sqlglot.dialects.clickhouse

 16class ClickHouse(Dialect):
 17    INDEX_OFFSET = 1
 18    NORMALIZE_FUNCTIONS: bool | str = False
 19    NULL_ORDERING = "nulls_are_last"
 20    SUPPORTS_USER_DEFINED_TYPES = False
 21    SAFE_DIVISION = True
 22    LOG_BASE_FIRST: bool | None = None
 23    FORCE_EARLY_ALIAS_REF_EXPANSION = True
 24    PRESERVE_ORIGINAL_NAMES = True
 25    NUMBERS_CAN_BE_UNDERSCORE_SEPARATED = True
 26    IDENTIFIERS_CAN_START_WITH_DIGIT = True
 27    HEX_STRING_IS_INTEGER_TYPE = True
 28
 29    # https://github.com/ClickHouse/ClickHouse/issues/33935#issue-1112165779
 30    NORMALIZATION_STRATEGY = NormalizationStrategy.CASE_SENSITIVE
 31
 32    EXPRESSION_METADATA = EXPRESSION_METADATA.copy()
 33
 34    UNESCAPED_SEQUENCES = {
 35        "\\0": "\0",
 36    }
 37
 38    CREATABLE_KIND_MAPPING = {"DATABASE": "SCHEMA"}
 39
 40    SET_OP_DISTINCT_BY_DEFAULT: dict[type[exp.Expr], bool | None] = {
 41        exp.Except: False,
 42        exp.Intersect: False,
 43        exp.Union: None,
 44    }
 45
 46    def generate_values_aliases(self, expression: exp.Values) -> list[exp.Identifier]:
 47        # Clickhouse allows VALUES to have an embedded structure e.g:
 48        # VALUES('person String, place String', ('Noah', 'Paris'), ...)
 49        # In this case, we don't want to qualify the columns
 50        values = expression.expressions[0].expressions
 51
 52        structure = (
 53            values[0]
 54            if (len(values) > 1 and values[0].is_string and isinstance(values[1], exp.Tuple))
 55            else None
 56        )
 57        if structure:
 58            # Split each column definition into the column name e.g:
 59            # 'person String, place String' -> ['person', 'place']
 60            structure_coldefs = [coldef.strip() for coldef in structure.name.split(",")]
 61            column_aliases = [
 62                exp.to_identifier(coldef.split(" ")[0]) for coldef in structure_coldefs
 63            ]
 64        else:
 65            # Default column aliases in CH are "c1", "c2", etc.
 66            column_aliases = [
 67                exp.to_identifier(f"c{i + 1}") for i in range(len(values[0].expressions))
 68            ]
 69
 70        return column_aliases
 71
 72    class Tokenizer(tokens.Tokenizer):
 73        COMMENTS = ["--", "#", "#!", ("/*", "*/")]
 74        IDENTIFIERS = ['"', "`"]
 75        IDENTIFIER_ESCAPES = ["\\"]
 76        STRING_ESCAPES = ["'", "\\"]
 77        BIT_STRINGS = [("0b", "")]
 78        HEX_STRINGS = [("0x", ""), ("0X", "")]
 79        HEREDOC_STRINGS = ["$"]
 80
 81        KEYWORDS = {
 82            **tokens.Tokenizer.KEYWORDS,
 83            ".:": TokenType.DOTCOLON,
 84            ".^": TokenType.DOTCARET,
 85            "ATTACH": TokenType.COMMAND,
 86            "DATE32": TokenType.DATE32,
 87            "DETACH": TokenType.DETACH,
 88            "DATETIME64": TokenType.DATETIME64,
 89            "DICTIONARY": TokenType.DICTIONARY,
 90            "DYNAMIC": TokenType.DYNAMIC,
 91            "ENUM8": TokenType.ENUM8,
 92            "ENUM16": TokenType.ENUM16,
 93            "EXCHANGE": TokenType.COMMAND,
 94            "FINAL": TokenType.FINAL,
 95            "FIXEDSTRING": TokenType.FIXEDSTRING,
 96            "FLOAT32": TokenType.FLOAT,
 97            "FLOAT64": TokenType.DOUBLE,
 98            "GLOBAL": TokenType.GLOBAL,
 99            "LOWCARDINALITY": TokenType.LOWCARDINALITY,
100            "MAP": TokenType.MAP,
101            "NESTED": TokenType.NESTED,
102            "NOTHING": TokenType.NOTHING,
103            "SAMPLE": TokenType.TABLE_SAMPLE,
104            "TUPLE": TokenType.STRUCT,
105            "UINT16": TokenType.USMALLINT,
106            "UINT32": TokenType.UINT,
107            "UINT64": TokenType.UBIGINT,
108            "UINT8": TokenType.UTINYINT,
109            "IPV4": TokenType.IPV4,
110            "IPV6": TokenType.IPV6,
111            "POINT": TokenType.POINT,
112            "RING": TokenType.RING,
113            "LINESTRING": TokenType.LINESTRING,
114            "MULTILINESTRING": TokenType.MULTILINESTRING,
115            "POLYGON": TokenType.POLYGON,
116            "MULTIPOLYGON": TokenType.MULTIPOLYGON,
117            "AGGREGATEFUNCTION": TokenType.AGGREGATEFUNCTION,
118            "SIMPLEAGGREGATEFUNCTION": TokenType.SIMPLEAGGREGATEFUNCTION,
119            "SYSTEM": TokenType.COMMAND,
120            "PREWHERE": TokenType.PREWHERE,
121        }
122
123        KEYWORDS.pop("/*+")
124
125        SINGLE_TOKENS = {
126            **tokens.Tokenizer.SINGLE_TOKENS,
127            "$": TokenType.HEREDOC_STRING,
128        }
129
130    Parser = ClickHouseParser
131
132    Generator = ClickHouseGenerator

 72    class Tokenizer(tokens.Tokenizer):
 73        COMMENTS = ["--", "#", "#!", ("/*", "*/")]
 74        IDENTIFIERS = ['"', "`"]
 75        IDENTIFIER_ESCAPES = ["\\"]
 76        STRING_ESCAPES = ["'", "\\"]
 77        BIT_STRINGS = [("0b", "")]
 78        HEX_STRINGS = [("0x", ""), ("0X", "")]
 79        HEREDOC_STRINGS = ["$"]
 80
 81        KEYWORDS = {
 82            **tokens.Tokenizer.KEYWORDS,
 83            ".:": TokenType.DOTCOLON,
 84            ".^": TokenType.DOTCARET,
 85            "ATTACH": TokenType.COMMAND,
 86            "DATE32": TokenType.DATE32,
 87            "DETACH": TokenType.DETACH,
 88            "DATETIME64": TokenType.DATETIME64,
 89            "DICTIONARY": TokenType.DICTIONARY,
 90            "DYNAMIC": TokenType.DYNAMIC,
 91            "ENUM8": TokenType.ENUM8,
 92            "ENUM16": TokenType.ENUM16,
 93            "EXCHANGE": TokenType.COMMAND,
 94            "FINAL": TokenType.FINAL,
 95            "FIXEDSTRING": TokenType.FIXEDSTRING,
 96            "FLOAT32": TokenType.FLOAT,
 97            "FLOAT64": TokenType.DOUBLE,
 98            "GLOBAL": TokenType.GLOBAL,
 99            "LOWCARDINALITY": TokenType.LOWCARDINALITY,
100            "MAP": TokenType.MAP,
101            "NESTED": TokenType.NESTED,
102            "NOTHING": TokenType.NOTHING,
103            "SAMPLE": TokenType.TABLE_SAMPLE,
104            "TUPLE": TokenType.STRUCT,
105            "UINT16": TokenType.USMALLINT,
106            "UINT32": TokenType.UINT,
107            "UINT64": TokenType.UBIGINT,
108            "UINT8": TokenType.UTINYINT,
109            "IPV4": TokenType.IPV4,
110            "IPV6": TokenType.IPV6,
111            "POINT": TokenType.POINT,
112            "RING": TokenType.RING,
113            "LINESTRING": TokenType.LINESTRING,
114            "MULTILINESTRING": TokenType.MULTILINESTRING,
115            "POLYGON": TokenType.POLYGON,
116            "MULTIPOLYGON": TokenType.MULTIPOLYGON,
117            "AGGREGATEFUNCTION": TokenType.AGGREGATEFUNCTION,
118            "SIMPLEAGGREGATEFUNCTION": TokenType.SIMPLEAGGREGATEFUNCTION,
119            "SYSTEM": TokenType.COMMAND,
120            "PREWHERE": TokenType.PREWHERE,
121        }
122
123        KEYWORDS.pop("/*+")
124
125        SINGLE_TOKENS = {
126            **tokens.Tokenizer.SINGLE_TOKENS,
127            "$": TokenType.HEREDOC_STRING,
128        }