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    COLUMN_ALIAS_PRECEDENCE = {
 25        exp.Distinct: AliasPrecedence.SOURCE,
 26        exp.Having: AliasPrecedence.SOURCE,
 27        exp.Order: AliasPrecedence.SOURCE,
 28        exp.Qualify: AliasPrecedence.SOURCE,
 29    }
 30    TIME_FORMAT = "'YYYY-MM-DD HH24:MI:SS'"
 31    TABLESAMPLE_SIZE_IS_PERCENT = True
 32    TABLES_REFERENCEABLE_AS_COLUMNS = True
 33    OUTER_COLUMNS_SHADOW_TABLE_ALIASES = True
 34
 35    DEFAULT_FUNCTIONS_COLUMN_NAMES = {
 36        exp.ExplodingGenerateSeries: "generate_series",
 37    }
 38
 39    TIME_MAPPING = {
 40        "d": "%u",  # 1-based day of week
 41        "D": "%u",  # 1-based day of week
 42        "dd": "%d",  # day of month
 43        "DD": "%d",  # day of month
 44        "ddd": "%j",  # zero padded day of year
 45        "DDD": "%j",  # zero padded day of year
 46        "FMDD": "%-d",  # - is no leading zero for Python; same for FM in postgres
 47        "FMDDD": "%-j",  # day of year
 48        "FMHH12": "%-I",  # 9
 49        "FMHH24": "%-H",  # 9
 50        "FMMI": "%-M",  # Minute
 51        "FMMM": "%-m",  # 1
 52        "FMSS": "%-S",  # Second
 53        "HH12": "%I",  # 09
 54        "HH24": "%H",  # 09
 55        "iw": "%V",  # ISO 8601 week of year
 56        "IW": "%V",  # ISO 8601 week of year
 57        "iyyy": "%G",  # ISO 8601 week-numbering year
 58        "IYYY": "%G",  # ISO 8601 week-numbering year
 59        "mi": "%M",  # zero padded minute
 60        "MI": "%M",  # zero padded minute
 61        "mm": "%m",  # 01
 62        "MM": "%m",  # 01
 63        "OF": "%z",  # utc offset
 64        "ss": "%S",  # zero padded second
 65        "SS": "%S",  # zero padded second
 66        "TMDay": "%A",  # TM is locale dependent
 67        "TMDy": "%a",
 68        "TMMon": "%b",  # Sep
 69        "TMMonth": "%B",  # September
 70        "day": "%Aenlower",  # tuesday
 71        "dy": "%aenlower",  # tue
 72        "TZ": "%Z",  # uppercase timezone name
 73        "US": "%f",  # zero padded microsecond
 74        "ww": "%U",  # 1-based week of year
 75        "WW": "%U",  # 1-based week of year
 76        "yy": "%y",  # 15
 77        "YY": "%y",  # 15
 78        "yyy": "%Ythree",  # 015
 79        "YYY": "%Ythree",  # 015
 80        "yyyy": "%Y",  # 2015
 81        "YYYY": "%Y",  # 2015
 82    }
 83
 84    UNESCAPED_SEQUENCES = {"\\a": "a"}
 85
 86    class Tokenizer(tokens.Tokenizer):
 87        NUMERIC_ESCAPES = {
 88            "x": (16, 1, 2, 0xFF),
 89            "u": (16, 4, 4, 0xFFFF),
 90            "U": (16, 8, 8, 0x10FFFF),
 91            "0": (8, 1, 3, 0o777),
 92        }
 93        NUMERIC_ESCAPES_ARE_BYTES = True
 94        DROP_UNKNOWN_ESCAPES = True
 95        BIT_STRINGS = [("b'", "'"), ("B'", "'")]
 96        HEX_STRINGS = [("x'", "'"), ("X'", "'")]
 97        BYTE_STRINGS = [("e'", "'"), ("E'", "'")]
 98        UNICODE_STRINGS = [("U&'", "'"), ("u&'", "'")]
 99        BYTE_STRING_ESCAPES = ["'", "\\"]
100        HEREDOC_STRINGS = ["$"]
101
102        HEREDOC_TAG_IS_IDENTIFIER = True
103        HEREDOC_STRING_ALTERNATIVE = TokenType.PARAMETER
104
105        COMMANDS = {*tokens.Tokenizer.COMMANDS, TokenType.LOCK}
106
107        KEYWORDS = {
108            **tokens.Tokenizer.KEYWORDS,
109            "~": TokenType.RLIKE,
110            "@@": TokenType.DAT,
111            "@?": TokenType.AT_QMARK,
112            "@>": TokenType.AT_GT,
113            "<@": TokenType.LT_AT,
114            "?&": TokenType.QMARK_AMP,
115            "?|": TokenType.QMARK_PIPE,
116            "#-": TokenType.HASH_DASH,
117            "|/": TokenType.PIPE_SLASH,
118            "||/": TokenType.DPIPE_SLASH,
119            "^@": TokenType.CARET_AT,
120            "BEGIN": TokenType.BEGIN,
121            "BIGSERIAL": TokenType.BIGSERIAL,
122            "CSTRING": TokenType.PSEUDO_TYPE,
123            "DECLARE": TokenType.COMMAND,
124            "DO": TokenType.COMMAND,
125            "EXEC": TokenType.COMMAND,
126            "HSTORE": TokenType.HSTORE,
127            "INT8": TokenType.BIGINT,
128            "MONEY": TokenType.MONEY,
129            "NAME": TokenType.NAME,
130            "OID": TokenType.OBJECT_IDENTIFIER,
131            "ONLY": TokenType.ONLY,
132            "POINT": TokenType.POINT,
133            "REFRESH": TokenType.COMMAND,
134            "REINDEX": TokenType.COMMAND,
135            "RESET": TokenType.COMMAND,
136            "SERIAL": TokenType.SERIAL,
137            "SMALLSERIAL": TokenType.SMALLSERIAL,
138            "TEMP": TokenType.TEMPORARY,
139            "TYPE": TokenType.TYPE,
140            "REGCLASS": TokenType.OBJECT_IDENTIFIER,
141            "REGCOLLATION": TokenType.OBJECT_IDENTIFIER,
142            "REGCONFIG": TokenType.OBJECT_IDENTIFIER,
143            "REGDICTIONARY": TokenType.OBJECT_IDENTIFIER,
144            "REGNAMESPACE": TokenType.OBJECT_IDENTIFIER,
145            "REGOPER": TokenType.OBJECT_IDENTIFIER,
146            "REGOPERATOR": TokenType.OBJECT_IDENTIFIER,
147            "REGPROC": TokenType.OBJECT_IDENTIFIER,
148            "REGPROCEDURE": TokenType.OBJECT_IDENTIFIER,
149            "REGROLE": TokenType.OBJECT_IDENTIFIER,
150            "REGTYPE": TokenType.OBJECT_IDENTIFIER,
151            "FLOAT": TokenType.DOUBLE,
152            "XML": TokenType.XML,
153            "VARIADIC": TokenType.VARIADIC,
154            "INOUT": TokenType.INOUT,
155        }
156        KEYWORDS.pop("/*+")
157        KEYWORDS.pop("DIV")
158
159        SINGLE_TOKENS = {
160            **tokens.Tokenizer.SINGLE_TOKENS,
161            "$": TokenType.HEREDOC_STRING,
162        }
163
164        VAR_SINGLE_TOKENS = {"$"}
165
166    Parser = PostgresParser
167
168    Generator = PostgresGenerator

 86    class Tokenizer(tokens.Tokenizer):
 87        NUMERIC_ESCAPES = {
 88            "x": (16, 1, 2, 0xFF),
 89            "u": (16, 4, 4, 0xFFFF),
 90            "U": (16, 8, 8, 0x10FFFF),
 91            "0": (8, 1, 3, 0o777),
 92        }
 93        NUMERIC_ESCAPES_ARE_BYTES = True
 94        DROP_UNKNOWN_ESCAPES = True
 95        BIT_STRINGS = [("b'", "'"), ("B'", "'")]
 96        HEX_STRINGS = [("x'", "'"), ("X'", "'")]
 97        BYTE_STRINGS = [("e'", "'"), ("E'", "'")]
 98        UNICODE_STRINGS = [("U&'", "'"), ("u&'", "'")]
 99        BYTE_STRING_ESCAPES = ["'", "\\"]
100        HEREDOC_STRINGS = ["$"]
101
102        HEREDOC_TAG_IS_IDENTIFIER = True
103        HEREDOC_STRING_ALTERNATIVE = TokenType.PARAMETER
104
105        COMMANDS = {*tokens.Tokenizer.COMMANDS, TokenType.LOCK}
106
107        KEYWORDS = {
108            **tokens.Tokenizer.KEYWORDS,
109            "~": TokenType.RLIKE,
110            "@@": TokenType.DAT,
111            "@?": TokenType.AT_QMARK,
112            "@>": TokenType.AT_GT,
113            "<@": TokenType.LT_AT,
114            "?&": TokenType.QMARK_AMP,
115            "?|": TokenType.QMARK_PIPE,
116            "#-": TokenType.HASH_DASH,
117            "|/": TokenType.PIPE_SLASH,
118            "||/": TokenType.DPIPE_SLASH,
119            "^@": TokenType.CARET_AT,
120            "BEGIN": TokenType.BEGIN,
121            "BIGSERIAL": TokenType.BIGSERIAL,
122            "CSTRING": TokenType.PSEUDO_TYPE,
123            "DECLARE": TokenType.COMMAND,
124            "DO": TokenType.COMMAND,
125            "EXEC": TokenType.COMMAND,
126            "HSTORE": TokenType.HSTORE,
127            "INT8": TokenType.BIGINT,
128            "MONEY": TokenType.MONEY,
129            "NAME": TokenType.NAME,
130            "OID": TokenType.OBJECT_IDENTIFIER,
131            "ONLY": TokenType.ONLY,
132            "POINT": TokenType.POINT,
133            "REFRESH": TokenType.COMMAND,
134            "REINDEX": TokenType.COMMAND,
135            "RESET": TokenType.COMMAND,
136            "SERIAL": TokenType.SERIAL,
137            "SMALLSERIAL": TokenType.SMALLSERIAL,
138            "TEMP": TokenType.TEMPORARY,
139            "TYPE": TokenType.TYPE,
140            "REGCLASS": TokenType.OBJECT_IDENTIFIER,
141            "REGCOLLATION": TokenType.OBJECT_IDENTIFIER,
142            "REGCONFIG": TokenType.OBJECT_IDENTIFIER,
143            "REGDICTIONARY": TokenType.OBJECT_IDENTIFIER,
144            "REGNAMESPACE": TokenType.OBJECT_IDENTIFIER,
145            "REGOPER": TokenType.OBJECT_IDENTIFIER,
146            "REGOPERATOR": TokenType.OBJECT_IDENTIFIER,
147            "REGPROC": TokenType.OBJECT_IDENTIFIER,
148            "REGPROCEDURE": TokenType.OBJECT_IDENTIFIER,
149            "REGROLE": TokenType.OBJECT_IDENTIFIER,
150            "REGTYPE": TokenType.OBJECT_IDENTIFIER,
151            "FLOAT": TokenType.DOUBLE,
152            "XML": TokenType.XML,
153            "VARIADIC": TokenType.VARIADIC,
154            "INOUT": TokenType.INOUT,
155        }
156        KEYWORDS.pop("/*+")
157        KEYWORDS.pop("DIV")
158
159        SINGLE_TOKENS = {
160            **tokens.Tokenizer.SINGLE_TOKENS,
161            "$": TokenType.HEREDOC_STRING,
162        }
163
164        VAR_SINGLE_TOKENS = {"$"}