sqlglot.dialects.postgres

 12class Postgres(Dialect):
 13    EXPRESSION_METADATA = EXPRESSION_METADATA.copy()
 14    INDEX_OFFSET = 1
 15    ASCII_ONLY_NORMALIZATION = True
 16    # Normalizing `x IS NOT NULL` to `NOT x IS NULL` is unsafe due to row values,
 17    # e.g. `ROW(1, NULL) IS NOT NULL` is false whereas `NOT ROW(1, NULL) IS NULL` is true
 18    NORMALIZE_NOT_NULL = False
 19    TYPED_DIVISION = True
 20    CONCAT_COALESCE = True
 21    CONCAT_WS_COALESCE = True
 22    NULL_ORDERING = "nulls_are_large"
 23    SUPPORTS_LIMIT_ALL = True
 24    TIME_FORMAT = "'YYYY-MM-DD HH24:MI:SS'"
 25    TABLESAMPLE_SIZE_IS_PERCENT = True
 26    TABLES_REFERENCEABLE_AS_COLUMNS = True
 27
 28    DEFAULT_FUNCTIONS_COLUMN_NAMES = {
 29        exp.ExplodingGenerateSeries: "generate_series",
 30    }
 31
 32    TIME_MAPPING = {
 33        "d": "%u",  # 1-based day of week
 34        "D": "%u",  # 1-based day of week
 35        "dd": "%d",  # day of month
 36        "DD": "%d",  # day of month
 37        "ddd": "%j",  # zero padded day of year
 38        "DDD": "%j",  # zero padded day of year
 39        "FMDD": "%-d",  # - is no leading zero for Python; same for FM in postgres
 40        "FMDDD": "%-j",  # day of year
 41        "FMHH12": "%-I",  # 9
 42        "FMHH24": "%-H",  # 9
 43        "FMMI": "%-M",  # Minute
 44        "FMMM": "%-m",  # 1
 45        "FMSS": "%-S",  # Second
 46        "HH12": "%I",  # 09
 47        "HH24": "%H",  # 09
 48        "mi": "%M",  # zero padded minute
 49        "MI": "%M",  # zero padded minute
 50        "mm": "%m",  # 01
 51        "MM": "%m",  # 01
 52        "OF": "%z",  # utc offset
 53        "ss": "%S",  # zero padded second
 54        "SS": "%S",  # zero padded second
 55        "TMDay": "%A",  # TM is locale dependent
 56        "TMDy": "%a",
 57        "TMMon": "%b",  # Sep
 58        "TMMonth": "%B",  # September
 59        "day": "%Aenlower",  # tuesday
 60        "dy": "%aenlower",  # tue
 61        "TZ": "%Z",  # uppercase timezone name
 62        "US": "%f",  # zero padded microsecond
 63        "ww": "%U",  # 1-based week of year
 64        "WW": "%U",  # 1-based week of year
 65        "yy": "%y",  # 15
 66        "YY": "%y",  # 15
 67        "yyy": "%Ythree",  # 015
 68        "YYY": "%Ythree",  # 015
 69        "yyyy": "%Y",  # 2015
 70        "YYYY": "%Y",  # 2015
 71    }
 72
 73    UNESCAPED_SEQUENCES = {"\\a": "a"}
 74
 75    class Tokenizer(tokens.Tokenizer):
 76        NUMERIC_ESCAPES = {
 77            "x": (16, 1, 2, 0xFF),
 78            "u": (16, 4, 4, 0xFFFF),
 79            "U": (16, 8, 8, 0x10FFFF),
 80            "0": (8, 1, 3, 0o777),
 81        }
 82        NUMERIC_ESCAPES_ARE_BYTES = True
 83        DROP_UNKNOWN_ESCAPES = True
 84        BIT_STRINGS = [("b'", "'"), ("B'", "'")]
 85        HEX_STRINGS = [("x'", "'"), ("X'", "'")]
 86        BYTE_STRINGS = [("e'", "'"), ("E'", "'")]
 87        UNICODE_STRINGS = [("U&'", "'"), ("u&'", "'")]
 88        BYTE_STRING_ESCAPES = ["'", "\\"]
 89        HEREDOC_STRINGS = ["$"]
 90
 91        HEREDOC_TAG_IS_IDENTIFIER = True
 92        HEREDOC_STRING_ALTERNATIVE = TokenType.PARAMETER
 93
 94        COMMANDS = {*tokens.Tokenizer.COMMANDS, TokenType.LOCK}
 95
 96        KEYWORDS = {
 97            **tokens.Tokenizer.KEYWORDS,
 98            "~": TokenType.RLIKE,
 99            "@@": TokenType.DAT,
100            "@?": TokenType.AT_QMARK,
101            "@>": TokenType.AT_GT,
102            "<@": TokenType.LT_AT,
103            "?&": TokenType.QMARK_AMP,
104            "?|": TokenType.QMARK_PIPE,
105            "#-": TokenType.HASH_DASH,
106            "|/": TokenType.PIPE_SLASH,
107            "||/": TokenType.DPIPE_SLASH,
108            "^@": TokenType.CARET_AT,
109            "BEGIN": TokenType.BEGIN,
110            "BIGSERIAL": TokenType.BIGSERIAL,
111            "CSTRING": TokenType.PSEUDO_TYPE,
112            "DECLARE": TokenType.COMMAND,
113            "DO": TokenType.COMMAND,
114            "EXEC": TokenType.COMMAND,
115            "HSTORE": TokenType.HSTORE,
116            "INT8": TokenType.BIGINT,
117            "MONEY": TokenType.MONEY,
118            "NAME": TokenType.NAME,
119            "OID": TokenType.OBJECT_IDENTIFIER,
120            "ONLY": TokenType.ONLY,
121            "POINT": TokenType.POINT,
122            "REFRESH": TokenType.COMMAND,
123            "REINDEX": TokenType.COMMAND,
124            "RESET": TokenType.COMMAND,
125            "SERIAL": TokenType.SERIAL,
126            "SMALLSERIAL": TokenType.SMALLSERIAL,
127            "TEMP": TokenType.TEMPORARY,
128            "TYPE": TokenType.TYPE,
129            "REGCLASS": TokenType.OBJECT_IDENTIFIER,
130            "REGCOLLATION": TokenType.OBJECT_IDENTIFIER,
131            "REGCONFIG": TokenType.OBJECT_IDENTIFIER,
132            "REGDICTIONARY": TokenType.OBJECT_IDENTIFIER,
133            "REGNAMESPACE": TokenType.OBJECT_IDENTIFIER,
134            "REGOPER": TokenType.OBJECT_IDENTIFIER,
135            "REGOPERATOR": TokenType.OBJECT_IDENTIFIER,
136            "REGPROC": TokenType.OBJECT_IDENTIFIER,
137            "REGPROCEDURE": TokenType.OBJECT_IDENTIFIER,
138            "REGROLE": TokenType.OBJECT_IDENTIFIER,
139            "REGTYPE": TokenType.OBJECT_IDENTIFIER,
140            "FLOAT": TokenType.DOUBLE,
141            "XML": TokenType.XML,
142            "VARIADIC": TokenType.VARIADIC,
143            "INOUT": TokenType.INOUT,
144        }
145        KEYWORDS.pop("/*+")
146        KEYWORDS.pop("DIV")
147
148        SINGLE_TOKENS = {
149            **tokens.Tokenizer.SINGLE_TOKENS,
150            "$": TokenType.HEREDOC_STRING,
151        }
152
153        VAR_SINGLE_TOKENS = {"$"}
154
155    Parser = PostgresParser
156
157    Generator = PostgresGenerator

 75    class Tokenizer(tokens.Tokenizer):
 76        NUMERIC_ESCAPES = {
 77            "x": (16, 1, 2, 0xFF),
 78            "u": (16, 4, 4, 0xFFFF),
 79            "U": (16, 8, 8, 0x10FFFF),
 80            "0": (8, 1, 3, 0o777),
 81        }
 82        NUMERIC_ESCAPES_ARE_BYTES = True
 83        DROP_UNKNOWN_ESCAPES = True
 84        BIT_STRINGS = [("b'", "'"), ("B'", "'")]
 85        HEX_STRINGS = [("x'", "'"), ("X'", "'")]
 86        BYTE_STRINGS = [("e'", "'"), ("E'", "'")]
 87        UNICODE_STRINGS = [("U&'", "'"), ("u&'", "'")]
 88        BYTE_STRING_ESCAPES = ["'", "\\"]
 89        HEREDOC_STRINGS = ["$"]
 90
 91        HEREDOC_TAG_IS_IDENTIFIER = True
 92        HEREDOC_STRING_ALTERNATIVE = TokenType.PARAMETER
 93
 94        COMMANDS = {*tokens.Tokenizer.COMMANDS, TokenType.LOCK}
 95
 96        KEYWORDS = {
 97            **tokens.Tokenizer.KEYWORDS,
 98            "~": TokenType.RLIKE,
 99            "@@": TokenType.DAT,
100            "@?": TokenType.AT_QMARK,
101            "@>": TokenType.AT_GT,
102            "<@": TokenType.LT_AT,
103            "?&": TokenType.QMARK_AMP,
104            "?|": TokenType.QMARK_PIPE,
105            "#-": TokenType.HASH_DASH,
106            "|/": TokenType.PIPE_SLASH,
107            "||/": TokenType.DPIPE_SLASH,
108            "^@": TokenType.CARET_AT,
109            "BEGIN": TokenType.BEGIN,
110            "BIGSERIAL": TokenType.BIGSERIAL,
111            "CSTRING": TokenType.PSEUDO_TYPE,
112            "DECLARE": TokenType.COMMAND,
113            "DO": TokenType.COMMAND,
114            "EXEC": TokenType.COMMAND,
115            "HSTORE": TokenType.HSTORE,
116            "INT8": TokenType.BIGINT,
117            "MONEY": TokenType.MONEY,
118            "NAME": TokenType.NAME,
119            "OID": TokenType.OBJECT_IDENTIFIER,
120            "ONLY": TokenType.ONLY,
121            "POINT": TokenType.POINT,
122            "REFRESH": TokenType.COMMAND,
123            "REINDEX": TokenType.COMMAND,
124            "RESET": TokenType.COMMAND,
125            "SERIAL": TokenType.SERIAL,
126            "SMALLSERIAL": TokenType.SMALLSERIAL,
127            "TEMP": TokenType.TEMPORARY,
128            "TYPE": TokenType.TYPE,
129            "REGCLASS": TokenType.OBJECT_IDENTIFIER,
130            "REGCOLLATION": TokenType.OBJECT_IDENTIFIER,
131            "REGCONFIG": TokenType.OBJECT_IDENTIFIER,
132            "REGDICTIONARY": TokenType.OBJECT_IDENTIFIER,
133            "REGNAMESPACE": TokenType.OBJECT_IDENTIFIER,
134            "REGOPER": TokenType.OBJECT_IDENTIFIER,
135            "REGOPERATOR": TokenType.OBJECT_IDENTIFIER,
136            "REGPROC": TokenType.OBJECT_IDENTIFIER,
137            "REGPROCEDURE": TokenType.OBJECT_IDENTIFIER,
138            "REGROLE": TokenType.OBJECT_IDENTIFIER,
139            "REGTYPE": TokenType.OBJECT_IDENTIFIER,
140            "FLOAT": TokenType.DOUBLE,
141            "XML": TokenType.XML,
142            "VARIADIC": TokenType.VARIADIC,
143            "INOUT": TokenType.INOUT,
144        }
145        KEYWORDS.pop("/*+")
146        KEYWORDS.pop("DIV")
147
148        SINGLE_TOKENS = {
149            **tokens.Tokenizer.SINGLE_TOKENS,
150            "$": TokenType.HEREDOC_STRING,
151        }
152
153        VAR_SINGLE_TOKENS = {"$"}