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