summaryrefslogtreecommitdiffstats
path: root/sqlglot/dialects/teradata.py
blob: a79eaeb11c230ff49fa3adaa17a3c9ef649345c8 (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
from __future__ import annotations

import typing as t

from sqlglot import exp, generator, parser, tokens, transforms
from sqlglot.dialects.dialect import (
    Dialect,
    format_time_lambda,
    max_or_greatest,
    min_or_least,
)
from sqlglot.tokens import TokenType


class Teradata(Dialect):
    class Tokenizer(tokens.Tokenizer):
        # https://docs.teradata.com/r/Teradata-Database-SQL-Functions-Operators-Expressions-and-Predicates/March-2017/Comparison-Operators-and-Functions/Comparison-Operators/ANSI-Compliance
        KEYWORDS = {
            **tokens.Tokenizer.KEYWORDS,
            "BYTEINT": TokenType.SMALLINT,
            "SEL": TokenType.SELECT,
            "INS": TokenType.INSERT,
            "MOD": TokenType.MOD,
            "LT": TokenType.LT,
            "LE": TokenType.LTE,
            "GT": TokenType.GT,
            "GE": TokenType.GTE,
            "^=": TokenType.NEQ,
            "NE": TokenType.NEQ,
            "NOT=": TokenType.NEQ,
            "ST_GEOMETRY": TokenType.GEOMETRY,
        }

        # teradata does not support % for modulus
        SINGLE_TOKENS = {**tokens.Tokenizer.SINGLE_TOKENS}
        SINGLE_TOKENS.pop("%")

    class Parser(parser.Parser):
        CHARSET_TRANSLATORS = {
            "GRAPHIC_TO_KANJISJIS",
            "GRAPHIC_TO_LATIN",
            "GRAPHIC_TO_UNICODE",
            "GRAPHIC_TO_UNICODE_PadSpace",
            "KANJI1_KanjiEBCDIC_TO_UNICODE",
            "KANJI1_KanjiEUC_TO_UNICODE",
            "KANJI1_KANJISJIS_TO_UNICODE",
            "KANJI1_SBC_TO_UNICODE",
            "KANJISJIS_TO_GRAPHIC",
            "KANJISJIS_TO_LATIN",
            "KANJISJIS_TO_UNICODE",
            "LATIN_TO_GRAPHIC",
            "LATIN_TO_KANJISJIS",
            "LATIN_TO_UNICODE",
            "LOCALE_TO_UNICODE",
            "UNICODE_TO_GRAPHIC",
            "UNICODE_TO_GRAPHIC_PadGraphic",
            "UNICODE_TO_GRAPHIC_VarGraphic",
            "UNICODE_TO_KANJI1_KanjiEBCDIC",
            "UNICODE_TO_KANJI1_KanjiEUC",
            "UNICODE_TO_KANJI1_KANJISJIS",
            "UNICODE_TO_KANJI1_SBC",
            "UNICODE_TO_KANJISJIS",
            "UNICODE_TO_LATIN",
            "UNICODE_TO_LOCALE",
            "UNICODE_TO_UNICODE_FoldSpace",
            "UNICODE_TO_UNICODE_Fullwidth",
            "UNICODE_TO_UNICODE_Halfwidth",
            "UNICODE_TO_UNICODE_NFC",
            "UNICODE_TO_UNICODE_NFD",
            "UNICODE_TO_UNICODE_NFKC",
            "UNICODE_TO_UNICODE_NFKD",
        }

        FUNC_TOKENS = {*parser.Parser.FUNC_TOKENS}
        FUNC_TOKENS.remove(TokenType.REPLACE)

        STATEMENT_PARSERS = {
            **parser.Parser.STATEMENT_PARSERS,  # type: ignore
            TokenType.REPLACE: lambda self: self._parse_create(),
        }

        FUNCTION_PARSERS = {
            **parser.Parser.FUNCTION_PARSERS,  # type: ignore
            "RANGE_N": lambda self: self._parse_rangen(),
            "TRANSLATE": lambda self: self._parse_translate(self.STRICT_CAST),
        }

        def _parse_translate(self, strict: bool) -> exp.Expression:
            this = self._parse_conjunction()

            if not self._match(TokenType.USING):
                self.raise_error("Expected USING in TRANSLATE")

            if self._match_texts(self.CHARSET_TRANSLATORS):
                charset_split = self._prev.text.split("_TO_")
                to = self.expression(exp.CharacterSet, this=charset_split[1])
            else:
                self.raise_error("Expected a character set translator after USING in TRANSLATE")

            return self.expression(exp.Cast if strict else exp.TryCast, this=this, to=to)

        # FROM before SET in Teradata UPDATE syntax
        # https://docs.teradata.com/r/Enterprise_IntelliFlex_VMware/Teradata-VantageTM-SQL-Data-Manipulation-Language-17.20/Statement-Syntax/UPDATE/UPDATE-Syntax-Basic-Form-FROM-Clause
        def _parse_update(self) -> exp.Expression:
            return self.expression(
                exp.Update,
                **{  # type: ignore
                    "this": self._parse_table(alias_tokens=self.UPDATE_ALIAS_TOKENS),
                    "from": self._parse_from(),
                    "expressions": self._match(TokenType.SET)
                    and self._parse_csv(self._parse_equality),
                    "where": self._parse_where(),
                },
            )

        def _parse_rangen(self):
            this = self._parse_id_var()
            self._match(TokenType.BETWEEN)

            expressions = self._parse_csv(self._parse_conjunction)
            each = self._match_text_seq("EACH") and self._parse_conjunction()

            return self.expression(exp.RangeN, this=this, expressions=expressions, each=each)

        def _parse_cast(self, strict: bool) -> exp.Expression:
            cast = t.cast(exp.Cast, super()._parse_cast(strict))
            if cast.to.this == exp.DataType.Type.DATE and self._match(TokenType.FORMAT):
                return format_time_lambda(exp.TimeToStr, "teradata")(
                    [cast.this, self._parse_string()]
                )
            return cast

    class Generator(generator.Generator):
        JOIN_HINTS = False
        TABLE_HINTS = False

        TYPE_MAPPING = {
            **generator.Generator.TYPE_MAPPING,  # type: ignore
            exp.DataType.Type.GEOMETRY: "ST_GEOMETRY",
        }

        PROPERTIES_LOCATION = {
            **generator.Generator.PROPERTIES_LOCATION,  # type: ignore
            exp.PartitionedByProperty: exp.Properties.Location.POST_INDEX,
        }

        TRANSFORMS = {
            **generator.Generator.TRANSFORMS,
            exp.Max: max_or_greatest,
            exp.Min: min_or_least,
            exp.Select: transforms.preprocess([transforms.eliminate_distinct_on]),
            exp.TimeToStr: lambda self, e: f"CAST({self.sql(e, 'this')} AS DATE FORMAT {self.format_time(e)})",
            exp.ToChar: lambda self, e: self.function_fallback_sql(e),
        }

        def partitionedbyproperty_sql(self, expression: exp.PartitionedByProperty) -> str:
            return f"PARTITION BY {self.sql(expression, 'this')}"

        # FROM before SET in Teradata UPDATE syntax
        # https://docs.teradata.com/r/Enterprise_IntelliFlex_VMware/Teradata-VantageTM-SQL-Data-Manipulation-Language-17.20/Statement-Syntax/UPDATE/UPDATE-Syntax-Basic-Form-FROM-Clause
        def update_sql(self, expression: exp.Update) -> str:
            this = self.sql(expression, "this")
            from_sql = self.sql(expression, "from")
            set_sql = self.expressions(expression, flat=True)
            where_sql = self.sql(expression, "where")
            sql = f"UPDATE {this}{from_sql} SET {set_sql}{where_sql}"
            return self.prepend_ctes(expression, sql)

        def mod_sql(self, expression: exp.Mod) -> str:
            return self.binary(expression, "MOD")

        def datatype_sql(self, expression: exp.DataType) -> str:
            type_sql = super().datatype_sql(expression)
            prefix_sql = expression.args.get("prefix")
            return f"SYSUDTLIB.{type_sql}" if prefix_sql else type_sql

        def rangen_sql(self, expression: exp.RangeN) -> str:
            this = self.sql(expression, "this")
            expressions_sql = self.expressions(expression)
            each_sql = self.sql(expression, "each")
            each_sql = f" EACH {each_sql}" if each_sql else ""

            return f"RANGE_N({this} BETWEEN {expressions_sql}{each_sql})"