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    LOG_BASE_FIRST: bool | None = None
 22    FORCE_EARLY_ALIAS_REF_EXPANSION = True
 23    ORIGINAL_NAME_META_KEY = "clickhouse_name"
 24    NUMBERS_CAN_BE_UNDERSCORE_SEPARATED = True
 25    IDENTIFIERS_CAN_START_WITH_DIGIT = True
 26    HEX_STRING_IS_INTEGER_TYPE = True
 27
 28    # https://github.com/ClickHouse/ClickHouse/issues/33935#issue-1112165779
 29    NORMALIZATION_STRATEGY = NormalizationStrategy.CASE_SENSITIVE
 30
 31    EXPRESSION_METADATA = EXPRESSION_METADATA.copy()
 32
 33    UNESCAPED_SEQUENCES = {
 34        "\\0": "\0",
 35        "\\e": "\x1b",
 36        "\\N": "",
 37        "\\/": "/",
 38        '\\"': '"',
 39        "\\=": "=",
 40        "\\`": "`",
 41        **{"\\" + chr(i): chr(i) for i in range(1, 32)},
 42    }
 43
 44    CREATABLE_KIND_MAPPING = {"DATABASE": "SCHEMA"}
 45
 46    SET_OP_DISTINCT_BY_DEFAULT: dict[type[exp.Expr], bool | None] = {
 47        exp.Except: False,
 48        exp.Intersect: False,
 49        exp.Union: None,
 50    }
 51
 52    def generate_values_aliases(self, expression: exp.Values) -> list[exp.Identifier]:
 53        # Clickhouse allows VALUES to have an embedded structure e.g:
 54        # VALUES('person String, place String', ('Noah', 'Paris'), ...)
 55        # In this case, we don't want to qualify the columns
 56        values = expression.expressions[0].expressions
 57
 58        structure = (
 59            values[0]
 60            if (len(values) > 1 and values[0].is_string and isinstance(values[1], exp.Tuple))
 61            else None
 62        )
 63        if structure:
 64            # Split each column definition into the column name e.g:
 65            # 'person String, place String' -> ['person', 'place']
 66            structure_coldefs = [coldef.strip() for coldef in structure.name.split(",")]
 67            column_aliases = [
 68                exp.to_identifier(coldef.split(" ")[0]) for coldef in structure_coldefs
 69            ]
 70        else:
 71            # Default column aliases in CH are "c1", "c2", etc.
 72            column_aliases = [
 73                exp.to_identifier(f"c{i + 1}") for i in range(len(values[0].expressions))
 74            ]
 75
 76        return column_aliases
 77
 78    class Tokenizer(tokens.Tokenizer):
 79        NUMERIC_ESCAPES = {"x": (16, 2, 2, 0xFF)}
 80        NUMERIC_ESCAPES_ARE_BYTES = True
 81        COMMENTS = ["--", "#", "#!", ("/*", "*/")]
 82        COMMENTS_TERMINATE_AT_NEWLINE_ONLY = True
 83        IDENTIFIERS = ['"', "`"]
 84        IDENTIFIER_ESCAPES = ["\\"]
 85        STRING_ESCAPES = ["'", "\\"]
 86        BIT_STRINGS = [("0b", "")]
 87        HEX_STRINGS = [("0x", ""), ("0X", "")]
 88        HEREDOC_STRINGS = ["$"]
 89
 90        KEYWORDS = {
 91            **tokens.Tokenizer.KEYWORDS,
 92            ".:": TokenType.DOTCOLON,
 93            ".^": TokenType.DOTCARET,
 94            "ATTACH": TokenType.COMMAND,
 95            "DATE32": TokenType.DATE32,
 96            "DETACH": TokenType.DETACH,
 97            "DATETIME64": TokenType.DATETIME64,
 98            "DICTIONARY": TokenType.DICTIONARY,
 99            "DYNAMIC": TokenType.DYNAMIC,
100            "ENUM8": TokenType.ENUM8,
101            "ENUM16": TokenType.ENUM16,
102            "EXCHANGE": TokenType.COMMAND,
103            "FINAL": TokenType.FINAL,
104            "FIXEDSTRING": TokenType.FIXEDSTRING,
105            "FLOAT32": TokenType.FLOAT,
106            "FLOAT64": TokenType.DOUBLE,
107            "GLOBAL": TokenType.GLOBAL,
108            "LOWCARDINALITY": TokenType.LOWCARDINALITY,
109            "MAP": TokenType.MAP,
110            "NESTED": TokenType.NESTED,
111            "NOTHING": TokenType.NOTHING,
112            "SAMPLE": TokenType.TABLE_SAMPLE,
113            "TUPLE": TokenType.STRUCT,
114            "UINT16": TokenType.USMALLINT,
115            "UINT32": TokenType.UINT,
116            "UINT64": TokenType.UBIGINT,
117            "UINT8": TokenType.UTINYINT,
118            "IPV4": TokenType.IPV4,
119            "IPV6": TokenType.IPV6,
120            "POINT": TokenType.POINT,
121            "PROJECTION": TokenType.PROJECTION,
122            "RING": TokenType.RING,
123            "LINESTRING": TokenType.LINESTRING,
124            "MULTILINESTRING": TokenType.MULTILINESTRING,
125            "POLYGON": TokenType.POLYGON,
126            "MULTIPOLYGON": TokenType.MULTIPOLYGON,
127            "AGGREGATEFUNCTION": TokenType.AGGREGATEFUNCTION,
128            "SIMPLEAGGREGATEFUNCTION": TokenType.SIMPLEAGGREGATEFUNCTION,
129            "SYSTEM": TokenType.COMMAND,
130            "PREWHERE": TokenType.PREWHERE,
131        }
132
133        KEYWORDS.pop("/*+")
134
135        SINGLE_TOKENS = {
136            **tokens.Tokenizer.SINGLE_TOKENS,
137            "$": TokenType.HEREDOC_STRING,
138        }
139
140    Parser = ClickHouseParser
141
142    Generator = ClickHouseGenerator

 78    class Tokenizer(tokens.Tokenizer):
 79        NUMERIC_ESCAPES = {"x": (16, 2, 2, 0xFF)}
 80        NUMERIC_ESCAPES_ARE_BYTES = True
 81        COMMENTS = ["--", "#", "#!", ("/*", "*/")]
 82        COMMENTS_TERMINATE_AT_NEWLINE_ONLY = True
 83        IDENTIFIERS = ['"', "`"]
 84        IDENTIFIER_ESCAPES = ["\\"]
 85        STRING_ESCAPES = ["'", "\\"]
 86        BIT_STRINGS = [("0b", "")]
 87        HEX_STRINGS = [("0x", ""), ("0X", "")]
 88        HEREDOC_STRINGS = ["$"]
 89
 90        KEYWORDS = {
 91            **tokens.Tokenizer.KEYWORDS,
 92            ".:": TokenType.DOTCOLON,
 93            ".^": TokenType.DOTCARET,
 94            "ATTACH": TokenType.COMMAND,
 95            "DATE32": TokenType.DATE32,
 96            "DETACH": TokenType.DETACH,
 97            "DATETIME64": TokenType.DATETIME64,
 98            "DICTIONARY": TokenType.DICTIONARY,
 99            "DYNAMIC": TokenType.DYNAMIC,
100            "ENUM8": TokenType.ENUM8,
101            "ENUM16": TokenType.ENUM16,
102            "EXCHANGE": TokenType.COMMAND,
103            "FINAL": TokenType.FINAL,
104            "FIXEDSTRING": TokenType.FIXEDSTRING,
105            "FLOAT32": TokenType.FLOAT,
106            "FLOAT64": TokenType.DOUBLE,
107            "GLOBAL": TokenType.GLOBAL,
108            "LOWCARDINALITY": TokenType.LOWCARDINALITY,
109            "MAP": TokenType.MAP,
110            "NESTED": TokenType.NESTED,
111            "NOTHING": TokenType.NOTHING,
112            "SAMPLE": TokenType.TABLE_SAMPLE,
113            "TUPLE": TokenType.STRUCT,
114            "UINT16": TokenType.USMALLINT,
115            "UINT32": TokenType.UINT,
116            "UINT64": TokenType.UBIGINT,
117            "UINT8": TokenType.UTINYINT,
118            "IPV4": TokenType.IPV4,
119            "IPV6": TokenType.IPV6,
120            "POINT": TokenType.POINT,
121            "PROJECTION": TokenType.PROJECTION,
122            "RING": TokenType.RING,
123            "LINESTRING": TokenType.LINESTRING,
124            "MULTILINESTRING": TokenType.MULTILINESTRING,
125            "POLYGON": TokenType.POLYGON,
126            "MULTIPOLYGON": TokenType.MULTIPOLYGON,
127            "AGGREGATEFUNCTION": TokenType.AGGREGATEFUNCTION,
128            "SIMPLEAGGREGATEFUNCTION": TokenType.SIMPLEAGGREGATEFUNCTION,
129            "SYSTEM": TokenType.COMMAND,
130            "PREWHERE": TokenType.PREWHERE,
131        }
132
133        KEYWORDS.pop("/*+")
134
135        SINGLE_TOKENS = {
136            **tokens.Tokenizer.SINGLE_TOKENS,
137            "$": TokenType.HEREDOC_STRING,
138        }