File: linginfo.c

package info (click to toggle)
hspell 1.4-3
  • links: PTS, VCS
  • area: main
  • in suites: buster
  • size: 1,556 kB
  • sloc: ansic: 2,808; perl: 1,989; makefile: 209; sh: 109; awk: 15
file content (330 lines) | stat: -rw-r--r-- 8,338 bytes parent folder | download | duplicates (4)
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
/* Copyright (C) 2003-2004 Nadav Har'El and Dan Kenigsberg */

#include <stdio.h>
#include <stdlib.h>
#include <string.h>
#include "linginfo.h"

#include "dmask.c"


/* For an explanation of this bizarre set of definitions, see the comment
   in dict_radix.c, before a similar set. */
#ifdef HAVE_ZLIB
#define BUFFERED_ZLIB
#undef FILE
#undef pclose
#undef pclose
#undef getc
#ifdef BUFFERED_ZLIB
#include "gzbuffered.h"
#undef gzopen
#undef gzdopen
#define FILE void /* void* can be either normal FILE* or gzbFile*. Eek. */
#define popen(path,mode) gzb_open(path,mode)
#define gzopen(path,mode) gzb_open(path,mode)
#define gzdopen(path,mode) gzb_dopen(path,mode)
#define pclose(f) (gzb_close((gzbFile *)(f)))
#define getc(f) (gzb_getc(((gzbFile *)(f))))
#define fgets(s,n,f) (gzb_gets((s),(n),((gzbFile *)(f))))
#else
#include <zlib.h>
#define FILE void    /* FILE* is void*, a.k.a. voidp or gzFile */
#define pclose(f) (gzclose((f)))
#define popen(path,mode) (gzopen((path),(mode)))
#define getc(f) (gzgetc((f)))
#define fgets(s,n,f) (gzgets((s),(n),(f)))
#endif
#undef fgetc
#define fgetc(f) getc(f)
#endif /* HAVE_ZLIB */

static char *flat, **lookup;
static int lookuplen;
extern int hspell_debug;

static int dcode2dmask(const char *dcode) {
	int i = dcode[0]-'A'+(dcode[1]-'A')*26;
	return dmasks[i];
}

static char *dmask2text(char *s, int dmask) {
	char *c;
	s[0]=0;
	switch(dmask & D_TYPEMASK) {
		case D_NOUN: c=""; break;
		case D_VERB: c=""; break;
		case D_ADJ: c=""; break;
	        case 0: c="x"; break;
	        default: c="";
	}
	strcat(s,c);
	/* In few cases, both masculine and faminine are possible */
	if(dmask & D_GENDERMASK & D_MASCULINE) { strcat(s,","); }
	if(dmask & D_GENDERMASK & D_FEMININE) { strcat(s,","); }

	switch(dmask & D_GUFMASK) {
		case D_FIRST: c=",1"; break;
		case D_SECOND: c=",2"; break;
		case D_THIRD: c=",3"; break;
		default: c="";
	}
	strcat(s,c);
	switch(dmask & D_NUMMASK) {
		case D_SINGULAR: c=","; break;
		case D_DOUBLE: c=","; break;
		case D_PLURAL: c=","; break;
		default: c="";
	}
	strcat(s,c);
	switch(dmask & D_TENSEMASK) {
		case D_PAST: c=","; break;
		case D_PRESENT: c=","; break;
		case D_FUTURE: c=","; break;
		case D_IMPERATIVE: c=","; break;
		case D_INFINITIVE: c=","; break;
		case D_BINFINITIVE: c=",,"; break;
		default: c="";
	}
	strcat(s,c);
	if (dmask & D_SPECNOUN) {strcat(s,",");}
	if (dmask & D_OSMICHUT) {strcat(s,",");}
	if (dmask & D_OMASK) {
	strcat(s,",/");
	switch(dmask & D_OGENDERMASK) {
		case D_OMASCULINE: c=""; break;
		case D_OFEMININE: c=""; break;
		default: c="";
	}
	strcat(s,c);
	switch(dmask & D_OGUFMASK) {
		case D_OFIRST: c=",1"; break;
		case D_OSECOND: c=",2"; break;
		case D_OTHIRD: c=",3"; break;
		default: c="";
	}
	strcat(s,c);
	switch(dmask & D_ONUMMASK) {
		case D_OSINGULAR: c=","; break;
		case D_ODOUBLE: c=","; break;
		case D_OPLURAL: c=","; break;
		default: c="";
	}
	strcat(s,c);

	}

	return s;
}

char *linginfo_desc2text(char *text, const char *desc, int i) {
	int dmask;
	if (desc[i*2]==0) return 0;
	dmask = dcode2dmask(&desc[i*2]);
	dmask2text(text,dmask);
	return text;
}

/* find the prefixes required by a word according to its details */
static int linginfo_dmask2ps(int dmask) {
	int specifier;
	if ((dmask&D_TYPEMASK)==D_VERB) {
		if ((dmask&D_TENSEMASK)==D_IMPERATIVE) {
			specifier = PS_IMPER;
		} else if ((dmask&D_TENSEMASK)!=D_PRESENT) {
			specifier = PS_VERB;
		} else if (dmask & D_OSMICHUT || dmask & D_OMASK) {
			specifier = PS_NONDEF;
		} else specifier = PS_ALL;
/* TODO I feel that this may lead to a bug with  and other infinitives that
 * did not loose their initial lamed.  I should correct this all the way from
 * woo.pl */
		if ((dmask&D_TENSEMASK)==D_INFINITIVE) specifier = PS_L;
		else if ((dmask&D_TENSEMASK)==D_BINFINITIVE) specifier = PS_B;
	} else if (((dmask&D_TYPEMASK)==D_NOUN) || ((dmask&D_TYPEMASK) == D_ADJ)) {
		if (dmask & D_OSMICHUT || dmask & D_OMASK
		    || dmask & D_SPECNOUN) {
			specifier = PS_NONDEF;
		} else {
			specifier = PS_ALL;
		}
	} else specifier = PS_ALL;
	return specifier;
}

int linginfo_desc2ps(const char *desc, int i) {
	int dmask;
	if (desc[i*2]==0) return 0;
	dmask = dcode2dmask(&desc[i*2]);
	return linginfo_dmask2ps(dmask);
}

char *linginfo_stem2text(const char *stem, int i) {
	int wp;
	if (stem[i*3]==0) return 0;
   	wp = stem[i*3]-33+(stem[i*3+1]-33)*94+
		(stem[i*3+2]-33)*94*94;
	return lookup[wp];
}

/* currently linginfo_init reopens the words file, reinterprets it, and stores
 * it flat in memory. If it sounds silly to you, you probably can hear. */
int linginfo_init(const char *dir) {
	FILE *fp,*fpstems,*fpdesc;
	char *current;
	char s[1024],stem[100],desc[100];
	int i=0,j;
	int flatsize;

	snprintf(s,sizeof(s),"%s.sizes",dir);
	if(!(fp=fopen(s,"r"))){
		fprintf(stderr,"Hspell: can't open %s.\n",s);
		return 0;
	}
	fscanf(fp,"%*d %*d %*d"); /* ignore non linginfo sizes */
	if(fscanf(fp,"%d %d",&flatsize,&lookuplen)!=2){
		fprintf(stderr,"Hspell: can't read from %s.\n",s);
		return 0;
	}
	fclose(fp);

	current = flat = (char *)malloc(flatsize);
	lookup = (char **)malloc(sizeof(char *)*lookuplen);
	if (!current || !lookup) {
		fprintf (stderr, "Hspell: alloc failed\n");
		return 0;
	}

	/* read dictionary into memory */

	/* TODO: have better quoting for filename, or use zlib directly */
#ifdef HAVE_ZLIB
	snprintf(s,sizeof(s),"%s",dir);
#else
	snprintf(s,sizeof(s),"gzip -dc '%s'",dir);
#endif
	if(!(fp=popen(s,"r"))){
		fprintf(stderr,"Hspell: can't open %s.\n",s);
		return 0;
	}
#ifdef HAVE_ZLIB
	snprintf(s,sizeof(s),"%s.stems",dir);
#else
	snprintf(s,sizeof(s),"gzip -dc '%s.stems'",dir);
#endif
	if(!(fpstems=popen(s,"r"))){
		fprintf(stderr,"Hspell: can't open %s.\n",s);
		pclose(fp);
		return 0;
	}
#ifdef HAVE_ZLIB
	snprintf(s,sizeof(s),"%s.desc",dir);
#else
	snprintf(s,sizeof(s),"gzip -dc '%s.desc'",dir);
#endif
	if(!(fpdesc=popen(s,"r"))){
		fprintf(stderr,"Hspell: can't open %s.\n",s);
		pclose(fp);
		pclose(fpstems);
		return 0;
	}


	/* The following code for reading wunzip'ed word list is copied from
	 * wunzip.c and repeats what was done dict_radix.c's do_read_dict(). It
	 * would be much nicer to read the word list only once. */
	{
	char sbuf[256];
	int slen=0;
	int c,n;
	while(1){
		c=fgetc(fp);
		if((c>='0' && c<='9') || c==EOF){
			/* new word - output old word first */
			sbuf[slen]='\0';
			lookup[i++] = current;
			for(j=0; j<=slen; j++) current++[0]=sbuf[j];
			if (!fgets(stem,sizeof(stem),fpstems)) {
				fprintf(stderr, "Hspell: linginfo: unexpected end of file in stems file\n");
				return 0;
			}
			if (!fgets(desc,sizeof(desc),fpdesc)) {
				fprintf(stderr, "Hspell: linginfo: unexpected end of file in description file\n");
				return 0;
			}
			for (j=0; desc[j]!='\n' && desc[j]!=0; j++) {
				current++[0]=desc[j];
			}
			current++[0]=0;
			for (j=0; stem[j]!='\n' && stem[j]!=0; j++) {
				current++[0]=stem[j];
			}
			current++[0]=0;
			if (c==EOF) break;

			/* and read how much to go back */
			n=0;
			do {
				/* base 10... */
				n*=10;
				n+=(c-'0');
			} while ((c=fgetc(fp))!=EOF && c>='0' && c<='9');
			slen-=n;
			if(slen<0 || slen >= sizeof(sbuf)-1){
				fprintf(stderr,"Hspell: bad backlength %d... giving up.\n", slen);
				return 0;
			}
			/* we got a new letter c - continue the loop */

		}
		/* word letter - add it */
		if(slen>=sizeof(sbuf)-1){
			fprintf(stderr,"Hspell: word too long... giving up.\n");
			return 0;
		}
		sbuf[slen++]=c;
	}
	}

	pclose(fp);
	pclose(fpstems);
	pclose(fpdesc);

	if (hspell_debug) {
		fprintf (stderr, "linginfo: finished reading %d words and stems\n",i);
	}
	return 1;
}

int linginfo_lookup(const char *word, char **desc, char **stem)
{
	int res,i=0,bottom=0,top=lookuplen;
	while (top>=bottom) {
		if (i==(top-bottom)/2 + bottom) {
			return 0;
		}
		i=(top-bottom)/2 + bottom;
		if (hspell_debug) fprintf(stderr,"bot=%d i=%d top=%d) %s\n",bottom,i,top, lookup[i]);
		res = strcmp(lookup[i],word);
		if (res>0) {top=i;}
		else if (res<0) {bottom=i;}
		else {
			int len,desclen;
		        len = strlen(lookup[i]);
			*desc = lookup[i]+len+1;
			desclen = strlen(*desc);
			*stem = *desc+desclen+1;
			return 1;
		}
	}
	return 0;
}

int linginfo_free(void) {
	if (lookup) {
		free(lookup);
		free(flat);
	}
	return 1;
}