1 /* PSPP - a program for statistical analysis.
2 Copyright (C) 1997-9, 2000, 2005, 2009, 2010, 2011 Free Software Foundation, Inc.
4 This program is free software: you can redistribute it and/or modify
5 it under the terms of the GNU General Public License as published by
6 the Free Software Foundation, either version 3 of the License, or
7 (at your option) any later version.
9 This program is distributed in the hope that it will be useful,
10 but WITHOUT ANY WARRANTY; without even the implied warranty of
11 MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
12 GNU General Public License for more details.
14 You should have received a copy of the GNU General Public License
15 along with this program. If not, see <http://www.gnu.org/licenses/>. */
18 This file is concerned with the definition of the PSPP syntax, NOT the
19 action of scanning/parsing code .
24 #include "data/identifier.h"
29 #include "libpspp/assertion.h"
31 #include "gl/c-ctype.h"
34 #define _(msgid) gettext (msgid)
38 /* Returns TYPE as a string, e.g. "ID" for T_ID. */
40 token_type_to_name (enum token_type type)
44 #define TOKEN_TYPE(TYPE) case T_##TYPE: return #TYPE;
49 return "unknown token type";
53 /* Returns an ASCII string that yields TOKEN if it appeared in a syntax file,
54 as a statically allocated constant string. This function returns NULL for
55 tokens that don't have any fixed string representation, such as identifier
58 token_type_to_string (enum token_type token)
151 /* Recognizing identifiers. */
154 is_ascii_id1 (unsigned char c)
156 return c_isalpha (c) || c == '@' || c == '#' || c == '$';
160 is_ascii_idn (unsigned char c)
162 return is_ascii_id1 (c) || isdigit (c) || c == '.' || c == '_';
165 /* Returns true if C may be the first byte in an identifier in the current
168 (PSPP is transitioning to using Unicode internally for syntax, so please
169 use lex_uc_is_id1() instead, if possible.) */
173 return is_ascii_id1 (c) || (unsigned char) c >= 128;
176 /* Returns true if C may be a byte in an identifier other than the first.
178 (PSPP is transitioning to using Unicode internally for syntax, so please
179 use lex_uc_is_idn() instead, if possible.) */
183 return is_ascii_idn (c) || (unsigned char) c >= 128;
186 /* Returns true if Unicode code point UC may be the first character in an
187 identifier in the current locale. */
189 lex_uc_is_id1 (ucs4_t uc)
191 return is_ascii_id1 (uc) || (uc >= 0x80 && uc_is_property_id_start (uc));
194 /* Returns true if Unicode code point UC may be a character in an identifier
195 other than the first. */
197 lex_uc_is_idn (ucs4_t uc)
199 return (is_ascii_id1 (uc) || isdigit (uc) || uc == '.' || uc == '_'
200 || (uc >= 0x80 && uc_is_property_id_continue (uc)));
203 /* Returns true if Unicode code point UC is a space that separates tokens. */
205 lex_uc_is_space (ucs4_t uc)
207 /* These are all of the Unicode characters in category Zs, Zl, or Zp. */
208 return (uc == ' ' || (uc <= 0x000d && uc >= 0x0009)
210 && (uc == 0xa0 || uc == 0x85 || uc == 0x1680 || uc == 0x180e
211 || (uc >= 0x2000 && uc <= 0x200a)
212 || uc == 0x2028 || uc == 0x2029 || uc == 0x202f
213 || uc == 0x205f || uc == 0x3000)));
217 /* Returns the length of the longest prefix of STRING that forms
218 a valid identifier. Returns zero if STRING does not begin
219 with a valid identifier. */
221 lex_id_get_length (struct substring string)
224 if (!ss_is_empty (string) && lex_is_id1 (ss_first (string)))
227 while (length < ss_length (string)
228 && lex_is_idn (ss_at (string, length)))
234 /* Comparing identifiers. */
236 /* Returns true if TOKEN is a case-insensitive match for KEYWORD.
238 Keywords match if one of the following is true: KEYWORD and
239 TOKEN are identical, or TOKEN is at least 3 characters long
240 and those characters are identical to KEYWORD. (Letters that
241 differ only in case are considered identical.)
243 KEYWORD must be ASCII, but TOKEN may be ASCII or UTF-8. */
245 lex_id_match (struct substring keyword, struct substring token)
247 return lex_id_match_n (keyword, token, 3);
250 /* Returns true if TOKEN is a case-insensitive match for at least
251 the first N characters of KEYWORD.
253 KEYWORD must be ASCII, but TOKEN may be ASCII or UTF-8. */
255 lex_id_match_n (struct substring keyword, struct substring token, size_t n)
257 size_t token_len = ss_length (token);
258 size_t keyword_len = ss_length (keyword);
260 if (token_len >= n && token_len < keyword_len)
261 return ss_equals_case (ss_head (keyword, token_len), token);
263 return ss_equals_case (keyword, token);
266 /* Table of keywords. */
270 const struct substring identifier;
273 static const struct keyword keywords[] =
275 { T_AND, SS_LITERAL_INITIALIZER ("AND") },
276 { T_OR, SS_LITERAL_INITIALIZER ("OR") },
277 { T_NOT, SS_LITERAL_INITIALIZER ("NOT") },
278 { T_EQ, SS_LITERAL_INITIALIZER ("EQ") },
279 { T_GE, SS_LITERAL_INITIALIZER ("GE") },
280 { T_GT, SS_LITERAL_INITIALIZER ("GT") },
281 { T_LE, SS_LITERAL_INITIALIZER ("LE") },
282 { T_LT, SS_LITERAL_INITIALIZER ("LT") },
283 { T_NE, SS_LITERAL_INITIALIZER ("NE") },
284 { T_ALL, SS_LITERAL_INITIALIZER ("ALL") },
285 { T_BY, SS_LITERAL_INITIALIZER ("BY") },
286 { T_TO, SS_LITERAL_INITIALIZER ("TO") },
287 { T_WITH, SS_LITERAL_INITIALIZER ("WITH") },
289 static const size_t keyword_cnt = sizeof keywords / sizeof *keywords;
291 /* Returns true if TOKEN is representable as a keyword. */
293 lex_is_keyword (enum token_type token)
295 const struct keyword *kw;
296 for (kw = keywords; kw < &keywords[keyword_cnt]; kw++)
297 if (kw->token == token)
302 /* Returns the proper token type, either T_ID or a reserved
303 keyword enum, for ID. */
305 lex_id_to_token (struct substring id)
307 if (ss_length (id) >= 2 && ss_length (id) <= 4)
309 const struct keyword *kw;
310 for (kw = keywords; kw < &keywords[keyword_cnt]; kw++)
311 if (ss_equals_case (kw->identifier, id))