summaryrefslogtreecommitdiffstats
path: root/sqlglot/dialects/redshift.py
blob: 7b98ed450f6f6f0ab7c8536e737581d07b0bcf7a (plain)
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
384
385
386
387
388
389
390
391
392
393
394
395
396
397
398
from __future__ import annotations

import typing as t

from sqlglot import exp, transforms
from sqlglot.dialects.dialect import (
    NormalizationStrategy,
    concat_to_dpipe_sql,
    concat_ws_to_dpipe_sql,
    date_delta_sql,
    generatedasidentitycolumnconstraint_sql,
    json_extract_segments,
    no_tablesample_sql,
    rename_func,
)
from sqlglot.dialects.postgres import Postgres
from sqlglot.helper import seq_get
from sqlglot.tokens import TokenType

if t.TYPE_CHECKING:
    from sqlglot._typing import E


def _build_date_delta(expr_type: t.Type[E]) -> t.Callable[[t.List], E]:
    def _builder(args: t.List) -> E:
        expr = expr_type(this=seq_get(args, 2), expression=seq_get(args, 1), unit=seq_get(args, 0))
        if expr_type is exp.TsOrDsAdd:
            expr.set("return_type", exp.DataType.build("TIMESTAMP"))

        return expr

    return _builder


class Redshift(Postgres):
    # https://docs.aws.amazon.com/redshift/latest/dg/r_names.html
    NORMALIZATION_STRATEGY = NormalizationStrategy.CASE_INSENSITIVE

    SUPPORTS_USER_DEFINED_TYPES = False
    INDEX_OFFSET = 0

    TIME_FORMAT = "'YYYY-MM-DD HH:MI:SS'"
    TIME_MAPPING = {
        **Postgres.TIME_MAPPING,
        "MON": "%b",
        "HH": "%H",
    }

    class Parser(Postgres.Parser):
        FUNCTIONS = {
            **Postgres.Parser.FUNCTIONS,
            "ADD_MONTHS": lambda args: exp.TsOrDsAdd(
                this=seq_get(args, 0),
                expression=seq_get(args, 1),
                unit=exp.var("month"),
                return_type=exp.DataType.build("TIMESTAMP"),
            ),
            "DATEADD": _build_date_delta(exp.TsOrDsAdd),
            "DATE_ADD": _build_date_delta(exp.TsOrDsAdd),
            "DATEDIFF": _build_date_delta(exp.TsOrDsDiff),
            "DATE_DIFF": _build_date_delta(exp.TsOrDsDiff),
            "GETDATE": exp.CurrentTimestamp.from_arg_list,
            "LISTAGG": exp.GroupConcat.from_arg_list,
            "STRTOL": exp.FromBase.from_arg_list,
        }

        NO_PAREN_FUNCTION_PARSERS = {
            **Postgres.Parser.NO_PAREN_FUNCTION_PARSERS,
            "APPROXIMATE": lambda self: self._parse_approximate_count(),
            "SYSDATE": lambda self: self.expression(exp.CurrentTimestamp, transaction=True),
        }

        SUPPORTS_IMPLICIT_UNNEST = True

        def _parse_table(
            self,
            schema: bool = False,
            joins: bool = False,
            alias_tokens: t.Optional[t.Collection[TokenType]] = None,
            parse_bracket: bool = False,
            is_db_reference: bool = False,
            parse_partition: bool = False,
        ) -> t.Optional[exp.Expression]:
            # Redshift supports UNPIVOTing SUPER objects, e.g. `UNPIVOT foo.obj[0] AS val AT attr`
            unpivot = self._match(TokenType.UNPIVOT)
            table = super()._parse_table(
                schema=schema,
                joins=joins,
                alias_tokens=alias_tokens,
                parse_bracket=parse_bracket,
                is_db_reference=is_db_reference,
            )

            return self.expression(exp.Pivot, this=table, unpivot=True) if unpivot else table

        def _parse_convert(
            self, strict: bool, safe: t.Optional[bool] = None
        ) -> t.Optional[exp.Expression]:
            to = self._parse_types()
            self._match(TokenType.COMMA)
            this = self._parse_bitwise()
            return self.expression(exp.TryCast, this=this, to=to, safe=safe)

        def _parse_approximate_count(self) -> t.Optional[exp.ApproxDistinct]:
            index = self._index - 1
            func = self._parse_function()

            if isinstance(func, exp.Count) and isinstance(func.this, exp.Distinct):
                return self.expression(exp.ApproxDistinct, this=seq_get(func.this.expressions, 0))
            self._retreat(index)
            return None

    class Tokenizer(Postgres.Tokenizer):
        BIT_STRINGS = []
        HEX_STRINGS = []
        STRING_ESCAPES = ["\\", "'"]

        KEYWORDS = {
            **Postgres.Tokenizer.KEYWORDS,
            "HLLSKETCH": TokenType.HLLSKETCH,
            "SUPER": TokenType.SUPER,
            "TOP": TokenType.TOP,
            "UNLOAD": TokenType.COMMAND,
            "VARBYTE": TokenType.VARBINARY,
        }
        KEYWORDS.pop("VALUES")

        # Redshift allows # to appear as a table identifier prefix
        SINGLE_TOKENS = Postgres.Tokenizer.SINGLE_TOKENS.copy()
        SINGLE_TOKENS.pop("#")

    class Generator(Postgres.Generator):
        LOCKING_READS_SUPPORTED = False
        QUERY_HINTS = False
        VALUES_AS_TABLE = False
        TZ_TO_WITH_TIME_ZONE = True
        NVL2_SUPPORTED = True
        LAST_DAY_SUPPORTS_DATE_PART = False
        CAN_IMPLEMENT_ARRAY_ANY = False
        MULTI_ARG_DISTINCT = True

        TYPE_MAPPING = {
            **Postgres.Generator.TYPE_MAPPING,
            exp.DataType.Type.BINARY: "VARBYTE",
            exp.DataType.Type.INT: "INTEGER",
            exp.DataType.Type.TIMETZ: "TIME",
            exp.DataType.Type.TIMESTAMPTZ: "TIMESTAMP",
            exp.DataType.Type.VARBINARY: "VARBYTE",
            exp.DataType.Type.ROWVERSION: "VARBYTE",
        }

        TRANSFORMS = {
            **Postgres.Generator.TRANSFORMS,
            exp.Concat: concat_to_dpipe_sql,
            exp.ConcatWs: concat_ws_to_dpipe_sql,
            exp.ApproxDistinct: lambda self,
            e: f"APPROXIMATE COUNT(DISTINCT {self.sql(e, 'this')})",
            exp.CurrentTimestamp: lambda self, e: (
                "SYSDATE" if e.args.get("transaction") else "GETDATE()"
            ),
            exp.DateAdd: date_delta_sql("DATEADD"),
            exp.DateDiff: date_delta_sql("DATEDIFF"),
            exp.DistKeyProperty: lambda self, e: self.func("DISTKEY", e.this),
            exp.DistStyleProperty: lambda self, e: self.naked_property(e),
            exp.FromBase: rename_func("STRTOL"),
            exp.GeneratedAsIdentityColumnConstraint: generatedasidentitycolumnconstraint_sql,
            exp.JSONExtract: json_extract_segments("JSON_EXTRACT_PATH_TEXT"),
            exp.JSONExtractScalar: json_extract_segments("JSON_EXTRACT_PATH_TEXT"),
            exp.GroupConcat: rename_func("LISTAGG"),
            exp.ParseJSON: rename_func("JSON_PARSE"),
            exp.Select: transforms.preprocess(
                [
                    transforms.eliminate_distinct_on,
                    transforms.eliminate_semi_and_anti_joins,
                    transforms.unqualify_unnest,
                ]
            ),
            exp.SortKeyProperty: lambda self,
            e: f"{'COMPOUND ' if e.args['compound'] else ''}SORTKEY({self.format_args(*e.this)})",
            exp.StartsWith: lambda self,
            e: f"{self.sql(e.this)} LIKE {self.sql(e.expression)} || '%'",
            exp.TableSample: no_tablesample_sql,
            exp.TsOrDsAdd: date_delta_sql("DATEADD"),
            exp.TsOrDsDiff: date_delta_sql("DATEDIFF"),
            exp.UnixToTime: lambda self,
            e: f"(TIMESTAMP 'epoch' + {self.sql(e.this)} * INTERVAL '1 SECOND')",
        }

        # Postgres maps exp.Pivot to no_pivot_sql, but Redshift support pivots
        TRANSFORMS.pop(exp.Pivot)

        # Redshift uses the POW | POWER (expr1, expr2) syntax instead of expr1 ^ expr2 (postgres)
        TRANSFORMS.pop(exp.Pow)

        # Redshift supports ANY_VALUE(..)
        TRANSFORMS.pop(exp.AnyValue)

        # Redshift supports LAST_DAY(..)
        TRANSFORMS.pop(exp.LastDay)

        RESERVED_KEYWORDS = {
            "aes128",
            "aes256",
            "all",
            "allowoverwrite",
            "analyse",
            "analyze",
            "and",
            "any",
            "array",
            "as",
            "asc",
            "authorization",
            "az64",
            "backup",
            "between",
            "binary",
            "blanksasnull",
            "both",
            "bytedict",
            "bzip2",
            "case",
            "cast",
            "check",
            "collate",
            "column",
            "constraint",
            "create",
            "credentials",
            "cross",
            "current_date",
            "current_time",
            "current_timestamp",
            "current_user",
            "current_user_id",
            "default",
            "deferrable",
            "deflate",
            "defrag",
            "delta",
            "delta32k",
            "desc",
            "disable",
            "distinct",
            "do",
            "else",
            "emptyasnull",
            "enable",
            "encode",
            "encrypt     ",
            "encryption",
            "end",
            "except",
            "explicit",
            "false",
            "for",
            "foreign",
            "freeze",
            "from",
            "full",
            "globaldict256",
            "globaldict64k",
            "grant",
            "group",
            "gzip",
            "having",
            "identity",
            "ignore",
            "ilike",
            "in",
            "initially",
            "inner",
            "intersect",
            "interval",
            "into",
            "is",
            "isnull",
            "join",
            "leading",
            "left",
            "like",
            "limit",
            "localtime",
            "localtimestamp",
            "lun",
            "luns",
            "lzo",
            "lzop",
            "minus",
            "mostly16",
            "mostly32",
            "mostly8",
            "natural",
            "new",
            "not",
            "notnull",
            "null",
            "nulls",
            "off",
            "offline",
            "offset",
            "oid",
            "old",
            "on",
            "only",
            "open",
            "or",
            "order",
            "outer",
            "overlaps",
            "parallel",
            "partition",
            "percent",
            "permissions",
            "pivot",
            "placing",
            "primary",
            "raw",
            "readratio",
            "recover",
            "references",
            "rejectlog",
            "resort",
            "respect",
            "restore",
            "right",
            "select",
            "session_user",
            "similar",
            "snapshot",
            "some",
            "sysdate",
            "system",
            "table",
            "tag",
            "tdes",
            "text255",
            "text32k",
            "then",
            "timestamp",
            "to",
            "top",
            "trailing",
            "true",
            "truncatecolumns",
            "type",
            "union",
            "unique",
            "unnest",
            "unpivot",
            "user",
            "using",
            "verbose",
            "wallet",
            "when",
            "where",
            "with",
            "without",
        }

        def unnest_sql(self, expression: exp.Unnest) -> str:
            args = expression.expressions
            num_args = len(args)

            if num_args > 1:
                self.unsupported(f"Unsupported number of arguments in UNNEST: {num_args}")
                return ""

            arg = self.sql(seq_get(args, 0))
            alias = self.expressions(expression.args.get("alias"), key="columns", flat=True)
            return f"{arg} AS {alias}" if alias else arg

        def with_properties(self, properties: exp.Properties) -> str:
            """Redshift doesn't have `WITH` as part of their with_properties so we remove it"""
            return self.properties(properties, prefix=" ", suffix="")

        def cast_sql(self, expression: exp.Cast, safe_prefix: t.Optional[str] = None) -> str:
            if expression.is_type(exp.DataType.Type.JSON):
                # Redshift doesn't support a JSON type, so casting to it is treated as a noop
                return self.sql(expression, "this")

            return super().cast_sql(expression, safe_prefix=safe_prefix)

        def datatype_sql(self, expression: exp.DataType) -> str:
            """
            Redshift converts the `TEXT` data type to `VARCHAR(255)` by default when people more generally mean
            VARCHAR of max length which is `VARCHAR(max)` in Redshift. Therefore if we get a `TEXT` data type
            without precision we convert it to `VARCHAR(max)` and if it does have precision then we just convert
            `TEXT` to `VARCHAR`.
            """
            if expression.is_type("text"):
                expression.set("this", exp.DataType.Type.VARCHAR)
                precision = expression.args.get("expressions")

                if not precision:
                    expression.append("expressions", exp.var("MAX"))

            return super().datatype_sql(expression)