Edit on GitHub

sqlglot.dialects.sqlite

View Source

  1from __future__ import annotations
  2
  3from sqlglot import exp, generator, parser, tokens
  4from sqlglot.dialects.dialect import (
  5    Dialect,
  6    arrow_json_extract_scalar_sql,
  7    arrow_json_extract_sql,
  8    no_ilike_sql,
  9    no_tablesample_sql,
 10    no_trycast_sql,
 11    rename_func,
 12)
 13from sqlglot.tokens import TokenType
 14
 15
 16# https://www.sqlite.org/lang_aggfunc.html#group_concat
 17def _group_concat_sql(self, expression):
 18    this = expression.this
 19    distinct = expression.find(exp.Distinct)
 20    if distinct:
 21        this = distinct.expressions[0]
 22        distinct = "DISTINCT "
 23
 24    if isinstance(expression.this, exp.Order):
 25        self.unsupported("SQLite GROUP_CONCAT doesn't support ORDER BY.")
 26        if expression.this.this and not distinct:
 27            this = expression.this.this
 28
 29    separator = expression.args.get("separator")
 30    return f"GROUP_CONCAT({distinct or ''}{self.format_args(this, separator)})"
 31
 32
 33def _date_add_sql(self, expression):
 34    modifier = expression.expression
 35    modifier = expression.name if modifier.is_string else self.sql(modifier)
 36    unit = expression.args.get("unit")
 37    modifier = f"'{modifier} {unit.name}'" if unit else f"'{modifier}'"
 38    return self.func("DATE", expression.this, modifier)
 39
 40
 41class SQLite(Dialect):
 42    class Tokenizer(tokens.Tokenizer):
 43        IDENTIFIERS = ['"', ("[", "]"), "`"]
 44        HEX_STRINGS = [("x'", "'"), ("X'", "'"), ("0x", ""), ("0X", "")]
 45
 46        KEYWORDS = {
 47            **tokens.Tokenizer.KEYWORDS,
 48        }
 49
 50    class Parser(parser.Parser):
 51        FUNCTIONS = {
 52            **parser.Parser.FUNCTIONS,  # type: ignore
 53            "EDITDIST3": exp.Levenshtein.from_arg_list,
 54        }
 55
 56    class Generator(generator.Generator):
 57        TYPE_MAPPING = {
 58            **generator.Generator.TYPE_MAPPING,  # type: ignore
 59            exp.DataType.Type.BOOLEAN: "INTEGER",
 60            exp.DataType.Type.TINYINT: "INTEGER",
 61            exp.DataType.Type.SMALLINT: "INTEGER",
 62            exp.DataType.Type.INT: "INTEGER",
 63            exp.DataType.Type.BIGINT: "INTEGER",
 64            exp.DataType.Type.FLOAT: "REAL",
 65            exp.DataType.Type.DOUBLE: "REAL",
 66            exp.DataType.Type.DECIMAL: "REAL",
 67            exp.DataType.Type.CHAR: "TEXT",
 68            exp.DataType.Type.NCHAR: "TEXT",
 69            exp.DataType.Type.VARCHAR: "TEXT",
 70            exp.DataType.Type.NVARCHAR: "TEXT",
 71            exp.DataType.Type.BINARY: "BLOB",
 72            exp.DataType.Type.VARBINARY: "BLOB",
 73        }
 74
 75        TOKEN_MAPPING = {
 76            TokenType.AUTO_INCREMENT: "AUTOINCREMENT",
 77        }
 78
 79        TRANSFORMS = {
 80            **generator.Generator.TRANSFORMS,  # type: ignore
 81            exp.DateAdd: _date_add_sql,
 82            exp.ILike: no_ilike_sql,
 83            exp.JSONExtract: arrow_json_extract_sql,
 84            exp.JSONExtractScalar: arrow_json_extract_scalar_sql,
 85            exp.JSONBExtract: arrow_json_extract_sql,
 86            exp.JSONBExtractScalar: arrow_json_extract_scalar_sql,
 87            exp.Levenshtein: rename_func("EDITDIST3"),
 88            exp.TableSample: no_tablesample_sql,
 89            exp.DateStrToDate: lambda self, e: self.sql(e, "this"),
 90            exp.TimeStrToTime: lambda self, e: self.sql(e, "this"),
 91            exp.TryCast: no_trycast_sql,
 92            exp.GroupConcat: _group_concat_sql,
 93        }
 94
 95        def datediff_sql(self, expression: exp.DateDiff) -> str:
 96            unit = expression.args.get("unit")
 97            unit = unit.name.upper() if unit else "DAY"
 98
 99            sql = f"(JULIANDAY({self.sql(expression, 'this')}) - JULIANDAY({self.sql(expression, 'expression')}))"
100
101            if unit == "MONTH":
102                sql = f"{sql} / 30.0"
103            elif unit == "YEAR":
104                sql = f"{sql} / 365.0"
105            elif unit == "HOUR":
106                sql = f"{sql} * 24.0"
107            elif unit == "MINUTE":
108                sql = f"{sql} * 1440.0"
109            elif unit == "SECOND":
110                sql = f"{sql} * 86400.0"
111            elif unit == "MILLISECOND":
112                sql = f"{sql} * 86400000.0"
113            elif unit == "MICROSECOND":
114                sql = f"{sql} * 86400000000.0"
115            elif unit == "NANOSECOND":
116                sql = f"{sql} * 8640000000000.0"
117            else:
118                self.unsupported("DATEDIFF unsupported for '{unit}'.")
119
120            return f"CAST({sql} AS INTEGER)"
121
122        def fetch_sql(self, expression):
123            return self.limit_sql(exp.Limit(expression=expression.args.get("count")))
124
125        def least_sql(self, expression):
126            if len(expression.expressions) > 1:
127                return rename_func("MIN")(self, expression)
128
129            return self.expressions(expression)
130
131        def transaction_sql(self, expression):
132            this = expression.this
133            this = f" {this}" if this else ""
134            return f"BEGIN{this} TRANSACTION"

class SQLite(sqlglot.dialects.dialect.Dialect): View Source

 42class SQLite(Dialect):
 43    class Tokenizer(tokens.Tokenizer):
 44        IDENTIFIERS = ['"', ("[", "]"), "`"]
 45        HEX_STRINGS = [("x'", "'"), ("X'", "'"), ("0x", ""), ("0X", "")]
 46
 47        KEYWORDS = {
 48            **tokens.Tokenizer.KEYWORDS,
 49        }
 50
 51    class Parser(parser.Parser):
 52        FUNCTIONS = {
 53            **parser.Parser.FUNCTIONS,  # type: ignore
 54            "EDITDIST3": exp.Levenshtein.from_arg_list,
 55        }
 56
 57    class Generator(generator.Generator):
 58        TYPE_MAPPING = {
 59            **generator.Generator.TYPE_MAPPING,  # type: ignore
 60            exp.DataType.Type.BOOLEAN: "INTEGER",
 61            exp.DataType.Type.TINYINT: "INTEGER",
 62            exp.DataType.Type.SMALLINT: "INTEGER",
 63            exp.DataType.Type.INT: "INTEGER",
 64            exp.DataType.Type.BIGINT: "INTEGER",
 65            exp.DataType.Type.FLOAT: "REAL",
 66            exp.DataType.Type.DOUBLE: "REAL",
 67            exp.DataType.Type.DECIMAL: "REAL",
 68            exp.DataType.Type.CHAR: "TEXT",
 69            exp.DataType.Type.NCHAR: "TEXT",
 70            exp.DataType.Type.VARCHAR: "TEXT",
 71            exp.DataType.Type.NVARCHAR: "TEXT",
 72            exp.DataType.Type.BINARY: "BLOB",
 73            exp.DataType.Type.VARBINARY: "BLOB",
 74        }
 75
 76        TOKEN_MAPPING = {
 77            TokenType.AUTO_INCREMENT: "AUTOINCREMENT",
 78        }
 79
 80        TRANSFORMS = {
 81            **generator.Generator.TRANSFORMS,  # type: ignore
 82            exp.DateAdd: _date_add_sql,
 83            exp.ILike: no_ilike_sql,
 84            exp.JSONExtract: arrow_json_extract_sql,
 85            exp.JSONExtractScalar: arrow_json_extract_scalar_sql,
 86            exp.JSONBExtract: arrow_json_extract_sql,
 87            exp.JSONBExtractScalar: arrow_json_extract_scalar_sql,
 88            exp.Levenshtein: rename_func("EDITDIST3"),
 89            exp.TableSample: no_tablesample_sql,
 90            exp.DateStrToDate: lambda self, e: self.sql(e, "this"),
 91            exp.TimeStrToTime: lambda self, e: self.sql(e, "this"),
 92            exp.TryCast: no_trycast_sql,
 93            exp.GroupConcat: _group_concat_sql,
 94        }
 95
 96        def datediff_sql(self, expression: exp.DateDiff) -> str:
 97            unit = expression.args.get("unit")
 98            unit = unit.name.upper() if unit else "DAY"
 99
100            sql = f"(JULIANDAY({self.sql(expression, 'this')}) - JULIANDAY({self.sql(expression, 'expression')}))"
101
102            if unit == "MONTH":
103                sql = f"{sql} / 30.0"
104            elif unit == "YEAR":
105                sql = f"{sql} / 365.0"
106            elif unit == "HOUR":
107                sql = f"{sql} * 24.0"
108            elif unit == "MINUTE":
109                sql = f"{sql} * 1440.0"
110            elif unit == "SECOND":
111                sql = f"{sql} * 86400.0"
112            elif unit == "MILLISECOND":
113                sql = f"{sql} * 86400000.0"
114            elif unit == "MICROSECOND":
115                sql = f"{sql} * 86400000000.0"
116            elif unit == "NANOSECOND":
117                sql = f"{sql} * 8640000000000.0"
118            else:
119                self.unsupported("DATEDIFF unsupported for '{unit}'.")
120
121            return f"CAST({sql} AS INTEGER)"
122
123        def fetch_sql(self, expression):
124            return self.limit_sql(exp.Limit(expression=expression.args.get("count")))
125
126        def least_sql(self, expression):
127            if len(expression.expressions) > 1:
128                return rename_func("MIN")(self, expression)
129
130            return self.expressions(expression)
131
132        def transaction_sql(self, expression):
133            this = expression.this
134            this = f" {this}" if this else ""
135            return f"BEGIN{this} TRANSACTION"

Inherited Members

sqlglot.dialects.dialect.Dialect: get_or_raise; format_time; parse; parse_into; generate; transpile; parser; generator

class SQLite.Tokenizer(sqlglot.tokens.Tokenizer): View Source

43    class Tokenizer(tokens.Tokenizer):
44        IDENTIFIERS = ['"', ("[", "]"), "`"]
45        HEX_STRINGS = [("x'", "'"), ("X'", "'"), ("0x", ""), ("0X", "")]
46
47        KEYWORDS = {
48            **tokens.Tokenizer.KEYWORDS,
49        }

Inherited Members

sqlglot.tokens.Tokenizer: reset; tokenize

class SQLite.Parser(sqlglot.parser.Parser): View Source

51    class Parser(parser.Parser):
52        FUNCTIONS = {
53            **parser.Parser.FUNCTIONS,  # type: ignore
54            "EDITDIST3": exp.Levenshtein.from_arg_list,
55        }

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

Arguments:

error_level: the desired error level. Default: ErrorLevel.RAISE
error_message_context: determines the amount of context to capture from a query string when displaying the error message (in number of characters). Default: 50.
index_offset: Index offset for arrays eg ARRAY[0] vs ARRAY[1] as the head of a list. Default: 0
alias_post_tablesample: If the table alias comes after tablesample. Default: False
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
null_ordering: Indicates the default null ordering method to use if not explicitly set. Options are "nulls_are_small", "nulls_are_large", "nulls_are_last". Default: "nulls_are_small"

Inherited Members

sqlglot.parser.Parser: Parser; reset; parse; parse_into; check_errors; raise_error; expression; validate_expression

class SQLite.Generator(sqlglot.generator.Generator): View Source

 57    class Generator(generator.Generator):
 58        TYPE_MAPPING = {
 59            **generator.Generator.TYPE_MAPPING,  # type: ignore
 60            exp.DataType.Type.BOOLEAN: "INTEGER",
 61            exp.DataType.Type.TINYINT: "INTEGER",
 62            exp.DataType.Type.SMALLINT: "INTEGER",
 63            exp.DataType.Type.INT: "INTEGER",
 64            exp.DataType.Type.BIGINT: "INTEGER",
 65            exp.DataType.Type.FLOAT: "REAL",
 66            exp.DataType.Type.DOUBLE: "REAL",
 67            exp.DataType.Type.DECIMAL: "REAL",
 68            exp.DataType.Type.CHAR: "TEXT",
 69            exp.DataType.Type.NCHAR: "TEXT",
 70            exp.DataType.Type.VARCHAR: "TEXT",
 71            exp.DataType.Type.NVARCHAR: "TEXT",
 72            exp.DataType.Type.BINARY: "BLOB",
 73            exp.DataType.Type.VARBINARY: "BLOB",
 74        }
 75
 76        TOKEN_MAPPING = {
 77            TokenType.AUTO_INCREMENT: "AUTOINCREMENT",
 78        }
 79
 80        TRANSFORMS = {
 81            **generator.Generator.TRANSFORMS,  # type: ignore
 82            exp.DateAdd: _date_add_sql,
 83            exp.ILike: no_ilike_sql,
 84            exp.JSONExtract: arrow_json_extract_sql,
 85            exp.JSONExtractScalar: arrow_json_extract_scalar_sql,
 86            exp.JSONBExtract: arrow_json_extract_sql,
 87            exp.JSONBExtractScalar: arrow_json_extract_scalar_sql,
 88            exp.Levenshtein: rename_func("EDITDIST3"),
 89            exp.TableSample: no_tablesample_sql,
 90            exp.DateStrToDate: lambda self, e: self.sql(e, "this"),
 91            exp.TimeStrToTime: lambda self, e: self.sql(e, "this"),
 92            exp.TryCast: no_trycast_sql,
 93            exp.GroupConcat: _group_concat_sql,
 94        }
 95
 96        def datediff_sql(self, expression: exp.DateDiff) -> str:
 97            unit = expression.args.get("unit")
 98            unit = unit.name.upper() if unit else "DAY"
 99
100            sql = f"(JULIANDAY({self.sql(expression, 'this')}) - JULIANDAY({self.sql(expression, 'expression')}))"
101
102            if unit == "MONTH":
103                sql = f"{sql} / 30.0"
104            elif unit == "YEAR":
105                sql = f"{sql} / 365.0"
106            elif unit == "HOUR":
107                sql = f"{sql} * 24.0"
108            elif unit == "MINUTE":
109                sql = f"{sql} * 1440.0"
110            elif unit == "SECOND":
111                sql = f"{sql} * 86400.0"
112            elif unit == "MILLISECOND":
113                sql = f"{sql} * 86400000.0"
114            elif unit == "MICROSECOND":
115                sql = f"{sql} * 86400000000.0"
116            elif unit == "NANOSECOND":
117                sql = f"{sql} * 8640000000000.0"
118            else:
119                self.unsupported("DATEDIFF unsupported for '{unit}'.")
120
121            return f"CAST({sql} AS INTEGER)"
122
123        def fetch_sql(self, expression):
124            return self.limit_sql(exp.Limit(expression=expression.args.get("count")))
125
126        def least_sql(self, expression):
127            if len(expression.expressions) > 1:
128                return rename_func("MIN")(self, expression)
129
130            return self.expressions(expression)
131
132        def transaction_sql(self, expression):
133            this = expression.this
134            this = f" {this}" if this else ""
135            return f"BEGIN{this} TRANSACTION"

Generator interprets the given syntax tree and produces a SQL string as an output.

Arguments:

time_mapping (dict): the dictionary of custom time mappings in which the key represents a python time format and the output the target time format
time_trie (trie): a trie of the time_mapping keys
pretty (bool): if set to True the returned string will be formatted. Default: False.
quote_start (str): specifies which starting character to use to delimit quotes. Default: '.
quote_end (str): specifies which ending character to use to delimit quotes. Default: '.
identifier_start (str): specifies which starting character to use to delimit identifiers. Default: ".
identifier_end (str): specifies which ending character to use to delimit identifiers. Default: ".
identify (bool): if set to True all identifiers will be delimited by the corresponding character.
normalize (bool): if set to True all identifiers will lower cased
string_escape (str): specifies a string escape character. Default: '.
identifier_escape (str): specifies an identifier escape character. Default: ".
pad (int): determines padding in a formatted string. Default: 2.
indent (int): determines the size of indentation in a formatted string. Default: 4.
unnest_column_only (bool): if true unnest table aliases are considered only as column aliases
normalize_functions (str): normalize function names, "upper", "lower", or None Default: "upper"
alias_post_tablesample (bool): if the table alias comes after tablesample Default: False
unsupported_level (ErrorLevel): determines the generator's behavior when it encounters unsupported expressions. Default ErrorLevel.WARN.
null_ordering (str): Indicates the default null ordering method to use if not explicitly set. Options are "nulls_are_small", "nulls_are_large", "nulls_are_last". Default: "nulls_are_small"
max_unsupported (int): 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 (bool): if the the comma is leading or trailing in select statements Default: False
max_text_width: The max number of characters in a segment before creating new lines in pretty mode. The default is on the smaller end because the length only represents a segment and not the true line length. Default: 80
comments: Whether or not to preserve comments in the output SQL code. Default: True

def datediff_sql(self, expression: sqlglot.expressions.DateDiff) -> str: View Source

 96        def datediff_sql(self, expression: exp.DateDiff) -> str:
 97            unit = expression.args.get("unit")
 98            unit = unit.name.upper() if unit else "DAY"
 99
100            sql = f"(JULIANDAY({self.sql(expression, 'this')}) - JULIANDAY({self.sql(expression, 'expression')}))"
101
102            if unit == "MONTH":
103                sql = f"{sql} / 30.0"
104            elif unit == "YEAR":
105                sql = f"{sql} / 365.0"
106            elif unit == "HOUR":
107                sql = f"{sql} * 24.0"
108            elif unit == "MINUTE":
109                sql = f"{sql} * 1440.0"
110            elif unit == "SECOND":
111                sql = f"{sql} * 86400.0"
112            elif unit == "MILLISECOND":
113                sql = f"{sql} * 86400000.0"
114            elif unit == "MICROSECOND":
115                sql = f"{sql} * 86400000000.0"
116            elif unit == "NANOSECOND":
117                sql = f"{sql} * 8640000000000.0"
118            else:
119                self.unsupported("DATEDIFF unsupported for '{unit}'.")
120
121            return f"CAST({sql} AS INTEGER)"

def fetch_sql(self, expression): View Source

123        def fetch_sql(self, expression):
124            return self.limit_sql(exp.Limit(expression=expression.args.get("count")))

def least_sql(self, expression): View Source

126        def least_sql(self, expression):
127            if len(expression.expressions) > 1:
128                return rename_func("MIN")(self, expression)
129
130            return self.expressions(expression)

def transaction_sql(self, expression): View Source

132        def transaction_sql(self, expression):
133            this = expression.this
134            this = f" {this}" if this else ""
135            return f"BEGIN{this} TRANSACTION"

Inherited Members

sqlglot.generator.Generator: Generator; generate; unsupported; sep; seg; pad_comment; maybe_comment; wrap; no_identify; normalize_func; indent; sql; uncache_sql; cache_sql; characterset_sql; column_sql; columndef_sql; columnconstraint_sql; autoincrementcolumnconstraint_sql; compresscolumnconstraint_sql; generatedasidentitycolumnconstraint_sql; notnullcolumnconstraint_sql; primarykeycolumnconstraint_sql; uniquecolumnconstraint_sql; create_sql; describe_sql; prepend_ctes; with_sql; cte_sql; tablealias_sql; bitstring_sql; hexstring_sql; datatype_sql; directory_sql; delete_sql; drop_sql; except_sql; except_op; filter_sql; hint_sql; index_sql; identifier_sql; national_sql; partition_sql; properties_sql; root_properties; properties; with_properties; locate_properties; property_sql; likeproperty_sql; fallbackproperty_sql; journalproperty_sql; freespaceproperty_sql; afterjournalproperty_sql; checksumproperty_sql; mergeblockratioproperty_sql; datablocksizeproperty_sql; blockcompressionproperty_sql; isolatedloadingproperty_sql; lockingproperty_sql; withdataproperty_sql; insert_sql; intersect_sql; intersect_op; introducer_sql; pseudotype_sql; returning_sql; rowformatdelimitedproperty_sql; table_sql; tablesample_sql; pivot_sql; tuple_sql; update_sql; values_sql; var_sql; into_sql; from_sql; group_sql; having_sql; join_sql; lambda_sql; lateral_sql; limit_sql; offset_sql; lock_sql; literal_sql; loaddata_sql; null_sql; boolean_sql; order_sql; cluster_sql; distribute_sql; sort_sql; ordered_sql; matchrecognize_sql; query_modifiers; select_sql; schema_sql; star_sql; structkwarg_sql; parameter_sql; sessionparameter_sql; placeholder_sql; subquery_sql; qualify_sql; union_sql; union_op; unnest_sql; where_sql; window_sql; partition_by_sql; window_spec_sql; withingroup_sql; between_sql; bracket_sql; all_sql; any_sql; exists_sql; case_sql; constraint_sql; extract_sql; trim_sql; concat_sql; check_sql; foreignkey_sql; primarykey_sql; unique_sql; if_sql; in_sql; in_unnest_op; interval_sql; return_sql; reference_sql; anonymous_sql; paren_sql; neg_sql; not_sql; alias_sql; aliases_sql; attimezone_sql; add_sql; and_sql; connector_sql; bitwiseand_sql; bitwiseleftshift_sql; bitwisenot_sql; bitwiseor_sql; bitwiserightshift_sql; bitwisexor_sql; cast_sql; currentdate_sql; collate_sql; command_sql; comment_sql; commit_sql; rollback_sql; altercolumn_sql; renametable_sql; altertable_sql; droppartition_sql; addconstraint_sql; distinct_sql; ignorenulls_sql; respectnulls_sql; intdiv_sql; dpipe_sql; div_sql; floatdiv_sql; overlaps_sql; distance_sql; dot_sql; eq_sql; escape_sql; glob_sql; gt_sql; gte_sql; ilike_sql; is_sql; like_sql; similarto_sql; lt_sql; lte_sql; mod_sql; mul_sql; neq_sql; nullsafeeq_sql; nullsafeneq_sql; or_sql; slice_sql; sub_sql; trycast_sql; use_sql; binary; function_fallback_sql; func; format_args; text_width; format_time; expressions; op_expressions; naked_property; set_operation; tag_sql; token_sql; userdefinedfunction_sql; joinhint_sql; kwarg_sql; when_sql; merge_sql