Edit on GitHub

sqlglot.dialects.fabric

View Source

  1from __future__ import annotations
  2
  3
  4from sqlglot import exp, transforms
  5from sqlglot.dialects.dialect import NormalizationStrategy
  6from sqlglot.dialects.tsql import TSQL
  7from sqlglot.tokens import TokenType
  8
  9
 10def _cap_data_type_precision(expression: exp.DataType, max_precision: int = 6) -> exp.DataType:
 11    """
 12    Cap the precision of to a maximum of `max_precision` digits.
 13    If no precision is specified, default to `max_precision`.
 14    """
 15
 16    precision_param = expression.find(exp.DataTypeParam)
 17
 18    if precision_param and precision_param.this.is_int:
 19        current_precision = precision_param.this.to_py()
 20        target_precision = min(current_precision, max_precision)
 21    else:
 22        target_precision = max_precision
 23
 24    return exp.DataType(
 25        this=expression.this,
 26        expressions=[exp.DataTypeParam(this=exp.Literal.number(target_precision))],
 27    )
 28
 29
 30def _add_default_precision_to_varchar(expression: exp.Expression) -> exp.Expression:
 31    """Transform function to add VARCHAR(MAX) or CHAR(MAX) for cross-dialect conversion."""
 32    if (
 33        isinstance(expression, exp.Create)
 34        and expression.kind == "TABLE"
 35        and isinstance(expression.this, exp.Schema)
 36    ):
 37        for column in expression.this.expressions:
 38            if isinstance(column, exp.ColumnDef):
 39                column_type = column.kind
 40                if (
 41                    isinstance(column_type, exp.DataType)
 42                    and column_type.this in (exp.DataType.Type.VARCHAR, exp.DataType.Type.CHAR)
 43                    and not column_type.expressions
 44                ):
 45                    # For transpilation, VARCHAR/CHAR without precision becomes VARCHAR(MAX)/CHAR(MAX)
 46                    column_type.set("expressions", [exp.var("MAX")])
 47
 48    return expression
 49
 50
 51class Fabric(TSQL):
 52    """
 53    Microsoft Fabric Data Warehouse dialect that inherits from T-SQL.
 54
 55    Microsoft Fabric is a cloud-based analytics platform that provides a unified
 56    data warehouse experience. While it shares much of T-SQL's syntax, it has
 57    specific differences and limitations that this dialect addresses.
 58
 59    Key differences from T-SQL:
 60    - Case-sensitive identifiers (unlike T-SQL which is case-insensitive)
 61    - Limited data type support with mappings to supported alternatives
 62    - Temporal types (DATETIME2, DATETIMEOFFSET, TIME) limited to 6 digits precision
 63    - Certain legacy types (MONEY, SMALLMONEY, etc.) are not supported
 64    - Unicode types (NCHAR, NVARCHAR) are mapped to non-unicode equivalents
 65
 66    References:
 67    - Data Types: https://learn.microsoft.com/en-us/fabric/data-warehouse/data-types
 68    - T-SQL Surface Area: https://learn.microsoft.com/en-us/fabric/data-warehouse/tsql-surface-area
 69    """
 70
 71    # Fabric is case-sensitive unlike T-SQL which is case-insensitive
 72    NORMALIZATION_STRATEGY = NormalizationStrategy.CASE_SENSITIVE
 73
 74    class Tokenizer(TSQL.Tokenizer):
 75        # Override T-SQL tokenizer to handle TIMESTAMP differently
 76        # In T-SQL, TIMESTAMP is a synonym for ROWVERSION, but in Fabric we want it to be a datetime type
 77        # Also add UTINYINT keyword mapping since T-SQL doesn't have it
 78        KEYWORDS = {
 79            **TSQL.Tokenizer.KEYWORDS,
 80            "TIMESTAMP": TokenType.TIMESTAMP,
 81            "UTINYINT": TokenType.UTINYINT,
 82        }
 83
 84    class Parser(TSQL.Parser):
 85        def _parse_create(self) -> exp.Create | exp.Command:
 86            create = super()._parse_create()
 87
 88            if isinstance(create, exp.Create):
 89                # Transform VARCHAR/CHAR without precision to VARCHAR(1)/CHAR(1)
 90                if create.kind == "TABLE" and isinstance(create.this, exp.Schema):
 91                    for column in create.this.expressions:
 92                        if isinstance(column, exp.ColumnDef):
 93                            column_type = column.kind
 94                            if (
 95                                isinstance(column_type, exp.DataType)
 96                                and column_type.this
 97                                in (exp.DataType.Type.VARCHAR, exp.DataType.Type.CHAR)
 98                                and not column_type.expressions
 99                            ):
100                                # Add default precision of 1 to VARCHAR/CHAR without precision
101                                # When n isn't specified in a data definition or variable declaration statement, the default length is 1.
102                                # https://learn.microsoft.com/en-us/sql/t-sql/data-types/char-and-varchar-transact-sql?view=sql-server-ver17#remarks
103                                column_type.set("expressions", [exp.Literal.number("1")])
104
105            return create
106
107    class Generator(TSQL.Generator):
108        # Fabric-specific type mappings - override T-SQL types that aren't supported
109        # Reference: https://learn.microsoft.com/en-us/fabric/data-warehouse/data-types
110        TYPE_MAPPING = {
111            **TSQL.Generator.TYPE_MAPPING,
112            exp.DataType.Type.DATETIME: "DATETIME2",
113            exp.DataType.Type.DECIMAL: "DECIMAL",
114            exp.DataType.Type.IMAGE: "VARBINARY",
115            exp.DataType.Type.INT: "INT",
116            exp.DataType.Type.JSON: "VARCHAR",
117            exp.DataType.Type.MONEY: "DECIMAL",
118            exp.DataType.Type.NCHAR: "CHAR",
119            exp.DataType.Type.NVARCHAR: "VARCHAR",
120            exp.DataType.Type.ROWVERSION: "ROWVERSION",
121            exp.DataType.Type.SMALLDATETIME: "DATETIME2",
122            exp.DataType.Type.SMALLMONEY: "DECIMAL",
123            exp.DataType.Type.TIMESTAMP: "DATETIME2",
124            exp.DataType.Type.TIMESTAMPNTZ: "DATETIME2",
125            exp.DataType.Type.TIMESTAMPTZ: "DATETIME2",
126            exp.DataType.Type.TINYINT: "SMALLINT",
127            exp.DataType.Type.UTINYINT: "SMALLINT",
128            exp.DataType.Type.UUID: "UNIQUEIDENTIFIER",
129            exp.DataType.Type.XML: "VARCHAR",
130        }
131
132        TRANSFORMS = {
133            **TSQL.Generator.TRANSFORMS,
134            exp.Create: transforms.preprocess([_add_default_precision_to_varchar]),
135        }
136
137        def datatype_sql(self, expression: exp.DataType) -> str:
138            # Check if this is a temporal type that needs precision handling. Fabric limits temporal
139            # types to max 6 digits precision. When no precision is specified, we default to 6 digits.
140            if (
141                expression.is_type(*exp.DataType.TEMPORAL_TYPES)
142                and expression.this != exp.DataType.Type.DATE
143            ):
144                # Create a new expression with the capped precision
145                expression = _cap_data_type_precision(expression)
146
147            return super().datatype_sql(expression)
148
149        def cast_sql(self, expression: exp.Cast, safe_prefix: str | None = None) -> str:
150            # Cast to DATETIMEOFFSET if inside an AT TIME ZONE expression
151            # https://learn.microsoft.com/en-us/sql/t-sql/data-types/datetimeoffset-transact-sql#microsoft-fabric-support
152            if expression.is_type(exp.DataType.Type.TIMESTAMPTZ):
153                at_time_zone = expression.find_ancestor(exp.AtTimeZone, exp.Select)
154
155                # Return normal cast, if the expression is not in an AT TIME ZONE context
156                if not isinstance(at_time_zone, exp.AtTimeZone):
157                    return super().cast_sql(expression, safe_prefix)
158
159                # Get the precision from the original TIMESTAMPTZ cast and cap it to 6
160                capped_data_type = _cap_data_type_precision(expression.to, max_precision=6)
161                precision = capped_data_type.find(exp.DataTypeParam)
162                precision_value = (
163                    precision.this.to_py() if precision and precision.this.is_int else 6
164                )
165
166                # Do the cast explicitly to bypass sqlglot's default handling
167                datetimeoffset = f"CAST({expression.this} AS DATETIMEOFFSET({precision_value}))"
168
169                return self.sql(datetimeoffset)
170
171            return super().cast_sql(expression, safe_prefix)
172
173        def attimezone_sql(self, expression: exp.AtTimeZone) -> str:
174            # Wrap the AT TIME ZONE expression in a cast to DATETIME2 if it contains a TIMESTAMPTZ
175            ## https://learn.microsoft.com/en-us/sql/t-sql/data-types/datetimeoffset-transact-sql#microsoft-fabric-support
176            timestamptz_cast = expression.find(exp.Cast)
177            if timestamptz_cast and timestamptz_cast.to.is_type(exp.DataType.Type.TIMESTAMPTZ):
178                # Get the precision from the original TIMESTAMPTZ cast and cap it to 6
179                data_type = timestamptz_cast.to
180                capped_data_type = _cap_data_type_precision(data_type, max_precision=6)
181                precision_param = capped_data_type.find(exp.DataTypeParam)
182                precision = precision_param.this.to_py() if precision_param else 6
183
184                # Generate the AT TIME ZONE expression (which will handle the inner cast conversion)
185                at_time_zone_sql = super().attimezone_sql(expression)
186
187                # Wrap it in an outer cast to DATETIME2
188                return f"CAST({at_time_zone_sql} AS DATETIME2({precision}))"
189
190            return super().attimezone_sql(expression)
191
192        def unixtotime_sql(self, expression: exp.UnixToTime) -> str:
193            scale = expression.args.get("scale")
194            timestamp = expression.this
195
196            if scale not in (None, exp.UnixToTime.SECONDS):
197                self.unsupported(f"UnixToTime scale {scale} is not supported by Fabric")
198                return ""
199
200            # Convert unix timestamp (seconds) to microseconds and round to avoid decimals
201            microseconds = timestamp * exp.Literal.number("1e6")
202            rounded = exp.func("round", microseconds, 0)
203            rounded_ms_as_bigint = exp.cast(rounded, exp.DataType.Type.BIGINT)
204
205            # Create the base datetime as '1970-01-01' cast to DATETIME2(6)
206            epoch_start = exp.cast("'1970-01-01'", "datetime2(6)", dialect="fabric")
207
208            dateadd = exp.DateAdd(
209                this=epoch_start,
210                expression=rounded_ms_as_bigint,
211                unit=exp.Literal.string("MICROSECONDS"),
212            )
213            return self.sql(dateadd)

class Fabric(sqlglot.dialects.tsql.TSQL): View Source

 52class Fabric(TSQL):
 53    """
 54    Microsoft Fabric Data Warehouse dialect that inherits from T-SQL.
 55
 56    Microsoft Fabric is a cloud-based analytics platform that provides a unified
 57    data warehouse experience. While it shares much of T-SQL's syntax, it has
 58    specific differences and limitations that this dialect addresses.
 59
 60    Key differences from T-SQL:
 61    - Case-sensitive identifiers (unlike T-SQL which is case-insensitive)
 62    - Limited data type support with mappings to supported alternatives
 63    - Temporal types (DATETIME2, DATETIMEOFFSET, TIME) limited to 6 digits precision
 64    - Certain legacy types (MONEY, SMALLMONEY, etc.) are not supported
 65    - Unicode types (NCHAR, NVARCHAR) are mapped to non-unicode equivalents
 66
 67    References:
 68    - Data Types: https://learn.microsoft.com/en-us/fabric/data-warehouse/data-types
 69    - T-SQL Surface Area: https://learn.microsoft.com/en-us/fabric/data-warehouse/tsql-surface-area
 70    """
 71
 72    # Fabric is case-sensitive unlike T-SQL which is case-insensitive
 73    NORMALIZATION_STRATEGY = NormalizationStrategy.CASE_SENSITIVE
 74
 75    class Tokenizer(TSQL.Tokenizer):
 76        # Override T-SQL tokenizer to handle TIMESTAMP differently
 77        # In T-SQL, TIMESTAMP is a synonym for ROWVERSION, but in Fabric we want it to be a datetime type
 78        # Also add UTINYINT keyword mapping since T-SQL doesn't have it
 79        KEYWORDS = {
 80            **TSQL.Tokenizer.KEYWORDS,
 81            "TIMESTAMP": TokenType.TIMESTAMP,
 82            "UTINYINT": TokenType.UTINYINT,
 83        }
 84
 85    class Parser(TSQL.Parser):
 86        def _parse_create(self) -> exp.Create | exp.Command:
 87            create = super()._parse_create()
 88
 89            if isinstance(create, exp.Create):
 90                # Transform VARCHAR/CHAR without precision to VARCHAR(1)/CHAR(1)
 91                if create.kind == "TABLE" and isinstance(create.this, exp.Schema):
 92                    for column in create.this.expressions:
 93                        if isinstance(column, exp.ColumnDef):
 94                            column_type = column.kind
 95                            if (
 96                                isinstance(column_type, exp.DataType)
 97                                and column_type.this
 98                                in (exp.DataType.Type.VARCHAR, exp.DataType.Type.CHAR)
 99                                and not column_type.expressions
100                            ):
101                                # Add default precision of 1 to VARCHAR/CHAR without precision
102                                # When n isn't specified in a data definition or variable declaration statement, the default length is 1.
103                                # https://learn.microsoft.com/en-us/sql/t-sql/data-types/char-and-varchar-transact-sql?view=sql-server-ver17#remarks
104                                column_type.set("expressions", [exp.Literal.number("1")])
105
106            return create
107
108    class Generator(TSQL.Generator):
109        # Fabric-specific type mappings - override T-SQL types that aren't supported
110        # Reference: https://learn.microsoft.com/en-us/fabric/data-warehouse/data-types
111        TYPE_MAPPING = {
112            **TSQL.Generator.TYPE_MAPPING,
113            exp.DataType.Type.DATETIME: "DATETIME2",
114            exp.DataType.Type.DECIMAL: "DECIMAL",
115            exp.DataType.Type.IMAGE: "VARBINARY",
116            exp.DataType.Type.INT: "INT",
117            exp.DataType.Type.JSON: "VARCHAR",
118            exp.DataType.Type.MONEY: "DECIMAL",
119            exp.DataType.Type.NCHAR: "CHAR",
120            exp.DataType.Type.NVARCHAR: "VARCHAR",
121            exp.DataType.Type.ROWVERSION: "ROWVERSION",
122            exp.DataType.Type.SMALLDATETIME: "DATETIME2",
123            exp.DataType.Type.SMALLMONEY: "DECIMAL",
124            exp.DataType.Type.TIMESTAMP: "DATETIME2",
125            exp.DataType.Type.TIMESTAMPNTZ: "DATETIME2",
126            exp.DataType.Type.TIMESTAMPTZ: "DATETIME2",
127            exp.DataType.Type.TINYINT: "SMALLINT",
128            exp.DataType.Type.UTINYINT: "SMALLINT",
129            exp.DataType.Type.UUID: "UNIQUEIDENTIFIER",
130            exp.DataType.Type.XML: "VARCHAR",
131        }
132
133        TRANSFORMS = {
134            **TSQL.Generator.TRANSFORMS,
135            exp.Create: transforms.preprocess([_add_default_precision_to_varchar]),
136        }
137
138        def datatype_sql(self, expression: exp.DataType) -> str:
139            # Check if this is a temporal type that needs precision handling. Fabric limits temporal
140            # types to max 6 digits precision. When no precision is specified, we default to 6 digits.
141            if (
142                expression.is_type(*exp.DataType.TEMPORAL_TYPES)
143                and expression.this != exp.DataType.Type.DATE
144            ):
145                # Create a new expression with the capped precision
146                expression = _cap_data_type_precision(expression)
147
148            return super().datatype_sql(expression)
149
150        def cast_sql(self, expression: exp.Cast, safe_prefix: str | None = None) -> str:
151            # Cast to DATETIMEOFFSET if inside an AT TIME ZONE expression
152            # https://learn.microsoft.com/en-us/sql/t-sql/data-types/datetimeoffset-transact-sql#microsoft-fabric-support
153            if expression.is_type(exp.DataType.Type.TIMESTAMPTZ):
154                at_time_zone = expression.find_ancestor(exp.AtTimeZone, exp.Select)
155
156                # Return normal cast, if the expression is not in an AT TIME ZONE context
157                if not isinstance(at_time_zone, exp.AtTimeZone):
158                    return super().cast_sql(expression, safe_prefix)
159
160                # Get the precision from the original TIMESTAMPTZ cast and cap it to 6
161                capped_data_type = _cap_data_type_precision(expression.to, max_precision=6)
162                precision = capped_data_type.find(exp.DataTypeParam)
163                precision_value = (
164                    precision.this.to_py() if precision and precision.this.is_int else 6
165                )
166
167                # Do the cast explicitly to bypass sqlglot's default handling
168                datetimeoffset = f"CAST({expression.this} AS DATETIMEOFFSET({precision_value}))"
169
170                return self.sql(datetimeoffset)
171
172            return super().cast_sql(expression, safe_prefix)
173
174        def attimezone_sql(self, expression: exp.AtTimeZone) -> str:
175            # Wrap the AT TIME ZONE expression in a cast to DATETIME2 if it contains a TIMESTAMPTZ
176            ## https://learn.microsoft.com/en-us/sql/t-sql/data-types/datetimeoffset-transact-sql#microsoft-fabric-support
177            timestamptz_cast = expression.find(exp.Cast)
178            if timestamptz_cast and timestamptz_cast.to.is_type(exp.DataType.Type.TIMESTAMPTZ):
179                # Get the precision from the original TIMESTAMPTZ cast and cap it to 6
180                data_type = timestamptz_cast.to
181                capped_data_type = _cap_data_type_precision(data_type, max_precision=6)
182                precision_param = capped_data_type.find(exp.DataTypeParam)
183                precision = precision_param.this.to_py() if precision_param else 6
184
185                # Generate the AT TIME ZONE expression (which will handle the inner cast conversion)
186                at_time_zone_sql = super().attimezone_sql(expression)
187
188                # Wrap it in an outer cast to DATETIME2
189                return f"CAST({at_time_zone_sql} AS DATETIME2({precision}))"
190
191            return super().attimezone_sql(expression)
192
193        def unixtotime_sql(self, expression: exp.UnixToTime) -> str:
194            scale = expression.args.get("scale")
195            timestamp = expression.this
196
197            if scale not in (None, exp.UnixToTime.SECONDS):
198                self.unsupported(f"UnixToTime scale {scale} is not supported by Fabric")
199                return ""
200
201            # Convert unix timestamp (seconds) to microseconds and round to avoid decimals
202            microseconds = timestamp * exp.Literal.number("1e6")
203            rounded = exp.func("round", microseconds, 0)
204            rounded_ms_as_bigint = exp.cast(rounded, exp.DataType.Type.BIGINT)
205
206            # Create the base datetime as '1970-01-01' cast to DATETIME2(6)
207            epoch_start = exp.cast("'1970-01-01'", "datetime2(6)", dialect="fabric")
208
209            dateadd = exp.DateAdd(
210                this=epoch_start,
211                expression=rounded_ms_as_bigint,
212                unit=exp.Literal.string("MICROSECONDS"),
213            )
214            return self.sql(dateadd)

Microsoft Fabric Data Warehouse dialect that inherits from T-SQL.

Microsoft Fabric is a cloud-based analytics platform that provides a unified data warehouse experience. While it shares much of T-SQL's syntax, it has specific differences and limitations that this dialect addresses.

Key differences from T-SQL:

Case-sensitive identifiers (unlike T-SQL which is case-insensitive)
Limited data type support with mappings to supported alternatives
Temporal types (DATETIME2, DATETIMEOFFSET, TIME) limited to 6 digits precision
Certain legacy types (MONEY, SMALLMONEY, etc.) are not supported
Unicode types (NCHAR, NVARCHAR) are mapped to non-unicode equivalents

References:

Data Types: https://learn.microsoft.com/en-us/fabric/data-warehouse/data-types
T-SQL Surface Area: https://learn.microsoft.com/en-us/fabric/data-warehouse/tsql-surface-area

NORMALIZATION_STRATEGY = <NormalizationStrategy.CASE_SENSITIVE: 'CASE_SENSITIVE'>

Specifies the strategy according to which identifiers should be normalized.

SUPPORTS_COLUMN_JOIN_MARKS = False

Whether the old-style outer join (+) syntax is supported.

INITCAP_SUPPORTS_CUSTOM_DELIMITERS = False

tokenizer_class = <class 'Fabric.Tokenizer'>

jsonpath_tokenizer_class = <class 'sqlglot.dialects.dialect.JSONPathTokenizer'>

parser_class = <class 'Fabric.Parser'>

generator_class = <class 'Fabric.Generator'>

TIME_TRIE: Dict = {'y': {'e': {'a': {'r': {0: True}}}, 0: True, 'y': {'y': {'y': {0: True}}, 0: True}}, 'd': {'a': {'y': {'o': {'f': {'y': {'e': {'a': {'r': {0: True}}}}}}, 0: True}}, 'y': {0: True}, 'w': {0: True}, 'd': {'d': {'d': {0: True}}, 0: True}, 0: True}, 'w': {'e': {'e': {'k': {0: True, 'd': {'a': {'y': {0: True}}}}}}, 'w': {0: True}, 'k': {0: True}}, 'i': {'s': {'o': {'w': {'k': {0: True}, 'w': {0: True}}, '_': {'w': {'e': {'e': {'k': {0: True}}}}}}}}, 'h': {'o': {'u': {'r': {0: True}}}, 'h': {0: True}, 0: True}, 'm': {'i': {'n': {'u': {'t': {'e': {0: True}}}}, 0: True, 'l': {'l': {'i': {'s': {'e': {'c': {'o': {'n': {'d': {0: True}}}}}}}}}}, 's': {0: True}, 'o': {'n': {'t': {'h': {0: True}}}}, 'm': {0: True}, 0: True}, 'n': {0: True}, 's': {'e': {'c': {'o': {'n': {'d': {0: True}}}}}, 's': {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}, 'H': {'H': {0: True}, 0: True}, 'f': {'f': {'f': {'f': {'f': {'f': {0: True}}}}}}}

FORMAT_TRIE: Dict = {'y': {'e': {'a': {'r': {0: True}}}, 0: True, 'y': {'y': {'y': {0: True}}, 0: True}}, 'd': {'a': {'y': {'o': {'f': {'y': {'e': {'a': {'r': {0: True}}}}}}, 0: True}}, 'y': {0: True}, 'w': {0: True}, 'd': {'d': {'d': {0: True}}, 0: True}, 0: True}, 'w': {'e': {'e': {'k': {0: True, 'd': {'a': {'y': {0: True}}}}}}, 'w': {0: True}, 'k': {0: True}}, 'i': {'s': {'o': {'w': {'k': {0: True}, 'w': {0: True}}, '_': {'w': {'e': {'e': {'k': {0: True}}}}}}}}, 'h': {'o': {'u': {'r': {0: True}}}, 'h': {0: True}, 0: True}, 'm': {'i': {'n': {'u': {'t': {'e': {0: True}}}}, 0: True, 'l': {'l': {'i': {'s': {'e': {'c': {'o': {'n': {'d': {0: True}}}}}}}}}}, 's': {0: True}, 'o': {'n': {'t': {'h': {0: True}}}}, 'm': {0: True}, 0: True}, 'n': {0: True}, 's': {'e': {'c': {'o': {'n': {'d': {0: True}}}}}, 's': {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}, 'H': {'H': {0: True}, 0: True}, 'f': {'f': {'f': {'f': {'f': {'f': {0: True}}}}}}}

INVERSE_TIME_MAPPING: Dict[str, str] = {'%Y': 'yyyy', '%j': 'dayofyear', '%d': 'dd', '%W': 'wk', '%V': 'iso_week', '%h': 'hour', '%I': 'hh', '%M': 'mm', '%S': 'ss', '%-S': 's', '%f': 'ffffff', '%w': 'dw', '%m': 'MM', '%-M': 'm', '%y': 'yy', '%B': 'MMMM', '%b': 'MMM', '%-m': 'M', '%A': 'dddd', '%-d': 'd', '%H': 'HH', '%-H': 'H', '%-I': 'h'}

INVERSE_TIME_TRIE: Dict = {'%': {'Y': {0: True}, 'j': {0: True}, 'd': {0: True}, 'W': {0: True}, 'V': {0: True}, 'h': {0: True}, 'I': {0: True}, 'M': {0: True}, 'S': {0: True}, '-': {'S': {0: True}, 'M': {0: True}, 'm': {0: True}, 'd': {0: True}, 'H': {0: True}, 'I': {0: True}}, 'f': {0: True}, 'w': {0: True}, 'm': {0: True}, 'y': {0: True}, 'B': {0: True}, 'b': {0: True}, 'A': {0: True}, 'H': {0: True}}}

INVERSE_FORMAT_MAPPING: Dict[str, str] = {}

INVERSE_FORMAT_TRIE: Dict = {}

INVERSE_CREATABLE_KIND_MAPPING: dict[str, str] = {}

ESCAPED_SEQUENCES: Dict[str, str] = {}

QUOTE_START = "'"

QUOTE_END = "'"

IDENTIFIER_START = '['

IDENTIFIER_END = ']'

VALID_INTERVAL_UNITS: Set[str] = {'WOY', 'DECADE', 'YR', 'QUARTERS', 'MON', 'DAYOFWEEK', 'TZH', 'MS', 'MILLENNIUM', 'TZ', 'NANOSECOND', 'MONTHS', 'EPOCH_MICROSECONDS', 'WEEKOFYEAR_ISO', 'ISOWW', 'C', 'SECONDS', 'WEEKOFYEAR', 'NS', 'EPOCH_MILLISECOND', 'TZOFFSET', 'Y', 'WK', 'WY', 'USECOND', 'EPOCH_MICROSECOND', 'MILLISEC', 'MONTH', 'YEAR', 'EPOCH_NANOSECONDS', 'MILS', 'DOY', 'NANOSEC', 'N', 'DOW_ISO', 'CENT', 'DAYS', 'MINS', 'SEC', 'MILLISECS', 'MI', 'Q', 'MICROSECS', 'DAYOFMONTH', 'WEEKOFYEARISO', 'DECADES', 'SS', 'S', 'YRS', 'YYYY', 'WEEK', 'MINUTES', 'WEEKDAY', 'EPOCH', 'HOUR', 'YEARS', 'NSEC', 'MSECOND', 'USEC', 'DAYOFWEEK_ISO', 'CENTURIES', 'QTRS', 'MM', 'D', 'MILLISECONDS', 'DAYOFYEAR', 'WEEKISO', 'NANOSECS', 'CENTS', 'WEEKDAY_ISO', 'DAYOFWEEKISO', 'EPOCH_NANOSECOND', 'MICROSECONDS', 'DAY OF YEAR', 'EPOCH_MILLISECONDS', 'QQ', 'USECONDS', 'MCS', 'TIMEZONE_MINUTE', 'YY', 'HR', 'EPOCH_SECONDS', 'MILLENIA', 'ISOWK', 'W', 'MICROSECOND', 'DECS', 'MINUTE', 'CENTURY', 'MIN', 'YYY', 'MSECONDS', 'HH', 'DD', 'QTR', 'WW', 'TIMEZONE_HOUR', 'DAY OF WEEK', 'DW', 'QUARTER', 'SECS', 'NSECONDS', 'DW_ISO', 'DAY', 'DY', 'SECOND', 'MILLISECON', 'MICROSEC', 'EPOCH_SECOND', 'HRS', 'HOURS', 'ISO_WEEK', 'MIL', 'TZM', 'MSEC', 'H', 'DOW', 'MSECS', 'MILLISECOND', 'US', 'MONS', 'DEC', 'WEEK_ISO', 'NSECOND', 'USECS', 'M'}

BIT_START: Optional[str] = None

BIT_END: Optional[str] = None

HEX_START: Optional[str] = '0x'

HEX_END: Optional[str] = ''

BYTE_START: Optional[str] = None

BYTE_END: Optional[str] = None

UNICODE_START: Optional[str] = None

UNICODE_END: Optional[str] = None

STRINGS_SUPPORT_ESCAPED_SEQUENCES = False

BYTE_STRINGS_SUPPORT_ESCAPED_SEQUENCES = False

Inherited Members

sqlglot.dialects.tsql.TSQL: SUPPORTS_SEMI_ANTI_JOIN; LOG_BASE_FIRST; TYPED_DIVISION; CONCAT_COALESCE; ALTER_TABLE_ADD_REQUIRED_FOR_EACH_COLUMN; TIME_FORMAT; EXPRESSION_METADATA; DATE_PART_MAPPING; TIME_MAPPING; CONVERT_FORMAT_MAPPING; FORMAT_TIME_MAPPING

class Fabric.Tokenizer(sqlglot.dialects.tsql.TSQL.Tokenizer): View Source

75    class Tokenizer(TSQL.Tokenizer):
76        # Override T-SQL tokenizer to handle TIMESTAMP differently
77        # In T-SQL, TIMESTAMP is a synonym for ROWVERSION, but in Fabric we want it to be a datetime type
78        # Also add UTINYINT keyword mapping since T-SQL doesn't have it
79        KEYWORDS = {
80            **TSQL.Tokenizer.KEYWORDS,
81            "TIMESTAMP": TokenType.TIMESTAMP,
82            "UTINYINT": TokenType.UTINYINT,
83        }

KEYWORDS = {'{%': <TokenType.BLOCK_START: 70>, '{%+': <TokenType.BLOCK_START: 70>, '{%-': <TokenType.BLOCK_START: 70>, '%}': <TokenType.BLOCK_END: 71>, '+%}': <TokenType.BLOCK_END: 71>, '-%}': <TokenType.BLOCK_END: 71>, '{{+': <TokenType.BLOCK_START: 70>, '{{-': <TokenType.BLOCK_START: 70>, '+}}': <TokenType.BLOCK_END: 71>, '-}}': <TokenType.BLOCK_END: 71>, '&<': <TokenType.AMP_LT: 60>, '&>': <TokenType.AMP_GT: 61>, '==': <TokenType.EQ: 27>, '::': <TokenType.DCOLON: 13>, '?::': <TokenType.QDCOLON: 360>, '||': <TokenType.DPIPE: 36>, '|>': <TokenType.PIPE_GT: 37>, '>=': <TokenType.GTE: 25>, '<=': <TokenType.LTE: 23>, '<>': <TokenType.NEQ: 28>, '!=': <TokenType.NEQ: 28>, ':=': <TokenType.COLON_EQ: 30>, '<=>': <TokenType.NULLSAFE_EQ: 29>, '->': <TokenType.ARROW: 44>, '->>': <TokenType.DARROW: 45>, '=>': <TokenType.FARROW: 46>, '#>': <TokenType.HASH_ARROW: 48>, '#>>': <TokenType.DHASH_ARROW: 49>, '<->': <TokenType.LR_ARROW: 50>, '&&': <TokenType.DAMP: 59>, '??': <TokenType.DQMARK: 17>, '~~~': <TokenType.GLOB: 282>, '~~': <TokenType.LIKE: 312>, '~~*': <TokenType.ILIKE: 290>, '~*': <TokenType.IRLIKE: 301>, '-|-': <TokenType.ADJACENT: 62>, 'ALL': <TokenType.ALL: 216>, 'AND': <TokenType.AND: 33>, 'ANTI': <TokenType.ANTI: 217>, 'ANY': <TokenType.ANY: 218>, 'ASC': <TokenType.ASC: 221>, 'AS': <TokenType.ALIAS: 214>, 'ASOF': <TokenType.ASOF: 222>, 'AUTOINCREMENT': <TokenType.AUTO_INCREMENT: 224>, 'AUTO_INCREMENT': <TokenType.AUTO_INCREMENT: 224>, 'BEGIN': <TokenType.BEGIN: 225>, 'BETWEEN': <TokenType.BETWEEN: 226>, 'CACHE': <TokenType.CACHE: 228>, 'UNCACHE': <TokenType.UNCACHE: 400>, 'CASE': <TokenType.CASE: 229>, 'CHARACTER SET': <TokenType.CHARACTER_SET: 230>, 'CLUSTER BY': <TokenType.CLUSTER_BY: 231>, 'COLLATE': <TokenType.COLLATE: 232>, 'COLUMN': <TokenType.COLUMN: 78>, 'COMMIT': <TokenType.COMMIT: 235>, 'CONNECT BY': <TokenType.CONNECT_BY: 236>, 'CONSTRAINT': <TokenType.CONSTRAINT: 237>, 'COPY': <TokenType.COPY: 238>, 'CREATE': <TokenType.CREATE: 239>, 'CROSS': <TokenType.CROSS: 240>, 'CUBE': <TokenType.CUBE: 241>, 'CURRENT_DATE': <TokenType.CURRENT_DATE: 242>, 'CURRENT_SCHEMA': <TokenType.CURRENT_SCHEMA: 244>, 'CURRENT_TIME': <TokenType.CURRENT_TIME: 245>, 'CURRENT_TIMESTAMP': <TokenType.CURRENT_TIMESTAMP: 246>, 'CURRENT_USER': <TokenType.CURRENT_USER: 247>, 'CURRENT_CATALOG': <TokenType.CURRENT_CATALOG: 249>, 'DATABASE': <TokenType.DATABASE: 77>, 'DEFAULT': <TokenType.DEFAULT: 251>, 'DELETE': <TokenType.DELETE: 252>, 'DESC': <TokenType.DESC: 253>, 'DESCRIBE': <TokenType.DESCRIBE: 254>, 'DISTINCT': <TokenType.DISTINCT: 257>, 'DISTRIBUTE BY': <TokenType.DISTRIBUTE_BY: 258>, 'DIV': <TokenType.DIV: 259>, 'DROP': <TokenType.DROP: 260>, 'ELSE': <TokenType.ELSE: 261>, 'END': <TokenType.END: 262>, 'ENUM': <TokenType.ENUM: 201>, 'ESCAPE': <TokenType.ESCAPE: 263>, 'EXCEPT': <TokenType.EXCEPT: 264>, 'EXECUTE': <TokenType.EXECUTE: 265>, 'EXISTS': <TokenType.EXISTS: 266>, 'FALSE': <TokenType.FALSE: 267>, 'FETCH': <TokenType.FETCH: 268>, 'FILTER': <TokenType.FILTER: 271>, 'FILE': <TokenType.FILE: 269>, 'FIRST': <TokenType.FIRST: 273>, 'FULL': <TokenType.FULL: 279>, 'FUNCTION': <TokenType.FUNCTION: 280>, 'FOR': <TokenType.FOR: 274>, 'FOREIGN KEY': <TokenType.FOREIGN_KEY: 276>, 'FORMAT': <TokenType.FORMAT: 277>, 'FROM': <TokenType.FROM: 278>, 'GEOGRAPHY': <TokenType.GEOGRAPHY: 168>, 'GEOMETRY': <TokenType.GEOMETRY: 171>, 'GLOB': <TokenType.GLOB: 282>, 'GROUP BY': <TokenType.GROUP_BY: 285>, 'GROUPING SETS': <TokenType.GROUPING_SETS: 286>, 'HAVING': <TokenType.HAVING: 287>, 'ILIKE': <TokenType.ILIKE: 290>, 'IN': <TokenType.IN: 291>, 'INDEX': <TokenType.INDEX: 292>, 'INET': <TokenType.INET: 196>, 'INNER': <TokenType.INNER: 294>, 'INSERT': <TokenType.INSERT: 295>, 'INTERVAL': <TokenType.INTERVAL: 298>, 'INTERSECT': <TokenType.INTERSECT: 297>, 'INTO': <TokenType.INTO: 299>, 'IS': <TokenType.IS: 302>, 'ISNULL': <TokenType.ISNULL: 303>, 'JOIN': <TokenType.JOIN: 304>, 'KEEP': <TokenType.KEEP: 306>, 'KILL': <TokenType.KILL: 308>, 'LATERAL': <TokenType.LATERAL: 310>, 'LEFT': <TokenType.LEFT: 311>, 'LIKE': <TokenType.LIKE: 312>, 'LIMIT': <TokenType.LIMIT: 313>, 'LOAD': <TokenType.LOAD: 315>, 'LOCALTIME': <TokenType.LOCALTIME: 175>, 'LOCALTIMESTAMP': <TokenType.LOCALTIMESTAMP: 176>, 'LOCK': <TokenType.LOCK: 316>, 'MERGE': <TokenType.MERGE: 322>, 'NAMESPACE': <TokenType.NAMESPACE: 426>, 'NATURAL': <TokenType.NATURAL: 325>, 'NEXT': <TokenType.NEXT: 326>, 'NOT': <TokenType.NOT: 26>, 'NOTNULL': <TokenType.NOTNULL: 328>, 'NULL': <TokenType.NULL: 329>, 'OBJECT': <TokenType.OBJECT: 195>, 'OFFSET': <TokenType.OFFSET: 331>, 'ON': <TokenType.ON: 332>, 'OR': <TokenType.OR: 34>, 'XOR': <TokenType.XOR: 63>, 'ORDER BY': <TokenType.ORDER_BY: 335>, 'ORDINALITY': <TokenType.ORDINALITY: 338>, 'OUT': <TokenType.OUT: 339>, 'OUTER': <TokenType.OUTER: 341>, 'OVER': <TokenType.OVER: 342>, 'OVERLAPS': <TokenType.OVERLAPS: 343>, 'OVERWRITE': <TokenType.OVERWRITE: 344>, 'PARTITION': <TokenType.PARTITION: 345>, 'PARTITION BY': <TokenType.PARTITION_BY: 346>, 'PARTITIONED BY': <TokenType.PARTITION_BY: 346>, 'PARTITIONED_BY': <TokenType.PARTITION_BY: 346>, 'PERCENT': <TokenType.PERCENT: 347>, 'PIVOT': <TokenType.PIVOT: 348>, 'PRAGMA': <TokenType.PRAGMA: 351>, 'PRIMARY KEY': <TokenType.PRIMARY_KEY: 353>, 'PROCEDURE': <TokenType.PROCEDURE: 354>, 'OPERATOR': <TokenType.OPERATOR: 334>, 'QUALIFY': <TokenType.QUALIFY: 358>, 'RANGE': <TokenType.RANGE: 361>, 'RECURSIVE': <TokenType.RECURSIVE: 362>, 'REGEXP': <TokenType.RLIKE: 370>, 'RENAME': <TokenType.RENAME: 364>, 'REPLACE': <TokenType.REPLACE: 365>, 'RETURNING': <TokenType.RETURNING: 366>, 'REFERENCES': <TokenType.REFERENCES: 368>, 'RIGHT': <TokenType.RIGHT: 369>, 'RLIKE': <TokenType.RLIKE: 370>, 'ROLLBACK': <TokenType.ROLLBACK: 371>, 'ROLLUP': <TokenType.ROLLUP: 372>, 'ROW': <TokenType.ROW: 373>, 'ROWS': <TokenType.ROWS: 374>, 'SCHEMA': <TokenType.SCHEMA: 80>, 'SELECT': <TokenType.SELECT: 375>, 'SEMI': <TokenType.SEMI: 376>, 'SESSION': <TokenType.SESSION: 56>, 'SESSION_USER': <TokenType.SESSION_USER: 58>, 'SET': <TokenType.SET: 380>, 'SETTINGS': <TokenType.SETTINGS: 381>, 'SHOW': <TokenType.SHOW: 382>, 'SIMILAR TO': <TokenType.SIMILAR_TO: 383>, 'SOME': <TokenType.SOME: 384>, 'SORT BY': <TokenType.SORT_BY: 385>, 'START WITH': <TokenType.START_WITH: 387>, 'STRAIGHT_JOIN': <TokenType.STRAIGHT_JOIN: 389>, 'TABLE': <TokenType.TABLE: 81>, 'TABLESAMPLE': <TokenType.TABLE_SAMPLE: 392>, 'TEMP': <TokenType.TEMPORARY: 394>, 'TEMPORARY': <TokenType.TEMPORARY: 394>, 'THEN': <TokenType.THEN: 396>, 'TRUE': <TokenType.TRUE: 397>, 'TRUNCATE': <TokenType.TRUNCATE: 398>, 'TRIGGER': <TokenType.TRIGGER: 399>, 'UNION': <TokenType.UNION: 401>, 'UNKNOWN': <TokenType.UNKNOWN: 210>, 'UNNEST': <TokenType.UNNEST: 402>, 'UNPIVOT': <TokenType.UNPIVOT: 403>, 'UPDATE': <TokenType.UPDATE: 404>, 'USE': <TokenType.USE: 405>, 'USING': <TokenType.USING: 406>, 'UUID': <TokenType.UUID: 167>, 'VALUES': <TokenType.VALUES: 407>, 'VIEW': <TokenType.VIEW: 409>, 'VOLATILE': <TokenType.VOLATILE: 411>, 'WHEN': <TokenType.WHEN: 412>, 'WHERE': <TokenType.WHERE: 413>, 'WINDOW': <TokenType.WINDOW: 414>, 'WITH': <TokenType.WITH: 415>, 'APPLY': <TokenType.APPLY: 219>, 'ARRAY': <TokenType.ARRAY: 220>, 'BIT': <TokenType.BIT: 93>, 'BOOL': <TokenType.BOOLEAN: 94>, 'BOOLEAN': <TokenType.BOOLEAN: 94>, 'BYTE': <TokenType.TINYINT: 95>, 'MEDIUMINT': <TokenType.MEDIUMINT: 99>, 'INT1': <TokenType.TINYINT: 95>, 'TINYINT': <TokenType.UTINYINT: 96>, 'INT16': <TokenType.SMALLINT: 97>, 'SHORT': <TokenType.SMALLINT: 97>, 'SMALLINT': <TokenType.SMALLINT: 97>, 'HUGEINT': <TokenType.INT128: 106>, 'UHUGEINT': <TokenType.UINT128: 107>, 'INT2': <TokenType.SMALLINT: 97>, 'INTEGER': <TokenType.INT: 101>, 'INT': <TokenType.INT: 101>, 'INT4': <TokenType.INT: 101>, 'INT32': <TokenType.INT: 101>, 'INT64': <TokenType.BIGINT: 103>, 'INT128': <TokenType.INT128: 106>, 'INT256': <TokenType.INT256: 108>, 'LONG': <TokenType.BIGINT: 103>, 'BIGINT': <TokenType.BIGINT: 103>, 'INT8': <TokenType.TINYINT: 95>, 'UINT': <TokenType.UINT: 102>, 'UINT128': <TokenType.UINT128: 107>, 'UINT256': <TokenType.UINT256: 109>, 'DEC': <TokenType.DECIMAL: 113>, 'DECIMAL': <TokenType.DECIMAL: 113>, 'DECIMAL32': <TokenType.DECIMAL32: 114>, 'DECIMAL64': <TokenType.DECIMAL64: 115>, 'DECIMAL128': <TokenType.DECIMAL128: 116>, 'DECIMAL256': <TokenType.DECIMAL256: 117>, 'DECFLOAT': <TokenType.DECFLOAT: 118>, 'BIGDECIMAL': <TokenType.BIGDECIMAL: 120>, 'BIGNUMERIC': <TokenType.BIGDECIMAL: 120>, 'BIGNUM': <TokenType.BIGNUM: 105>, 'LIST': <TokenType.LIST: 314>, 'MAP': <TokenType.MAP: 317>, 'NULLABLE': <TokenType.NULLABLE: 170>, 'NUMBER': <TokenType.DECIMAL: 113>, 'NUMERIC': <TokenType.DECIMAL: 113>, 'FIXED': <TokenType.DECIMAL: 113>, 'REAL': <TokenType.FLOAT: 110>, 'FLOAT': <TokenType.FLOAT: 110>, 'FLOAT4': <TokenType.FLOAT: 110>, 'FLOAT8': <TokenType.DOUBLE: 111>, 'DOUBLE': <TokenType.DOUBLE: 111>, 'DOUBLE PRECISION': <TokenType.DOUBLE: 111>, 'JSON': <TokenType.JSON: 137>, 'JSONB': <TokenType.JSONB: 138>, 'CHAR': <TokenType.CHAR: 121>, 'CHARACTER': <TokenType.CHAR: 121>, 'CHAR VARYING': <TokenType.VARCHAR: 123>, 'CHARACTER VARYING': <TokenType.VARCHAR: 123>, 'NCHAR': <TokenType.NCHAR: 122>, 'VARCHAR': <TokenType.VARCHAR: 123>, 'VARCHAR2': <TokenType.VARCHAR: 123>, 'NVARCHAR': <TokenType.NVARCHAR: 124>, 'NVARCHAR2': <TokenType.NVARCHAR: 124>, 'BPCHAR': <TokenType.BPCHAR: 125>, 'STR': <TokenType.TEXT: 126>, 'STRING': <TokenType.TEXT: 126>, 'TEXT': <TokenType.TEXT: 126>, 'LONGTEXT': <TokenType.LONGTEXT: 128>, 'MEDIUMTEXT': <TokenType.MEDIUMTEXT: 127>, 'TINYTEXT': <TokenType.TINYTEXT: 133>, 'CLOB': <TokenType.TEXT: 126>, 'LONGVARCHAR': <TokenType.TEXT: 126>, 'BINARY': <TokenType.BINARY: 135>, 'BLOB': <TokenType.VARBINARY: 136>, 'LONGBLOB': <TokenType.LONGBLOB: 131>, 'MEDIUMBLOB': <TokenType.MEDIUMBLOB: 130>, 'TINYBLOB': <TokenType.TINYBLOB: 132>, 'BYTEA': <TokenType.VARBINARY: 136>, 'VARBINARY': <TokenType.VARBINARY: 136>, 'TIME': <TokenType.TIME: 139>, 'TIMETZ': <TokenType.TIMETZ: 140>, 'TIME_NS': <TokenType.TIME_NS: 141>, 'TIMESTAMP': <TokenType.TIMESTAMP: 142>, 'TIMESTAMPTZ': <TokenType.TIMESTAMPTZ: 143>, 'TIMESTAMPLTZ': <TokenType.TIMESTAMPLTZ: 144>, 'TIMESTAMP_LTZ': <TokenType.TIMESTAMPLTZ: 144>, 'TIMESTAMPNTZ': <TokenType.TIMESTAMPNTZ: 145>, 'TIMESTAMP_NTZ': <TokenType.TIMESTAMPNTZ: 145>, 'DATE': <TokenType.DATE: 153>, 'DATETIME': <TokenType.DATETIME: 149>, 'INT4RANGE': <TokenType.INT4RANGE: 155>, 'INT4MULTIRANGE': <TokenType.INT4MULTIRANGE: 156>, 'INT8RANGE': <TokenType.INT8RANGE: 157>, 'INT8MULTIRANGE': <TokenType.INT8MULTIRANGE: 158>, 'NUMRANGE': <TokenType.NUMRANGE: 159>, 'NUMMULTIRANGE': <TokenType.NUMMULTIRANGE: 160>, 'TSRANGE': <TokenType.TSRANGE: 161>, 'TSMULTIRANGE': <TokenType.TSMULTIRANGE: 162>, 'TSTZRANGE': <TokenType.TSTZRANGE: 163>, 'TSTZMULTIRANGE': <TokenType.TSTZMULTIRANGE: 164>, 'DATERANGE': <TokenType.DATERANGE: 165>, 'DATEMULTIRANGE': <TokenType.DATEMULTIRANGE: 166>, 'UNIQUE': <TokenType.UNIQUE: 416>, 'VECTOR': <TokenType.VECTOR: 211>, 'STRUCT': <TokenType.STRUCT: 390>, 'SEQUENCE': <TokenType.SEQUENCE: 378>, 'VARIANT': <TokenType.VARIANT: 194>, 'ALTER': <TokenType.ALTER: 215>, 'ANALYZE': <TokenType.ANALYZE: 425>, 'CALL': <TokenType.COMMAND: 233>, 'COMMENT': <TokenType.COMMENT: 234>, 'EXPLAIN': <TokenType.COMMAND: 233>, 'GRANT': <TokenType.GRANT: 284>, 'REVOKE': <TokenType.REVOKE: 367>, 'OPTIMIZE': <TokenType.COMMAND: 233>, 'PREPARE': <TokenType.COMMAND: 233>, 'VACUUM': <TokenType.COMMAND: 233>, 'USER-DEFINED': <TokenType.USERDEFINED: 189>, 'FOR VERSION': <TokenType.VERSION_SNAPSHOT: 420>, 'FOR TIMESTAMP': <TokenType.TIMESTAMP_SNAPSHOT: 421>, 'CLUSTERED INDEX': <TokenType.INDEX: 292>, 'DATETIME2': <TokenType.DATETIME2: 150>, 'DATETIMEOFFSET': <TokenType.TIMESTAMPTZ: 143>, 'DECLARE': <TokenType.DECLARE: 250>, 'EXEC': <TokenType.EXECUTE: 265>, 'FOR SYSTEM_TIME': <TokenType.TIMESTAMP_SNAPSHOT: 421>, 'GO': <TokenType.COMMAND: 233>, 'IMAGE': <TokenType.IMAGE: 193>, 'MONEY': <TokenType.MONEY: 190>, 'NONCLUSTERED INDEX': <TokenType.INDEX: 292>, 'NTEXT': <TokenType.TEXT: 126>, 'OPTION': <TokenType.OPTION: 422>, 'OUTPUT': <TokenType.RETURNING: 366>, 'PRINT': <TokenType.COMMAND: 233>, 'PROC': <TokenType.PROCEDURE: 354>, 'ROWVERSION': <TokenType.ROWVERSION: 192>, 'SMALLDATETIME': <TokenType.SMALLDATETIME: 152>, 'SMALLMONEY': <TokenType.SMALLMONEY: 191>, 'SQL_VARIANT': <TokenType.VARIANT: 194>, 'SYSTEM_USER': <TokenType.CURRENT_USER: 247>, 'TOP': <TokenType.TOP: 395>, 'UNIQUEIDENTIFIER': <TokenType.UUID: 167>, 'UPDATE STATISTICS': <TokenType.COMMAND: 233>, 'XML': <TokenType.XML: 187>, 'UTINYINT': <TokenType.UTINYINT: 96>}

BYTE_STRING_ESCAPES: ClassVar[List[str]] = ["'"]

Inherited Members

sqlglot.tokens.Tokenizer: Tokenizer; SINGLE_TOKENS; BIT_STRINGS; BYTE_STRINGS; RAW_STRINGS; HEREDOC_STRINGS; UNICODE_STRINGS; STRING_ESCAPES; ESCAPE_FOLLOW_CHARS; IDENTIFIER_ESCAPES; HEREDOC_TAG_IS_IDENTIFIER; HEREDOC_STRING_ALTERNATIVE; STRING_ESCAPES_ALLOWED_IN_RAW_STRINGS; NESTED_COMMENTS; HINT_START; TOKENS_PRECEDING_HINT; COMMAND_PREFIX_TOKENS; NUMERIC_LITERALS; COMMENTS; dialect; tokenize; sql; size; tokens
sqlglot.dialects.tsql.TSQL.Tokenizer: IDENTIFIERS; QUOTES; HEX_STRINGS; VAR_SINGLE_TOKENS; COMMANDS

class Fabric.Parser(sqlglot.dialects.tsql.TSQL.Parser): View Source

 85    class Parser(TSQL.Parser):
 86        def _parse_create(self) -> exp.Create | exp.Command:
 87            create = super()._parse_create()
 88
 89            if isinstance(create, exp.Create):
 90                # Transform VARCHAR/CHAR without precision to VARCHAR(1)/CHAR(1)
 91                if create.kind == "TABLE" and isinstance(create.this, exp.Schema):
 92                    for column in create.this.expressions:
 93                        if isinstance(column, exp.ColumnDef):
 94                            column_type = column.kind
 95                            if (
 96                                isinstance(column_type, exp.DataType)
 97                                and column_type.this
 98                                in (exp.DataType.Type.VARCHAR, exp.DataType.Type.CHAR)
 99                                and not column_type.expressions
100                            ):
101                                # Add default precision of 1 to VARCHAR/CHAR without precision
102                                # When n isn't specified in a data definition or variable declaration statement, the default length is 1.
103                                # https://learn.microsoft.com/en-us/sql/t-sql/data-types/char-and-varchar-transact-sql?view=sql-server-ver17#remarks
104                                column_type.set("expressions", [exp.Literal.number("1")])
105
106            return create

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: 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

ID_VAR_TOKENS = {<TokenType.SESSION: 56>, <TokenType.SESSION_USER: 58>, <TokenType.IDENTIFIER: 76>, <TokenType.DATABASE: 77>, <TokenType.COLUMN: 78>, <TokenType.SCHEMA: 80>, <TokenType.TABLE: 81>, <TokenType.WAREHOUSE: 82>, <TokenType.STAGE: 83>, <TokenType.STREAMLIT: 84>, <TokenType.VAR: 85>, <TokenType.BIT: 93>, <TokenType.BOOLEAN: 94>, <TokenType.TINYINT: 95>, <TokenType.UTINYINT: 96>, <TokenType.SMALLINT: 97>, <TokenType.USMALLINT: 98>, <TokenType.MEDIUMINT: 99>, <TokenType.UMEDIUMINT: 100>, <TokenType.INT: 101>, <TokenType.UINT: 102>, <TokenType.BIGINT: 103>, <TokenType.UBIGINT: 104>, <TokenType.BIGNUM: 105>, <TokenType.INT128: 106>, <TokenType.UINT128: 107>, <TokenType.INT256: 108>, <TokenType.UINT256: 109>, <TokenType.FLOAT: 110>, <TokenType.DOUBLE: 111>, <TokenType.UDOUBLE: 112>, <TokenType.DECIMAL: 113>, <TokenType.DECIMAL32: 114>, <TokenType.DECIMAL64: 115>, <TokenType.DECIMAL128: 116>, <TokenType.DECIMAL256: 117>, <TokenType.DECFLOAT: 118>, <TokenType.UDECIMAL: 119>, <TokenType.BIGDECIMAL: 120>, <TokenType.CHAR: 121>, <TokenType.NCHAR: 122>, <TokenType.VARCHAR: 123>, <TokenType.NVARCHAR: 124>, <TokenType.BPCHAR: 125>, <TokenType.TEXT: 126>, <TokenType.MEDIUMTEXT: 127>, <TokenType.LONGTEXT: 128>, <TokenType.BLOB: 129>, <TokenType.MEDIUMBLOB: 130>, <TokenType.LONGBLOB: 131>, <TokenType.TINYBLOB: 132>, <TokenType.TINYTEXT: 133>, <TokenType.NAME: 134>, <TokenType.BINARY: 135>, <TokenType.VARBINARY: 136>, <TokenType.JSON: 137>, <TokenType.JSONB: 138>, <TokenType.TIME: 139>, <TokenType.TIMETZ: 140>, <TokenType.TIME_NS: 141>, <TokenType.TIMESTAMP: 142>, <TokenType.TIMESTAMPTZ: 143>, <TokenType.TIMESTAMPLTZ: 144>, <TokenType.TIMESTAMPNTZ: 145>, <TokenType.TIMESTAMP_S: 146>, <TokenType.TIMESTAMP_MS: 147>, <TokenType.TIMESTAMP_NS: 148>, <TokenType.DATETIME: 149>, <TokenType.DATETIME2: 150>, <TokenType.DATETIME64: 151>, <TokenType.SMALLDATETIME: 152>, <TokenType.DATE: 153>, <TokenType.DATE32: 154>, <TokenType.INT4RANGE: 155>, <TokenType.INT4MULTIRANGE: 156>, <TokenType.INT8RANGE: 157>, <TokenType.INT8MULTIRANGE: 158>, <TokenType.NUMRANGE: 159>, <TokenType.NUMMULTIRANGE: 160>, <TokenType.TSRANGE: 161>, <TokenType.TSMULTIRANGE: 162>, <TokenType.TSTZRANGE: 163>, <TokenType.TSTZMULTIRANGE: 164>, <TokenType.DATERANGE: 165>, <TokenType.DATEMULTIRANGE: 166>, <TokenType.UUID: 167>, <TokenType.GEOGRAPHY: 168>, <TokenType.GEOGRAPHYPOINT: 169>, <TokenType.NULLABLE: 170>, <TokenType.GEOMETRY: 171>, <TokenType.POINT: 172>, <TokenType.RING: 173>, <TokenType.LINESTRING: 174>, <TokenType.LOCALTIME: 175>, <TokenType.LOCALTIMESTAMP: 176>, <TokenType.MULTILINESTRING: 178>, <TokenType.POLYGON: 179>, <TokenType.MULTIPOLYGON: 180>, <TokenType.HLLSKETCH: 181>, <TokenType.HSTORE: 182>, <TokenType.SUPER: 183>, <TokenType.SERIAL: 184>, <TokenType.SMALLSERIAL: 185>, <TokenType.BIGSERIAL: 186>, <TokenType.XML: 187>, <TokenType.YEAR: 188>, <TokenType.USERDEFINED: 189>, <TokenType.MONEY: 190>, <TokenType.SMALLMONEY: 191>, <TokenType.ROWVERSION: 192>, <TokenType.IMAGE: 193>, <TokenType.VARIANT: 194>, <TokenType.OBJECT: 195>, <TokenType.INET: 196>, <TokenType.IPADDRESS: 197>, <TokenType.IPPREFIX: 198>, <TokenType.IPV4: 199>, <TokenType.IPV6: 200>, <TokenType.ENUM: 201>, <TokenType.ENUM8: 202>, <TokenType.ENUM16: 203>, <TokenType.FIXEDSTRING: 204>, <TokenType.LOWCARDINALITY: 205>, <TokenType.NESTED: 206>, <TokenType.AGGREGATEFUNCTION: 207>, <TokenType.SIMPLEAGGREGATEFUNCTION: 208>, <TokenType.TDIGEST: 209>, <TokenType.UNKNOWN: 210>, <TokenType.VECTOR: 211>, <TokenType.DYNAMIC: 212>, <TokenType.VOID: 213>, <TokenType.ALL: 216>, <TokenType.ANTI: 217>, <TokenType.ANY: 218>, <TokenType.APPLY: 219>, <TokenType.ARRAY: 220>, <TokenType.ASC: 221>, <TokenType.ASOF: 222>, <TokenType.ATTACH: 223>, <TokenType.AUTO_INCREMENT: 224>, <TokenType.CACHE: 228>, <TokenType.CASE: 229>, <TokenType.COLLATE: 232>, <TokenType.COMMAND: 233>, <TokenType.COMMENT: 234>, <TokenType.COMMIT: 235>, <TokenType.CONSTRAINT: 237>, <TokenType.COPY: 238>, <TokenType.CUBE: 241>, <TokenType.CURRENT_DATE: 242>, <TokenType.CURRENT_DATETIME: 243>, <TokenType.CURRENT_SCHEMA: 244>, <TokenType.CURRENT_TIME: 245>, <TokenType.CURRENT_TIMESTAMP: 246>, <TokenType.CURRENT_USER: 247>, <TokenType.CURRENT_ROLE: 248>, <TokenType.CURRENT_CATALOG: 249>, <TokenType.DEFAULT: 251>, <TokenType.DELETE: 252>, <TokenType.DESC: 253>, <TokenType.DESCRIBE: 254>, <TokenType.DETACH: 255>, <TokenType.DICTIONARY: 256>, <TokenType.DIV: 259>, <TokenType.END: 262>, <TokenType.ESCAPE: 263>, <TokenType.EXECUTE: 265>, <TokenType.EXISTS: 266>, <TokenType.FALSE: 267>, <TokenType.FILE: 269>, <TokenType.FILE_FORMAT: 270>, <TokenType.FILTER: 271>, <TokenType.FINAL: 272>, <TokenType.FIRST: 273>, <TokenType.FOREIGN_KEY: 276>, <TokenType.FORMAT: 277>, <TokenType.FULL: 279>, <TokenType.FUNCTION: 280>, <TokenType.GET: 281>, <TokenType.INDEX: 292>, <TokenType.INTERVAL: 298>, <TokenType.IS: 302>, <TokenType.ISNULL: 303>, <TokenType.KEEP: 306>, <TokenType.KILL: 308>, <TokenType.LEFT: 311>, <TokenType.LIMIT: 313>, <TokenType.LIST: 314>, <TokenType.LOAD: 315>, <TokenType.LOCK: 316>, <TokenType.MAP: 317>, <TokenType.MATCH: 318>, <TokenType.MERGE: 322>, <TokenType.MODEL: 324>, <TokenType.NATURAL: 325>, <TokenType.NEXT: 326>, <TokenType.NOTHING: 327>, <TokenType.NULL: 329>, <TokenType.OBJECT_IDENTIFIER: 330>, <TokenType.OFFSET: 331>, <TokenType.OPERATOR: 334>, <TokenType.ORDINALITY: 338>, <TokenType.INOUT: 340>, <TokenType.OVER: 342>, <TokenType.OVERLAPS: 343>, <TokenType.OVERWRITE: 344>, <TokenType.PARTITION: 345>, <TokenType.PERCENT: 347>, <TokenType.PIVOT: 348>, <TokenType.PRAGMA: 351>, <TokenType.PROCEDURE: 354>, <TokenType.PSEUDO_TYPE: 356>, <TokenType.PUT: 357>, <TokenType.RANGE: 361>, <TokenType.RECURSIVE: 362>, <TokenType.REFRESH: 363>, <TokenType.RENAME: 364>, <TokenType.REPLACE: 365>, <TokenType.REFERENCES: 368>, <TokenType.RIGHT: 369>, <TokenType.ROLLUP: 372>, <TokenType.ROW: 373>, <TokenType.ROWS: 374>, <TokenType.SEMI: 376>, <TokenType.SEQUENCE: 378>, <TokenType.SET: 380>, <TokenType.SETTINGS: 381>, <TokenType.SHOW: 382>, <TokenType.SOME: 384>, <TokenType.STORAGE_INTEGRATION: 388>, <TokenType.STRAIGHT_JOIN: 389>, <TokenType.STRUCT: 390>, <TokenType.TAG: 393>, <TokenType.TEMPORARY: 394>, <TokenType.TOP: 395>, <TokenType.TRUE: 397>, <TokenType.TRUNCATE: 398>, <TokenType.TRIGGER: 399>, <TokenType.UNNEST: 402>, <TokenType.UNPIVOT: 403>, <TokenType.UPDATE: 404>, <TokenType.USE: 405>, <TokenType.VIEW: 409>, <TokenType.SEMANTIC_VIEW: 410>, <TokenType.VOLATILE: 411>, <TokenType.WINDOW: 414>, <TokenType.UNIQUE: 416>, <TokenType.SINK: 423>, <TokenType.SOURCE: 424>, <TokenType.ANALYZE: 425>, <TokenType.NAMESPACE: 426>, <TokenType.EXPORT: 427>}

TABLE_ALIAS_TOKENS = {<TokenType.SESSION: 56>, <TokenType.IDENTIFIER: 76>, <TokenType.DATABASE: 77>, <TokenType.COLUMN: 78>, <TokenType.SCHEMA: 80>, <TokenType.TABLE: 81>, <TokenType.WAREHOUSE: 82>, <TokenType.STAGE: 83>, <TokenType.STREAMLIT: 84>, <TokenType.VAR: 85>, <TokenType.BIT: 93>, <TokenType.BOOLEAN: 94>, <TokenType.TINYINT: 95>, <TokenType.UTINYINT: 96>, <TokenType.SMALLINT: 97>, <TokenType.USMALLINT: 98>, <TokenType.MEDIUMINT: 99>, <TokenType.UMEDIUMINT: 100>, <TokenType.INT: 101>, <TokenType.UINT: 102>, <TokenType.BIGINT: 103>, <TokenType.UBIGINT: 104>, <TokenType.BIGNUM: 105>, <TokenType.INT128: 106>, <TokenType.UINT128: 107>, <TokenType.INT256: 108>, <TokenType.UINT256: 109>, <TokenType.FLOAT: 110>, <TokenType.DOUBLE: 111>, <TokenType.UDOUBLE: 112>, <TokenType.DECIMAL: 113>, <TokenType.DECIMAL32: 114>, <TokenType.DECIMAL64: 115>, <TokenType.DECIMAL128: 116>, <TokenType.DECIMAL256: 117>, <TokenType.DECFLOAT: 118>, <TokenType.UDECIMAL: 119>, <TokenType.BIGDECIMAL: 120>, <TokenType.CHAR: 121>, <TokenType.NCHAR: 122>, <TokenType.VARCHAR: 123>, <TokenType.NVARCHAR: 124>, <TokenType.BPCHAR: 125>, <TokenType.TEXT: 126>, <TokenType.MEDIUMTEXT: 127>, <TokenType.LONGTEXT: 128>, <TokenType.BLOB: 129>, <TokenType.MEDIUMBLOB: 130>, <TokenType.LONGBLOB: 131>, <TokenType.TINYBLOB: 132>, <TokenType.TINYTEXT: 133>, <TokenType.NAME: 134>, <TokenType.BINARY: 135>, <TokenType.VARBINARY: 136>, <TokenType.JSON: 137>, <TokenType.JSONB: 138>, <TokenType.TIME: 139>, <TokenType.TIMETZ: 140>, <TokenType.TIME_NS: 141>, <TokenType.TIMESTAMP: 142>, <TokenType.TIMESTAMPTZ: 143>, <TokenType.TIMESTAMPLTZ: 144>, <TokenType.TIMESTAMPNTZ: 145>, <TokenType.TIMESTAMP_S: 146>, <TokenType.TIMESTAMP_MS: 147>, <TokenType.TIMESTAMP_NS: 148>, <TokenType.DATETIME: 149>, <TokenType.DATETIME2: 150>, <TokenType.DATETIME64: 151>, <TokenType.SMALLDATETIME: 152>, <TokenType.DATE: 153>, <TokenType.DATE32: 154>, <TokenType.INT4RANGE: 155>, <TokenType.INT4MULTIRANGE: 156>, <TokenType.INT8RANGE: 157>, <TokenType.INT8MULTIRANGE: 158>, <TokenType.NUMRANGE: 159>, <TokenType.NUMMULTIRANGE: 160>, <TokenType.TSRANGE: 161>, <TokenType.TSMULTIRANGE: 162>, <TokenType.TSTZRANGE: 163>, <TokenType.TSTZMULTIRANGE: 164>, <TokenType.DATERANGE: 165>, <TokenType.DATEMULTIRANGE: 166>, <TokenType.UUID: 167>, <TokenType.GEOGRAPHY: 168>, <TokenType.GEOGRAPHYPOINT: 169>, <TokenType.NULLABLE: 170>, <TokenType.GEOMETRY: 171>, <TokenType.POINT: 172>, <TokenType.RING: 173>, <TokenType.LINESTRING: 174>, <TokenType.LOCALTIME: 175>, <TokenType.LOCALTIMESTAMP: 176>, <TokenType.MULTILINESTRING: 178>, <TokenType.POLYGON: 179>, <TokenType.MULTIPOLYGON: 180>, <TokenType.HLLSKETCH: 181>, <TokenType.HSTORE: 182>, <TokenType.SUPER: 183>, <TokenType.SERIAL: 184>, <TokenType.SMALLSERIAL: 185>, <TokenType.BIGSERIAL: 186>, <TokenType.XML: 187>, <TokenType.YEAR: 188>, <TokenType.USERDEFINED: 189>, <TokenType.MONEY: 190>, <TokenType.SMALLMONEY: 191>, <TokenType.ROWVERSION: 192>, <TokenType.IMAGE: 193>, <TokenType.VARIANT: 194>, <TokenType.OBJECT: 195>, <TokenType.INET: 196>, <TokenType.IPADDRESS: 197>, <TokenType.IPPREFIX: 198>, <TokenType.IPV4: 199>, <TokenType.IPV6: 200>, <TokenType.ENUM: 201>, <TokenType.ENUM8: 202>, <TokenType.ENUM16: 203>, <TokenType.FIXEDSTRING: 204>, <TokenType.LOWCARDINALITY: 205>, <TokenType.NESTED: 206>, <TokenType.AGGREGATEFUNCTION: 207>, <TokenType.SIMPLEAGGREGATEFUNCTION: 208>, <TokenType.TDIGEST: 209>, <TokenType.UNKNOWN: 210>, <TokenType.VECTOR: 211>, <TokenType.DYNAMIC: 212>, <TokenType.VOID: 213>, <TokenType.ALL: 216>, <TokenType.ANTI: 217>, <TokenType.ANY: 218>, <TokenType.APPLY: 219>, <TokenType.ARRAY: 220>, <TokenType.ASC: 221>, <TokenType.ATTACH: 223>, <TokenType.AUTO_INCREMENT: 224>, <TokenType.CACHE: 228>, <TokenType.CASE: 229>, <TokenType.COLLATE: 232>, <TokenType.COMMAND: 233>, <TokenType.COMMENT: 234>, <TokenType.COMMIT: 235>, <TokenType.CONSTRAINT: 237>, <TokenType.COPY: 238>, <TokenType.CUBE: 241>, <TokenType.CURRENT_DATE: 242>, <TokenType.CURRENT_DATETIME: 243>, <TokenType.CURRENT_SCHEMA: 244>, <TokenType.CURRENT_TIME: 245>, <TokenType.CURRENT_TIMESTAMP: 246>, <TokenType.CURRENT_USER: 247>, <TokenType.CURRENT_ROLE: 248>, <TokenType.DEFAULT: 251>, <TokenType.DELETE: 252>, <TokenType.DESC: 253>, <TokenType.DESCRIBE: 254>, <TokenType.DETACH: 255>, <TokenType.DICTIONARY: 256>, <TokenType.DIV: 259>, <TokenType.END: 262>, <TokenType.ESCAPE: 263>, <TokenType.EXECUTE: 265>, <TokenType.EXISTS: 266>, <TokenType.FALSE: 267>, <TokenType.FILE: 269>, <TokenType.FILE_FORMAT: 270>, <TokenType.FILTER: 271>, <TokenType.FINAL: 272>, <TokenType.FIRST: 273>, <TokenType.FOREIGN_KEY: 276>, <TokenType.FORMAT: 277>, <TokenType.FUNCTION: 280>, <TokenType.GET: 281>, <TokenType.INDEX: 292>, <TokenType.INTERVAL: 298>, <TokenType.IS: 302>, <TokenType.ISNULL: 303>, <TokenType.KEEP: 306>, <TokenType.KILL: 308>, <TokenType.LIMIT: 313>, <TokenType.LIST: 314>, <TokenType.LOAD: 315>, <TokenType.MAP: 317>, <TokenType.MATCH: 318>, <TokenType.MERGE: 322>, <TokenType.MODEL: 324>, <TokenType.NEXT: 326>, <TokenType.NOTHING: 327>, <TokenType.NULL: 329>, <TokenType.OBJECT_IDENTIFIER: 330>, <TokenType.OFFSET: 331>, <TokenType.OPERATOR: 334>, <TokenType.ORDINALITY: 338>, <TokenType.INOUT: 340>, <TokenType.OVER: 342>, <TokenType.OVERLAPS: 343>, <TokenType.OVERWRITE: 344>, <TokenType.PARTITION: 345>, <TokenType.PERCENT: 347>, <TokenType.PIVOT: 348>, <TokenType.PRAGMA: 351>, <TokenType.PROCEDURE: 354>, <TokenType.PSEUDO_TYPE: 356>, <TokenType.PUT: 357>, <TokenType.RANGE: 361>, <TokenType.RECURSIVE: 362>, <TokenType.REFRESH: 363>, <TokenType.RENAME: 364>, <TokenType.REPLACE: 365>, <TokenType.REFERENCES: 368>, <TokenType.ROLLUP: 372>, <TokenType.ROW: 373>, <TokenType.ROWS: 374>, <TokenType.SEMI: 376>, <TokenType.SEQUENCE: 378>, <TokenType.SET: 380>, <TokenType.SETTINGS: 381>, <TokenType.SHOW: 382>, <TokenType.SOME: 384>, <TokenType.STORAGE_INTEGRATION: 388>, <TokenType.STRAIGHT_JOIN: 389>, <TokenType.STRUCT: 390>, <TokenType.TAG: 393>, <TokenType.TEMPORARY: 394>, <TokenType.TOP: 395>, <TokenType.TRUE: 397>, <TokenType.TRUNCATE: 398>, <TokenType.TRIGGER: 399>, <TokenType.UNNEST: 402>, <TokenType.UNPIVOT: 403>, <TokenType.UPDATE: 404>, <TokenType.USE: 405>, <TokenType.VIEW: 409>, <TokenType.SEMANTIC_VIEW: 410>, <TokenType.VOLATILE: 411>, <TokenType.UNIQUE: 416>, <TokenType.SINK: 423>, <TokenType.SOURCE: 424>, <TokenType.ANALYZE: 425>, <TokenType.NAMESPACE: 426>, <TokenType.EXPORT: 427>}

NO_PAREN_FUNCTIONS = {<TokenType.CURRENT_DATE: 242>: <class 'sqlglot.expressions.CurrentDate'>, <TokenType.CURRENT_DATETIME: 243>: <class 'sqlglot.expressions.CurrentDate'>, <TokenType.CURRENT_TIME: 245>: <class 'sqlglot.expressions.CurrentTime'>, <TokenType.CURRENT_TIMESTAMP: 246>: <class 'sqlglot.expressions.CurrentTimestamp'>, <TokenType.CURRENT_USER: 247>: <class 'sqlglot.expressions.CurrentUser'>, <TokenType.CURRENT_ROLE: 248>: <class 'sqlglot.expressions.CurrentRole'>, <TokenType.SESSION_USER: 58>: <class 'sqlglot.expressions.SessionUser'>}

SHOW_TRIE: Dict = {}

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

Inherited Members

sqlglot.parser.Parser: Parser; STRUCT_TYPE_TOKENS; NESTED_TYPE_TOKENS; ENUM_TYPE_TOKENS; AGGREGATE_TYPE_TOKENS; TYPE_TOKENS; SIGNED_TO_UNSIGNED_TYPE_TOKEN; SUBQUERY_PREDICATES; RESERVED_TOKENS; DB_CREATABLES; CREATABLES; TRIGGER_EVENTS; ALTERABLES; COLON_PLACEHOLDER_TOKENS; ARRAY_CONSTRUCTORS; TRIM_TYPES; FUNC_TOKENS; CONJUNCTION; ASSIGNMENT; DISJUNCTION; EQUALITY; COMPARISON; BITWISE; TERM; FACTOR; EXPONENT; TIMES; TIMESTAMPS; SET_OPERATIONS; JOIN_METHODS; JOIN_SIDES; JOIN_KINDS; LAMBDAS; CAST_COLUMN_OPERATORS; EXPRESSION_PARSERS; UNARY_PARSERS; STRING_PARSERS; NUMERIC_PARSERS; PRIMARY_PARSERS; PLACEHOLDER_PARSERS; PIPE_SYNTAX_TRANSFORM_PARSERS; PROPERTY_PARSERS; CONSTRAINT_PARSERS; ALTER_PARSERS; ALTER_ALTER_PARSERS; SCHEMA_UNNAMED_CONSTRAINTS; INVALID_FUNC_NAME_TOKENS; FUNCTIONS_WITH_ALIASED_ARGS; KEY_VALUE_DEFINITIONS; QUERY_MODIFIER_TOKENS; SET_PARSERS; SHOW_PARSERS; TYPE_LITERAL_PARSERS; TYPE_CONVERTERS; DDL_SELECT_TOKENS; PRE_VOLATILE_TOKENS; TRANSACTION_KIND; TRANSACTION_CHARACTERISTICS; CONFLICT_ACTIONS; TRIGGER_TIMING; TRIGGER_DEFERRABLE; CREATE_SEQUENCE; ISOLATED_LOADING_OPTIONS; USABLES; CAST_ACTIONS; SCHEMA_BINDING_OPTIONS; EXECUTE_AS_OPTIONS; KEY_CONSTRAINT_OPTIONS; WINDOW_EXCLUDE_OPTIONS; INSERT_ALTERNATIVES; CLONE_KEYWORDS; HISTORICAL_DATA_PREFIX; HISTORICAL_DATA_KIND; OPCLASS_FOLLOW_KEYWORDS; OPTYPE_FOLLOW_TOKENS; TABLE_INDEX_HINT_TOKENS; VIEW_ATTRIBUTES; WINDOW_ALIAS_TOKENS; WINDOW_BEFORE_PAREN_TOKENS; WINDOW_SIDES; JSON_KEY_VALUE_SEPARATOR_TOKENS; FETCH_TOKENS; ADD_CONSTRAINT_TOKENS; DISTINCT_TOKENS; UNNEST_OFFSET_ALIAS_TOKENS; SELECT_START_TOKENS; COPY_INTO_VARLEN_OPTIONS; IS_JSON_PREDICATE_KIND; ON_CONDITION_TOKENS; PRIVILEGE_FOLLOW_TOKENS; DESCRIBE_STYLES; SET_ASSIGNMENT_DELIMITERS; ANALYZE_STYLES; ANALYZE_EXPRESSION_PARSERS; PARTITION_KEYWORDS; AMBIGUOUS_ALIAS_TOKENS; OPERATION_MODIFIERS; RECURSIVE_CTE_SEARCH_KIND; MODIFIABLES; STRICT_CAST; PREFIXED_PIVOT_COLUMNS; IDENTIFY_PIVOT_STRINGS; TABLESAMPLE_CSV; DEFAULT_SAMPLING_METHOD; TRIM_PATTERN_FIRST; MODIFIERS_ATTACHED_TO_SET_OP; JSON_ARROWS_REQUIRE_JSON_TYPE; COLON_IS_VARIANT_EXTRACT; VALUES_FOLLOWED_BY_PAREN; SUPPORTS_IMPLICIT_UNNEST; INTERVAL_SPANS; SUPPORTS_PARTITION_SELECTION; WRAPPED_TRANSFORM_COLUMN_CONSTRAINT; OPTIONAL_ALIAS_TOKEN_CTE; ALTER_RENAME_REQUIRES_COLUMN; ALTER_TABLE_PARTITIONS; JOINS_HAVE_EQUAL_PRECEDENCE; ZONE_AWARE_TIMESTAMP_CONSTRUCTOR; MAP_KEYS_ARE_ARBITRARY_EXPRESSIONS; JSON_EXTRACT_REQUIRES_JSON_EXPRESSION; ADD_JOIN_ON_TRUE; SUPPORTS_OMITTED_INTERVAL_SPAN_UNIT; raise_error; validate_expression; reset; errors; error_level; error_message_context; max_errors; dialect; sql; parse; parse_into; check_errors; expression; parse_set_operation; build_cast
sqlglot.dialects.tsql.TSQL.Parser: SET_REQUIRES_ASSIGNMENT_DELIMITER; LOG_DEFAULTS_TO_LN; STRING_ALIASES; NO_PAREN_IF_COMMANDS; QUERY_MODIFIER_PARSERS; ALIAS_TOKENS; COMMENT_TABLE_ALIAS_TOKENS; UPDATE_ALIAS_TOKENS; FUNCTIONS; JOIN_HINTS; PROCEDURE_OPTIONS; COLUMN_DEFINITION_MODES; RETURNS_TABLE_TOKENS; STATEMENT_PARSERS; RANGE_PARSERS; NO_PAREN_FUNCTION_PARSERS; FUNCTION_PARSERS; COLUMN_OPERATORS; SET_OP_MODIFIERS; ODBC_DATETIME_LITERALS

class Fabric.Generator(sqlglot.dialects.tsql.TSQL.Generator): View Source

108    class Generator(TSQL.Generator):
109        # Fabric-specific type mappings - override T-SQL types that aren't supported
110        # Reference: https://learn.microsoft.com/en-us/fabric/data-warehouse/data-types
111        TYPE_MAPPING = {
112            **TSQL.Generator.TYPE_MAPPING,
113            exp.DataType.Type.DATETIME: "DATETIME2",
114            exp.DataType.Type.DECIMAL: "DECIMAL",
115            exp.DataType.Type.IMAGE: "VARBINARY",
116            exp.DataType.Type.INT: "INT",
117            exp.DataType.Type.JSON: "VARCHAR",
118            exp.DataType.Type.MONEY: "DECIMAL",
119            exp.DataType.Type.NCHAR: "CHAR",
120            exp.DataType.Type.NVARCHAR: "VARCHAR",
121            exp.DataType.Type.ROWVERSION: "ROWVERSION",
122            exp.DataType.Type.SMALLDATETIME: "DATETIME2",
123            exp.DataType.Type.SMALLMONEY: "DECIMAL",
124            exp.DataType.Type.TIMESTAMP: "DATETIME2",
125            exp.DataType.Type.TIMESTAMPNTZ: "DATETIME2",
126            exp.DataType.Type.TIMESTAMPTZ: "DATETIME2",
127            exp.DataType.Type.TINYINT: "SMALLINT",
128            exp.DataType.Type.UTINYINT: "SMALLINT",
129            exp.DataType.Type.UUID: "UNIQUEIDENTIFIER",
130            exp.DataType.Type.XML: "VARCHAR",
131        }
132
133        TRANSFORMS = {
134            **TSQL.Generator.TRANSFORMS,
135            exp.Create: transforms.preprocess([_add_default_precision_to_varchar]),
136        }
137
138        def datatype_sql(self, expression: exp.DataType) -> str:
139            # Check if this is a temporal type that needs precision handling. Fabric limits temporal
140            # types to max 6 digits precision. When no precision is specified, we default to 6 digits.
141            if (
142                expression.is_type(*exp.DataType.TEMPORAL_TYPES)
143                and expression.this != exp.DataType.Type.DATE
144            ):
145                # Create a new expression with the capped precision
146                expression = _cap_data_type_precision(expression)
147
148            return super().datatype_sql(expression)
149
150        def cast_sql(self, expression: exp.Cast, safe_prefix: str | None = None) -> str:
151            # Cast to DATETIMEOFFSET if inside an AT TIME ZONE expression
152            # https://learn.microsoft.com/en-us/sql/t-sql/data-types/datetimeoffset-transact-sql#microsoft-fabric-support
153            if expression.is_type(exp.DataType.Type.TIMESTAMPTZ):
154                at_time_zone = expression.find_ancestor(exp.AtTimeZone, exp.Select)
155
156                # Return normal cast, if the expression is not in an AT TIME ZONE context
157                if not isinstance(at_time_zone, exp.AtTimeZone):
158                    return super().cast_sql(expression, safe_prefix)
159
160                # Get the precision from the original TIMESTAMPTZ cast and cap it to 6
161                capped_data_type = _cap_data_type_precision(expression.to, max_precision=6)
162                precision = capped_data_type.find(exp.DataTypeParam)
163                precision_value = (
164                    precision.this.to_py() if precision and precision.this.is_int else 6
165                )
166
167                # Do the cast explicitly to bypass sqlglot's default handling
168                datetimeoffset = f"CAST({expression.this} AS DATETIMEOFFSET({precision_value}))"
169
170                return self.sql(datetimeoffset)
171
172            return super().cast_sql(expression, safe_prefix)
173
174        def attimezone_sql(self, expression: exp.AtTimeZone) -> str:
175            # Wrap the AT TIME ZONE expression in a cast to DATETIME2 if it contains a TIMESTAMPTZ
176            ## https://learn.microsoft.com/en-us/sql/t-sql/data-types/datetimeoffset-transact-sql#microsoft-fabric-support
177            timestamptz_cast = expression.find(exp.Cast)
178            if timestamptz_cast and timestamptz_cast.to.is_type(exp.DataType.Type.TIMESTAMPTZ):
179                # Get the precision from the original TIMESTAMPTZ cast and cap it to 6
180                data_type = timestamptz_cast.to
181                capped_data_type = _cap_data_type_precision(data_type, max_precision=6)
182                precision_param = capped_data_type.find(exp.DataTypeParam)
183                precision = precision_param.this.to_py() if precision_param else 6
184
185                # Generate the AT TIME ZONE expression (which will handle the inner cast conversion)
186                at_time_zone_sql = super().attimezone_sql(expression)
187
188                # Wrap it in an outer cast to DATETIME2
189                return f"CAST({at_time_zone_sql} AS DATETIME2({precision}))"
190
191            return super().attimezone_sql(expression)
192
193        def unixtotime_sql(self, expression: exp.UnixToTime) -> str:
194            scale = expression.args.get("scale")
195            timestamp = expression.this
196
197            if scale not in (None, exp.UnixToTime.SECONDS):
198                self.unsupported(f"UnixToTime scale {scale} is not supported by Fabric")
199                return ""
200
201            # Convert unix timestamp (seconds) to microseconds and round to avoid decimals
202            microseconds = timestamp * exp.Literal.number("1e6")
203            rounded = exp.func("round", microseconds, 0)
204            rounded_ms_as_bigint = exp.cast(rounded, exp.DataType.Type.BIGINT)
205
206            # Create the base datetime as '1970-01-01' cast to DATETIME2(6)
207            epoch_start = exp.cast("'1970-01-01'", "datetime2(6)", dialect="fabric")
208
209            dateadd = exp.DateAdd(
210                this=epoch_start,
211                expression=rounded_ms_as_bigint,
212                unit=exp.Literal.string("MICROSECONDS"),
213            )
214            return self.sql(dateadd)

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

Arguments:

pretty: Whether 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: Always quote except for specials cases. 'safe': Only quote identifiers that are case insensitive.
normalize: Whether to normalize identifiers to lowercase. Default: False.
pad: The pad size in a formatted string. For example, this affects the indentation of a projection in a query, relative to its nesting level. Default: 2.
indent: The indentation size in a formatted string. For example, this affects the indentation of subqueries and filters under a WHERE clause. Default: 2.
normalize_functions: How to normalize 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: Whether 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 to preserve comments in the output SQL code. Default: True

TYPE_MAPPING = {<Type.DATETIME2: 'DATETIME2'>: 'DATETIME2', <Type.MEDIUMTEXT: 'MEDIUMTEXT'>: 'TEXT', <Type.LONGTEXT: 'LONGTEXT'>: 'TEXT', <Type.TINYTEXT: 'TINYTEXT'>: 'TEXT', <Type.BLOB: 'BLOB'>: 'VARBINARY', <Type.MEDIUMBLOB: 'MEDIUMBLOB'>: 'BLOB', <Type.LONGBLOB: 'LONGBLOB'>: 'BLOB', <Type.TINYBLOB: 'TINYBLOB'>: 'BLOB', <Type.INET: 'INET'>: 'INET', <Type.ROWVERSION: 'ROWVERSION'>: 'ROWVERSION', <Type.SMALLDATETIME: 'SMALLDATETIME'>: 'DATETIME2', <Type.BOOLEAN: 'BOOLEAN'>: 'BIT', <Type.DECIMAL: 'DECIMAL'>: 'DECIMAL', <Type.DOUBLE: 'DOUBLE'>: 'FLOAT', <Type.INT: 'INT'>: 'INT', <Type.TEXT: 'TEXT'>: 'VARCHAR(MAX)', <Type.TIMESTAMP: 'TIMESTAMP'>: 'DATETIME2', <Type.TIMESTAMPNTZ: 'TIMESTAMPNTZ'>: 'DATETIME2', <Type.TIMESTAMPTZ: 'TIMESTAMPTZ'>: 'DATETIME2', <Type.UTINYINT: 'UTINYINT'>: 'SMALLINT', <Type.VARIANT: 'VARIANT'>: 'SQL_VARIANT', <Type.UUID: 'UUID'>: 'UNIQUEIDENTIFIER', <Type.DATETIME: 'DATETIME'>: 'DATETIME2', <Type.IMAGE: 'IMAGE'>: 'VARBINARY', <Type.JSON: 'JSON'>: 'VARCHAR', <Type.MONEY: 'MONEY'>: 'DECIMAL', <Type.NCHAR: 'NCHAR'>: 'CHAR', <Type.NVARCHAR: 'NVARCHAR'>: 'VARCHAR', <Type.SMALLMONEY: 'SMALLMONEY'>: 'DECIMAL', <Type.TINYINT: 'TINYINT'>: 'SMALLINT', <Type.XML: 'XML'>: 'VARCHAR'}

TRANSFORMS = {<class 'sqlglot.expressions.JSONPathKey'>: <function <lambda>>, <class 'sqlglot.expressions.JSONPathRoot'>: <function <lambda>>, <class 'sqlglot.expressions.JSONPathSubscript'>: <function <lambda>>, <class 'sqlglot.expressions.Adjacent'>: <function Generator.<lambda>>, <class 'sqlglot.expressions.AllowedValuesProperty'>: <function Generator.<lambda>>, <class 'sqlglot.expressions.AnalyzeColumns'>: <function Generator.<lambda>>, <class 'sqlglot.expressions.AnalyzeWith'>: <function Generator.<lambda>>, <class 'sqlglot.expressions.ArrayContainsAll'>: <function Generator.<lambda>>, <class 'sqlglot.expressions.ArrayOverlaps'>: <function Generator.<lambda>>, <class 'sqlglot.expressions.AutoRefreshProperty'>: <function Generator.<lambda>>, <class 'sqlglot.expressions.BackupProperty'>: <function Generator.<lambda>>, <class 'sqlglot.expressions.CaseSpecificColumnConstraint'>: <function Generator.<lambda>>, <class 'sqlglot.expressions.Ceil'>: <function rename_func.<locals>.<lambda>>, <class 'sqlglot.expressions.CharacterSetColumnConstraint'>: <function Generator.<lambda>>, <class 'sqlglot.expressions.CharacterSetProperty'>: <function Generator.<lambda>>, <class 'sqlglot.expressions.ClusteredColumnConstraint'>: <function Generator.<lambda>>, <class 'sqlglot.expressions.CollateColumnConstraint'>: <function Generator.<lambda>>, <class 'sqlglot.expressions.CommentColumnConstraint'>: <function Generator.<lambda>>, <class 'sqlglot.expressions.ConnectByRoot'>: <function Generator.<lambda>>, <class 'sqlglot.expressions.ConvertToCharset'>: <function Generator.<lambda>>, <class 'sqlglot.expressions.CopyGrantsProperty'>: <function Generator.<lambda>>, <class 'sqlglot.expressions.CredentialsProperty'>: <function Generator.<lambda>>, <class 'sqlglot.expressions.CurrentCatalog'>: <function Generator.<lambda>>, <class 'sqlglot.expressions.SessionUser'>: <function Generator.<lambda>>, <class 'sqlglot.expressions.DateFormatColumnConstraint'>: <function Generator.<lambda>>, <class 'sqlglot.expressions.DefaultColumnConstraint'>: <function Generator.<lambda>>, <class 'sqlglot.expressions.DynamicProperty'>: <function Generator.<lambda>>, <class 'sqlglot.expressions.EmptyProperty'>: <function Generator.<lambda>>, <class 'sqlglot.expressions.EncodeColumnConstraint'>: <function Generator.<lambda>>, <class 'sqlglot.expressions.EndStatement'>: <function Generator.<lambda>>, <class 'sqlglot.expressions.EnviromentProperty'>: <function Generator.<lambda>>, <class 'sqlglot.expressions.EphemeralColumnConstraint'>: <function Generator.<lambda>>, <class 'sqlglot.expressions.ExcludeColumnConstraint'>: <function Generator.<lambda>>, <class 'sqlglot.expressions.ExecuteAsProperty'>: <function Generator.<lambda>>, <class 'sqlglot.expressions.Except'>: <function Generator.<lambda>>, <class 'sqlglot.expressions.ExternalProperty'>: <function Generator.<lambda>>, <class 'sqlglot.expressions.Floor'>: <function Generator.<lambda>>, <class 'sqlglot.expressions.Get'>: <function Generator.<lambda>>, <class 'sqlglot.expressions.GlobalProperty'>: <function Generator.<lambda>>, <class 'sqlglot.expressions.HeapProperty'>: <function Generator.<lambda>>, <class 'sqlglot.expressions.IcebergProperty'>: <function Generator.<lambda>>, <class 'sqlglot.expressions.InheritsProperty'>: <function Generator.<lambda>>, <class 'sqlglot.expressions.InlineLengthColumnConstraint'>: <function Generator.<lambda>>, <class 'sqlglot.expressions.InputModelProperty'>: <function Generator.<lambda>>, <class 'sqlglot.expressions.Intersect'>: <function Generator.<lambda>>, <class 'sqlglot.expressions.IntervalSpan'>: <function Generator.<lambda>>, <class 'sqlglot.expressions.Int64'>: <function Generator.<lambda>>, <class 'sqlglot.expressions.JSONBContainsAnyTopKeys'>: <function Generator.<lambda>>, <class 'sqlglot.expressions.JSONBContainsAllTopKeys'>: <function Generator.<lambda>>, <class 'sqlglot.expressions.JSONBDeleteAtPath'>: <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.NetFunc'>: <function Generator.<lambda>>, <class 'sqlglot.expressions.NonClusteredColumnConstraint'>: <function Generator.<lambda>>, <class 'sqlglot.expressions.NoPrimaryIndexProperty'>: <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.Operator'>: <function Generator.<lambda>>, <class 'sqlglot.expressions.OutputModelProperty'>: <function Generator.<lambda>>, <class 'sqlglot.expressions.ExtendsLeft'>: <function Generator.<lambda>>, <class 'sqlglot.expressions.ExtendsRight'>: <function Generator.<lambda>>, <class 'sqlglot.expressions.PathColumnConstraint'>: <function Generator.<lambda>>, <class 'sqlglot.expressions.PartitionedByBucket'>: <function Generator.<lambda>>, <class 'sqlglot.expressions.PartitionByTruncate'>: <function Generator.<lambda>>, <class 'sqlglot.expressions.PivotAny'>: <function Generator.<lambda>>, <class 'sqlglot.expressions.PositionalColumn'>: <function Generator.<lambda>>, <class 'sqlglot.expressions.ProjectionPolicyColumnConstraint'>: <function Generator.<lambda>>, <class 'sqlglot.expressions.ZeroFillColumnConstraint'>: <function Generator.<lambda>>, <class 'sqlglot.expressions.Put'>: <function Generator.<lambda>>, <class 'sqlglot.expressions.RemoteWithConnectionModelProperty'>: <function Generator.<lambda>>, <class 'sqlglot.expressions.SafeFunc'>: <function Generator.<lambda>>, <class 'sqlglot.expressions.SampleProperty'>: <function Generator.<lambda>>, <class 'sqlglot.expressions.SecureProperty'>: <function Generator.<lambda>>, <class 'sqlglot.expressions.SecurityProperty'>: <function Generator.<lambda>>, <class 'sqlglot.expressions.SetConfigProperty'>: <function Generator.<lambda>>, <class 'sqlglot.expressions.SetProperty'>: <function Generator.<lambda>>, <class 'sqlglot.expressions.SettingsProperty'>: <function Generator.<lambda>>, <class 'sqlglot.expressions.SharingProperty'>: <function Generator.<lambda>>, <class 'sqlglot.expressions.SqlReadWriteProperty'>: <function Generator.<lambda>>, <class 'sqlglot.expressions.SqlSecurityProperty'>: <function Generator.<lambda>>, <class 'sqlglot.expressions.StabilityProperty'>: <function Generator.<lambda>>, <class 'sqlglot.expressions.Stream'>: <function Generator.<lambda>>, <class 'sqlglot.expressions.StreamingTableProperty'>: <function Generator.<lambda>>, <class 'sqlglot.expressions.StrictProperty'>: <function Generator.<lambda>>, <class 'sqlglot.expressions.SwapTable'>: <function Generator.<lambda>>, <class 'sqlglot.expressions.TableColumn'>: <function Generator.<lambda>>, <class 'sqlglot.expressions.Tags'>: <function Generator.<lambda>>, <class 'sqlglot.expressions.TemporaryProperty'>: <function TSQL.Generator.<lambda>>, <class 'sqlglot.expressions.TitleColumnConstraint'>: <function Generator.<lambda>>, <class 'sqlglot.expressions.ToMap'>: <function Generator.<lambda>>, <class 'sqlglot.expressions.ToTableProperty'>: <function Generator.<lambda>>, <class 'sqlglot.expressions.TransformModelProperty'>: <function Generator.<lambda>>, <class 'sqlglot.expressions.TransientProperty'>: <function Generator.<lambda>>, <class 'sqlglot.expressions.TriggerExecute'>: <function Generator.<lambda>>, <class 'sqlglot.expressions.Union'>: <function Generator.<lambda>>, <class 'sqlglot.expressions.UnloggedProperty'>: <function Generator.<lambda>>, <class 'sqlglot.expressions.UsingTemplateProperty'>: <function Generator.<lambda>>, <class 'sqlglot.expressions.UsingData'>: <function Generator.<lambda>>, <class 'sqlglot.expressions.UppercaseColumnConstraint'>: <function Generator.<lambda>>, <class 'sqlglot.expressions.UtcDate'>: <function Generator.<lambda>>, <class 'sqlglot.expressions.UtcTime'>: <function Generator.<lambda>>, <class 'sqlglot.expressions.UtcTimestamp'>: <function Generator.<lambda>>, <class 'sqlglot.expressions.Variadic'>: <function Generator.<lambda>>, <class 'sqlglot.expressions.VarMap'>: <function Generator.<lambda>>, <class 'sqlglot.expressions.ViewAttributeProperty'>: <function Generator.<lambda>>, <class 'sqlglot.expressions.VolatileProperty'>: <function Generator.<lambda>>, <class 'sqlglot.expressions.WithJournalTableProperty'>: <function Generator.<lambda>>, <class 'sqlglot.expressions.WithProcedureOptions'>: <function Generator.<lambda>>, <class 'sqlglot.expressions.WithSchemaBindingProperty'>: <function Generator.<lambda>>, <class 'sqlglot.expressions.WithOperator'>: <function Generator.<lambda>>, <class 'sqlglot.expressions.ForceProperty'>: <function Generator.<lambda>>, <class 'sqlglot.expressions.AnyValue'>: <function any_value_to_max_sql>, <class 'sqlglot.expressions.Atan2'>: <function rename_func.<locals>.<lambda>>, <class 'sqlglot.expressions.ArrayToString'>: <function rename_func.<locals>.<lambda>>, <class 'sqlglot.expressions.AutoIncrementColumnConstraint'>: <function TSQL.Generator.<lambda>>, <class 'sqlglot.expressions.Chr'>: <function rename_func.<locals>.<lambda>>, <class 'sqlglot.expressions.DateAdd'>: <function date_delta_sql.<locals>._delta_sql>, <class 'sqlglot.expressions.CTE'>: <function preprocess.<locals>._to_sql>, <class 'sqlglot.expressions.CurrentDate'>: <function rename_func.<locals>.<lambda>>, <class 'sqlglot.expressions.CurrentTimestamp'>: <function rename_func.<locals>.<lambda>>, <class 'sqlglot.expressions.CurrentTimestampLTZ'>: <function rename_func.<locals>.<lambda>>, <class 'sqlglot.expressions.DateStrToDate'>: <function datestrtodate_sql>, <class 'sqlglot.expressions.GeneratedAsIdentityColumnConstraint'>: <function generatedasidentitycolumnconstraint_sql>, <class 'sqlglot.expressions.GroupConcat'>: <function _string_agg_sql>, <class 'sqlglot.expressions.If'>: <function rename_func.<locals>.<lambda>>, <class 'sqlglot.expressions.JSONExtract'>: <function _json_extract_sql>, <class 'sqlglot.expressions.JSONExtractScalar'>: <function _json_extract_sql>, <class 'sqlglot.expressions.LastDay'>: <function TSQL.Generator.<lambda>>, <class 'sqlglot.expressions.Ln'>: <function rename_func.<locals>.<lambda>>, <class 'sqlglot.expressions.Max'>: <function max_or_greatest>, <class 'sqlglot.expressions.MD5'>: <function TSQL.Generator.<lambda>>, <class 'sqlglot.expressions.Min'>: <function min_or_least>, <class 'sqlglot.expressions.NumberToStr'>: <function _format_sql>, <class 'sqlglot.expressions.Repeat'>: <function rename_func.<locals>.<lambda>>, <class 'sqlglot.expressions.CurrentSchema'>: <function rename_func.<locals>.<lambda>>, <class 'sqlglot.expressions.Select'>: <function preprocess.<locals>._to_sql>, <class 'sqlglot.expressions.Stddev'>: <function rename_func.<locals>.<lambda>>, <class 'sqlglot.expressions.StrPosition'>: <function TSQL.Generator.<lambda>>, <class 'sqlglot.expressions.Subquery'>: <function preprocess.<locals>._to_sql>, <class 'sqlglot.expressions.SHA'>: <function TSQL.Generator.<lambda>>, <class 'sqlglot.expressions.SHA1Digest'>: <function TSQL.Generator.<lambda>>, <class 'sqlglot.expressions.SHA2'>: <function TSQL.Generator.<lambda>>, <class 'sqlglot.expressions.TimeStrToTime'>: <function _timestrtotime_sql>, <class 'sqlglot.expressions.TimeToStr'>: <function _format_sql>, <class 'sqlglot.expressions.Trim'>: <function trim_sql>, <class 'sqlglot.expressions.TsOrDsAdd'>: <function date_delta_sql.<locals>._delta_sql>, <class 'sqlglot.expressions.TsOrDsDiff'>: <function date_delta_sql.<locals>._delta_sql>, <class 'sqlglot.expressions.TimestampTrunc'>: <function TSQL.Generator.<lambda>>, <class 'sqlglot.expressions.Trunc'>: <function TSQL.Generator.<lambda>>, <class 'sqlglot.expressions.Uuid'>: <function TSQL.Generator.<lambda>>, <class 'sqlglot.expressions.DateFromParts'>: <function rename_func.<locals>.<lambda>>, <class 'sqlglot.expressions.Create'>: <function preprocess.<locals>._to_sql>}

def datatype_sql(self, expression: sqlglot.expressions.DataType) -> str: View Source

138        def datatype_sql(self, expression: exp.DataType) -> str:
139            # Check if this is a temporal type that needs precision handling. Fabric limits temporal
140            # types to max 6 digits precision. When no precision is specified, we default to 6 digits.
141            if (
142                expression.is_type(*exp.DataType.TEMPORAL_TYPES)
143                and expression.this != exp.DataType.Type.DATE
144            ):
145                # Create a new expression with the capped precision
146                expression = _cap_data_type_precision(expression)
147
148            return super().datatype_sql(expression)

def cast_sql( self, expression: sqlglot.expressions.Cast, safe_prefix: str | None = None) -> str: View Source

150        def cast_sql(self, expression: exp.Cast, safe_prefix: str | None = None) -> str:
151            # Cast to DATETIMEOFFSET if inside an AT TIME ZONE expression
152            # https://learn.microsoft.com/en-us/sql/t-sql/data-types/datetimeoffset-transact-sql#microsoft-fabric-support
153            if expression.is_type(exp.DataType.Type.TIMESTAMPTZ):
154                at_time_zone = expression.find_ancestor(exp.AtTimeZone, exp.Select)
155
156                # Return normal cast, if the expression is not in an AT TIME ZONE context
157                if not isinstance(at_time_zone, exp.AtTimeZone):
158                    return super().cast_sql(expression, safe_prefix)
159
160                # Get the precision from the original TIMESTAMPTZ cast and cap it to 6
161                capped_data_type = _cap_data_type_precision(expression.to, max_precision=6)
162                precision = capped_data_type.find(exp.DataTypeParam)
163                precision_value = (
164                    precision.this.to_py() if precision and precision.this.is_int else 6
165                )
166
167                # Do the cast explicitly to bypass sqlglot's default handling
168                datetimeoffset = f"CAST({expression.this} AS DATETIMEOFFSET({precision_value}))"
169
170                return self.sql(datetimeoffset)
171
172            return super().cast_sql(expression, safe_prefix)

def attimezone_sql(self, expression: sqlglot.expressions.AtTimeZone) -> str: View Source

174        def attimezone_sql(self, expression: exp.AtTimeZone) -> str:
175            # Wrap the AT TIME ZONE expression in a cast to DATETIME2 if it contains a TIMESTAMPTZ
176            ## https://learn.microsoft.com/en-us/sql/t-sql/data-types/datetimeoffset-transact-sql#microsoft-fabric-support
177            timestamptz_cast = expression.find(exp.Cast)
178            if timestamptz_cast and timestamptz_cast.to.is_type(exp.DataType.Type.TIMESTAMPTZ):
179                # Get the precision from the original TIMESTAMPTZ cast and cap it to 6
180                data_type = timestamptz_cast.to
181                capped_data_type = _cap_data_type_precision(data_type, max_precision=6)
182                precision_param = capped_data_type.find(exp.DataTypeParam)
183                precision = precision_param.this.to_py() if precision_param else 6
184
185                # Generate the AT TIME ZONE expression (which will handle the inner cast conversion)
186                at_time_zone_sql = super().attimezone_sql(expression)
187
188                # Wrap it in an outer cast to DATETIME2
189                return f"CAST({at_time_zone_sql} AS DATETIME2({precision}))"
190
191            return super().attimezone_sql(expression)

def unixtotime_sql(self, expression: sqlglot.expressions.UnixToTime) -> str: View Source

193        def unixtotime_sql(self, expression: exp.UnixToTime) -> str:
194            scale = expression.args.get("scale")
195            timestamp = expression.this
196
197            if scale not in (None, exp.UnixToTime.SECONDS):
198                self.unsupported(f"UnixToTime scale {scale} is not supported by Fabric")
199                return ""
200
201            # Convert unix timestamp (seconds) to microseconds and round to avoid decimals
202            microseconds = timestamp * exp.Literal.number("1e6")
203            rounded = exp.func("round", microseconds, 0)
204            rounded_ms_as_bigint = exp.cast(rounded, exp.DataType.Type.BIGINT)
205
206            # Create the base datetime as '1970-01-01' cast to DATETIME2(6)
207            epoch_start = exp.cast("'1970-01-01'", "datetime2(6)", dialect="fabric")
208
209            dateadd = exp.DateAdd(
210                this=epoch_start,
211                expression=rounded_ms_as_bigint,
212                unit=exp.Literal.string("MICROSECONDS"),
213            )
214            return self.sql(dateadd)

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

TRY_SUPPORTED = False

SUPPORTS_UESCAPE = False

SUPPORTS_DECODE_CASE = False

AFTER_HAVING_MODIFIER_TRANSFORMS = {'windows': <function Generator.<lambda>>, 'qualify': <function Generator.<lambda>>}

Inherited Members

sqlglot.generator.Generator: Generator; IGNORE_NULLS_IN_FUNC; LOCKING_READS_SUPPORTED; WRAP_DERIVED_VALUES; CREATE_FUNCTION_RETURN_AS; MATCHED_BY_SOURCE; SINGLE_STRING_INTERVAL; INTERVAL_ALLOWS_PLURAL_FORM; LIMIT_ONLY_LITERALS; RENAME_TABLE_WITH_DB; GROUPINGS_SEP; INDEX_ON; INOUT_SEPARATOR; JOIN_HINTS; DIRECTED_JOINS; TABLE_HINTS; QUERY_HINT_SEP; IS_BOOL_ALLOWED; DUPLICATE_KEY_UPDATE_WITH_SET; EXTRACT_ALLOWS_QUOTES; TZ_TO_WITH_TIME_ZONE; VALUES_AS_TABLE; UNNEST_WITH_ORDINALITY; AGGREGATE_FILTER_SUPPORTED; SEMI_ANTI_JOIN_WITH_SIDE; SUPPORTS_TABLE_COPY; TABLESAMPLE_REQUIRES_PARENS; TABLESAMPLE_SIZE_IS_ROWS; TABLESAMPLE_KEYWORDS; TABLESAMPLE_WITH_METHOD; COLLATE_IS_FUNC; DATA_TYPE_SPECIFIERS_ALLOWED; LAST_DAY_SUPPORTS_DATE_PART; SUPPORTS_TABLE_ALIAS_COLUMNS; UNPIVOT_ALIASES_ARE_IDENTIFIERS; JSON_KEY_VALUE_PAIR_SEP; INSERT_OVERWRITE; SUPPORTS_UNLOGGED_TABLES; SUPPORTS_CREATE_TABLE_LIKE; LIKE_PROPERTY_INSIDE_SCHEMA; MULTI_ARG_DISTINCT; JSON_TYPE_REQUIRED_FOR_EXTRACTION; JSON_PATH_SINGLE_QUOTE_ESCAPE; CAN_IMPLEMENT_ARRAY_ANY; SUPPORTS_WINDOW_EXCLUDE; COPY_PARAMS_ARE_WRAPPED; COPY_HAS_INTO_KEYWORD; UNICODE_SUBSTITUTE; STAR_EXCEPT; HEX_FUNC; WITH_PROPERTIES_PREFIX; QUOTE_JSON_PATH; PAD_FILL_PATTERN_IS_REQUIRED; SUPPORTS_EXPLODING_PROJECTIONS; ARRAY_CONCAT_IS_VAR_LEN; SUPPORTS_CONVERT_TIMEZONE; SUPPORTS_MEDIAN; SUPPORTS_UNIX_SECONDS; NORMALIZE_EXTRACT_DATE_PARTS; ARRAY_SIZE_NAME; ARRAY_SIZE_DIM_REQUIRED; SUPPORTS_BETWEEN_FLAGS; SUPPORTS_LIKE_QUANTIFIERS; MATCH_AGAINST_TABLE_PREFIX; SET_ASSIGNMENT_REQUIRES_VARIABLE_KEYWORD; DECLARE_DEFAULT_ASSIGNMENT; UPDATE_STATEMENT_SUPPORTS_FROM; STAR_EXCLUDE_REQUIRES_DERIVED_TABLE; UNSUPPORTED_TYPES; TIME_PART_SINGULARS; TOKEN_MAPPING; STRUCT_DELIMITER; PARAMETER_TOKEN; NAMED_PLACEHOLDER_TOKEN; EXPRESSION_PRECEDES_PROPERTIES_CREATABLES; RESERVED_KEYWORDS; WITH_SEPARATED_COMMENTS; EXCLUDE_COMMENTS; UNWRAPPED_INTERVAL_VALUES; PARAMETERIZABLE_TEXT_TYPES; RESPECT_IGNORE_NULLS_UNSUPPORTED_EXPRESSIONS; SAFE_JSON_PATH_KEY_RE; SENTINEL_LINE_BREAK; pretty; identify; normalize; pad; unsupported_level; max_unsupported; leading_comma; max_text_width; comments; dialect; normalize_functions; unsupported_messages; generate; preprocess; unsupported; sep; seg; sanitize_comment; maybe_comment; wrap; no_identify; normalize_func; indent; sql; uncache_sql; cache_sql; characterset_sql; column_parts; column_sql; pseudocolumn_sql; columnposition_sql; columnconstraint_sql; computedcolumnconstraint_sql; autoincrementcolumnconstraint_sql; compresscolumnconstraint_sql; generatedasidentitycolumnconstraint_sql; generatedasrowcolumnconstraint_sql; periodforsystemtimeconstraint_sql; notnullcolumnconstraint_sql; primarykeycolumnconstraint_sql; uniquecolumnconstraint_sql; inoutcolumnconstraint_sql; sequenceproperties_sql; triggerproperties_sql; triggerreferencing_sql; triggerevent_sql; clone_sql; describe_sql; heredoc_sql; prepend_ctes; with_sql; cte_sql; tablealias_sql; bitstring_sql; hexstring_sql; bytestring_sql; unicodestring_sql; rawstring_sql; datatypeparam_sql; directory_sql; delete_sql; set_operation; set_operations; fetch_sql; limitoptions_sql; filter_sql; hint_sql; indexparameters_sql; index_sql; hex_sql; lowerhex_sql; inputoutputformat_sql; national_sql; properties_sql; root_properties; properties; with_properties; locate_properties; property_name; property_sql; likeproperty_sql; fallbackproperty_sql; journalproperty_sql; freespaceproperty_sql; checksumproperty_sql; mergeblockratioproperty_sql; datablocksizeproperty_sql; blockcompressionproperty_sql; isolatedloadingproperty_sql; partitionboundspec_sql; partitionedofproperty_sql; lockingproperty_sql; withdataproperty_sql; withsystemversioningproperty_sql; insert_sql; introducer_sql; kill_sql; pseudotype_sql; objectidentifier_sql; onconflict_sql; rowformatdelimitedproperty_sql; withtablehint_sql; indextablehint_sql; historicaldata_sql; table_parts; table_sql; tablefromrows_sql; tablesample_sql; pivot_sql; tuple_sql; update_sql; values_sql; var_sql; from_sql; groupingsets_sql; rollup_sql; rollupindex_sql; rollupproperty_sql; cube_sql; group_sql; having_sql; connect_sql; prior_sql; join_sql; lambda_sql; lateral_sql; limit_sql; set_sql; queryband_sql; pragma_sql; lock_sql; literal_sql; escape_str; loaddata_sql; null_sql; booland_sql; boolor_sql; order_sql; withfill_sql; cluster_sql; distribute_sql; sort_sql; ordered_sql; matchrecognizemeasure_sql; matchrecognize_sql; query_modifiers; for_modifiers; offset_limit_modifiers; after_limit_modifiers; schema_sql; schema_columns_sql; star_sql; parameter_sql; sessionparameter_sql; placeholder_sql; subquery_sql; qualify_sql; unnest_sql; prewhere_sql; where_sql; window_sql; partition_by_sql; windowspec_sql; withingroup_sql; between_sql; bracket_offset_expressions; bracket_sql; all_sql; any_sql; exists_sql; case_sql; nextvaluefor_sql; trim_sql; convert_concat_args; concat_sql; concatws_sql; check_sql; foreignkey_sql; primarykey_sql; if_sql; matchagainst_sql; jsonkeyvalue_sql; jsonpath_sql; json_path_part; formatjson_sql; formatphrase_sql; jsonobject_sql; jsonobjectagg_sql; jsonarray_sql; jsonarrayagg_sql; jsoncolumndef_sql; jsonschema_sql; jsontable_sql; openjsoncolumndef_sql; openjson_sql; in_sql; in_unnest_op; interval_sql; return_sql; reference_sql; anonymous_sql; paren_sql; neg_sql; not_sql; alias_sql; pivotalias_sql; aliases_sql; atindex_sql; fromtimezone_sql; add_sql; and_sql; or_sql; xor_sql; connector_sql; bitwiseand_sql; bitwiseleftshift_sql; bitwisenot_sql; bitwiseor_sql; bitwiserightshift_sql; bitwisexor_sql; strtotime_sql; currentdate_sql; collate_sql; command_sql; comment_sql; mergetreettlaction_sql; mergetreettl_sql; altercolumn_sql; alterindex_sql; alterdiststyle_sql; altersortkey_sql; alterrename_sql; renamecolumn_sql; alterset_sql; altersession_sql; add_column_sql; droppartition_sql; addconstraint_sql; addpartition_sql; distinct_sql; ignorenulls_sql; respectnulls_sql; havingmax_sql; intdiv_sql; div_sql; safedivide_sql; overlaps_sql; distance_sql; dot_sql; eq_sql; propertyeq_sql; escape_sql; glob_sql; gt_sql; gte_sql; like_sql; ilike_sql; match_sql; similarto_sql; lt_sql; lte_sql; mod_sql; mul_sql; neq_sql; nullsafeeq_sql; nullsafeneq_sql; sub_sql; trycast_sql; jsoncast_sql; try_sql; log_sql; use_sql; binary; ceil_floor; function_fallback_sql; func; format_args; too_wide; format_time; expressions; op_expressions; naked_property; tag_sql; token_sql; userdefinedfunction_sql; joinhint_sql; kwarg_sql; when_sql; whens_sql; merge_sql; tochar_sql; tonumber_sql; dictproperty_sql; dictrange_sql; dictsubproperty_sql; duplicatekeyproperty_sql; uniquekeyproperty_sql; distributedbyproperty_sql; oncluster_sql; clusteredbyproperty_sql; anyvalue_sql; querytransform_sql; indexconstraintoption_sql; checkcolumnconstraint_sql; indexcolumnconstraint_sql; nvl2_sql; comprehension_sql; columnprefix_sql; opclass_sql; predict_sql; generateembedding_sql; mltranslate_sql; mlforecast_sql; featuresattime_sql; vectorsearch_sql; forin_sql; refresh_sql; toarray_sql; tsordstotime_sql; tsordstotimestamp_sql; tsordstodatetime_sql; tsordstodate_sql; unixdate_sql; lastday_sql; dateadd_sql; arrayany_sql; struct_sql; partitionrange_sql; truncatetable_sql; copyparameter_sql; credentials_sql; copy_sql; semicolon_sql; datadeletionproperty_sql; maskingpolicycolumnconstraint_sql; gapfill_sql; scoperesolution_sql; parsejson_sql; rand_sql; changes_sql; pad_sql; summarize_sql; explodinggenerateseries_sql; converttimezone_sql; json_sql; jsonvalue_sql; conditionalinsert_sql; multitableinserts_sql; oncondition_sql; jsonextractquote_sql; jsonexists_sql; arrayagg_sql; slice_sql; apply_sql; grant_sql; revoke_sql; grantprivilege_sql; grantprincipal_sql; columns_sql; overlay_sql; todouble_sql; string_sql; median_sql; overflowtruncatebehavior_sql; unixseconds_sql; arraysize_sql; attach_sql; detach_sql; attachoption_sql; watermarkcolumnconstraint_sql; encodeproperty_sql; includeproperty_sql; xmlelement_sql; xmlkeyvalueoption_sql; partitionbyrangeproperty_sql; partitionbyrangepropertydynamic_sql; unpivotcolumns_sql; analyzesample_sql; analyzestatistics_sql; analyzehistogram_sql; analyzedelete_sql; analyzelistchainedrows_sql; analyzevalidate_sql; analyze_sql; xmltable_sql; xmlnamespace_sql; export_sql; declare_sql; declareitem_sql; recursivewithsearch_sql; parameterizedagg_sql; anonymousaggfunc_sql; combinedaggfunc_sql; combinedparameterizedagg_sql; show_sql; install_sql; get_put_sql; translatecharacters_sql; decodecase_sql; semanticview_sql; getextract_sql; datefromunixdate_sql; space_sql; buildproperty_sql; refreshtriggerproperty_sql; modelattribute_sql; directorystage_sql; uuid_sql; initcap_sql; localtime_sql; localtimestamp_sql; weekstart_sql; chr_sql; block_sql
sqlglot.dialects.tsql.TSQL.Generator: LIMIT_IS_TOP; QUERY_HINTS; RETURNING_END; NVL2_SUPPORTED; ALTER_TABLE_INCLUDE_COLUMN_KEYWORD; LIMIT_FETCH; COMPUTED_COLUMN_WITH_TYPE; CTE_RECURSIVE_KEYWORD_REQUIRED; ENSURE_BOOLS; NULL_ORDERING_SUPPORTED; SUPPORTS_SINGLE_ARG_CONCAT; TABLESAMPLE_SEED_KEYWORD; SUPPORTS_SELECT_INTO; JSON_PATH_BRACKETED_KEY_SUPPORTED; SUPPORTS_TO_NUMBER; SET_OP_MODIFIERS; COPY_PARAMS_EQ_REQUIRED; PARSE_JSON_NAME; EXCEPT_INTERSECT_SUPPORT_ALL_CLAUSE; ALTER_SET_WRAPPED; ALTER_SET_TYPE; EXPRESSIONS_WITHOUT_NESTED_CTES; SUPPORTED_JSON_PATH_PARTS; PROPERTIES_LOCATION; scope_resolution; select_sql; convert_sql; queryoption_sql; lateral_op; splitpart_sql; extract_sql; timefromparts_sql; timestampfromparts_sql; setitem_sql; boolean_sql; is_sql; createable_sql; create_sql; into_sql; count_sql; datediff_sql; offset_sql; version_sql; returnsproperty_sql; returning_sql; transaction_sql; commit_sql; rollback_sql; identifier_sql; constraint_sql; length_sql; right_sql; left_sql; partition_sql; alter_sql; drop_sql; options_modifier; dpipe_sql; isascii_sql; columndef_sql; coalesce_sql; storedprocedure_sql; ifblock_sql; whileblock_sql; execute_sql; executesql_sql