1 | /**************************************************************************
|
---|
2 | *
|
---|
3 | * words.h -- Macros for parsing out words from the source text
|
---|
4 | * Copyright (C) 1994 Neil Sharman
|
---|
5 | *
|
---|
6 | * This program is free software; you can redistribute it and/or modify
|
---|
7 | * it under the terms of the GNU General Public License as published by
|
---|
8 | * the Free Software Foundation; either version 2 of the License, or
|
---|
9 | * (at your option) any later version.
|
---|
10 | *
|
---|
11 | * This program is distributed in the hope that it will be useful,
|
---|
12 | * but WITHOUT ANY WARRANTY; without even the implied warranty of
|
---|
13 | * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
|
---|
14 | * GNU General Public License for more details.
|
---|
15 | *
|
---|
16 | * You should have received a copy of the GNU General Public License
|
---|
17 | * along with this program; if not, write to the Free Software
|
---|
18 | * Foundation, Inc., 675 Mass Ave, Cambridge, MA 02139, USA.
|
---|
19 | *
|
---|
20 | * $Id: words.h 29327 2014-09-22 06:13:42Z ak19 $
|
---|
21 | *
|
---|
22 | **************************************************************************/
|
---|
23 |
|
---|
24 |
|
---|
25 | #include "sysfuncs.h"
|
---|
26 | #include "unitool.h"
|
---|
27 |
|
---|
28 | #include "environment.h"
|
---|
29 |
|
---|
30 | /*
|
---|
31 | * This has been cleaned up by Tim Shimmin.
|
---|
32 | */
|
---|
33 |
|
---|
34 | /*
|
---|
35 | * ---NOTE---
|
---|
36 | *
|
---|
37 | * "WORD" refers to a word in the compressed text.
|
---|
38 | * "STEM" or "STEM_WORD" refers to a word for indexing on
|
---|
39 | *
|
---|
40 | */
|
---|
41 |
|
---|
42 | #define MAXWORDLEN 15
|
---|
43 | /* Maximum length in bytes of any word or non-word. Note that
|
---|
44 | variations to MAXWORDLEN may have dramatic effects on the rest
|
---|
45 | of the program, as the length and the prefix match are packed
|
---|
46 | together into a four bit nibble, and there is not check that
|
---|
47 | this is possible, i.e., leave MAXWORDLEN alone... */
|
---|
48 |
|
---|
49 | #define MAXSTEMLEN 255
|
---|
50 | /* Maximum length in bytes of any stem. Note that
|
---|
51 | variations to MAXSTEMLEN may have dramatic effects on the rest
|
---|
52 | of the program, , i.e., leave MAXSTEMLEN alone... */
|
---|
53 |
|
---|
54 | /*#define MAXNUMERIC 4*/
|
---|
55 |
|
---|
56 | /* Maximum number of numeric characters permitted in a word.
|
---|
57 | This avoids long sequences of numbers creating just one
|
---|
58 | word occurrence for each number. At most 10,000 all numeric
|
---|
59 | words will be permitted. */
|
---|
60 |
|
---|
61 | /* [RPAP - Jan 97: Stem Index Change] */
|
---|
62 | #define MAXPARAMLEN 20
|
---|
63 | /* Maximum number of bytes to read for a parameter value for a
|
---|
64 | term in a query. */
|
---|
65 | #define WEIGHTPARAM '/'
|
---|
66 | #define STEMPARAM '#'
|
---|
67 |
|
---|
68 | /* [RJM 07/97: Ranked Required Terms] */
|
---|
69 | #define MUSTMATCHPARAM '+'
|
---|
70 |
|
---|
71 |
|
---|
72 | #define PESINAWORD(c) (isalnum(c) || ((c) >= 0x80 && (c) <= 0xff))
|
---|
73 | /* The definition of what characters are permitted in a word.
|
---|
74 | This macro is pessimistic, you cannot tell from a particular
|
---|
75 | byte above 0x80 whether it is a character or not. This function
|
---|
76 | is needed by various functions relating to huffman coding
|
---|
77 | where frequency counts need to be primed, it should not be
|
---|
78 | used in parsing the UTF-8 encoded input. */
|
---|
79 |
|
---|
80 | int inaword (const u_char *here, const u_char *end);
|
---|
81 | /* Takes the place of the old INAWORD macro. It determines
|
---|
82 | whether a given place in a UTF-8 encoded Unicode string
|
---|
83 | is part of a word. */
|
---|
84 |
|
---|
85 | int isaspace (const u_char *here, const u_char *end);
|
---|
86 | /* It determines whether a given place in a UTF-8 encoded
|
---|
87 | Unicode string is a unicode space. */
|
---|
88 |
|
---|
89 | u_char *skipspace(u_char *here, u_char *end);
|
---|
90 | /* Return a the UTF-8 encoded Unicode string with beginning
|
---|
91 | unicode spaces skipped. */
|
---|
92 |
|
---|
93 |
|
---|
94 | /* =========================================================================
|
---|
95 | * Macro: PARSE_WORD
|
---|
96 | * Description:
|
---|
97 | * Extract a word out for compressing text
|
---|
98 | * Input:
|
---|
99 | * s_in = string start in buffer
|
---|
100 | * end = string end in buffer
|
---|
101 | * Output:
|
---|
102 | * Word = extracted word with length in 1st byte
|
---|
103 | * s_in = ptr to next character in buffer yet to be processed
|
---|
104 | * ========================================================================= */
|
---|
105 | #define PARSE_WORD(Word, s_in, end) \
|
---|
106 | do { \
|
---|
107 | register int charlength = 0; \
|
---|
108 | register u_char *wptr = (Word)+1; \
|
---|
109 | register int length = 0; \
|
---|
110 | register int numeric = 0; \
|
---|
111 | unsigned short c; \
|
---|
112 | register int maxnumeric = IntEnv (GetEnv ("maxnumeric"), 4); \
|
---|
113 | \
|
---|
114 | charlength = parse_utf8_char((s_in),(end),&c); \
|
---|
115 | \
|
---|
116 | while (length+charlength <= MAXWORDLEN && charlength > 0 && \
|
---|
117 | (is_unicode_letter(c) || (is_unicode_digit(c) && \
|
---|
118 | ++numeric <= maxnumeric))) { \
|
---|
119 | while (charlength-- > 0) { \
|
---|
120 | *wptr++ = *(s_in)++; ++length; \
|
---|
121 | } \
|
---|
122 | charlength = parse_utf8_char((s_in),(end),&c); \
|
---|
123 | } \
|
---|
124 | *(Word) = length; \
|
---|
125 | }while(0)
|
---|
126 |
|
---|
127 | /*
|
---|
128 | #define PARSE_WORD(Word, s_in, end) \
|
---|
129 | do { \
|
---|
130 | register u_char *wptr = (Word)+1; \
|
---|
131 | register int length = 0; \
|
---|
132 | register int c = *(s_in); \
|
---|
133 | register int numeric = 0; \
|
---|
134 | \
|
---|
135 | while( length < MAXWORDLEN && INAWORD(c) && (s_in)<=(end)) \
|
---|
136 | { \
|
---|
137 | if ((numeric += INNUMBER(c)) > MAXNUMERIC) \
|
---|
138 | break; \
|
---|
139 | *wptr++ = c; \
|
---|
140 | ++length; \
|
---|
141 | c = *++(s_in); \
|
---|
142 | } \
|
---|
143 | *(Word) = length; \
|
---|
144 | }while(0)
|
---|
145 | */
|
---|
146 |
|
---|
147 |
|
---|
148 | /* =========================================================================
|
---|
149 | * Macro: PARSE_NON_WORD
|
---|
150 | * Description:
|
---|
151 | * Extract a non-word out for storing compressed text
|
---|
152 | * Input: as above
|
---|
153 | * Output: as above
|
---|
154 | * ========================================================================= */
|
---|
155 | #define PARSE_NON_WORD(Word, s_in, end) \
|
---|
156 | do { \
|
---|
157 | register int charlength = 0; \
|
---|
158 | register u_char *wptr = (Word)+1; \
|
---|
159 | register int length = 0; \
|
---|
160 | unsigned short c; \
|
---|
161 | \
|
---|
162 | charlength = parse_utf8_char((s_in),(end),&c); \
|
---|
163 | \
|
---|
164 | while (length+charlength <= MAXWORDLEN && charlength > 0 && \
|
---|
165 | !is_unicode_letdig(c)) { \
|
---|
166 | while (charlength-- > 0) { \
|
---|
167 | *wptr++ = *(s_in)++; ++length; \
|
---|
168 | } \
|
---|
169 | charlength = parse_utf8_char((s_in),(end),&c); \
|
---|
170 | } \
|
---|
171 | *(Word) = length; \
|
---|
172 | }while(0)
|
---|
173 |
|
---|
174 | /*
|
---|
175 | #define PARSE_NON_WORD(Word, s_in, end) \
|
---|
176 | do { \
|
---|
177 | register u_char *wptr = (Word)+1; \
|
---|
178 | register int length = 0; \
|
---|
179 | register int c = *(s_in); \
|
---|
180 | \
|
---|
181 | while( length < MAXWORDLEN && !INAWORD(c) && (s_in)<=(end) ) \
|
---|
182 | { \
|
---|
183 | *wptr++ = c; \
|
---|
184 | ++length; \
|
---|
185 | c = *++(s_in); \
|
---|
186 | } \
|
---|
187 | *(Word) = length; \
|
---|
188 | }while(0)
|
---|
189 | */
|
---|
190 |
|
---|
191 |
|
---|
192 | /* =========================================================================
|
---|
193 | * Macro: PARSE_STEM_WORD
|
---|
194 | * Description:
|
---|
195 | * Extracts out Word.
|
---|
196 | * Input:
|
---|
197 | * s_in points to 1st letter in buffer to test
|
---|
198 | * end points to last letter in buffer
|
---|
199 | * Output:
|
---|
200 | * s_in is modified to move to next word
|
---|
201 | * Returns Word filled in with length in 1st byte.
|
---|
202 | * ========================================================================= */
|
---|
203 | #define PARSE_STEM_WORD(Word, s_in, end) \
|
---|
204 | do { \
|
---|
205 | register int charlength = 0; \
|
---|
206 | register u_char *wptr = (Word)+1; \
|
---|
207 | register int length = 0; \
|
---|
208 | register int numeric = 0; \
|
---|
209 | unsigned short c; \
|
---|
210 | register int maxnumeric = IntEnv (GetEnv ("maxnumeric"), 4); \
|
---|
211 | \
|
---|
212 | charlength = parse_utf8_char((s_in),(end),&c); \
|
---|
213 | \
|
---|
214 | while (length+charlength <= MAXSTEMLEN && charlength > 0 && \
|
---|
215 | (is_unicode_letter(c) || (is_unicode_digit(c) && \
|
---|
216 | ++numeric <= maxnumeric))) { \
|
---|
217 | while (charlength-- > 0) { \
|
---|
218 | *wptr++ = *(s_in)++; ++length; \
|
---|
219 | } \
|
---|
220 | charlength = parse_utf8_char((s_in),(end),&c); \
|
---|
221 | } \
|
---|
222 | *(Word) = length; \
|
---|
223 | }while(0)
|
---|
224 | /*
|
---|
225 | #define PARSE_STEM_WORD(Word, s_in, end) \
|
---|
226 | do \
|
---|
227 | { \
|
---|
228 | register u_char *wptr = (Word)+1; \
|
---|
229 | register int length = 0; \
|
---|
230 | register int c = *(s_in); \
|
---|
231 | register int numeric = 0; \
|
---|
232 | \
|
---|
233 | while ( length < MAXSTEMLEN && INAWORD(c) && (s_in)<=(end)) \
|
---|
234 | { \
|
---|
235 | if ((numeric += INNUMBER(c)) > MAXNUMERIC) \
|
---|
236 | break; \
|
---|
237 | *wptr++ = c; \
|
---|
238 | ++length; \
|
---|
239 | c = *++(s_in); \
|
---|
240 | } \
|
---|
241 | *(Word) = length; \
|
---|
242 | }while(0)
|
---|
243 | */
|
---|
244 |
|
---|
245 |
|
---|
246 | /* =========================================================================
|
---|
247 | * Macro: PARSE_NON_STEM_WORD
|
---|
248 | * Description:
|
---|
249 | * Eat up non-word. Do not store non-word.
|
---|
250 | * It is not needed in index only in text !
|
---|
251 | *
|
---|
252 | * Input: as above but no Word needed
|
---|
253 | * Output: as above
|
---|
254 | * ========================================================================= */
|
---|
255 | #define PARSE_NON_STEM_WORD(s_in, end) \
|
---|
256 | do { \
|
---|
257 | register int charlength = 0; \
|
---|
258 | unsigned short c; \
|
---|
259 | \
|
---|
260 | charlength = parse_utf8_char((s_in),(end),&c); \
|
---|
261 | \
|
---|
262 | while (charlength > 0 && !is_unicode_letdig(c)) { \
|
---|
263 | (s_in) += charlength; \
|
---|
264 | charlength = parse_utf8_char((s_in),(end),&c); \
|
---|
265 | } \
|
---|
266 | }while(0)
|
---|
267 |
|
---|
268 | /*
|
---|
269 | #define PARSE_NON_STEM_WORD(s_in, end) \
|
---|
270 | do \
|
---|
271 | { \
|
---|
272 | while (!INAWORD(*(s_in)) && (s_in)<=(end)) \
|
---|
273 | (s_in)++; \
|
---|
274 | }while(0)
|
---|
275 | */
|
---|
276 |
|
---|
277 |
|
---|
278 | /* =========================================================================
|
---|
279 | * Macro: PARSE_NON_STEM_WORD_OR_SGML_TAG
|
---|
280 | * Description:
|
---|
281 | * Like PARSE_NON_STEM_WORD but also eats up SGML tags
|
---|
282 | * Input: as above
|
---|
283 | * Output: as above
|
---|
284 | * ========================================================================= */
|
---|
285 | #define PARSE_NON_STEM_WORD_OR_SGML_TAG(s_in, end) \
|
---|
286 | do { \
|
---|
287 | register int charlength = 0; \
|
---|
288 | unsigned short c; \
|
---|
289 | \
|
---|
290 | charlength = parse_utf8_char((s_in),(end),&c); \
|
---|
291 | \
|
---|
292 | while (charlength > 0 && !is_unicode_letdig(c)) { \
|
---|
293 | if (c == '<') { \
|
---|
294 | while (charlength > 0 && c != '>') { \
|
---|
295 | (s_in) += charlength; \
|
---|
296 | charlength = parse_utf8_char((s_in),(end),&c); \
|
---|
297 | } \
|
---|
298 | } \
|
---|
299 | (s_in) += charlength; \
|
---|
300 | charlength = parse_utf8_char((s_in),(end),&c); \
|
---|
301 | } \
|
---|
302 | }while(0)
|
---|
303 |
|
---|
304 | /*
|
---|
305 | #define PARSE_NON_STEM_WORD_OR_SGML_TAG(s_in, end) \
|
---|
306 | do \
|
---|
307 | { \
|
---|
308 | register int c = *(s_in); \
|
---|
309 | \
|
---|
310 | while (!INAWORD(c) && (s_in)<=(end)) \
|
---|
311 | { \
|
---|
312 | if (c == '<') \
|
---|
313 | { \
|
---|
314 | while (c != '>' && (s_in)<=(end)) \
|
---|
315 | c = *++(s_in); \
|
---|
316 | } \
|
---|
317 | if ((s_in)<=(end)) \
|
---|
318 | c = *++(s_in); \
|
---|
319 | } \
|
---|
320 | }while(0)
|
---|
321 | */
|
---|
322 |
|
---|
323 |
|
---|
324 | /* =========================================================================
|
---|
325 | * Macro: PARSE_OPT_TERM_PARAM [RPAP - Jan 97: Stem Index Change]
|
---|
326 | * Description:
|
---|
327 | * Extracts out optional paramater for query term.
|
---|
328 | * Needed only in parsing the query line !
|
---|
329 | *
|
---|
330 | * Note: that this function has not been converted to use UTF-8
|
---|
331 | * as it should still work as it is (only uses ascii
|
---|
332 | * characters)
|
---|
333 | *
|
---|
334 | * Input: as above but no Word needed
|
---|
335 | * Output: as above
|
---|
336 | * ========================================================================= */
|
---|
337 | #define PARSE_OPT_TERM_PARAM(Param, type, s_in, end) \
|
---|
338 | do \
|
---|
339 | { \
|
---|
340 | register u_char *wptr = (Param); \
|
---|
341 | register int length = 0; \
|
---|
342 | register int c = *(s_in); \
|
---|
343 | \
|
---|
344 | if (c == WEIGHTPARAM || c == STEMPARAM) \
|
---|
345 | { \
|
---|
346 | type = c; \
|
---|
347 | c = *++(s_in); \
|
---|
348 | while( length < MAXPARAMLEN && isdigit(c) && (s_in)<=(end)) \
|
---|
349 | { \
|
---|
350 | *wptr++ = c; \
|
---|
351 | ++length; \
|
---|
352 | c = *++(s_in); \
|
---|
353 | } \
|
---|
354 | *wptr = '\0'; \
|
---|
355 | for (; isdigit(c) && (s_in)<=(end); c = *++(s_in)) \
|
---|
356 | ; \
|
---|
357 | } \
|
---|
358 | }while(0)
|
---|
359 |
|
---|
360 | /* =========================================================================
|
---|
361 | * Macro: PARSE_RANKED_NON_STEM_WORD [RJM 07/97: Ranked Required Terms]
|
---|
362 | * Description:
|
---|
363 | * Eat up non-word. Do not store non-word.
|
---|
364 | * If come across a match requirement store it in require_match
|
---|
365 | * It is not needed in index only in text !
|
---|
366 | *
|
---|
367 | * Input: as above
|
---|
368 | * Output: the requirement mode for the next term. -1=must not match,
|
---|
369 | * 0=optional match, 1=must match
|
---|
370 | * ========================================================================= */
|
---|
371 | #define PARSE_RANKED_NON_STEM_WORD(require_match, s_in, end) \
|
---|
372 | do { \
|
---|
373 | register int charlength = 0; \
|
---|
374 | unsigned short c; \
|
---|
375 | (require_match) = 0; \
|
---|
376 | \
|
---|
377 | charlength = parse_utf8_char((s_in),(end),&c); \
|
---|
378 | \
|
---|
379 | while (charlength > 0 && !is_unicode_letdig(c)) { \
|
---|
380 | if (c == MUSTMATCHPARAM) (require_match) = 1; \
|
---|
381 | (s_in) += charlength; \
|
---|
382 | charlength = parse_utf8_char((s_in),(end),&c); \
|
---|
383 | } \
|
---|
384 | }while(0)
|
---|
385 |
|
---|
386 | /*
|
---|
387 | #define PARSE_RANKED_NON_STEM_WORD(require_match, s_in, end) \
|
---|
388 | do { \
|
---|
389 | (require_match) = 0; \
|
---|
390 | while (!INAWORD(*(s_in)) && (s_in)<=(end)) { \
|
---|
391 | if (*(s_in) == MUSTMATCHPARAM) { \
|
---|
392 | (require_match) = 1; \
|
---|
393 | } \
|
---|
394 | (s_in)++; \
|
---|
395 | } \
|
---|
396 | } while (0)
|
---|
397 | */
|
---|