src/backend/parser/keywords.c

   1 /*-------------------------------------------------------------------------
   2  *
   3  * keywords.c
   4  *        lexical token lookup for reserved words in PostgreSQL
   5  *
   6  * Portions Copyright (c) 1996-2003, PostgreSQL Global Development Group
   7  * Portions Copyright (c) 1994, Regents of the University of California
   8  *
   9  *
  10  * IDENTIFICATION
  11  *        $Header: /cvsroot/pgsql/src/backend/parser/keywords.c,v 1.141 2003/08/04 02:40:01 momjian Exp $
  12  *
  13  *-------------------------------------------------------------------------
  14  */
  15 #include "postgres.h"
  16
  17 #include <ctype.h>
  18
  19 #include "nodes/parsenodes.h"
  20 #include "parser/keywords.h"
  21 #include "parser/parse.h"
  22
  23 /* NB: This file is also used by pg_dump. */
  24
  25 /*
  26  * List of (keyword-name, keyword-token-value) pairs.
  27  *
  28  * !!WARNING!!: This list must be sorted, because binary
  29  *               search is used to locate entries.
  30  */
  31 static const ScanKeyword ScanKeywords[] = {
  32         /* name, value */
  33         {"abort", ABORT_P},
  34         {"absolute", ABSOLUTE_P},
  35         {"access", ACCESS},
  36         {"action", ACTION},
  37         {"add", ADD},
  38         {"after", AFTER},
  39         {"aggregate", AGGREGATE},
  40         {"all", ALL},
  41         {"alter", ALTER},
  42         {"analyse", ANALYSE},           /* British spelling */
  43         {"analyze", ANALYZE},
  44         {"and", AND},
  45         {"any", ANY},
  46         {"array", ARRAY},
  47         {"as", AS},
  48         {"asc", ASC},
  49         {"assertion", ASSERTION},
  50         {"assignment", ASSIGNMENT},
  51         {"at", AT},
  52         {"authorization", AUTHORIZATION},
  53         {"backward", BACKWARD},
  54         {"before", BEFORE},
  55         {"begin", BEGIN_P},
  56         {"between", BETWEEN},
  57         {"bigint", BIGINT},
  58         {"binary", BINARY},
  59         {"bit", BIT},
  60         {"boolean", BOOLEAN_P},
  61         {"both", BOTH},
  62         {"by", BY},
  63         {"cache", CACHE},
  64         {"called", CALLED},
  65         {"cascade", CASCADE},
  66         {"case", CASE},
  67         {"cast", CAST},
  68         {"chain", CHAIN},
  69         {"char", CHAR_P},
  70         {"character", CHARACTER},
  71         {"characteristics", CHARACTERISTICS},
  72         {"check", CHECK},
  73         {"checkpoint", CHECKPOINT},
  74         {"class", CLASS},
  75         {"close", CLOSE},
  76         {"cluster", CLUSTER},
  77         {"coalesce", COALESCE},
  78         {"collate", COLLATE},
  79         {"column", COLUMN},
  80         {"comment", COMMENT},
  81         {"commit", COMMIT},
  82         {"committed", COMMITTED},
  83         {"constraint", CONSTRAINT},
  84         {"constraints", CONSTRAINTS},
  85         {"conversion", CONVERSION_P},
  86         {"convert", CONVERT},
  87         {"copy", COPY},
  88         {"create", CREATE},
  89         {"createdb", CREATEDB},
  90         {"createuser", CREATEUSER},
  91         {"cross", CROSS},
  92         {"current_date", CURRENT_DATE},
  93         {"current_time", CURRENT_TIME},
  94         {"current_timestamp", CURRENT_TIMESTAMP},
  95         {"current_user", CURRENT_USER},
  96         {"cursor", CURSOR},
  97         {"cycle", CYCLE},
  98         {"database", DATABASE},
  99         {"day", DAY_P},
 100         {"deallocate", DEALLOCATE},
 101         {"dec", DEC},
 102         {"decimal", DECIMAL_P},
 103         {"declare", DECLARE},
 104         {"default", DEFAULT},
 105         {"defaults", DEFAULTS},
 106         {"deferrable", DEFERRABLE},
 107         {"deferred", DEFERRED},
 108         {"definer", DEFINER},
 109         {"delete", DELETE_P},
 110         {"delimiter", DELIMITER},
 111         {"delimiters", DELIMITERS},
 112         {"desc", DESC},
 113         {"distinct", DISTINCT},
 114         {"do", DO},
 115         {"domain", DOMAIN_P},
 116         {"double", DOUBLE_P},
 117         {"drop", DROP},
 118         {"each", EACH},
 119         {"else", ELSE},
 120         {"encoding", ENCODING},
 121         {"encrypted", ENCRYPTED},
 122         {"end", END_P},
 123         {"escape", ESCAPE},
 124         {"except", EXCEPT},
 125         {"excluding", EXCLUDING},
 126         {"exclusive", EXCLUSIVE},
 127         {"execute", EXECUTE},
 128         {"exists", EXISTS},
 129         {"explain", EXPLAIN},
 130         {"external", EXTERNAL},
 131         {"extract", EXTRACT},
 132         {"false", FALSE_P},
 133         {"fetch", FETCH},
 134         {"first", FIRST_P},
 135         {"float", FLOAT_P},
 136         {"for", FOR},
 137         {"force", FORCE},
 138         {"foreign", FOREIGN},
 139         {"forward", FORWARD},
 140         {"freeze", FREEZE},
 141         {"from", FROM},
 142         {"full", FULL},
 143         {"function", FUNCTION},
 144         {"global", GLOBAL},
 145         {"grant", GRANT},
 146         {"group", GROUP_P},
 147         {"handler", HANDLER},
 148         {"having", HAVING},
 149         {"hold", HOLD},
 150         {"hour", HOUR_P},
 151         {"ilike", ILIKE},
 152         {"immediate", IMMEDIATE},
 153         {"immutable", IMMUTABLE},
 154         {"implicit", IMPLICIT_P},
 155         {"in", IN_P},
 156         {"including", INCLUDING},
 157         {"increment", INCREMENT},
 158         {"index", INDEX},
 159         {"inherits", INHERITS},
 160         {"initially", INITIALLY},
 161         {"inner", INNER_P},
 162         {"inout", INOUT},
 163         {"input", INPUT_P},
 164         {"insensitive", INSENSITIVE},
 165         {"insert", INSERT},
 166         {"instead", INSTEAD},
 167         {"int", INT_P},
 168         {"integer", INTEGER},
 169         {"intersect", INTERSECT},
 170         {"interval", INTERVAL},
 171         {"into", INTO},
 172         {"invoker", INVOKER},
 173         {"is", IS},
 174         {"isnull", ISNULL},
 175         {"isolation", ISOLATION},
 176         {"join", JOIN},
 177         {"key", KEY},
 178         {"lancompiler", LANCOMPILER},
 179         {"language", LANGUAGE},
 180         {"last", LAST_P},
 181         {"leading", LEADING},
 182         {"left", LEFT},
 183         {"level", LEVEL},
 184         {"like", LIKE},
 185         {"limit", LIMIT},
 186         {"listen", LISTEN},
 187         {"load", LOAD},
 188         {"local", LOCAL},
 189         {"localtime", LOCALTIME},
 190         {"localtimestamp", LOCALTIMESTAMP},
 191         {"location", LOCATION},
 192         {"lock", LOCK_P},
 193         {"match", MATCH},
 194         {"maxvalue", MAXVALUE},
 195         {"minute", MINUTE_P},
 196         {"minvalue", MINVALUE},
 197         {"mode", MODE},
 198         {"month", MONTH_P},
 199         {"move", MOVE},
 200         {"names", NAMES},
 201         {"national", NATIONAL},
 202         {"natural", NATURAL},
 203         {"nchar", NCHAR},
 204         {"new", NEW},
 205         {"next", NEXT},
 206         {"no", NO},
 207         {"nocreatedb", NOCREATEDB},
 208         {"nocreateuser", NOCREATEUSER},
 209         {"none", NONE},
 210         {"not", NOT},
 211         {"nothing", NOTHING},
 212         {"notify", NOTIFY},
 213         {"notnull", NOTNULL},
 214         {"null", NULL_P},
 215         {"nullif", NULLIF},
 216         {"numeric", NUMERIC},
 217         {"of", OF},
 218         {"off", OFF},
 219         {"offset", OFFSET},
 220         {"oids", OIDS},
 221         {"old", OLD},
 222         {"on", ON},
 223         {"only", ONLY},
 224         {"operator", OPERATOR},
 225         {"option", OPTION},
 226         {"or", OR},
 227         {"order", ORDER},
 228         {"out", OUT_P},
 229         {"outer", OUTER_P},
 230         {"overlaps", OVERLAPS},
 231         {"overlay", OVERLAY},
 232         {"owner", OWNER},
 233         {"partial", PARTIAL},
 234         {"password", PASSWORD},
 235         {"path", PATH_P},
 236         {"pendant", PENDANT},
 237         {"placing", PLACING},
 238         {"position", POSITION},
 239         {"precision", PRECISION},
 240         {"prepare", PREPARE},
 241         {"preserve", PRESERVE},
 242         {"primary", PRIMARY},
 243         {"prior", PRIOR},
 244         {"privileges", PRIVILEGES},
 245         {"procedural", PROCEDURAL},
 246         {"procedure", PROCEDURE},
 247         {"read", READ},
 248         {"real", REAL},
 249         {"recheck", RECHECK},
 250         {"references", REFERENCES},
 251         {"reindex", REINDEX},
 252         {"relative", RELATIVE_P},
 253         {"rename", RENAME},
 254         {"replace", REPLACE},
 255         {"reset", RESET},
 256         {"restart", RESTART},
 257         {"restrict", RESTRICT},
 258         {"returns", RETURNS},
 259         {"revoke", REVOKE},
 260         {"right", RIGHT},
 261         {"rollback", ROLLBACK},
 262         {"row", ROW},
 263         {"rows", ROWS},
 264         {"rule", RULE},
 265         {"schema", SCHEMA},
 266         {"scroll", SCROLL},
 267         {"second", SECOND_P},
 268         {"security", SECURITY},
 269         {"select", SELECT},
 270         {"sequence", SEQUENCE},
 271         {"serializable", SERIALIZABLE},
 272         {"session", SESSION},
 273         {"session_user", SESSION_USER},
 274         {"set", SET},
 275         {"setof", SETOF},
 276         {"share", SHARE},
 277         {"show", SHOW},
 278         {"similar", SIMILAR},
 279         {"simple", SIMPLE},
 280         {"smallint", SMALLINT},
 281         {"some", SOME},
 282         {"stable", STABLE},
 283         {"start", START},
 284         {"statement", STATEMENT},
 285         {"statistics", STATISTICS},
 286         {"stdin", STDIN},
 287         {"stdout", STDOUT},
 288         {"storage", STORAGE},
 289         {"strict", STRICT_P},
 290         {"substring", SUBSTRING},
 291         {"sysid", SYSID},
 292         {"table", TABLE},
 293         {"temp", TEMP},
 294         {"template", TEMPLATE},
 295         {"temporary", TEMPORARY},
 296         {"then", THEN},
 297         {"time", TIME},
 298         {"timestamp", TIMESTAMP},
 299         {"to", TO},
 300         {"toast", TOAST},
 301         {"trailing", TRAILING},
 302         {"transaction", TRANSACTION},
 303         {"treat", TREAT},
 304         {"trigger", TRIGGER},
 305         {"trim", TRIM},
 306         {"true", TRUE_P},
 307         {"truncate", TRUNCATE},
 308         {"trusted", TRUSTED},
 309         {"type", TYPE_P},
 310         {"unencrypted", UNENCRYPTED},
 311         {"union", UNION},
 312         {"unique", UNIQUE},
 313         {"unknown", UNKNOWN},
 314         {"unlisten", UNLISTEN},
 315         {"until", UNTIL},
 316         {"update", UPDATE},
 317         {"usage", USAGE},
 318         {"user", USER},
 319         {"using", USING},
 320         {"vacuum", VACUUM},
 321         {"valid", VALID},
 322         {"validator", VALIDATOR},
 323         {"values", VALUES},
 324         {"varchar", VARCHAR},
 325         {"varying", VARYING},
 326         {"verbose", VERBOSE},
 327         {"version", VERSION},
 328         {"view", VIEW},
 329         {"volatile", VOLATILE},
 330         {"when", WHEN},
 331         {"where", WHERE},
 332         {"with", WITH},
 333         {"without", WITHOUT},
 334         {"work", WORK},
 335         {"write", WRITE},
 336         {"year", YEAR_P},
 337         {"zone", ZONE},
 338 };
 339
 340 /*
 341  * ScanKeywordLookup - see if a given word is a keyword
 342  *
 343  * Returns a pointer to the ScanKeyword table entry, or NULL if no match.
 344  *
 345  * The match is done case-insensitively.  Note that we deliberately use a
 346  * dumbed-down case conversion that will only translate 'A'-'Z' into 'a'-'z',
 347  * even if we are in a locale where tolower() would produce more or different
 348  * translations.  This is to conform to the SQL99 spec, which says that
 349  * keywords are to be matched in this way even though non-keyword identifiers
 350  * receive a different case-normalization mapping.
 351  */
 352 const ScanKeyword *
 353 ScanKeywordLookup(const char *text)
 354 {
 355         int                     len,
 356                                 i;
 357         char            word[NAMEDATALEN];
 358         const ScanKeyword *low;
 359         const ScanKeyword *high;
 360
 361         len = strlen(text);
 362         /* We assume all keywords are shorter than NAMEDATALEN. */
 363         if (len >= NAMEDATALEN)
 364                 return NULL;
 365
 366         /*
 367          * Apply an ASCII-only downcasing.      We must not use tolower() since it
 368          * may produce the wrong translation in some locales (eg, Turkish),
 369          * and we don't trust isupper() very much either.  In an ASCII-based
 370          * encoding the tests against A and Z are sufficient, but we also
 371          * check isupper() so that we will work correctly under EBCDIC.  The
 372          * actual case conversion step should work for either ASCII or EBCDIC.
 373          */
 374         for (i = 0; i < len; i++)
 375         {
 376                 char            ch = text[i];
 377
 378                 if (ch >= 'A' && ch <= 'Z' && isupper((unsigned char) ch))
 379                         ch += 'a' - 'A';
 380                 word[i] = ch;
 381         }
 382         word[len] = '\0';
 383
 384         /*
 385          * Now do a binary search using plain strcmp() comparison.
 386          */
 387         low = &ScanKeywords[0];
 388         high = endof(ScanKeywords) - 1;
 389         while (low <= high)
 390         {
 391                 const ScanKeyword *middle;
 392                 int                     difference;
 393
 394                 middle = low + (high - low) / 2;
 395                 difference = strcmp(middle->name, word);
 396                 if (difference == 0)
 397                         return middle;
 398                 else if (difference < 0)
 399                         low = middle + 1;
 400                 else
 401                         high = middle - 1;
 402         }
 403
 404         return NULL;
 405 }