parser.<lambda>>, 'DAY_OF_WEEK': <bound method Func.from_arg_list of <class 'sqlglot.expressions.DayOfWeek'>>, 'DAYOFWEEK': <function Spark2.Parser.<lambda>>, 'DAY_OF_YEAR': <bound method Func.from_arg_list of <class 'sqlglot.expressions.DayOfYear'>>, 'DAYOFYEAR': <function Spark2.Parser.<lambda>>, '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'>>, 'EXTRACT': <bound method Func.from_arg_list of <class 'sqlglot.expressions.Extract'>>, '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'>>, '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'>>, '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>>, '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'>>, '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'>>, 'POWER': <bound method Func.from_arg_list of <class 'sqlglot.expressions.Pow'>>, 'POW': <bound method Func.from_arg_list of <class 'sqlglot.expressions.Pow'>>, '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_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'>>, '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_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'>>, '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_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'>>, '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': <function Spark2.Parser.<lambda>>, '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'>>, '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>, 'MAP_FROM_ARRAYS': <bound method Func.from_arg_list of <class 'sqlglot.expressions.Map'>>, 'TO_UNIX_TIMESTAMP': <bound method Func.from_arg_list of <class 'sqlglot.expressions.StrToUnix'>>, 'SHIFTLEFT': <function Spark2.Parser.<lambda>>, 'SHIFTRIGHT': <function Spark2.Parser.<lambda>>, 'APPROX_PERCENTILE': <bound method Func.from_arg_list of <class 'sqlglot.expressions.ApproxQuantile'>>, 'IIF': <bound method Func.from_arg_list of <class 'sqlglot.expressions.If'>>, 'AGGREGATE': <bound method Func.from_arg_list of <class 'sqlglot.expressions.Reduce'>>, 'TRUNC': <function Spark2.Parser.<lambda>>, 'BOOLEAN': <function _parse_as_cast.<locals>.<lambda>>, 'DOUBLE': <function _parse_as_cast.<locals>.<lambda>>, 'FLOAT': <function _parse_as_cast.<locals>.<lambda>>, 'INT': <function _parse_as_cast.<locals>.<lambda>>, 'STRING': <function _parse_as_cast.<locals>.<lambda>>, 'TIMESTAMP': <function _parse_as_cast.<locals>.<lambda>>}
FUNCTION_PARSERS = {'ANY_VALUE': <function Parser.<lambda>>, 'CAST': <function Parser.<lambda>>, 'CONCAT': <function Parser.<lambda>>, 'CONVERT': <function Parser.<lambda>>, 'DECODE': <function Parser.<lambda>>, 'EXTRACT': <function Parser.<lambda>>, 'JSON_OBJECT': <function Parser.<lambda>>, 'LOG': <function Parser.<lambda>>, 'MATCH': <function Parser.<lambda>>, 'OPENJSON': <function Parser.<lambda>>, 'POSITION': <function Parser.<lambda>>, 'SAFE_CAST': <function Parser.<lambda>>, 'STRING_AGG': <function Parser.<lambda>>, 'SUBSTRING': <function Parser.<lambda>>, 'TRIM': <function Parser.<lambda>>, 'TRY_CAST': <function Parser.<lambda>>, 'TRY_CONVERT': <function Parser.<lambda>>, 'BROADCAST': <function Spark2.Parser.<lambda>>, 'BROADCASTJOIN': <function Spark2.Parser.<lambda>>, 'MAPJOIN': <function Spark2.Parser.<lambda>>, 'MERGE': <function Spark2.Parser.<lambda>>, 'SHUFFLEMERGE': <function Spark2.Parser.<lambda>>, 'MERGEJOIN': <function Spark2.Parser.<lambda>>, 'SHUFFLE_HASH': <function Spark2.Parser.<lambda>>, 'SHUFFLE_REPLICATE_NL': <function Spark2.Parser.<lambda>>}
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_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}}}}}
class Spark2.Generator(sqlglot.dialects.hive.Hive.Generator):
176    class Generator(Hive.Generator):
177        QUERY_HINTS = True
178
179        TYPE_MAPPING = {
180            **Hive.Generator.TYPE_MAPPING,
181        }
182
183        PROPERTIES_LOCATION = {
184            **Hive.Generator.PROPERTIES_LOCATION,
185            exp.EngineProperty: exp.Properties.Location.UNSUPPORTED,
186            exp.AutoIncrementProperty: exp.Properties.Location.UNSUPPORTED,
187            exp.CharacterSetProperty: exp.Properties.Location.UNSUPPORTED,
188            exp.CollateProperty: exp.Properties.Location.UNSUPPORTED,
189        }
190
191        TRANSFORMS = {
192            **Hive.Generator.TRANSFORMS,
193            exp.ApproxDistinct: rename_func("APPROX_COUNT_DISTINCT"),
194            exp.ArraySum: lambda self, e: f"AGGREGATE({self.sql(e, 'this')}, 0, (acc, x) -> acc + x, acc -> acc)",
195            exp.AtTimeZone: lambda self, e: f"FROM_UTC_TIMESTAMP({self.sql(e, 'this')}, {self.sql(e, 'zone')})",
196            exp.BitwiseLeftShift: rename_func("SHIFTLEFT"),
197            exp.BitwiseRightShift: rename_func("SHIFTRIGHT"),
198            exp.Create: _create_sql,
199            exp.DateFromParts: rename_func("MAKE_DATE"),
200            exp.DateTrunc: lambda self, e: self.func("TRUNC", e.this, e.args.get("unit")),
201            exp.DayOfMonth: rename_func("DAYOFMONTH"),
202            exp.DayOfWeek: rename_func("DAYOFWEEK"),
203            exp.DayOfYear: rename_func("DAYOFYEAR"),
204            exp.FileFormatProperty: lambda self, e: f"USING {e.name.upper()}",
205            exp.From: transforms.preprocess([_unalias_pivot]),
206            exp.LogicalAnd: rename_func("BOOL_AND"),
207            exp.LogicalOr: rename_func("BOOL_OR"),
208            exp.Map: _map_sql,
209            exp.Pivot: transforms.preprocess([_unqualify_pivot_columns]),
210            exp.Reduce: rename_func("AGGREGATE"),
211            exp.StrToDate: _str_to_date,
212            exp.StrToTime: lambda self, e: f"TO_TIMESTAMP({self.sql(e, 'this')}, {self.format_time(e)})",
213            exp.TimestampTrunc: lambda self, e: self.func(
214                "DATE_TRUNC", exp.Literal.string(e.text("unit")), e.this
215            ),
216            exp.Trim: trim_sql,
217            exp.UnixToTime: _unix_to_time_sql,
218            exp.VariancePop: rename_func("VAR_POP"),
219            exp.WeekOfYear: rename_func("WEEKOFYEAR"),
220            exp.WithinGroup: transforms.preprocess(
221                [transforms.remove_within_group_for_percentiles]
222            ),
223        }
224        TRANSFORMS.pop(exp.ArrayJoin)
225        TRANSFORMS.pop(exp.ArraySort)
226        TRANSFORMS.pop(exp.ILike)
227        TRANSFORMS.pop(exp.Left)
228        TRANSFORMS.pop(exp.Right)
229
230        WRAP_DERIVED_VALUES = False
231        CREATE_FUNCTION_RETURN_AS = False
232
233        def cast_sql(self, expression: exp.Cast, safe_prefix: t.Optional[str] = None) -> str:
234            if isinstance(expression.this, exp.Cast) and expression.this.is_type("json"):
235                schema = f"'{self.sql(expression, 'to')}'"
236                return self.func("FROM_JSON", expression.this.this, schema)
237            if expression.is_type("json"):
238                return self.func("TO_JSON", expression.this)
239
240            return super(Hive.Generator, self).cast_sql(expression, safe_prefix=safe_prefix)
241
242        def columndef_sql(self, expression: exp.ColumnDef, sep: str = " ") -> str:
243            return super().columndef_sql(
244                expression,
245                sep=": "
246                if isinstance(expression.parent, exp.DataType)
247                and expression.parent.is_type("struct")
248                else sep,
249            )

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
QUERY_HINTS = True
TYPE_MAPPING = {<Type.NCHAR: 'NCHAR'>: 'CHAR', <Type.NVARCHAR: 'NVARCHAR'>: 'VARCHAR', <Type.MEDIUMTEXT: 'MEDIUMTEXT'>: 'TEXT', <Type.LONGTEXT: 'LONGTEXT'>: 'TEXT', <Type.MEDIUMBLOB: 'MEDIUMBLOB'>: 'BLOB', <Type.LONGBLOB: 'LONGBLOB'>: 'BLOB', <Type.INET: 'INET'>: 'INET', <Type.TEXT: 'TEXT'>: 'STRING', <Type.DATETIME: 'DATETIME'>: 'TIMESTAMP', <Type.VARBINARY: 'VARBINARY'>: 'BINARY', <Type.TIMESTAMPTZ: 'TIMESTAMPTZ'>: 'TIMESTAMP', <Type.BIT: 'BIT'>: 'BOOLEAN'}
PROPERTIES_LOCATION = {<class 'sqlglot.expressions.AlgorithmProperty'>: <Location.POST_CREATE: 'POST_CREATE'>, <class 'sqlglot.expressions.AutoIncrementProperty'>: <Location.UNSUPPORTED: 'UNSUPPORTED'>, <class 'sqlglot.expressions.BlockCompressionProperty'>: <Location.POST_NAME: 'POST_NAME'>, <class 'sqlglot.expressions.CharacterSetProperty'>: <Location.UNSUPPORTED: 'UNSUPPORTED'>, <class 'sqlglot.expressions.ChecksumProperty'>: <Location.POST_NAME: 'POST_NAME'>, <class 'sqlglot.expressions.CollateProperty'>: <Location.UNSUPPORTED: 'UNSUPPORTED'>, <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.UNSUPPORTED: 'UNSUPPORTED'>, <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.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.OnCommitProperty'>: <Location.POST_EXPRESSION: 'POST_EXPRESSION'>, <class 'sqlglot.expressions.Order'>: <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.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.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.MergeTreeTTL'>: <Location.POST_SCHEMA: 'POST_SCHEMA'>, <class 'sqlglot.expressions.VolatileProperty'>: <Location.UNSUPPORTED: 'UNSUPPORTED'>, <class 'sqlglot.expressions.WithDataProperty'>: <Location.POST_EXPRESSION: 'POST_EXPRESSION'>, <class 'sqlglot.expressions.WithJournalTableProperty'>: <Location.POST_NAME: 'POST_NAME'>}
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.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.InlineLengthColumnConstraint'>: <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.OnCommitProperty'>: <function Generator.<lambda>>, <class 'sqlglot.expressions.OnUpdateColumnConstraint'>: <function Generator.<lambda>>, <class 'sqlglot.expressions.PathColumnConstraint'>: <function Generator.<lambda>>, <class 'sqlglot.expressions.ReturnsProperty'>: <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.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.ApproxDistinct'>: <function rename_func.<locals>.<lambda>>, <class 'sqlglot.expressions.ArrayConcat'>: <function rename_func.<locals>.<lambda>>, <class 'sqlglot.expressions.ArraySize'>: <function rename_func.<locals>.<lambda>>, <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 Spark2.Generator.<lambda>>, <class 'sqlglot.expressions.FromBase64'>: <function rename_func.<locals>.<lambda>>, <class 'sqlglot.expressions.If'>: <function if_sql>, <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.Map'>: <function _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.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.RegexpLike'>: <function Hive.Generator.<lambda>>, <class 'sqlglot.expressions.RegexpSplit'>: <function rename_func.<locals>.<lambda>>, <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>, <class 'sqlglot.expressions.StrToTime'>: <function Spark2.Generator.<lambda>>, <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 _unix_to_time_sql>, <class 'sqlglot.expressions.UnixToTimeStr'>: <function rename_func.<locals>.<lambda>>, <class 'sqlglot.expressions.PartitionedByProperty'>: <function Hive.Generator.<lambda>>, <class 'sqlglot.expressions.RowFormatSerdeProperty'>: <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.ArraySum'>: <function Spark2.Generator.<lambda>>, <class 'sqlglot.expressions.AtTimeZone'>: <function Spark2.Generator.<lambda>>, <class 'sqlglot.expressions.BitwiseLeftShift'>: <function rename_func.<locals>.<lambda>>, <class 'sqlglot.expressions.BitwiseRightShift'>: <function rename_func.<locals>.<lambda>>, <class 'sqlglot.expressions.DateFromParts'>: <function rename_func.<locals>.<lambda>>, <class 'sqlglot.expressions.DateTrunc'>: <function Spark2.Generator.<lambda>>, <class 'sqlglot.expressions.DayOfMonth'>: <function rename_func.<locals>.<lambda>>, <class 'sqlglot.expressions.DayOfWeek'>: <function rename_func.<locals>.<lambda>>, <class 'sqlglot.expressions.DayOfYear'>: <function rename_func.<locals>.<lambda>>, <class 'sqlglot.expressions.From'>: <function preprocess.<locals>._to_sql>, <class 'sqlglot.expressions.LogicalAnd'>: <function rename_func.<locals>.<lambda>>, <class 'sqlglot.expressions.LogicalOr'>: <function rename_func.<locals>.<lambda>>, <class 'sqlglot.expressions.Pivot'>: <function preprocess.<locals>._to_sql>, <class 'sqlglot.expressions.Reduce'>: <function rename_func.<locals>.<lambda>>, <class 'sqlglot.expressions.TimestampTrunc'>: <function Spark2.Generator.<lambda>>, <class 'sqlglot.expressions.Trim'>: <function trim_sql>, <class 'sqlglot.expressions.VariancePop'>: <function rename_func.<locals>.<lambda>>, <class 'sqlglot.expressions.WeekOfYear'>: <function rename_func.<locals>.<lambda>>, <class 'sqlglot.expressions.WithinGroup'>: <function preprocess.<locals>._to_sql>}
WRAP_DERIVED_VALUES = False
CREATE_FUNCTION_RETURN_AS = False
def cast_sql( self, expression: sqlglot.expressions.Cast, safe_prefix: Optional[str] = None) -> str:
233        def cast_sql(self, expression: exp.Cast, safe_prefix: t.Optional[str] = None) -> str:
234            if isinstance(expression.this, exp.Cast) and expression.this.is_type("json"):
235                schema = f"'{self.sql(expression, 'to')}'"
236                return self.func("FROM_JSON", expression.this.this, schema)
237            if expression.is_type("json"):
238                return self.func("TO_JSON", expression.this)
239
240            return super(Hive.Generator, self).cast_sql(expression, safe_prefix=safe_prefix)
def columndef_sql(self, expression: sqlglot.expressions.ColumnDef, sep: str = ' ') -> str:
242        def columndef_sql(self, expression: exp.ColumnDef, sep: str = " ") -> str:
243            return super().columndef_sql(
244                expression,
245                sep=": "
246                if isinstance(expression.parent, exp.DataType)
247                and expression.parent.is_type("struct")
248                else sep,
249            )
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}}}
@classmethod
def can_identify(text: str, identify: str | bool = 'safe') -> bool:
246    @classmethod
247    def can_identify(cls, text: str, identify: str | bool = "safe") -> bool:
248        """Checks if text can be identified given an identify option.
249
250        Args:
251            text: The text to check.
252            identify:
253                "always" or `True`: Always returns true.
254                "safe": True if the identifier is case-insensitive.
255
256        Returns:
257            Whether or not the given text can be identified.
258        """
259        if identify is True or identify == "always":
260            return True
261
262        if identify == "safe":
263            return not cls.case_sensitive(text)
264
265        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 = '`'
STRING_ESCAPE = '\\'
IDENTIFIER_ESCAPE = '"'
BIT_START: Optional[str] = None
BIT_END: Optional[str] = None
HEX_START: Optional[str] = "X'"
HEX_END: Optional[str] = "'"
BYTE_START: Optional[str] = None
BYTE_END: Optional[str] = None
Inherited Members
sqlglot.generator.Generator
Generator
NULL_ORDERING_SUPPORTED
LOCKING_READS_SUPPORTED
EXPLICIT_UNION
MATCHED_BY_SOURCE
SINGLE_STRING_INTERVAL
INTERVAL_ALLOWS_PLURAL_FORM
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
STAR_MAPPING
TIME_PART_SINGULARS
TOKEN_MAPPING
STRUCT_DELIMITER
PARAMETER_TOKEN
RESERVED_KEYWORDS
WITH_SEPARATED_COMMENTS
UNWRAPPED_INTERVAL_VALUES
SENTINEL_LINE_BREAK
INDEX_OFFSET
UNNEST_COLUMN_ONLY
STRICT_STRING_CONCAT
NORMALIZE_FUNCTIONS
NULL_ORDERING
ESCAPE_LINE_BREAK
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
columnconstraint_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
datatypesize_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_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
pseudotype_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
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_sql
schema_columns_sql
star_sql
parameter_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
constraint_sql
nextvaluefor_sql
extract_sql
trim_sql
safeconcat_sql
check_sql
foreignkey_sql
primarykey_sql
if_sql
matchagainst_sql
jsonkeyvalue_sql
jsonobject_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
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
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
sqlglot.dialects.hive.Hive.Generator
LIMIT_FETCH
TABLESAMPLE_WITH_METHOD
TABLESAMPLE_SIZE_IS_PERCENT
JOIN_HINTS
TABLE_HINTS
INDEX_ON
arrayagg_sql
with_properties
datatype_sql
ALIAS_POST_TABLESAMPLE
IDENTIFIERS_CAN_START_WITH_DIGIT
class Spark2.Tokenizer(sqlglot.dialects.hive.Hive.Tokenizer):
251    class Tokenizer(Hive.Tokenizer):
252        HEX_STRINGS = [("X'", "'")]
HEX_STRINGS = [("X'", "'")]