Edit on GitHub

sqlglot.dialects.mysql

  1from __future__ import annotations
  2
  3import typing as t
  4
  5from sqlglot import exp, generator, parser, tokens, transforms
  6from sqlglot.dialects.dialect import (
  7    Dialect,
  8    arrow_json_extract_scalar_sql,
  9    datestrtodate_sql,
 10    format_time_lambda,
 11    locate_to_strposition,
 12    max_or_greatest,
 13    min_or_least,
 14    no_ilike_sql,
 15    no_paren_current_date_sql,
 16    no_pivot_sql,
 17    no_tablesample_sql,
 18    no_trycast_sql,
 19    parse_date_delta_with_interval,
 20    rename_func,
 21    strposition_to_locate_sql,
 22)
 23from sqlglot.helper import seq_get
 24from sqlglot.tokens import TokenType
 25
 26
 27def _show_parser(*args: t.Any, **kwargs: t.Any) -> t.Callable[[MySQL.Parser], exp.Show]:
 28    def _parse(self: MySQL.Parser) -> exp.Show:
 29        return self._parse_show_mysql(*args, **kwargs)
 30
 31    return _parse
 32
 33
 34def _date_trunc_sql(self: generator.Generator, expression: exp.DateTrunc) -> str:
 35    expr = self.sql(expression, "this")
 36    unit = expression.text("unit")
 37
 38    if unit == "day":
 39        return f"DATE({expr})"
 40
 41    if unit == "week":
 42        concat = f"CONCAT(YEAR({expr}), ' ', WEEK({expr}, 1), ' 1')"
 43        date_format = "%Y %u %w"
 44    elif unit == "month":
 45        concat = f"CONCAT(YEAR({expr}), ' ', MONTH({expr}), ' 1')"
 46        date_format = "%Y %c %e"
 47    elif unit == "quarter":
 48        concat = f"CONCAT(YEAR({expr}), ' ', QUARTER({expr}) * 3 - 2, ' 1')"
 49        date_format = "%Y %c %e"
 50    elif unit == "year":
 51        concat = f"CONCAT(YEAR({expr}), ' 1 1')"
 52        date_format = "%Y %c %e"
 53    else:
 54        self.unsupported(f"Unexpected interval unit: {unit}")
 55        return f"DATE({expr})"
 56
 57    return f"STR_TO_DATE({concat}, '{date_format}')"
 58
 59
 60def _str_to_date(args: t.List) -> exp.StrToDate:
 61    date_format = MySQL.format_time(seq_get(args, 1))
 62    return exp.StrToDate(this=seq_get(args, 0), format=date_format)
 63
 64
 65def _str_to_date_sql(self: generator.Generator, expression: exp.StrToDate | exp.StrToTime) -> str:
 66    date_format = self.format_time(expression)
 67    return f"STR_TO_DATE({self.sql(expression.this)}, {date_format})"
 68
 69
 70def _trim_sql(self: generator.Generator, expression: exp.Trim) -> str:
 71    target = self.sql(expression, "this")
 72    trim_type = self.sql(expression, "position")
 73    remove_chars = self.sql(expression, "expression")
 74
 75    # Use TRIM/LTRIM/RTRIM syntax if the expression isn't mysql-specific
 76    if not remove_chars:
 77        return self.trim_sql(expression)
 78
 79    trim_type = f"{trim_type} " if trim_type else ""
 80    remove_chars = f"{remove_chars} " if remove_chars else ""
 81    from_part = "FROM " if trim_type or remove_chars else ""
 82    return f"TRIM({trim_type}{remove_chars}{from_part}{target})"
 83
 84
 85def _date_add_sql(kind: str) -> t.Callable[[generator.Generator, exp.DateAdd | exp.DateSub], str]:
 86    def func(self: generator.Generator, expression: exp.DateAdd | exp.DateSub) -> str:
 87        this = self.sql(expression, "this")
 88        unit = expression.text("unit").upper() or "DAY"
 89        return (
 90            f"DATE_{kind}({this}, {self.sql(exp.Interval(this=expression.expression, unit=unit))})"
 91        )
 92
 93    return func
 94
 95
 96class MySQL(Dialect):
 97    time_format = "'%Y-%m-%d %T'"
 98
 99    # https://prestodb.io/docs/current/functions/datetime.html#mysql-date-functions
100    time_mapping = {
101        "%M": "%B",
102        "%c": "%-m",
103        "%e": "%-d",
104        "%h": "%I",
105        "%i": "%M",
106        "%s": "%S",
107        "%S": "%S",
108        "%u": "%W",
109        "%k": "%-H",
110        "%l": "%-I",
111        "%T": "%H:%M:%S",
112        "%W": "%a",
113    }
114
115    class Tokenizer(tokens.Tokenizer):
116        QUOTES = ["'", '"']
117        COMMENTS = ["--", "#", ("/*", "*/")]
118        IDENTIFIERS = ["`"]
119        STRING_ESCAPES = ["'", "\\"]
120        BIT_STRINGS = [("b'", "'"), ("B'", "'"), ("0b", "")]
121        HEX_STRINGS = [("x'", "'"), ("X'", "'"), ("0x", "")]
122
123        KEYWORDS = {
124            **tokens.Tokenizer.KEYWORDS,
125            "CHARSET": TokenType.CHARACTER_SET,
126            "LONGBLOB": TokenType.LONGBLOB,
127            "LONGTEXT": TokenType.LONGTEXT,
128            "MEDIUMBLOB": TokenType.MEDIUMBLOB,
129            "MEDIUMTEXT": TokenType.MEDIUMTEXT,
130            "SEPARATOR": TokenType.SEPARATOR,
131            "START": TokenType.BEGIN,
132            "_ARMSCII8": TokenType.INTRODUCER,
133            "_ASCII": TokenType.INTRODUCER,
134            "_BIG5": TokenType.INTRODUCER,
135            "_BINARY": TokenType.INTRODUCER,
136            "_CP1250": TokenType.INTRODUCER,
137            "_CP1251": TokenType.INTRODUCER,
138            "_CP1256": TokenType.INTRODUCER,
139            "_CP1257": TokenType.INTRODUCER,
140            "_CP850": TokenType.INTRODUCER,
141            "_CP852": TokenType.INTRODUCER,
142            "_CP866": TokenType.INTRODUCER,
143            "_CP932": TokenType.INTRODUCER,
144            "_DEC8": TokenType.INTRODUCER,
145            "_EUCJPMS": TokenType.INTRODUCER,
146            "_EUCKR": TokenType.INTRODUCER,
147            "_GB18030": TokenType.INTRODUCER,
148            "_GB2312": TokenType.INTRODUCER,
149            "_GBK": TokenType.INTRODUCER,
150            "_GEOSTD8": TokenType.INTRODUCER,
151            "_GREEK": TokenType.INTRODUCER,
152            "_HEBREW": TokenType.INTRODUCER,
153            "_HP8": TokenType.INTRODUCER,
154            "_KEYBCS2": TokenType.INTRODUCER,
155            "_KOI8R": TokenType.INTRODUCER,
156            "_KOI8U": TokenType.INTRODUCER,
157            "_LATIN1": TokenType.INTRODUCER,
158            "_LATIN2": TokenType.INTRODUCER,
159            "_LATIN5": TokenType.INTRODUCER,
160            "_LATIN7": TokenType.INTRODUCER,
161            "_MACCE": TokenType.INTRODUCER,
162            "_MACROMAN": TokenType.INTRODUCER,
163            "_SJIS": TokenType.INTRODUCER,
164            "_SWE7": TokenType.INTRODUCER,
165            "_TIS620": TokenType.INTRODUCER,
166            "_UCS2": TokenType.INTRODUCER,
167            "_UJIS": TokenType.INTRODUCER,
168            # https://dev.mysql.com/doc/refman/8.0/en/string-literals.html
169            "_UTF8": TokenType.INTRODUCER,
170            "_UTF16": TokenType.INTRODUCER,
171            "_UTF16LE": TokenType.INTRODUCER,
172            "_UTF32": TokenType.INTRODUCER,
173            "_UTF8MB3": TokenType.INTRODUCER,
174            "_UTF8MB4": TokenType.INTRODUCER,
175            "@@": TokenType.SESSION_PARAMETER,
176        }
177
178        COMMANDS = tokens.Tokenizer.COMMANDS - {TokenType.SHOW}
179
180    class Parser(parser.Parser):
181        FUNC_TOKENS = {*parser.Parser.FUNC_TOKENS, TokenType.SCHEMA, TokenType.DATABASE}
182
183        FUNCTIONS = {
184            **parser.Parser.FUNCTIONS,
185            "DATE_ADD": parse_date_delta_with_interval(exp.DateAdd),
186            "DATE_FORMAT": format_time_lambda(exp.TimeToStr, "mysql"),
187            "DATE_SUB": parse_date_delta_with_interval(exp.DateSub),
188            "INSTR": lambda args: exp.StrPosition(substr=seq_get(args, 1), this=seq_get(args, 0)),
189            "LEFT": lambda args: exp.Substring(
190                this=seq_get(args, 0), start=exp.Literal.number(1), length=seq_get(args, 1)
191            ),
192            "LOCATE": locate_to_strposition,
193            "STR_TO_DATE": _str_to_date,
194        }
195
196        FUNCTION_PARSERS = {
197            **parser.Parser.FUNCTION_PARSERS,
198            "GROUP_CONCAT": lambda self: self.expression(
199                exp.GroupConcat,
200                this=self._parse_lambda(),
201                separator=self._match(TokenType.SEPARATOR) and self._parse_field(),
202            ),
203        }
204
205        STATEMENT_PARSERS = {
206            **parser.Parser.STATEMENT_PARSERS,
207            TokenType.SHOW: lambda self: self._parse_show(),
208        }
209
210        SHOW_PARSERS = {
211            "BINARY LOGS": _show_parser("BINARY LOGS"),
212            "MASTER LOGS": _show_parser("BINARY LOGS"),
213            "BINLOG EVENTS": _show_parser("BINLOG EVENTS"),
214            "CHARACTER SET": _show_parser("CHARACTER SET"),
215            "CHARSET": _show_parser("CHARACTER SET"),
216            "COLLATION": _show_parser("COLLATION"),
217            "FULL COLUMNS": _show_parser("COLUMNS", target="FROM", full=True),
218            "COLUMNS": _show_parser("COLUMNS", target="FROM"),
219            "CREATE DATABASE": _show_parser("CREATE DATABASE", target=True),
220            "CREATE EVENT": _show_parser("CREATE EVENT", target=True),
221            "CREATE FUNCTION": _show_parser("CREATE FUNCTION", target=True),
222            "CREATE PROCEDURE": _show_parser("CREATE PROCEDURE", target=True),
223            "CREATE TABLE": _show_parser("CREATE TABLE", target=True),
224            "CREATE TRIGGER": _show_parser("CREATE TRIGGER", target=True),
225            "CREATE VIEW": _show_parser("CREATE VIEW", target=True),
226            "DATABASES": _show_parser("DATABASES"),
227            "ENGINE": _show_parser("ENGINE", target=True),
228            "STORAGE ENGINES": _show_parser("ENGINES"),
229            "ENGINES": _show_parser("ENGINES"),
230            "ERRORS": _show_parser("ERRORS"),
231            "EVENTS": _show_parser("EVENTS"),
232            "FUNCTION CODE": _show_parser("FUNCTION CODE", target=True),
233            "FUNCTION STATUS": _show_parser("FUNCTION STATUS"),
234            "GRANTS": _show_parser("GRANTS", target="FOR"),
235            "INDEX": _show_parser("INDEX", target="FROM"),
236            "MASTER STATUS": _show_parser("MASTER STATUS"),
237            "OPEN TABLES": _show_parser("OPEN TABLES"),
238            "PLUGINS": _show_parser("PLUGINS"),
239            "PROCEDURE CODE": _show_parser("PROCEDURE CODE", target=True),
240            "PROCEDURE STATUS": _show_parser("PROCEDURE STATUS"),
241            "PRIVILEGES": _show_parser("PRIVILEGES"),
242            "FULL PROCESSLIST": _show_parser("PROCESSLIST", full=True),
243            "PROCESSLIST": _show_parser("PROCESSLIST"),
244            "PROFILE": _show_parser("PROFILE"),
245            "PROFILES": _show_parser("PROFILES"),
246            "RELAYLOG EVENTS": _show_parser("RELAYLOG EVENTS"),
247            "REPLICAS": _show_parser("REPLICAS"),
248            "SLAVE HOSTS": _show_parser("REPLICAS"),
249            "REPLICA STATUS": _show_parser("REPLICA STATUS"),
250            "SLAVE STATUS": _show_parser("REPLICA STATUS"),
251            "GLOBAL STATUS": _show_parser("STATUS", global_=True),
252            "SESSION STATUS": _show_parser("STATUS"),
253            "STATUS": _show_parser("STATUS"),
254            "TABLE STATUS": _show_parser("TABLE STATUS"),
255            "FULL TABLES": _show_parser("TABLES", full=True),
256            "TABLES": _show_parser("TABLES"),
257            "TRIGGERS": _show_parser("TRIGGERS"),
258            "GLOBAL VARIABLES": _show_parser("VARIABLES", global_=True),
259            "SESSION VARIABLES": _show_parser("VARIABLES"),
260            "VARIABLES": _show_parser("VARIABLES"),
261            "WARNINGS": _show_parser("WARNINGS"),
262        }
263
264        SET_PARSERS = {
265            **parser.Parser.SET_PARSERS,
266            "PERSIST": lambda self: self._parse_set_item_assignment("PERSIST"),
267            "PERSIST_ONLY": lambda self: self._parse_set_item_assignment("PERSIST_ONLY"),
268            "CHARACTER SET": lambda self: self._parse_set_item_charset("CHARACTER SET"),
269            "CHARSET": lambda self: self._parse_set_item_charset("CHARACTER SET"),
270            "NAMES": lambda self: self._parse_set_item_names(),
271        }
272
273        PROFILE_TYPES = {
274            "ALL",
275            "BLOCK IO",
276            "CONTEXT SWITCHES",
277            "CPU",
278            "IPC",
279            "MEMORY",
280            "PAGE FAULTS",
281            "SOURCE",
282            "SWAPS",
283        }
284
285        LOG_DEFAULTS_TO_LN = True
286
287        def _parse_show_mysql(
288            self,
289            this: str,
290            target: bool | str = False,
291            full: t.Optional[bool] = None,
292            global_: t.Optional[bool] = None,
293        ) -> exp.Show:
294            if target:
295                if isinstance(target, str):
296                    self._match_text_seq(target)
297                target_id = self._parse_id_var()
298            else:
299                target_id = None
300
301            log = self._parse_string() if self._match_text_seq("IN") else None
302
303            if this in {"BINLOG EVENTS", "RELAYLOG EVENTS"}:
304                position = self._parse_number() if self._match_text_seq("FROM") else None
305                db = None
306            else:
307                position = None
308                db = None
309
310                if self._match(TokenType.FROM):
311                    db = self._parse_id_var()
312                elif self._match(TokenType.DOT):
313                    db = target_id
314                    target_id = self._parse_id_var()
315
316            channel = self._parse_id_var() if self._match_text_seq("FOR", "CHANNEL") else None
317
318            like = self._parse_string() if self._match_text_seq("LIKE") else None
319            where = self._parse_where()
320
321            if this == "PROFILE":
322                types = self._parse_csv(lambda: self._parse_var_from_options(self.PROFILE_TYPES))
323                query = self._parse_number() if self._match_text_seq("FOR", "QUERY") else None
324                offset = self._parse_number() if self._match_text_seq("OFFSET") else None
325                limit = self._parse_number() if self._match_text_seq("LIMIT") else None
326            else:
327                types, query = None, None
328                offset, limit = self._parse_oldstyle_limit()
329
330            mutex = True if self._match_text_seq("MUTEX") else None
331            mutex = False if self._match_text_seq("STATUS") else mutex
332
333            return self.expression(
334                exp.Show,
335                this=this,
336                target=target_id,
337                full=full,
338                log=log,
339                position=position,
340                db=db,
341                channel=channel,
342                like=like,
343                where=where,
344                types=types,
345                query=query,
346                offset=offset,
347                limit=limit,
348                mutex=mutex,
349                **{"global": global_},  # type: ignore
350            )
351
352        def _parse_oldstyle_limit(
353            self,
354        ) -> t.Tuple[t.Optional[exp.Expression], t.Optional[exp.Expression]]:
355            limit = None
356            offset = None
357            if self._match_text_seq("LIMIT"):
358                parts = self._parse_csv(self._parse_number)
359                if len(parts) == 1:
360                    limit = parts[0]
361                elif len(parts) == 2:
362                    limit = parts[1]
363                    offset = parts[0]
364
365            return offset, limit
366
367        def _parse_set_item_charset(self, kind: str) -> exp.Expression:
368            this = self._parse_string() or self._parse_id_var()
369            return self.expression(exp.SetItem, this=this, kind=kind)
370
371        def _parse_set_item_names(self) -> exp.Expression:
372            charset = self._parse_string() or self._parse_id_var()
373            if self._match_text_seq("COLLATE"):
374                collate = self._parse_string() or self._parse_id_var()
375            else:
376                collate = None
377
378            return self.expression(
379                exp.SetItem,
380                this=charset,
381                collate=collate,
382                kind="NAMES",
383            )
384
385    class Generator(generator.Generator):
386        LOCKING_READS_SUPPORTED = True
387        NULL_ORDERING_SUPPORTED = False
388        JOIN_HINTS = False
389        TABLE_HINTS = False
390
391        TRANSFORMS = {
392            **generator.Generator.TRANSFORMS,
393            exp.CurrentDate: no_paren_current_date_sql,
394            exp.DateDiff: lambda self, e: self.func("DATEDIFF", e.this, e.expression),
395            exp.DateAdd: _date_add_sql("ADD"),
396            exp.DateStrToDate: datestrtodate_sql,
397            exp.DateSub: _date_add_sql("SUB"),
398            exp.DateTrunc: _date_trunc_sql,
399            exp.DayOfMonth: rename_func("DAYOFMONTH"),
400            exp.DayOfWeek: rename_func("DAYOFWEEK"),
401            exp.DayOfYear: rename_func("DAYOFYEAR"),
402            exp.GroupConcat: lambda self, e: f"""GROUP_CONCAT({self.sql(e, "this")} SEPARATOR {self.sql(e, "separator") or "','"})""",
403            exp.ILike: no_ilike_sql,
404            exp.JSONExtractScalar: arrow_json_extract_scalar_sql,
405            exp.Max: max_or_greatest,
406            exp.Min: min_or_least,
407            exp.NullSafeEQ: lambda self, e: self.binary(e, "<=>"),
408            exp.NullSafeNEQ: lambda self, e: self.not_sql(self.binary(e, "<=>")),
409            exp.Pivot: no_pivot_sql,
410            exp.Select: transforms.preprocess([transforms.eliminate_distinct_on]),
411            exp.StrPosition: strposition_to_locate_sql,
412            exp.StrToDate: _str_to_date_sql,
413            exp.StrToTime: _str_to_date_sql,
414            exp.TableSample: no_tablesample_sql,
415            exp.TimeStrToUnix: rename_func("UNIX_TIMESTAMP"),
416            exp.TimeToStr: lambda self, e: self.func("DATE_FORMAT", e.this, self.format_time(e)),
417            exp.Trim: _trim_sql,
418            exp.TryCast: no_trycast_sql,
419            exp.WeekOfYear: rename_func("WEEKOFYEAR"),
420        }
421
422        TYPE_MAPPING = generator.Generator.TYPE_MAPPING.copy()
423        TYPE_MAPPING.pop(exp.DataType.Type.MEDIUMTEXT)
424        TYPE_MAPPING.pop(exp.DataType.Type.LONGTEXT)
425        TYPE_MAPPING.pop(exp.DataType.Type.MEDIUMBLOB)
426        TYPE_MAPPING.pop(exp.DataType.Type.LONGBLOB)
427
428        PROPERTIES_LOCATION = {
429            **generator.Generator.PROPERTIES_LOCATION,
430            exp.TransientProperty: exp.Properties.Location.UNSUPPORTED,
431            exp.VolatileProperty: exp.Properties.Location.UNSUPPORTED,
432        }
433
434        LIMIT_FETCH = "LIMIT"
435
436        def show_sql(self, expression: exp.Show) -> str:
437            this = f" {expression.name}"
438            full = " FULL" if expression.args.get("full") else ""
439            global_ = " GLOBAL" if expression.args.get("global") else ""
440
441            target = self.sql(expression, "target")
442            target = f" {target}" if target else ""
443            if expression.name in {"COLUMNS", "INDEX"}:
444                target = f" FROM{target}"
445            elif expression.name == "GRANTS":
446                target = f" FOR{target}"
447
448            db = self._prefixed_sql("FROM", expression, "db")
449
450            like = self._prefixed_sql("LIKE", expression, "like")
451            where = self.sql(expression, "where")
452
453            types = self.expressions(expression, key="types")
454            types = f" {types}" if types else types
455            query = self._prefixed_sql("FOR QUERY", expression, "query")
456
457            if expression.name == "PROFILE":
458                offset = self._prefixed_sql("OFFSET", expression, "offset")
459                limit = self._prefixed_sql("LIMIT", expression, "limit")
460            else:
461                offset = ""
462                limit = self._oldstyle_limit_sql(expression)
463
464            log = self._prefixed_sql("IN", expression, "log")
465            position = self._prefixed_sql("FROM", expression, "position")
466
467            channel = self._prefixed_sql("FOR CHANNEL", expression, "channel")
468
469            if expression.name == "ENGINE":
470                mutex_or_status = " MUTEX" if expression.args.get("mutex") else " STATUS"
471            else:
472                mutex_or_status = ""
473
474            return f"SHOW{full}{global_}{this}{target}{types}{db}{query}{log}{position}{channel}{mutex_or_status}{like}{where}{offset}{limit}"
475
476        def _prefixed_sql(self, prefix: str, expression: exp.Expression, arg: str) -> str:
477            sql = self.sql(expression, arg)
478            if not sql:
479                return ""
480            return f" {prefix} {sql}"
481
482        def _oldstyle_limit_sql(self, expression: exp.Show) -> str:
483            limit = self.sql(expression, "limit")
484            offset = self.sql(expression, "offset")
485            if limit:
486                limit_offset = f"{offset}, {limit}" if offset else limit
487                return f" LIMIT {limit_offset}"
488            return ""
class MySQL(sqlglot.dialects.dialect.Dialect):
 97class MySQL(Dialect):
 98    time_format = "'%Y-%m-%d %T'"
 99
100    # https://prestodb.io/docs/current/functions/datetime.html#mysql-date-functions
101    time_mapping = {
102        "%M": "%B",
103        "%c": "%-m",
104        "%e": "%-d",
105        "%h": "%I",
106        "%i": "%M",
107        "%s": "%S",
108        "%S": "%S",
109        "%u": "%W",
110        "%k": "%-H",
111        "%l": "%-I",
112        "%T": "%H:%M:%S",
113        "%W": "%a",
114    }
115
116    class Tokenizer(tokens.Tokenizer):
117        QUOTES = ["'", '"']
118        COMMENTS = ["--", "#", ("/*", "*/")]
119        IDENTIFIERS = ["`"]
120        STRING_ESCAPES = ["'", "\\"]
121        BIT_STRINGS = [("b'", "'"), ("B'", "'"), ("0b", "")]
122        HEX_STRINGS = [("x'", "'"), ("X'", "'"), ("0x", "")]
123
124        KEYWORDS = {
125            **tokens.Tokenizer.KEYWORDS,
126            "CHARSET": TokenType.CHARACTER_SET,
127            "LONGBLOB": TokenType.LONGBLOB,
128            "LONGTEXT": TokenType.LONGTEXT,
129            "MEDIUMBLOB": TokenType.MEDIUMBLOB,
130            "MEDIUMTEXT": TokenType.MEDIUMTEXT,
131            "SEPARATOR": TokenType.SEPARATOR,
132            "START": TokenType.BEGIN,
133            "_ARMSCII8": TokenType.INTRODUCER,
134            "_ASCII": TokenType.INTRODUCER,
135            "_BIG5": TokenType.INTRODUCER,
136            "_BINARY": TokenType.INTRODUCER,
137            "_CP1250": TokenType.INTRODUCER,
138            "_CP1251": TokenType.INTRODUCER,
139            "_CP1256": TokenType.INTRODUCER,
140            "_CP1257": TokenType.INTRODUCER,
141            "_CP850": TokenType.INTRODUCER,
142            "_CP852": TokenType.INTRODUCER,
143            "_CP866": TokenType.INTRODUCER,
144            "_CP932": TokenType.INTRODUCER,
145            "_DEC8": TokenType.INTRODUCER,
146            "_EUCJPMS": TokenType.INTRODUCER,
147            "_EUCKR": TokenType.INTRODUCER,
148            "_GB18030": TokenType.INTRODUCER,
149            "_GB2312": TokenType.INTRODUCER,
150            "_GBK": TokenType.INTRODUCER,
151            "_GEOSTD8": TokenType.INTRODUCER,
152            "_GREEK": TokenType.INTRODUCER,
153            "_HEBREW": TokenType.INTRODUCER,
154            "_HP8": TokenType.INTRODUCER,
155            "_KEYBCS2": TokenType.INTRODUCER,
156            "_KOI8R": TokenType.INTRODUCER,
157            "_KOI8U": TokenType.INTRODUCER,
158            "_LATIN1": TokenType.INTRODUCER,
159            "_LATIN2": TokenType.INTRODUCER,
160            "_LATIN5": TokenType.INTRODUCER,
161            "_LATIN7": TokenType.INTRODUCER,
162            "_MACCE": TokenType.INTRODUCER,
163            "_MACROMAN": TokenType.INTRODUCER,
164            "_SJIS": TokenType.INTRODUCER,
165            "_SWE7": TokenType.INTRODUCER,
166            "_TIS620": TokenType.INTRODUCER,
167            "_UCS2": TokenType.INTRODUCER,
168            "_UJIS": TokenType.INTRODUCER,
169            # https://dev.mysql.com/doc/refman/8.0/en/string-literals.html
170            "_UTF8": TokenType.INTRODUCER,
171            "_UTF16": TokenType.INTRODUCER,
172            "_UTF16LE": TokenType.INTRODUCER,
173            "_UTF32": TokenType.INTRODUCER,
174            "_UTF8MB3": TokenType.INTRODUCER,
175            "_UTF8MB4": TokenType.INTRODUCER,
176            "@@": TokenType.SESSION_PARAMETER,
177        }
178
179        COMMANDS = tokens.Tokenizer.COMMANDS - {TokenType.SHOW}
180
181    class Parser(parser.Parser):
182        FUNC_TOKENS = {*parser.Parser.FUNC_TOKENS, TokenType.SCHEMA, TokenType.DATABASE}
183
184        FUNCTIONS = {
185            **parser.Parser.FUNCTIONS,
186            "DATE_ADD": parse_date_delta_with_interval(exp.DateAdd),
187            "DATE_FORMAT": format_time_lambda(exp.TimeToStr, "mysql"),
188            "DATE_SUB": parse_date_delta_with_interval(exp.DateSub),
189            "INSTR": lambda args: exp.StrPosition(substr=seq_get(args, 1), this=seq_get(args, 0)),
190            "LEFT": lambda args: exp.Substring(
191                this=seq_get(args, 0), start=exp.Literal.number(1), length=seq_get(args, 1)
192            ),
193            "LOCATE": locate_to_strposition,
194            "STR_TO_DATE": _str_to_date,
195        }
196
197        FUNCTION_PARSERS = {
198            **parser.Parser.FUNCTION_PARSERS,
199            "GROUP_CONCAT": lambda self: self.expression(
200                exp.GroupConcat,
201                this=self._parse_lambda(),
202                separator=self._match(TokenType.SEPARATOR) and self._parse_field(),
203            ),
204        }
205
206        STATEMENT_PARSERS = {
207            **parser.Parser.STATEMENT_PARSERS,
208            TokenType.SHOW: lambda self: self._parse_show(),
209        }
210
211        SHOW_PARSERS = {
212            "BINARY LOGS": _show_parser("BINARY LOGS"),
213            "MASTER LOGS": _show_parser("BINARY LOGS"),
214            "BINLOG EVENTS": _show_parser("BINLOG EVENTS"),
215            "CHARACTER SET": _show_parser("CHARACTER SET"),
216            "CHARSET": _show_parser("CHARACTER SET"),
217            "COLLATION": _show_parser("COLLATION"),
218            "FULL COLUMNS": _show_parser("COLUMNS", target="FROM", full=True),
219            "COLUMNS": _show_parser("COLUMNS", target="FROM"),
220            "CREATE DATABASE": _show_parser("CREATE DATABASE", target=True),
221            "CREATE EVENT": _show_parser("CREATE EVENT", target=True),
222            "CREATE FUNCTION": _show_parser("CREATE FUNCTION", target=True),
223            "CREATE PROCEDURE": _show_parser("CREATE PROCEDURE", target=True),
224            "CREATE TABLE": _show_parser("CREATE TABLE", target=True),
225            "CREATE TRIGGER": _show_parser("CREATE TRIGGER", target=True),
226            "CREATE VIEW": _show_parser("CREATE VIEW", target=True),
227            "DATABASES": _show_parser("DATABASES"),
228            "ENGINE": _show_parser("ENGINE", target=True),
229            "STORAGE ENGINES": _show_parser("ENGINES"),
230            "ENGINES": _show_parser("ENGINES"),
231            "ERRORS": _show_parser("ERRORS"),
232            "EVENTS": _show_parser("EVENTS"),
233            "FUNCTION CODE": _show_parser("FUNCTION CODE", target=True),
234            "FUNCTION STATUS": _show_parser("FUNCTION STATUS"),
235            "GRANTS": _show_parser("GRANTS", target="FOR"),
236            "INDEX": _show_parser("INDEX", target="FROM"),
237            "MASTER STATUS": _show_parser("MASTER STATUS"),
238            "OPEN TABLES": _show_parser("OPEN TABLES"),
239            "PLUGINS": _show_parser("PLUGINS"),
240            "PROCEDURE CODE": _show_parser("PROCEDURE CODE", target=True),
241            "PROCEDURE STATUS": _show_parser("PROCEDURE STATUS"),
242            "PRIVILEGES": _show_parser("PRIVILEGES"),
243            "FULL PROCESSLIST": _show_parser("PROCESSLIST", full=True),
244            "PROCESSLIST": _show_parser("PROCESSLIST"),
245            "PROFILE": _show_parser("PROFILE"),
246            "PROFILES": _show_parser("PROFILES"),
247            "RELAYLOG EVENTS": _show_parser("RELAYLOG EVENTS"),
248            "REPLICAS": _show_parser("REPLICAS"),
249            "SLAVE HOSTS": _show_parser("REPLICAS"),
250            "REPLICA STATUS": _show_parser("REPLICA STATUS"),
251            "SLAVE STATUS": _show_parser("REPLICA STATUS"),
252            "GLOBAL STATUS": _show_parser("STATUS", global_=True),
253            "SESSION STATUS": _show_parser("STATUS"),
254            "STATUS": _show_parser("STATUS"),
255            "TABLE STATUS": _show_parser("TABLE STATUS"),
256            "FULL TABLES": _show_parser("TABLES", full=True),
257            "TABLES": _show_parser("TABLES"),
258            "TRIGGERS": _show_parser("TRIGGERS"),
259            "GLOBAL VARIABLES": _show_parser("VARIABLES", global_=True),
260            "SESSION VARIABLES": _show_parser("VARIABLES"),
261            "VARIABLES": _show_parser("VARIABLES"),
262            "WARNINGS": _show_parser("WARNINGS"),
263        }
264
265        SET_PARSERS = {
266            **parser.Parser.SET_PARSERS,
267            "PERSIST": lambda self: self._parse_set_item_assignment("PERSIST"),
268            "PERSIST_ONLY": lambda self: self._parse_set_item_assignment("PERSIST_ONLY"),
269            "CHARACTER SET": lambda self: self._parse_set_item_charset("CHARACTER SET"),
270            "CHARSET": lambda self: self._parse_set_item_charset("CHARACTER SET"),
271            "NAMES": lambda self: self._parse_set_item_names(),
272        }
273
274        PROFILE_TYPES = {
275            "ALL",
276            "BLOCK IO",
277            "CONTEXT SWITCHES",
278            "CPU",
279            "IPC",
280            "MEMORY",
281            "PAGE FAULTS",
282            "SOURCE",
283            "SWAPS",
284        }
285
286        LOG_DEFAULTS_TO_LN = True
287
288        def _parse_show_mysql(
289            self,
290            this: str,
291            target: bool | str = False,
292            full: t.Optional[bool] = None,
293            global_: t.Optional[bool] = None,
294        ) -> exp.Show:
295            if target:
296                if isinstance(target, str):
297                    self._match_text_seq(target)
298                target_id = self._parse_id_var()
299            else:
300                target_id = None
301
302            log = self._parse_string() if self._match_text_seq("IN") else None
303
304            if this in {"BINLOG EVENTS", "RELAYLOG EVENTS"}:
305                position = self._parse_number() if self._match_text_seq("FROM") else None
306                db = None
307            else:
308                position = None
309                db = None
310
311                if self._match(TokenType.FROM):
312                    db = self._parse_id_var()
313                elif self._match(TokenType.DOT):
314                    db = target_id
315                    target_id = self._parse_id_var()
316
317            channel = self._parse_id_var() if self._match_text_seq("FOR", "CHANNEL") else None
318
319            like = self._parse_string() if self._match_text_seq("LIKE") else None
320            where = self._parse_where()
321
322            if this == "PROFILE":
323                types = self._parse_csv(lambda: self._parse_var_from_options(self.PROFILE_TYPES))
324                query = self._parse_number() if self._match_text_seq("FOR", "QUERY") else None
325                offset = self._parse_number() if self._match_text_seq("OFFSET") else None
326                limit = self._parse_number() if self._match_text_seq("LIMIT") else None
327            else:
328                types, query = None, None
329                offset, limit = self._parse_oldstyle_limit()
330
331            mutex = True if self._match_text_seq("MUTEX") else None
332            mutex = False if self._match_text_seq("STATUS") else mutex
333
334            return self.expression(
335                exp.Show,
336                this=this,
337                target=target_id,
338                full=full,
339                log=log,
340                position=position,
341                db=db,
342                channel=channel,
343                like=like,
344                where=where,
345                types=types,
346                query=query,
347                offset=offset,
348                limit=limit,
349                mutex=mutex,
350                **{"global": global_},  # type: ignore
351            )
352
353        def _parse_oldstyle_limit(
354            self,
355        ) -> t.Tuple[t.Optional[exp.Expression], t.Optional[exp.Expression]]:
356            limit = None
357            offset = None
358            if self._match_text_seq("LIMIT"):
359                parts = self._parse_csv(self._parse_number)
360                if len(parts) == 1:
361                    limit = parts[0]
362                elif len(parts) == 2:
363                    limit = parts[1]
364                    offset = parts[0]
365
366            return offset, limit
367
368        def _parse_set_item_charset(self, kind: str) -> exp.Expression:
369            this = self._parse_string() or self._parse_id_var()
370            return self.expression(exp.SetItem, this=this, kind=kind)
371
372        def _parse_set_item_names(self) -> exp.Expression:
373            charset = self._parse_string() or self._parse_id_var()
374            if self._match_text_seq("COLLATE"):
375                collate = self._parse_string() or self._parse_id_var()
376            else:
377                collate = None
378
379            return self.expression(
380                exp.SetItem,
381                this=charset,
382                collate=collate,
383                kind="NAMES",
384            )
385
386    class Generator(generator.Generator):
387        LOCKING_READS_SUPPORTED = True
388        NULL_ORDERING_SUPPORTED = False
389        JOIN_HINTS = False
390        TABLE_HINTS = False
391
392        TRANSFORMS = {
393            **generator.Generator.TRANSFORMS,
394            exp.CurrentDate: no_paren_current_date_sql,
395            exp.DateDiff: lambda self, e: self.func("DATEDIFF", e.this, e.expression),
396            exp.DateAdd: _date_add_sql("ADD"),
397            exp.DateStrToDate: datestrtodate_sql,
398            exp.DateSub: _date_add_sql("SUB"),
399            exp.DateTrunc: _date_trunc_sql,
400            exp.DayOfMonth: rename_func("DAYOFMONTH"),
401            exp.DayOfWeek: rename_func("DAYOFWEEK"),
402            exp.DayOfYear: rename_func("DAYOFYEAR"),
403            exp.GroupConcat: lambda self, e: f"""GROUP_CONCAT({self.sql(e, "this")} SEPARATOR {self.sql(e, "separator") or "','"})""",
404            exp.ILike: no_ilike_sql,
405            exp.JSONExtractScalar: arrow_json_extract_scalar_sql,
406            exp.Max: max_or_greatest,
407            exp.Min: min_or_least,
408            exp.NullSafeEQ: lambda self, e: self.binary(e, "<=>"),
409            exp.NullSafeNEQ: lambda self, e: self.not_sql(self.binary(e, "<=>")),
410            exp.Pivot: no_pivot_sql,
411            exp.Select: transforms.preprocess([transforms.eliminate_distinct_on]),
412            exp.StrPosition: strposition_to_locate_sql,
413            exp.StrToDate: _str_to_date_sql,
414            exp.StrToTime: _str_to_date_sql,
415            exp.TableSample: no_tablesample_sql,
416            exp.TimeStrToUnix: rename_func("UNIX_TIMESTAMP"),
417            exp.TimeToStr: lambda self, e: self.func("DATE_FORMAT", e.this, self.format_time(e)),
418            exp.Trim: _trim_sql,
419            exp.TryCast: no_trycast_sql,
420            exp.WeekOfYear: rename_func("WEEKOFYEAR"),
421        }
422
423        TYPE_MAPPING = generator.Generator.TYPE_MAPPING.copy()
424        TYPE_MAPPING.pop(exp.DataType.Type.MEDIUMTEXT)
425        TYPE_MAPPING.pop(exp.DataType.Type.LONGTEXT)
426        TYPE_MAPPING.pop(exp.DataType.Type.MEDIUMBLOB)
427        TYPE_MAPPING.pop(exp.DataType.Type.LONGBLOB)
428
429        PROPERTIES_LOCATION = {
430            **generator.Generator.PROPERTIES_LOCATION,
431            exp.TransientProperty: exp.Properties.Location.UNSUPPORTED,
432            exp.VolatileProperty: exp.Properties.Location.UNSUPPORTED,
433        }
434
435        LIMIT_FETCH = "LIMIT"
436
437        def show_sql(self, expression: exp.Show) -> str:
438            this = f" {expression.name}"
439            full = " FULL" if expression.args.get("full") else ""
440            global_ = " GLOBAL" if expression.args.get("global") else ""
441
442            target = self.sql(expression, "target")
443            target = f" {target}" if target else ""
444            if expression.name in {"COLUMNS", "INDEX"}:
445                target = f" FROM{target}"
446            elif expression.name == "GRANTS":
447                target = f" FOR{target}"
448
449            db = self._prefixed_sql("FROM", expression, "db")
450
451            like = self._prefixed_sql("LIKE", expression, "like")
452            where = self.sql(expression, "where")
453
454            types = self.expressions(expression, key="types")
455            types = f" {types}" if types else types
456            query = self._prefixed_sql("FOR QUERY", expression, "query")
457
458            if expression.name == "PROFILE":
459                offset = self._prefixed_sql("OFFSET", expression, "offset")
460                limit = self._prefixed_sql("LIMIT", expression, "limit")
461            else:
462                offset = ""
463                limit = self._oldstyle_limit_sql(expression)
464
465            log = self._prefixed_sql("IN", expression, "log")
466            position = self._prefixed_sql("FROM", expression, "position")
467
468            channel = self._prefixed_sql("FOR CHANNEL", expression, "channel")
469
470            if expression.name == "ENGINE":
471                mutex_or_status = " MUTEX" if expression.args.get("mutex") else " STATUS"
472            else:
473                mutex_or_status = ""
474
475            return f"SHOW{full}{global_}{this}{target}{types}{db}{query}{log}{position}{channel}{mutex_or_status}{like}{where}{offset}{limit}"
476
477        def _prefixed_sql(self, prefix: str, expression: exp.Expression, arg: str) -> str:
478            sql = self.sql(expression, arg)
479            if not sql:
480                return ""
481            return f" {prefix} {sql}"
482
483        def _oldstyle_limit_sql(self, expression: exp.Show) -> str:
484            limit = self.sql(expression, "limit")
485            offset = self.sql(expression, "offset")
486            if limit:
487                limit_offset = f"{offset}, {limit}" if offset else limit
488                return f" LIMIT {limit_offset}"
489            return ""
class MySQL.Tokenizer(sqlglot.tokens.Tokenizer):
116    class Tokenizer(tokens.Tokenizer):
117        QUOTES = ["'", '"']
118        COMMENTS = ["--", "#", ("/*", "*/")]
119        IDENTIFIERS = ["`"]
120        STRING_ESCAPES = ["'", "\\"]
121        BIT_STRINGS = [("b'", "'"), ("B'", "'"), ("0b", "")]
122        HEX_STRINGS = [("x'", "'"), ("X'", "'"), ("0x", "")]
123
124        KEYWORDS = {
125            **tokens.Tokenizer.KEYWORDS,
126            "CHARSET": TokenType.CHARACTER_SET,
127            "LONGBLOB": TokenType.LONGBLOB,
128            "LONGTEXT": TokenType.LONGTEXT,
129            "MEDIUMBLOB": TokenType.MEDIUMBLOB,
130            "MEDIUMTEXT": TokenType.MEDIUMTEXT,
131            "SEPARATOR": TokenType.SEPARATOR,
132            "START": TokenType.BEGIN,
133            "_ARMSCII8": TokenType.INTRODUCER,
134            "_ASCII": TokenType.INTRODUCER,
135            "_BIG5": TokenType.INTRODUCER,
136            "_BINARY": TokenType.INTRODUCER,
137            "_CP1250": TokenType.INTRODUCER,
138            "_CP1251": TokenType.INTRODUCER,
139            "_CP1256": TokenType.INTRODUCER,
140            "_CP1257": TokenType.INTRODUCER,
141            "_CP850": TokenType.INTRODUCER,
142            "_CP852": TokenType.INTRODUCER,
143            "_CP866": TokenType.INTRODUCER,
144            "_CP932": TokenType.INTRODUCER,
145            "_DEC8": TokenType.INTRODUCER,
146            "_EUCJPMS": TokenType.INTRODUCER,
147            "_EUCKR": TokenType.INTRODUCER,
148            "_GB18030": TokenType.INTRODUCER,
149            "_GB2312": TokenType.INTRODUCER,
150            "_GBK": TokenType.INTRODUCER,
151            "_GEOSTD8": TokenType.INTRODUCER,
152            "_GREEK": TokenType.INTRODUCER,
153            "_HEBREW": TokenType.INTRODUCER,
154            "_HP8": TokenType.INTRODUCER,
155            "_KEYBCS2": TokenType.INTRODUCER,
156            "_KOI8R": TokenType.INTRODUCER,
157            "_KOI8U": TokenType.INTRODUCER,
158            "_LATIN1": TokenType.INTRODUCER,
159            "_LATIN2": TokenType.INTRODUCER,
160            "_LATIN5": TokenType.INTRODUCER,
161            "_LATIN7": TokenType.INTRODUCER,
162            "_MACCE": TokenType.INTRODUCER,
163            "_MACROMAN": TokenType.INTRODUCER,
164            "_SJIS": TokenType.INTRODUCER,
165            "_SWE7": TokenType.INTRODUCER,
166            "_TIS620": TokenType.INTRODUCER,
167            "_UCS2": TokenType.INTRODUCER,
168            "_UJIS": TokenType.INTRODUCER,
169            # https://dev.mysql.com/doc/refman/8.0/en/string-literals.html
170            "_UTF8": TokenType.INTRODUCER,
171            "_UTF16": TokenType.INTRODUCER,
172            "_UTF16LE": TokenType.INTRODUCER,
173            "_UTF32": TokenType.INTRODUCER,
174            "_UTF8MB3": TokenType.INTRODUCER,
175            "_UTF8MB4": TokenType.INTRODUCER,
176            "@@": TokenType.SESSION_PARAMETER,
177        }
178
179        COMMANDS = tokens.Tokenizer.COMMANDS - {TokenType.SHOW}
class MySQL.Parser(sqlglot.parser.Parser):
181    class Parser(parser.Parser):
182        FUNC_TOKENS = {*parser.Parser.FUNC_TOKENS, TokenType.SCHEMA, TokenType.DATABASE}
183
184        FUNCTIONS = {
185            **parser.Parser.FUNCTIONS,
186            "DATE_ADD": parse_date_delta_with_interval(exp.DateAdd),
187            "DATE_FORMAT": format_time_lambda(exp.TimeToStr, "mysql"),
188            "DATE_SUB": parse_date_delta_with_interval(exp.DateSub),
189            "INSTR": lambda args: exp.StrPosition(substr=seq_get(args, 1), this=seq_get(args, 0)),
190            "LEFT": lambda args: exp.Substring(
191                this=seq_get(args, 0), start=exp.Literal.number(1), length=seq_get(args, 1)
192            ),
193            "LOCATE": locate_to_strposition,
194            "STR_TO_DATE": _str_to_date,
195        }
196
197        FUNCTION_PARSERS = {
198            **parser.Parser.FUNCTION_PARSERS,
199            "GROUP_CONCAT": lambda self: self.expression(
200                exp.GroupConcat,
201                this=self._parse_lambda(),
202                separator=self._match(TokenType.SEPARATOR) and self._parse_field(),
203            ),
204        }
205
206        STATEMENT_PARSERS = {
207            **parser.Parser.STATEMENT_PARSERS,
208            TokenType.SHOW: lambda self: self._parse_show(),
209        }
210
211        SHOW_PARSERS = {
212            "BINARY LOGS": _show_parser("BINARY LOGS"),
213            "MASTER LOGS": _show_parser("BINARY LOGS"),
214            "BINLOG EVENTS": _show_parser("BINLOG EVENTS"),
215            "CHARACTER SET": _show_parser("CHARACTER SET"),
216            "CHARSET": _show_parser("CHARACTER SET"),
217            "COLLATION": _show_parser("COLLATION"),
218            "FULL COLUMNS": _show_parser("COLUMNS", target="FROM", full=True),
219            "COLUMNS": _show_parser("COLUMNS", target="FROM"),
220            "CREATE DATABASE": _show_parser("CREATE DATABASE", target=True),
221            "CREATE EVENT": _show_parser("CREATE EVENT", target=True),
222            "CREATE FUNCTION": _show_parser("CREATE FUNCTION", target=True),
223            "CREATE PROCEDURE": _show_parser("CREATE PROCEDURE", target=True),
224            "CREATE TABLE": _show_parser("CREATE TABLE", target=True),
225            "CREATE TRIGGER": _show_parser("CREATE TRIGGER", target=True),
226            "CREATE VIEW": _show_parser("CREATE VIEW", target=True),
227            "DATABASES": _show_parser("DATABASES"),
228            "ENGINE": _show_parser("ENGINE", target=True),
229            "STORAGE ENGINES": _show_parser("ENGINES"),
230            "ENGINES": _show_parser("ENGINES"),
231            "ERRORS": _show_parser("ERRORS"),
232            "EVENTS": _show_parser("EVENTS"),
233            "FUNCTION CODE": _show_parser("FUNCTION CODE", target=True),
234            "FUNCTION STATUS": _show_parser("FUNCTION STATUS"),
235            "GRANTS": _show_parser("GRANTS", target="FOR"),
236            "INDEX": _show_parser("INDEX", target="FROM"),
237            "MASTER STATUS": _show_parser("MASTER STATUS"),
238            "OPEN TABLES": _show_parser("OPEN TABLES"),
239            "PLUGINS": _show_parser("PLUGINS"),
240            "PROCEDURE CODE": _show_parser("PROCEDURE CODE", target=True),
241            "PROCEDURE STATUS": _show_parser("PROCEDURE STATUS"),
242            "PRIVILEGES": _show_parser("PRIVILEGES"),
243            "FULL PROCESSLIST": _show_parser("PROCESSLIST", full=True),
244            "PROCESSLIST": _show_parser("PROCESSLIST"),
245            "PROFILE": _show_parser("PROFILE"),
246            "PROFILES": _show_parser("PROFILES"),
247            "RELAYLOG EVENTS": _show_parser("RELAYLOG EVENTS"),
248            "REPLICAS": _show_parser("REPLICAS"),
249            "SLAVE HOSTS": _show_parser("REPLICAS"),
250            "REPLICA STATUS": _show_parser("REPLICA STATUS"),
251            "SLAVE STATUS": _show_parser("REPLICA STATUS"),
252            "GLOBAL STATUS": _show_parser("STATUS", global_=True),
253            "SESSION STATUS": _show_parser("STATUS"),
254            "STATUS": _show_parser("STATUS"),
255            "TABLE STATUS": _show_parser("TABLE STATUS"),
256            "FULL TABLES": _show_parser("TABLES", full=True),
257            "TABLES": _show_parser("TABLES"),
258            "TRIGGERS": _show_parser("TRIGGERS"),
259            "GLOBAL VARIABLES": _show_parser("VARIABLES", global_=True),
260            "SESSION VARIABLES": _show_parser("VARIABLES"),
261            "VARIABLES": _show_parser("VARIABLES"),
262            "WARNINGS": _show_parser("WARNINGS"),
263        }
264
265        SET_PARSERS = {
266            **parser.Parser.SET_PARSERS,
267            "PERSIST": lambda self: self._parse_set_item_assignment("PERSIST"),
268            "PERSIST_ONLY": lambda self: self._parse_set_item_assignment("PERSIST_ONLY"),
269            "CHARACTER SET": lambda self: self._parse_set_item_charset("CHARACTER SET"),
270            "CHARSET": lambda self: self._parse_set_item_charset("CHARACTER SET"),
271            "NAMES": lambda self: self._parse_set_item_names(),
272        }
273
274        PROFILE_TYPES = {
275            "ALL",
276            "BLOCK IO",
277            "CONTEXT SWITCHES",
278            "CPU",
279            "IPC",
280            "MEMORY",
281            "PAGE FAULTS",
282            "SOURCE",
283            "SWAPS",
284        }
285
286        LOG_DEFAULTS_TO_LN = True
287
288        def _parse_show_mysql(
289            self,
290            this: str,
291            target: bool | str = False,
292            full: t.Optional[bool] = None,
293            global_: t.Optional[bool] = None,
294        ) -> exp.Show:
295            if target:
296                if isinstance(target, str):
297                    self._match_text_seq(target)
298                target_id = self._parse_id_var()
299            else:
300                target_id = None
301
302            log = self._parse_string() if self._match_text_seq("IN") else None
303
304            if this in {"BINLOG EVENTS", "RELAYLOG EVENTS"}:
305                position = self._parse_number() if self._match_text_seq("FROM") else None
306                db = None
307            else:
308                position = None
309                db = None
310
311                if self._match(TokenType.FROM):
312                    db = self._parse_id_var()
313                elif self._match(TokenType.DOT):
314                    db = target_id
315                    target_id = self._parse_id_var()
316
317            channel = self._parse_id_var() if self._match_text_seq("FOR", "CHANNEL") else None
318
319            like = self._parse_string() if self._match_text_seq("LIKE") else None
320            where = self._parse_where()
321
322            if this == "PROFILE":
323                types = self._parse_csv(lambda: self._parse_var_from_options(self.PROFILE_TYPES))
324                query = self._parse_number() if self._match_text_seq("FOR", "QUERY") else None
325                offset = self._parse_number() if self._match_text_seq("OFFSET") else None
326                limit = self._parse_number() if self._match_text_seq("LIMIT") else None
327            else:
328                types, query = None, None
329                offset, limit = self._parse_oldstyle_limit()
330
331            mutex = True if self._match_text_seq("MUTEX") else None
332            mutex = False if self._match_text_seq("STATUS") else mutex
333
334            return self.expression(
335                exp.Show,
336                this=this,
337                target=target_id,
338                full=full,
339                log=log,
340                position=position,
341                db=db,
342                channel=channel,
343                like=like,
344                where=where,
345                types=types,
346                query=query,
347                offset=offset,
348                limit=limit,
349                mutex=mutex,
350                **{"global": global_},  # type: ignore
351            )
352
353        def _parse_oldstyle_limit(
354            self,
355        ) -> t.Tuple[t.Optional[exp.Expression], t.Optional[exp.Expression]]:
356            limit = None
357            offset = None
358            if self._match_text_seq("LIMIT"):
359                parts = self._parse_csv(self._parse_number)
360                if len(parts) == 1:
361                    limit = parts[0]
362                elif len(parts) == 2:
363                    limit = parts[1]
364                    offset = parts[0]
365
366            return offset, limit
367
368        def _parse_set_item_charset(self, kind: str) -> exp.Expression:
369            this = self._parse_string() or self._parse_id_var()
370            return self.expression(exp.SetItem, this=this, kind=kind)
371
372        def _parse_set_item_names(self) -> exp.Expression:
373            charset = self._parse_string() or self._parse_id_var()
374            if self._match_text_seq("COLLATE"):
375                collate = self._parse_string() or self._parse_id_var()
376            else:
377                collate = None
378
379            return self.expression(
380                exp.SetItem,
381                this=charset,
382                collate=collate,
383                kind="NAMES",
384            )

Parser consumes a list of tokens produced by the sqlglot.tokens.Tokenizer and produces a parsed syntax tree.

Arguments:
  • error_level: the desired error level. Default: ErrorLevel.RAISE
  • error_message_context: determines the amount of context to capture from a query string when displaying the error message (in number of characters). Default: 50.
  • index_offset: Index offset for arrays eg ARRAY[0] vs ARRAY[1] as the head of a list. Default: 0
  • alias_post_tablesample: If the table alias comes after tablesample. Default: False
  • max_errors: Maximum number of error messages to include in a raised ParseError. This is only relevant if error_level is ErrorLevel.RAISE. Default: 3
  • null_ordering: Indicates the default null ordering method to use if not explicitly set. Options are "nulls_are_small", "nulls_are_large", "nulls_are_last". Default: "nulls_are_small"
class MySQL.Generator(sqlglot.generator.Generator):
386    class Generator(generator.Generator):
387        LOCKING_READS_SUPPORTED = True
388        NULL_ORDERING_SUPPORTED = False
389        JOIN_HINTS = False
390        TABLE_HINTS = False
391
392        TRANSFORMS = {
393            **generator.Generator.TRANSFORMS,
394            exp.CurrentDate: no_paren_current_date_sql,
395            exp.DateDiff: lambda self, e: self.func("DATEDIFF", e.this, e.expression),
396            exp.DateAdd: _date_add_sql("ADD"),
397            exp.DateStrToDate: datestrtodate_sql,
398            exp.DateSub: _date_add_sql("SUB"),
399            exp.DateTrunc: _date_trunc_sql,
400            exp.DayOfMonth: rename_func("DAYOFMONTH"),
401            exp.DayOfWeek: rename_func("DAYOFWEEK"),
402            exp.DayOfYear: rename_func("DAYOFYEAR"),
403            exp.GroupConcat: lambda self, e: f"""GROUP_CONCAT({self.sql(e, "this")} SEPARATOR {self.sql(e, "separator") or "','"})""",
404            exp.ILike: no_ilike_sql,
405            exp.JSONExtractScalar: arrow_json_extract_scalar_sql,
406            exp.Max: max_or_greatest,
407            exp.Min: min_or_least,
408            exp.NullSafeEQ: lambda self, e: self.binary(e, "<=>"),
409            exp.NullSafeNEQ: lambda self, e: self.not_sql(self.binary(e, "<=>")),
410            exp.Pivot: no_pivot_sql,
411            exp.Select: transforms.preprocess([transforms.eliminate_distinct_on]),
412            exp.StrPosition: strposition_to_locate_sql,
413            exp.StrToDate: _str_to_date_sql,
414            exp.StrToTime: _str_to_date_sql,
415            exp.TableSample: no_tablesample_sql,
416            exp.TimeStrToUnix: rename_func("UNIX_TIMESTAMP"),
417            exp.TimeToStr: lambda self, e: self.func("DATE_FORMAT", e.this, self.format_time(e)),
418            exp.Trim: _trim_sql,
419            exp.TryCast: no_trycast_sql,
420            exp.WeekOfYear: rename_func("WEEKOFYEAR"),
421        }
422
423        TYPE_MAPPING = generator.Generator.TYPE_MAPPING.copy()
424        TYPE_MAPPING.pop(exp.DataType.Type.MEDIUMTEXT)
425        TYPE_MAPPING.pop(exp.DataType.Type.LONGTEXT)
426        TYPE_MAPPING.pop(exp.DataType.Type.MEDIUMBLOB)
427        TYPE_MAPPING.pop(exp.DataType.Type.LONGBLOB)
428
429        PROPERTIES_LOCATION = {
430            **generator.Generator.PROPERTIES_LOCATION,
431            exp.TransientProperty: exp.Properties.Location.UNSUPPORTED,
432            exp.VolatileProperty: exp.Properties.Location.UNSUPPORTED,
433        }
434
435        LIMIT_FETCH = "LIMIT"
436
437        def show_sql(self, expression: exp.Show) -> str:
438            this = f" {expression.name}"
439            full = " FULL" if expression.args.get("full") else ""
440            global_ = " GLOBAL" if expression.args.get("global") else ""
441
442            target = self.sql(expression, "target")
443            target = f" {target}" if target else ""
444            if expression.name in {"COLUMNS", "INDEX"}:
445                target = f" FROM{target}"
446            elif expression.name == "GRANTS":
447                target = f" FOR{target}"
448
449            db = self._prefixed_sql("FROM", expression, "db")
450
451            like = self._prefixed_sql("LIKE", expression, "like")
452            where = self.sql(expression, "where")
453
454            types = self.expressions(expression, key="types")
455            types = f" {types}" if types else types
456            query = self._prefixed_sql("FOR QUERY", expression, "query")
457
458            if expression.name == "PROFILE":
459                offset = self._prefixed_sql("OFFSET", expression, "offset")
460                limit = self._prefixed_sql("LIMIT", expression, "limit")
461            else:
462                offset = ""
463                limit = self._oldstyle_limit_sql(expression)
464
465            log = self._prefixed_sql("IN", expression, "log")
466            position = self._prefixed_sql("FROM", expression, "position")
467
468            channel = self._prefixed_sql("FOR CHANNEL", expression, "channel")
469
470            if expression.name == "ENGINE":
471                mutex_or_status = " MUTEX" if expression.args.get("mutex") else " STATUS"
472            else:
473                mutex_or_status = ""
474
475            return f"SHOW{full}{global_}{this}{target}{types}{db}{query}{log}{position}{channel}{mutex_or_status}{like}{where}{offset}{limit}"
476
477        def _prefixed_sql(self, prefix: str, expression: exp.Expression, arg: str) -> str:
478            sql = self.sql(expression, arg)
479            if not sql:
480                return ""
481            return f" {prefix} {sql}"
482
483        def _oldstyle_limit_sql(self, expression: exp.Show) -> str:
484            limit = self.sql(expression, "limit")
485            offset = self.sql(expression, "offset")
486            if limit:
487                limit_offset = f"{offset}, {limit}" if offset else limit
488                return f" LIMIT {limit_offset}"
489            return ""

Generator interprets the given syntax tree and produces a SQL string as an output.

Arguments:
  • time_mapping (dict): the dictionary of custom time mappings in which the key represents a python time format and the output the target time format
  • time_trie (trie): a trie of the time_mapping keys
  • pretty (bool): if set to True the returned string will be formatted. Default: False.
  • quote_start (str): specifies which starting character to use to delimit quotes. Default: '.
  • quote_end (str): specifies which ending character to use to delimit quotes. Default: '.
  • identifier_start (str): specifies which starting character to use to delimit identifiers. Default: ".
  • identifier_end (str): specifies which ending character to use to delimit identifiers. Default: ".
  • bit_start (str): specifies which starting character to use to delimit bit literals. Default: None.
  • bit_end (str): specifies which ending character to use to delimit bit literals. Default: None.
  • hex_start (str): specifies which starting character to use to delimit hex literals. Default: None.
  • hex_end (str): specifies which ending character to use to delimit hex literals. Default: None.
  • byte_start (str): specifies which starting character to use to delimit byte literals. Default: None.
  • byte_end (str): specifies which ending character to use to delimit byte literals. Default: None.
  • raw_start (str): specifies which starting character to use to delimit raw literals. Default: None.
  • raw_end (str): specifies which ending character to use to delimit raw literals. Default: None.
  • identify (bool | str): 'always': always quote, 'safe': quote identifiers if they don't contain an upcase, True defaults to always.
  • normalize (bool): if set to True all identifiers will lower cased
  • string_escape (str): specifies a string escape character. Default: '.
  • identifier_escape (str): specifies an identifier escape character. Default: ".
  • pad (int): determines padding in a formatted string. Default: 2.
  • indent (int): determines the size of indentation in a formatted string. Default: 4.
  • unnest_column_only (bool): if true unnest table aliases are considered only as column aliases
  • normalize_functions (str): normalize function names, "upper", "lower", or None Default: "upper"
  • alias_post_tablesample (bool): if the table alias comes after tablesample Default: False
  • unsupported_level (ErrorLevel): determines the generator's behavior when it encounters unsupported expressions. Default ErrorLevel.WARN.
  • null_ordering (str): Indicates the default null ordering method to use if not explicitly set. Options are "nulls_are_small", "nulls_are_large", "nulls_are_last". Default: "nulls_are_small"
  • max_unsupported (int): Maximum number of unsupported messages to include in a raised UnsupportedError. This is only relevant if unsupported_level is ErrorLevel.RAISE. Default: 3
  • leading_comma (bool): if the the comma is leading or trailing in select statements Default: False
  • max_text_width: The max number of characters in a segment before creating new lines in pretty mode. The default is on the smaller end because the length only represents a segment and not the true line length. Default: 80
  • comments: Whether or not to preserve comments in the output SQL code. Default: True
def show_sql(self, expression: sqlglot.expressions.Show) -> str:
437        def show_sql(self, expression: exp.Show) -> str:
438            this = f" {expression.name}"
439            full = " FULL" if expression.args.get("full") else ""
440            global_ = " GLOBAL" if expression.args.get("global") else ""
441
442            target = self.sql(expression, "target")
443            target = f" {target}" if target else ""
444            if expression.name in {"COLUMNS", "INDEX"}:
445                target = f" FROM{target}"
446            elif expression.name == "GRANTS":
447                target = f" FOR{target}"
448
449            db = self._prefixed_sql("FROM", expression, "db")
450
451            like = self._prefixed_sql("LIKE", expression, "like")
452            where = self.sql(expression, "where")
453
454            types = self.expressions(expression, key="types")
455            types = f" {types}" if types else types
456            query = self._prefixed_sql("FOR QUERY", expression, "query")
457
458            if expression.name == "PROFILE":
459                offset = self._prefixed_sql("OFFSET", expression, "offset")
460                limit = self._prefixed_sql("LIMIT", expression, "limit")
461            else:
462                offset = ""
463                limit = self._oldstyle_limit_sql(expression)
464
465            log = self._prefixed_sql("IN", expression, "log")
466            position = self._prefixed_sql("FROM", expression, "position")
467
468            channel = self._prefixed_sql("FOR CHANNEL", expression, "channel")
469
470            if expression.name == "ENGINE":
471                mutex_or_status = " MUTEX" if expression.args.get("mutex") else " STATUS"
472            else:
473                mutex_or_status = ""
474
475            return f"SHOW{full}{global_}{this}{target}{types}{db}{query}{log}{position}{channel}{mutex_or_status}{like}{where}{offset}{limit}"
Inherited Members
sqlglot.generator.Generator
Generator
generate
unsupported
sep
seg
pad_comment
maybe_comment
wrap
no_identify
normalize_func
indent
sql
uncache_sql
cache_sql
characterset_sql
column_sql
columnposition_sql
columndef_sql
columnconstraint_sql
autoincrementcolumnconstraint_sql
compresscolumnconstraint_sql
generatedasidentitycolumnconstraint_sql
notnullcolumnconstraint_sql
primarykeycolumnconstraint_sql
uniquecolumnconstraint_sql
create_sql
clone_sql
describe_sql
prepend_ctes
with_sql
cte_sql
tablealias_sql
bitstring_sql
hexstring_sql
bytestring_sql
rawstring_sql
datatypesize_sql
datatype_sql
directory_sql
delete_sql
drop_sql
except_sql
except_op
fetch_sql
filter_sql
hint_sql
index_sql
identifier_sql
inputoutputformat_sql
national_sql
partition_sql
properties_sql
root_properties
properties
with_properties
locate_properties
property_sql
likeproperty_sql
fallbackproperty_sql
journalproperty_sql
freespaceproperty_sql
checksumproperty_sql
mergeblockratioproperty_sql
datablocksizeproperty_sql
blockcompressionproperty_sql
isolatedloadingproperty_sql
lockingproperty_sql
withdataproperty_sql
insert_sql
intersect_sql
intersect_op
introducer_sql
pseudotype_sql
onconflict_sql
returning_sql
rowformatdelimitedproperty_sql
table_sql
tablesample_sql
pivot_sql
tuple_sql
update_sql
values_sql
var_sql
into_sql
from_sql
group_sql
having_sql
join_sql
lambda_sql
lateral_sql
limit_sql
offset_sql
setitem_sql
set_sql
pragma_sql
lock_sql
literal_sql
loaddata_sql
null_sql
boolean_sql
order_sql
cluster_sql
distribute_sql
sort_sql
ordered_sql
matchrecognize_sql
query_modifiers
after_having_modifiers
after_limit_modifiers
select_sql
schema_sql
star_sql
parameter_sql
sessionparameter_sql
placeholder_sql
subquery_sql
qualify_sql
union_sql
union_op
unnest_sql
where_sql
window_sql
partition_by_sql
windowspec_sql
withingroup_sql
between_sql
bracket_sql
all_sql
any_sql
exists_sql
case_sql
constraint_sql
nextvaluefor_sql
extract_sql
trim_sql
concat_sql
check_sql
foreignkey_sql
primarykey_sql
unique_sql
if_sql
matchagainst_sql
jsonkeyvalue_sql
jsonobject_sql
openjsoncolumndef_sql
openjson_sql
in_sql
in_unnest_op
interval_sql
return_sql
reference_sql
anonymous_sql
paren_sql
neg_sql
not_sql
alias_sql
aliases_sql
attimezone_sql
add_sql
and_sql
connector_sql
bitwiseand_sql
bitwiseleftshift_sql
bitwisenot_sql
bitwiseor_sql
bitwiserightshift_sql
bitwisexor_sql
cast_sql
currentdate_sql
collate_sql
command_sql
comment_sql
mergetreettlaction_sql
mergetreettl_sql
transaction_sql
commit_sql
rollback_sql
altercolumn_sql
renametable_sql
altertable_sql
droppartition_sql
addconstraint_sql
distinct_sql
ignorenulls_sql
respectnulls_sql
intdiv_sql
dpipe_sql
div_sql
overlaps_sql
distance_sql
dot_sql
eq_sql
escape_sql
glob_sql
gt_sql
gte_sql
ilike_sql
ilikeany_sql
is_sql
like_sql
likeany_sql
similarto_sql
lt_sql
lte_sql
mod_sql
mul_sql
neq_sql
nullsafeeq_sql
nullsafeneq_sql
or_sql
slice_sql
sub_sql
trycast_sql
use_sql
binary
function_fallback_sql
func
format_args
text_width
format_time
expressions
op_expressions
naked_property
set_operation
tag_sql
token_sql
userdefinedfunction_sql
joinhint_sql
kwarg_sql
when_sql
merge_sql
tochar_sql