Edit on GitHub

sqlglot.dialects.spark2

View Source

  1from __future__ import annotations
  2
  3import typing as t
  4
  5from sqlglot import exp, transforms
  6from sqlglot.dialects.dialect import (
  7    binary_from_function,
  8    create_with_partitions_sql,
  9    format_time_lambda,
 10    pivot_column_names,
 11    rename_func,
 12    trim_sql,
 13)
 14from sqlglot.dialects.hive import Hive
 15from sqlglot.helper import seq_get
 16
 17
 18def _create_sql(self: Spark2.Generator, e: exp.Create) -> str:
 19    kind = e.args["kind"]
 20    properties = e.args.get("properties")
 21
 22    if (
 23        kind.upper() == "TABLE"
 24        and e.expression
 25        and any(
 26            isinstance(prop, exp.TemporaryProperty)
 27            for prop in (properties.expressions if properties else [])
 28        )
 29    ):
 30        return f"CREATE TEMPORARY VIEW {self.sql(e, 'this')} AS {self.sql(e, 'expression')}"
 31    return create_with_partitions_sql(self, e)
 32
 33
 34def _map_sql(self: Spark2.Generator, expression: exp.Map) -> str:
 35    keys = expression.args.get("keys")
 36    values = expression.args.get("values")
 37
 38    if not keys or not values:
 39        return "MAP()"
 40
 41    return f"MAP_FROM_ARRAYS({self.sql(keys)}, {self.sql(values)})"
 42
 43
 44def _parse_as_cast(to_type: str) -> t.Callable[[t.List], exp.Expression]:
 45    return lambda args: exp.Cast(this=seq_get(args, 0), to=exp.DataType.build(to_type))
 46
 47
 48def _str_to_date(self: Spark2.Generator, expression: exp.StrToDate) -> str:
 49    this = self.sql(expression, "this")
 50    time_format = self.format_time(expression)
 51    if time_format == Hive.DATE_FORMAT:
 52        return f"TO_DATE({this})"
 53    return f"TO_DATE({this}, {time_format})"
 54
 55
 56def _unix_to_time_sql(self: Spark2.Generator, expression: exp.UnixToTime) -> str:
 57    scale = expression.args.get("scale")
 58    timestamp = self.sql(expression, "this")
 59    if scale is None:
 60        return f"CAST(FROM_UNIXTIME({timestamp}) AS TIMESTAMP)"
 61    if scale == exp.UnixToTime.SECONDS:
 62        return f"TIMESTAMP_SECONDS({timestamp})"
 63    if scale == exp.UnixToTime.MILLIS:
 64        return f"TIMESTAMP_MILLIS({timestamp})"
 65    if scale == exp.UnixToTime.MICROS:
 66        return f"TIMESTAMP_MICROS({timestamp})"
 67
 68    raise ValueError("Improper scale for timestamp")
 69
 70
 71def _unalias_pivot(expression: exp.Expression) -> exp.Expression:
 72    """
 73    Spark doesn't allow PIVOT aliases, so we need to remove them and possibly wrap a
 74    pivoted source in a subquery with the same alias to preserve the query's semantics.
 75
 76    Example:
 77        >>> from sqlglot import parse_one
 78        >>> expr = parse_one("SELECT piv.x FROM tbl PIVOT (SUM(a) FOR b IN ('x')) piv")
 79        >>> print(_unalias_pivot(expr).sql(dialect="spark"))
 80        SELECT piv.x FROM (SELECT * FROM tbl PIVOT(SUM(a) FOR b IN ('x'))) AS piv
 81    """
 82    if isinstance(expression, exp.From) and expression.this.args.get("pivots"):
 83        pivot = expression.this.args["pivots"][0]
 84        if pivot.alias:
 85            alias = pivot.args["alias"].pop()
 86            return exp.From(
 87                this=expression.this.replace(
 88                    exp.select("*").from_(expression.this.copy()).subquery(alias=alias)
 89                )
 90            )
 91
 92    return expression
 93
 94
 95def _unqualify_pivot_columns(expression: exp.Expression) -> exp.Expression:
 96    """
 97    Spark doesn't allow the column referenced in the PIVOT's field to be qualified,
 98    so we need to unqualify it.
 99
100    Example:
101        >>> from sqlglot import parse_one
102        >>> expr = parse_one("SELECT * FROM tbl PIVOT (SUM(tbl.sales) FOR tbl.quarter IN ('Q1', 'Q2'))")
103        >>> print(_unqualify_pivot_columns(expr).sql(dialect="spark"))
104        SELECT * FROM tbl PIVOT(SUM(tbl.sales) FOR quarter IN ('Q1', 'Q1'))
105    """
106    if isinstance(expression, exp.Pivot):
107        expression.args["field"].transform(
108            lambda node: exp.column(node.output_name, quoted=node.this.quoted)
109            if isinstance(node, exp.Column)
110            else node,
111            copy=False,
112        )
113
114    return expression
115
116
117def _insert_sql(self: Spark2.Generator, expression: exp.Insert) -> str:
118    if expression.expression.args.get("with"):
119        expression = expression.copy()
120        expression.set("with", expression.expression.args.pop("with"))
121    return self.insert_sql(expression)
122
123
124class Spark2(Hive):
125    class Parser(Hive.Parser):
126        FUNCTIONS = {
127            **Hive.Parser.FUNCTIONS,
128            "AGGREGATE": exp.Reduce.from_arg_list,
129            "APPROX_PERCENTILE": exp.ApproxQuantile.from_arg_list,
130            "BOOLEAN": _parse_as_cast("boolean"),
131            "DATE": _parse_as_cast("date"),
132            "DATE_TRUNC": lambda args: exp.TimestampTrunc(
133                this=seq_get(args, 1), unit=exp.var(seq_get(args, 0))
134            ),
135            "DAYOFMONTH": lambda args: exp.DayOfMonth(this=exp.TsOrDsToDate(this=seq_get(args, 0))),
136            "DAYOFWEEK": lambda args: exp.DayOfWeek(this=exp.TsOrDsToDate(this=seq_get(args, 0))),
137            "DAYOFYEAR": lambda args: exp.DayOfYear(this=exp.TsOrDsToDate(this=seq_get(args, 0))),
138            "DOUBLE": _parse_as_cast("double"),
139            "FLOAT": _parse_as_cast("float"),
140            "IIF": exp.If.from_arg_list,
141            "INT": _parse_as_cast("int"),
142            "MAP_FROM_ARRAYS": exp.Map.from_arg_list,
143            "RLIKE": exp.RegexpLike.from_arg_list,
144            "SHIFTLEFT": binary_from_function(exp.BitwiseLeftShift),
145            "SHIFTRIGHT": binary_from_function(exp.BitwiseRightShift),
146            "STRING": _parse_as_cast("string"),
147            "TIMESTAMP": _parse_as_cast("timestamp"),
148            "TO_TIMESTAMP": lambda args: _parse_as_cast("timestamp")(args)
149            if len(args) == 1
150            else format_time_lambda(exp.StrToTime, "spark")(args),
151            "TO_UNIX_TIMESTAMP": exp.StrToUnix.from_arg_list,
152            "TRUNC": lambda args: exp.DateTrunc(unit=seq_get(args, 1), this=seq_get(args, 0)),
153            "WEEKOFYEAR": lambda args: exp.WeekOfYear(this=exp.TsOrDsToDate(this=seq_get(args, 0))),
154        }
155
156        FUNCTION_PARSERS = {
157            **Hive.Parser.FUNCTION_PARSERS,
158            "BROADCAST": lambda self: self._parse_join_hint("BROADCAST"),
159            "BROADCASTJOIN": lambda self: self._parse_join_hint("BROADCASTJOIN"),
160            "MAPJOIN": lambda self: self._parse_join_hint("MAPJOIN"),
161            "MERGE": lambda self: self._parse_join_hint("MERGE"),
162            "SHUFFLEMERGE": lambda self: self._parse_join_hint("SHUFFLEMERGE"),
163            "MERGEJOIN": lambda self: self._parse_join_hint("MERGEJOIN"),
164            "SHUFFLE_HASH": lambda self: self._parse_join_hint("SHUFFLE_HASH"),
165            "SHUFFLE_REPLICATE_NL": lambda self: self._parse_join_hint("SHUFFLE_REPLICATE_NL"),
166        }
167
168        def _parse_add_column(self) -> t.Optional[exp.Expression]:
169            return self._match_text_seq("ADD", "COLUMNS") and self._parse_schema()
170
171        def _parse_drop_column(self) -> t.Optional[exp.Drop | exp.Command]:
172            return self._match_text_seq("DROP", "COLUMNS") and self.expression(
173                exp.Drop, this=self._parse_schema(), kind="COLUMNS"
174            )
175
176        def _pivot_column_names(self, aggregations: t.List[exp.Expression]) -> t.List[str]:
177            if len(aggregations) == 1:
178                return [""]
179            return pivot_column_names(aggregations, dialect="spark")
180
181    class Generator(Hive.Generator):
182        QUERY_HINTS = True
183        NVL2_SUPPORTED = True
184
185        PROPERTIES_LOCATION = {
186            **Hive.Generator.PROPERTIES_LOCATION,
187            exp.EngineProperty: exp.Properties.Location.UNSUPPORTED,
188            exp.AutoIncrementProperty: exp.Properties.Location.UNSUPPORTED,
189            exp.CharacterSetProperty: exp.Properties.Location.UNSUPPORTED,
190            exp.CollateProperty: exp.Properties.Location.UNSUPPORTED,
191        }
192
193        TRANSFORMS = {
194            **Hive.Generator.TRANSFORMS,
195            exp.ApproxDistinct: rename_func("APPROX_COUNT_DISTINCT"),
196            exp.ArraySum: lambda self, e: f"AGGREGATE({self.sql(e, 'this')}, 0, (acc, x) -> acc + x, acc -> acc)",
197            exp.AtTimeZone: lambda self, e: f"FROM_UTC_TIMESTAMP({self.sql(e, 'this')}, {self.sql(e, 'zone')})",
198            exp.BitwiseLeftShift: rename_func("SHIFTLEFT"),
199            exp.BitwiseRightShift: rename_func("SHIFTRIGHT"),
200            exp.Create: _create_sql,
201            exp.DateFromParts: rename_func("MAKE_DATE"),
202            exp.DateTrunc: lambda self, e: self.func("TRUNC", e.this, e.args.get("unit")),
203            exp.DayOfMonth: rename_func("DAYOFMONTH"),
204            exp.DayOfWeek: rename_func("DAYOFWEEK"),
205            exp.DayOfYear: rename_func("DAYOFYEAR"),
206            exp.FileFormatProperty: lambda self, e: f"USING {e.name.upper()}",
207            exp.From: transforms.preprocess([_unalias_pivot]),
208            exp.Insert: _insert_sql,
209            exp.LogicalAnd: rename_func("BOOL_AND"),
210            exp.LogicalOr: rename_func("BOOL_OR"),
211            exp.Map: _map_sql,
212            exp.Pivot: transforms.preprocess([_unqualify_pivot_columns]),
213            exp.Reduce: rename_func("AGGREGATE"),
214            exp.RegexpReplace: lambda self, e: self.func(
215                "REGEXP_REPLACE",
216                e.this,
217                e.expression,
218                e.args["replacement"],
219                e.args.get("position"),
220            ),
221            exp.StrToDate: _str_to_date,
222            exp.StrToTime: lambda self, e: f"TO_TIMESTAMP({self.sql(e, 'this')}, {self.format_time(e)})",
223            exp.TimestampTrunc: lambda self, e: self.func(
224                "DATE_TRUNC", exp.Literal.string(e.text("unit")), e.this
225            ),
226            exp.Trim: trim_sql,
227            exp.UnixToTime: _unix_to_time_sql,
228            exp.VariancePop: rename_func("VAR_POP"),
229            exp.WeekOfYear: rename_func("WEEKOFYEAR"),
230            exp.WithinGroup: transforms.preprocess(
231                [transforms.remove_within_group_for_percentiles]
232            ),
233        }
234        TRANSFORMS.pop(exp.ArrayJoin)
235        TRANSFORMS.pop(exp.ArraySort)
236        TRANSFORMS.pop(exp.ILike)
237        TRANSFORMS.pop(exp.Left)
238        TRANSFORMS.pop(exp.MonthsBetween)
239        TRANSFORMS.pop(exp.Right)
240
241        WRAP_DERIVED_VALUES = False
242        CREATE_FUNCTION_RETURN_AS = False
243
244        def cast_sql(self, expression: exp.Cast, safe_prefix: t.Optional[str] = None) -> str:
245            if isinstance(expression.this, exp.Cast) and expression.this.is_type("json"):
246                schema = f"'{self.sql(expression, 'to')}'"
247                return self.func("FROM_JSON", expression.this.this, schema)
248            if expression.is_type("json"):
249                return self.func("TO_JSON", expression.this)
250
251            return super(Hive.Generator, self).cast_sql(expression, safe_prefix=safe_prefix)
252
253        def columndef_sql(self, expression: exp.ColumnDef, sep: str = " ") -> str:
254            return super().columndef_sql(
255                expression,
256                sep=": "
257                if isinstance(expression.parent, exp.DataType)
258                and expression.parent.is_type("struct")
259                else sep,
260            )
261
262    class Tokenizer(Hive.Tokenizer):
263        HEX_STRINGS = [("X'", "'")]

class Spark2(sqlglot.dialects.hive.Hive): View Source

125class Spark2(Hive):
126    class Parser(Hive.Parser):
127        FUNCTIONS = {
128            **Hive.Parser.FUNCTIONS,
129            "AGGREGATE": exp.Reduce.from_arg_list,
130            "APPROX_PERCENTILE": exp.ApproxQuantile.from_arg_list,
131            "BOOLEAN": _parse_as_cast("boolean"),
132            "DATE": _parse_as_cast("date"),
133            "DATE_TRUNC": lambda args: exp.TimestampTrunc(
134                this=seq_get(args, 1), unit=exp.var(seq_get(args, 0))
135            ),
136            "DAYOFMONTH": lambda args: exp.DayOfMonth(this=exp.TsOrDsToDate(this=seq_get(args, 0))),
137            "DAYOFWEEK": lambda args: exp.DayOfWeek(this=exp.TsOrDsToDate(this=seq_get(args, 0))),
138            "DAYOFYEAR": lambda args: exp.DayOfYear(this=exp.TsOrDsToDate(this=seq_get(args, 0))),
139            "DOUBLE": _parse_as_cast("double"),
140            "FLOAT": _parse_as_cast("float"),
141            "IIF": exp.If.from_arg_list,
142            "INT": _parse_as_cast("int"),
143            "MAP_FROM_ARRAYS": exp.Map.from_arg_list,
144            "RLIKE": exp.RegexpLike.from_arg_list,
145            "SHIFTLEFT": binary_from_function(exp.BitwiseLeftShift),
146            "SHIFTRIGHT": binary_from_function(exp.BitwiseRightShift),
147            "STRING": _parse_as_cast("string"),
148            "TIMESTAMP": _parse_as_cast("timestamp"),
149            "TO_TIMESTAMP": lambda args: _parse_as_cast("timestamp")(args)
150            if len(args) == 1
151            else format_time_lambda(exp.StrToTime, "spark")(args),
152            "TO_UNIX_TIMESTAMP": exp.StrToUnix.from_arg_list,
153            "TRUNC": lambda args: exp.DateTrunc(unit=seq_get(args, 1), this=seq_get(args, 0)),
154            "WEEKOFYEAR": lambda args: exp.WeekOfYear(this=exp.TsOrDsToDate(this=seq_get(args, 0))),
155        }
156
157        FUNCTION_PARSERS = {
158            **Hive.Parser.FUNCTION_PARSERS,
159            "BROADCAST": lambda self: self._parse_join_hint("BROADCAST"),
160            "BROADCASTJOIN": lambda self: self._parse_join_hint("BROADCASTJOIN"),
161            "MAPJOIN": lambda self: self._parse_join_hint("MAPJOIN"),
162            "MERGE": lambda self: self._parse_join_hint("MERGE"),
163            "SHUFFLEMERGE": lambda self: self._parse_join_hint("SHUFFLEMERGE"),
164            "MERGEJOIN": lambda self: self._parse_join_hint("MERGEJOIN"),
165            "SHUFFLE_HASH": lambda self: self._parse_join_hint("SHUFFLE_HASH"),
166            "SHUFFLE_REPLICATE_NL": lambda self: self._parse_join_hint("SHUFFLE_REPLICATE_NL"),
167        }
168
169        def _parse_add_column(self) -> t.Optional[exp.Expression]:
170            return self._match_text_seq("ADD", "COLUMNS") and self._parse_schema()
171
172        def _parse_drop_column(self) -> t.Optional[exp.Drop | exp.Command]:
173            return self._match_text_seq("DROP", "COLUMNS") and self.expression(
174                exp.Drop, this=self._parse_schema(), kind="COLUMNS"
175            )
176
177        def _pivot_column_names(self, aggregations: t.List[exp.Expression]) -> t.List[str]:
178            if len(aggregations) == 1:
179                return [""]
180            return pivot_column_names(aggregations, dialect="spark")
181
182    class Generator(Hive.Generator):
183        QUERY_HINTS = True
184        NVL2_SUPPORTED = True
185
186        PROPERTIES_LOCATION = {
187            **Hive.Generator.PROPERTIES_LOCATION,
188            exp.EngineProperty: exp.Properties.Location.UNSUPPORTED,
189            exp.AutoIncrementProperty: exp.Properties.Location.UNSUPPORTED,
190            exp.CharacterSetProperty: exp.Properties.Location.UNSUPPORTED,
191            exp.CollateProperty: exp.Properties.Location.UNSUPPORTED,
192        }
193
194        TRANSFORMS = {
195            **Hive.Generator.TRANSFORMS,
196            exp.ApproxDistinct: rename_func("APPROX_COUNT_DISTINCT"),
197            exp.ArraySum: lambda self, e: f"AGGREGATE({self.sql(e, 'this')}, 0, (acc, x) -> acc + x, acc -> acc)",
198            exp.AtTimeZone: lambda self, e: f"FROM_UTC_TIMESTAMP({self.sql(e, 'this')}, {self.sql(e, 'zone')})",
199            exp.BitwiseLeftShift: rename_func("SHIFTLEFT"),
200            exp.BitwiseRightShift: rename_func("SHIFTRIGHT"),
201            exp.Create: _create_sql,
202            exp.DateFromParts: rename_func("MAKE_DATE"),
203            exp.DateTrunc: lambda self, e: self.func("TRUNC", e.this, e.args.get("unit")),
204            exp.DayOfMonth: rename_func("DAYOFMONTH"),
205            exp.DayOfWeek: rename_func("DAYOFWEEK"),
206            exp.DayOfYear: rename_func("DAYOFYEAR"),
207            exp.FileFormatProperty: lambda self, e: f"USING {e.name.upper()}",
208            exp.From: transforms.preprocess([_unalias_pivot]),
209            exp.Insert: _insert_sql,
210            exp.LogicalAnd: rename_func("BOOL_AND"),
211            exp.LogicalOr: rename_func("BOOL_OR"),
212            exp.Map: _map_sql,
213            exp.Pivot: transforms.preprocess([_unqualify_pivot_columns]),
214            exp.Reduce: rename_func("AGGREGATE"),
215            exp.RegexpReplace: lambda self, e: self.func(
216                "REGEXP_REPLACE",
217                e.this,
218                e.expression,
219                e.args["replacement"],
220                e.args.get("position"),
221            ),
222            exp.StrToDate: _str_to_date,
223            exp.StrToTime: lambda self, e: f"TO_TIMESTAMP({self.sql(e, 'this')}, {self.format_time(e)})",
224            exp.TimestampTrunc: lambda self, e: self.func(
225                "DATE_TRUNC", exp.Literal.string(e.text("unit")), e.this
226            ),
227            exp.Trim: trim_sql,
228            exp.UnixToTime: _unix_to_time_sql,
229            exp.VariancePop: rename_func("VAR_POP"),
230            exp.WeekOfYear: rename_func("WEEKOFYEAR"),
231            exp.WithinGroup: transforms.preprocess(
232                [transforms.remove_within_group_for_percentiles]
233            ),
234        }
235        TRANSFORMS.pop(exp.ArrayJoin)
236        TRANSFORMS.pop(exp.ArraySort)
237        TRANSFORMS.pop(exp.ILike)
238        TRANSFORMS.pop(exp.Left)
239        TRANSFORMS.pop(exp.MonthsBetween)
240        TRANSFORMS.pop(exp.Right)
241
242        WRAP_DERIVED_VALUES = False
243        CREATE_FUNCTION_RETURN_AS = False
244
245        def cast_sql(self, expression: exp.Cast, safe_prefix: t.Optional[str] = None) -> str:
246            if isinstance(expression.this, exp.Cast) and expression.this.is_type("json"):
247                schema = f"'{self.sql(expression, 'to')}'"
248                return self.func("FROM_JSON", expression.this.this, schema)
249            if expression.is_type("json"):
250                return self.func("TO_JSON", expression.this)
251
252            return super(Hive.Generator, self).cast_sql(expression, safe_prefix=safe_prefix)
253
254        def columndef_sql(self, expression: exp.ColumnDef, sep: str = " ") -> str:
255            return super().columndef_sql(
256                expression,
257                sep=": "
258                if isinstance(expression.parent, exp.DataType)
259                and expression.parent.is_type("struct")
260                else sep,
261            )
262
263    class Tokenizer(Hive.Tokenizer):
264        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

Inherited Members

sqlglot.dialects.hive.Hive: ALIAS_POST_TABLESAMPLE; IDENTIFIERS_CAN_START_WITH_DIGIT; RESOLVES_IDENTIFIERS_AS_UPPERCASE; TIME_MAPPING; DATE_FORMAT; DATEINT_FORMAT; TIME_FORMAT
sqlglot.dialects.dialect.Dialect: INDEX_OFFSET; UNNEST_COLUMN_ONLY; DPIPE_IS_STRING_CONCAT; STRICT_STRING_CONCAT; NORMALIZE_FUNCTIONS; NULL_ORDERING; FORMAT_MAPPING; PSEUDOCOLUMNS; get_or_raise; format_time; normalize_identifier; case_sensitive; can_identify; quote_identifier; parse; parse_into; generate; transpile; tokenize; tokenizer; parser; generator

class Spark2.Parser(sqlglot.dialects.hive.Hive.Parser): View Source

126    class Parser(Hive.Parser):
127        FUNCTIONS = {
128            **Hive.Parser.FUNCTIONS,
129            "AGGREGATE": exp.Reduce.from_arg_list,
130            "APPROX_PERCENTILE": exp.ApproxQuantile.from_arg_list,
131            "BOOLEAN": _parse_as_cast("boolean"),
132            "DATE": _parse_as_cast("date"),
133            "DATE_TRUNC": lambda args: exp.TimestampTrunc(
134                this=seq_get(args, 1), unit=exp.var(seq_get(args, 0))
135            ),
136            "DAYOFMONTH": lambda args: exp.DayOfMonth(this=exp.TsOrDsToDate(this=seq_get(args, 0))),
137            "DAYOFWEEK": lambda args: exp.DayOfWeek(this=exp.TsOrDsToDate(this=seq_get(args, 0))),
138            "DAYOFYEAR": lambda args: exp.DayOfYear(this=exp.TsOrDsToDate(this=seq_get(args, 0))),
139            "DOUBLE": _parse_as_cast("double"),
140            "FLOAT": _parse_as_cast("float"),
141            "IIF": exp.If.from_arg_list,
142            "INT": _parse_as_cast("int"),
143            "MAP_FROM_ARRAYS": exp.Map.from_arg_list,
144            "RLIKE": exp.RegexpLike.from_arg_list,
145            "SHIFTLEFT": binary_from_function(exp.BitwiseLeftShift),
146            "SHIFTRIGHT": binary_from_function(exp.BitwiseRightShift),
147            "STRING": _parse_as_cast("string"),
148            "TIMESTAMP": _parse_as_cast("timestamp"),
149            "TO_TIMESTAMP": lambda args: _parse_as_cast("timestamp")(args)
150            if len(args) == 1
151            else format_time_lambda(exp.StrToTime, "spark")(args),
152            "TO_UNIX_TIMESTAMP": exp.StrToUnix.from_arg_list,
153            "TRUNC": lambda args: exp.DateTrunc(unit=seq_get(args, 1), this=seq_get(args, 0)),
154            "WEEKOFYEAR": lambda args: exp.WeekOfYear(this=exp.TsOrDsToDate(this=seq_get(args, 0))),
155        }
156
157        FUNCTION_PARSERS = {
158            **Hive.Parser.FUNCTION_PARSERS,
159            "BROADCAST": lambda self: self._parse_join_hint("BROADCAST"),
160            "BROADCASTJOIN": lambda self: self._parse_join_hint("BROADCASTJOIN"),
161            "MAPJOIN": lambda self: self._parse_join_hint("MAPJOIN"),
162            "MERGE": lambda self: self._parse_join_hint("MERGE"),
163            "SHUFFLEMERGE": lambda self: self._parse_join_hint("SHUFFLEMERGE"),
164            "MERGEJOIN": lambda self: self._parse_join_hint("MERGEJOIN"),
165            "SHUFFLE_HASH": lambda self: self._parse_join_hint("SHUFFLE_HASH"),
166            "SHUFFLE_REPLICATE_NL": lambda self: self._parse_join_hint("SHUFFLE_REPLICATE_NL"),
167        }
168
169        def _parse_add_column(self) -> t.Optional[exp.Expression]:
170            return self._match_text_seq("ADD", "COLUMNS") and self._parse_schema()
171
172        def _parse_drop_column(self) -> t.Optional[exp.Drop | exp.Command]:
173            return self._match_text_seq("DROP", "COLUMNS") and self.expression(
174                exp.Drop, this=self._parse_schema(), kind="COLUMNS"
175            )
176
177        def _pivot_column_names(self, aggregations: t.List[exp.Expression]) -> t.List[str]:
178            if len(aggregations) == 1:
179                return [""]
180            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_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'>>, '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'>>, 'FIRST': <bound method Func.from_arg_list of <class 'sqlglot.expressions.First'>>, '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'>>, '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': <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'>>, '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_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_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'>>, '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': <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'>>, '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>, 'AGGREGATE': <bound method Func.from_arg_list of <class 'sqlglot.expressions.Reduce'>>, 'APPROX_PERCENTILE': <bound method Func.from_arg_list of <class 'sqlglot.expressions.ApproxQuantile'>>, 'BOOLEAN': <function _parse_as_cast.<locals>.<lambda>>, 'DOUBLE': <function _parse_as_cast.<locals>.<lambda>>, 'FLOAT': <function _parse_as_cast.<locals>.<lambda>>, 'IIF': <bound method Func.from_arg_list of <class 'sqlglot.expressions.If'>>, 'INT': <function _parse_as_cast.<locals>.<lambda>>, 'MAP_FROM_ARRAYS': <bound method Func.from_arg_list of <class 'sqlglot.expressions.Map'>>, 'RLIKE': <bound method Func.from_arg_list of <class 'sqlglot.expressions.RegexpLike'>>, 'SHIFTLEFT': <function binary_from_function.<locals>.<lambda>>, 'SHIFTRIGHT': <function binary_from_function.<locals>.<lambda>>, 'STRING': <function _parse_as_cast.<locals>.<lambda>>, 'TIMESTAMP': <function _parse_as_cast.<locals>.<lambda>>, 'TO_TIMESTAMP': <function Spark2.Parser.<lambda>>, 'TO_UNIX_TIMESTAMP': <bound method Func.from_arg_list of <class 'sqlglot.expressions.StrToUnix'>>, 'TRUNC': <function Spark2.Parser.<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>>}

TOKENIZER_CLASS: Type[sqlglot.tokens.Tokenizer] = <class 'sqlglot.dialects.spark2.Spark2.Tokenizer'>

SHOW_TRIE: Dict = {}

SET_TRIE: Dict = {'GLOBAL': {0: True}, 'LOCAL': {0: True}, 'SESSION': {0: True}, 'TRANSACTION': {0: True}}

Inherited Members

sqlglot.parser.Parser: Parser; NO_PAREN_FUNCTIONS; STRUCT_TYPE_TOKENS; NESTED_TYPE_TOKENS; ENUM_TYPE_TOKENS; TYPE_TOKENS; 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; QUERY_MODIFIER_PARSERS; SET_PARSERS; SHOW_PARSERS; TYPE_LITERAL_PARSERS; MODIFIABLES; DDL_SELECT_TOKENS; PRE_VOLATILE_TOKENS; TRANSACTION_KIND; TRANSACTION_CHARACTERISTICS; INSERT_ALTERNATIVES; CLONE_KINDS; TABLE_INDEX_HINT_TOKENS; WINDOW_ALIAS_TOKENS; WINDOW_BEFORE_PAREN_TOKENS; WINDOW_SIDES; ADD_CONSTRAINT_TOKENS; DISTINCT_TOKENS; CONCAT_NULL_OUTPUTS_STRING; PREFIXED_PIVOT_COLUMNS; IDENTIFY_PIVOT_STRINGS; LOG_BASE_FIRST; ALTER_TABLE_ADD_COLUMN_KEYWORD; 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
sqlglot.dialects.hive.Hive.Parser: LOG_DEFAULTS_TO_LN; STRICT_CAST; SUPPORTS_USER_DEFINED_TYPES; NO_PAREN_FUNCTION_PARSERS; PROPERTY_PARSERS; ALIAS_POST_TABLESAMPLE; TIME_MAPPING

class Spark2.Generator(sqlglot.dialects.hive.Hive.Generator): View Source

182    class Generator(Hive.Generator):
183        QUERY_HINTS = True
184        NVL2_SUPPORTED = True
185
186        PROPERTIES_LOCATION = {
187            **Hive.Generator.PROPERTIES_LOCATION,
188            exp.EngineProperty: exp.Properties.Location.UNSUPPORTED,
189            exp.AutoIncrementProperty: exp.Properties.Location.UNSUPPORTED,
190            exp.CharacterSetProperty: exp.Properties.Location.UNSUPPORTED,
191            exp.CollateProperty: exp.Properties.Location.UNSUPPORTED,
192        }
193
194        TRANSFORMS = {
195            **Hive.Generator.TRANSFORMS,
196            exp.ApproxDistinct: rename_func("APPROX_COUNT_DISTINCT"),
197            exp.ArraySum: lambda self, e: f"AGGREGATE({self.sql(e, 'this')}, 0, (acc, x) -> acc + x, acc -> acc)",
198            exp.AtTimeZone: lambda self, e: f"FROM_UTC_TIMESTAMP({self.sql(e, 'this')}, {self.sql(e, 'zone')})",
199            exp.BitwiseLeftShift: rename_func("SHIFTLEFT"),
200            exp.BitwiseRightShift: rename_func("SHIFTRIGHT"),
201            exp.Create: _create_sql,
202            exp.DateFromParts: rename_func("MAKE_DATE"),
203            exp.DateTrunc: lambda self, e: self.func("TRUNC", e.this, e.args.get("unit")),
204            exp.DayOfMonth: rename_func("DAYOFMONTH"),
205            exp.DayOfWeek: rename_func("DAYOFWEEK"),
206            exp.DayOfYear: rename_func("DAYOFYEAR"),
207            exp.FileFormatProperty: lambda self, e: f"USING {e.name.upper()}",
208            exp.From: transforms.preprocess([_unalias_pivot]),
209            exp.Insert: _insert_sql,
210            exp.LogicalAnd: rename_func("BOOL_AND"),
211            exp.LogicalOr: rename_func("BOOL_OR"),
212            exp.Map: _map_sql,
213            exp.Pivot: transforms.preprocess([_unqualify_pivot_columns]),
214            exp.Reduce: rename_func("AGGREGATE"),
215            exp.RegexpReplace: lambda self, e: self.func(
216                "REGEXP_REPLACE",
217                e.this,
218                e.expression,
219                e.args["replacement"],
220                e.args.get("position"),
221            ),
222            exp.StrToDate: _str_to_date,
223            exp.StrToTime: lambda self, e: f"TO_TIMESTAMP({self.sql(e, 'this')}, {self.format_time(e)})",
224            exp.TimestampTrunc: lambda self, e: self.func(
225                "DATE_TRUNC", exp.Literal.string(e.text("unit")), e.this
226            ),
227            exp.Trim: trim_sql,
228            exp.UnixToTime: _unix_to_time_sql,
229            exp.VariancePop: rename_func("VAR_POP"),
230            exp.WeekOfYear: rename_func("WEEKOFYEAR"),
231            exp.WithinGroup: transforms.preprocess(
232                [transforms.remove_within_group_for_percentiles]
233            ),
234        }
235        TRANSFORMS.pop(exp.ArrayJoin)
236        TRANSFORMS.pop(exp.ArraySort)
237        TRANSFORMS.pop(exp.ILike)
238        TRANSFORMS.pop(exp.Left)
239        TRANSFORMS.pop(exp.MonthsBetween)
240        TRANSFORMS.pop(exp.Right)
241
242        WRAP_DERIVED_VALUES = False
243        CREATE_FUNCTION_RETURN_AS = False
244
245        def cast_sql(self, expression: exp.Cast, safe_prefix: t.Optional[str] = None) -> str:
246            if isinstance(expression.this, exp.Cast) and expression.this.is_type("json"):
247                schema = f"'{self.sql(expression, 'to')}'"
248                return self.func("FROM_JSON", expression.this.this, schema)
249            if expression.is_type("json"):
250                return self.func("TO_JSON", expression.this)
251
252            return super(Hive.Generator, self).cast_sql(expression, safe_prefix=safe_prefix)
253
254        def columndef_sql(self, expression: exp.ColumnDef, sep: str = " ") -> str:
255            return super().columndef_sql(
256                expression,
257                sep=": "
258                if isinstance(expression.parent, exp.DataType)
259                and expression.parent.is_type("struct")
260                else sep,
261            )

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

NVL2_SUPPORTED = True

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.HeapProperty'>: <Location.POST_WITH: 'POST_WITH'>, <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.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.ClusteredColumnConstraint'>: <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.HeapProperty'>: <function Generator.<lambda>>, <class 'sqlglot.expressions.InlineLengthColumnConstraint'>: <function Generator.<lambda>>, <class 'sqlglot.expressions.IntervalDayToSecondSpan'>: 'DAY TO SECOND', <class 'sqlglot.expressions.IntervalYearToMonthSpan'>: 'YEAR TO MONTH', <class 'sqlglot.expressions.LanguageProperty'>: <function Generator.<lambda>>, <class 'sqlglot.expressions.LocationProperty'>: <function Generator.<lambda>>, <class 'sqlglot.expressions.LogProperty'>: <function Generator.<lambda>>, <class 'sqlglot.expressions.MaterializedProperty'>: <function Generator.<lambda>>, <class 'sqlglot.expressions.NoPrimaryIndexProperty'>: <function Generator.<lambda>>, <class 'sqlglot.expressions.NonClusteredColumnConstraint'>: <function Generator.<lambda>>, <class 'sqlglot.expressions.NotForReplicationColumnConstraint'>: <function Generator.<lambda>>, <class 'sqlglot.expressions.OnCommitProperty'>: <function Generator.<lambda>>, <class 'sqlglot.expressions.OnProperty'>: <function Generator.<lambda>>, <class 'sqlglot.expressions.OnUpdateColumnConstraint'>: <function Generator.<lambda>>, <class 'sqlglot.expressions.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.AnyValue'>: <function rename_func.<locals>.<lambda>>, <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.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.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.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 Spark2.Generator.<lambda>>, <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.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.Insert'>: <function _insert_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: View Source

245        def cast_sql(self, expression: exp.Cast, safe_prefix: t.Optional[str] = None) -> str:
246            if isinstance(expression.this, exp.Cast) and expression.this.is_type("json"):
247                schema = f"'{self.sql(expression, 'to')}'"
248                return self.func("FROM_JSON", expression.this.this, schema)
249            if expression.is_type("json"):
250                return self.func("TO_JSON", expression.this)
251
252            return super(Hive.Generator, self).cast_sql(expression, safe_prefix=safe_prefix)

def columndef_sql(self, expression: sqlglot.expressions.ColumnDef, sep: str = ' ') -> str: View Source

254        def columndef_sql(self, expression: exp.ColumnDef, sep: str = " ") -> str:
255            return super().columndef_sql(
256                expression,
257                sep=": "
258                if isinstance(expression.parent, exp.DataType)
259                and expression.parent.is_type("struct")
260                else sep,
261            )

SELECT_KINDS: Tuple[str, ...] = ()

@classmethod

def can_identify(text: str, identify: str | bool = 'safe') -> bool: View Source

256    @classmethod
257    def can_identify(cls, text: str, identify: str | bool = "safe") -> bool:
258        """Checks if text can be identified given an identify option.
259
260        Args:
261            text: The text to check.
262            identify:
263                "always" or `True`: Always returns true.
264                "safe": True if the identifier is case-insensitive.
265
266        Returns:
267            Whether or not the given text can be identified.
268        """
269        if identify is True or identify == "always":
270            return True
271
272        if identify == "safe":
273            return not cls.case_sensitive(text)
274
275        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 'sqlglot.dialects.spark2.Spark2.Tokenizer'>

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; TZ_TO_WITH_TIME_ZONE; VALUES_AS_TABLE; ALTER_TABLE_ADD_COLUMN_KEYWORD; 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; 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_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; 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_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; querytransform_sql; indexconstraintoption_sql; indexcolumnconstraint_sql; nvl2_sql; comprehension_sql
sqlglot.dialects.hive.Hive.Generator: LIMIT_FETCH; TABLESAMPLE_WITH_METHOD; TABLESAMPLE_SIZE_IS_PERCENT; JOIN_HINTS; TABLE_HINTS; INDEX_ON; EXTRACT_ALLOWS_QUOTES; TYPE_MAPPING; rowformatserdeproperty_sql; arrayagg_sql; with_properties; datatype_sql; version_sql; ALIAS_POST_TABLESAMPLE; IDENTIFIERS_CAN_START_WITH_DIGIT

class Spark2.Tokenizer(sqlglot.dialects.hive.Hive.Tokenizer): View Source

263    class Tokenizer(Hive.Tokenizer):
264        HEX_STRINGS = [("X'", "'")]

HEX_STRINGS = [("X'", "'")]

Inherited Members

sqlglot.tokens.Tokenizer: SINGLE_TOKENS; BIT_STRINGS; BYTE_STRINGS; RAW_STRINGS; IDENTIFIER_ESCAPES; VAR_SINGLE_TOKENS; WHITE_SPACE; COMMANDS; COMMAND_PREFIX_TOKENS; COMMENTS; reset; tokenize; peek; size; sql; tokens
sqlglot.dialects.hive.Hive.Tokenizer: QUOTES; IDENTIFIERS; STRING_ESCAPES; ENCODE; KEYWORDS; NUMERIC_LITERALS; IDENTIFIERS_CAN_START_WITH_DIGIT