IDZEBRA  2.1.2
charmap.c
Go to the documentation of this file.
1 /* This file is part of the Zebra server.
2  Copyright (C) Index Data
3 
4 Zebra is free software; you can redistribute it and/or modify it under
5 the terms of the GNU General Public License as published by the Free
6 Software Foundation; either version 2, or (at your option) any later
7 version.
8 
9 Zebra is distributed in the hope that it will be useful, but WITHOUT ANY
10 WARRANTY; without even the implied warranty of MERCHANTABILITY or
11 FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License
12 for more details.
13 
14 You should have received a copy of the GNU General Public License
15 along with this program; if not, write to the Free Software
16 Foundation, Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA
17 
18 */
19 
20 
29 #if HAVE_CONFIG_H
30 #include <config.h>
31 #endif
32 #include <ctype.h>
33 #include <string.h>
34 #include <stdlib.h>
35 #include <assert.h>
36 
37 typedef unsigned ucs4_t;
38 
39 #include <charmap.h>
40 
41 #include <yaz/yaz-util.h>
42 
43 #define CHR_MAXSTR 1024
44 #define CHR_MAXEQUIV 32
45 
46 const unsigned char CHR_FIELD_BEGIN = '^';
47 
48 const char *CHR_UNKNOWN = "\001";
49 const char *CHR_SPACE = "\002";
50 const char *CHR_CUT = "\003";
51 const char *CHR_BASE = "\005"; /* CHECK CHR_BASE_CHAR as well */
52 
54 {
55  chr_t_entry *input; /* mapping table for input data */
56  chr_t_entry *q_input; /* mapping table for queries */
57  unsigned char *output[256]; /* return mapping - for display of registers */
58  int base_uppercase; /* Start of upper-case ordinals */
59  NMEM nmem;
60 };
61 
62 /*
63  * Character map trie node.
64  */
66 {
67  chr_t_entry **children; /* array of children */
68  unsigned char **target; /* target for this node, if any */
69 };
70 
71 /*
72  * General argument structure for callback functions (internal use only)
73  */
74 typedef struct chrwork
75 {
77  char string[CHR_MAXSTR+1];
78 } chrwork;
79 
80 /*
81  * Callback for equivalent stuff
82  */
83 typedef struct
84 {
85  NMEM nmem;
86  int no_eq;
87  char *eq[CHR_MAXEQUIV];
89 /*
90  * Add an entry to the character map.
91  */
93  const char *from, int len, char *to,
94  const char *from_0)
95 {
96  if (!from_0)
97  from_0 = from;
98  if (!root)
99  {
100  root = (chr_t_entry *) nmem_malloc(nmem, sizeof(*root));
101  root->children = 0;
102  root->target = 0;
103  }
104  if (!len)
105  {
106  if (!root->target || !root->target[0] ||
107  strcmp((const char *) root->target[0], to))
108  {
109  if (from_0 &&
110  root->target && root->target[0] && root->target[0][0] &&
111  strcmp((const char *) root->target[0], CHR_UNKNOWN))
112  {
113  yaz_log(YLOG_WARN, "duplicate entry for charmap from '%s'",
114  from_0);
115  }
116  root->target = (unsigned char **)
117  nmem_malloc(nmem, sizeof(*root->target)*2);
118  root->target[0] = (unsigned char *) nmem_strdup(nmem, to);
119  root->target[1] = 0;
120  }
121  }
122  else
123  {
124  if (!root->children)
125  {
126  int i;
127 
128  root->children = (chr_t_entry **)
129  nmem_malloc(nmem, sizeof(chr_t_entry*) * 256);
130  for (i = 0; i < 256; i++)
131  root->children[i] = 0;
132  }
133  if (!(root->children[(unsigned char) *from] =
134  set_map_string(root->children[(unsigned char) *from], nmem,
135  from + 1, len - 1, to, from_0)))
136  return 0;
137  }
138  return root;
139 }
140 
141 static chr_t_entry *find_entry_x(chr_t_entry *t, const char **from, int *len, int first)
142 {
143  chr_t_entry *res;
144 
145  while (*len <= 0)
146  { /* switch to next buffer */
147  if (*len < 0)
148  break;
149  from++;
150  len++;
151  }
152  if (*len > 0 && t->children)
153  {
154  const char *old_from = *from;
155  int old_len = *len;
156 
157  res = 0;
158 
159  if (first && t->children[CHR_FIELD_BEGIN])
160  {
161  if ((res = find_entry_x(t->children[CHR_FIELD_BEGIN], from, len, 0)) && res != t->children[CHR_FIELD_BEGIN])
162  return res;
163  else
164  res = 0;
165  /* otherwhise there was no match on beginning of field, move on */
166  }
167 
168  if (!res && t->children[(unsigned char) **from])
169  {
170  (*len)--;
171  (*from)++;
172  if ((res = find_entry_x(t->children[(unsigned char) *old_from],
173  from, len, 0)))
174  return res;
175  /* no match */
176  *len = old_len;
177  *from = old_from;
178  }
179  }
180  /* no children match. use ourselves, if we have a target */
181  return t->target ? t : 0;
182 }
183 
184 const char **chr_map_input_x(chrmaptab maptab, const char **from, int *len, int first)
185 {
186  chr_t_entry *t = maptab->input;
187  chr_t_entry *res;
188 
189  if (!(res = find_entry_x(t, from, len, first)))
190  abort();
191  return (const char **) (res->target);
192 }
193 
194 const char **chr_map_input(chrmaptab maptab, const char **from, int len, int first)
195 {
196  chr_t_entry *t = maptab->input;
197  chr_t_entry *res;
198  int len_tmp[2];
199 
200  len_tmp[0] = len;
201  len_tmp[1] = -1;
202  if (!(res = find_entry_x(t, from, len_tmp, first)))
203  abort();
204  return (const char **) (res->target);
205 }
206 
207 const char **chr_map_q_input(chrmaptab maptab,
208  const char **from, int len, int first)
209 {
210  chr_t_entry *t = maptab->q_input;
211  chr_t_entry *res;
212  int len_tmp[2];
213 
214  len_tmp[0] = len;
215  len_tmp[1] = -1;
216  if (!(res = find_entry_x(t, from, len_tmp, first)))
217  return 0;
218  return (const char **) (res->target);
219 }
220 
221 const char *chr_map_output(chrmaptab maptab, const char **from, int len)
222 {
223  unsigned char c = ** (unsigned char **) from;
224  const char *out = (const char*) maptab->output[c];
225 
226  if (out)
227  (*from)++;
228  return out;
229 }
230 
231 static int zebra_ucs4_strlen(ucs4_t *s)
232 {
233  int i = 0;
234  while (*s++)
235  i++;
236  return i;
237 }
238 
240 {
241  ucs4_t c;
242  ucs4_t i = 0;
243  char fmtstr[8];
244 
245  if (**s == '\\' && 1[*s])
246  {
247  (*s)++;
248  c = **s;
249  switch (c)
250  {
251  case '\\': c = '\\'; (*s)++; break;
252  case 'r': c = '\r'; (*s)++; break;
253  case 'n': c = '\n'; (*s)++; break;
254  case 't': c = '\t'; (*s)++; break;
255  case 's': c = ' '; (*s)++; break;
256  case 'x':
257  if (zebra_ucs4_strlen(*s) >= 3)
258  {
259  fmtstr[0] = (*s)[1];
260  fmtstr[1] = (*s)[2];
261  fmtstr[2] = 0;
262  sscanf(fmtstr, "%x", &i);
263  c = i;
264  *s += 3;
265  }
266  break;
267  case '0':
268  case '1':
269  case '2':
270  case '3':
271  case '4':
272  case '5':
273  case '6':
274  case '7':
275  case '8':
276  case '9':
277  if (zebra_ucs4_strlen(*s) >= 3)
278  {
279  fmtstr[0] = (*s)[0];
280  fmtstr[1] = (*s)[1];
281  fmtstr[2] = (*s)[2];
282  fmtstr[3] = 0;
283  sscanf(fmtstr, "%o", &i);
284  c = i;
285  *s += 3;
286  }
287  break;
288  case 'L':
289  if (zebra_ucs4_strlen(*s) >= 5)
290  {
291  fmtstr[0] = (*s)[1];
292  fmtstr[1] = (*s)[2];
293  fmtstr[2] = (*s)[3];
294  fmtstr[3] = (*s)[4];
295  fmtstr[4] = 0;
296  sscanf(fmtstr, "%x", &i);
297  c = i;
298  *s += 5;
299  }
300  break;
301  default:
302  (*s)++;
303  }
304  }
305  else
306  {
307  c = **s;
308  ++(*s);
309  }
310  yaz_log(YLOG_DEBUG, "out %d", c);
311  return c;
312 }
313 
314 /*
315  * Callback function.
316  * Add an entry to the value space.
317  */
318 static void fun_addentry(const char *s, void *data, int num)
319 {
320  chrmaptab tab = (chrmaptab) data;
321  char tmp[2];
322 
323  tmp[0] = num; tmp[1] = '\0';
324  tab->input = set_map_string(tab->input, tab->nmem, s, strlen(s), tmp, 0);
325  tab->output[num + tab->base_uppercase] =
326  (unsigned char *) nmem_strdup(tab->nmem, s);
327 }
328 
329 /*
330  * Callback function.
331  * Add a space-entry to the value space.
332  */
333 static void fun_addspace(const char *s, void *data, int num)
334 {
335  chrmaptab tab = (chrmaptab) data;
336  tab->input = set_map_string(tab->input, tab->nmem, s, strlen(s),
337  (char*) CHR_SPACE, 0);
338 }
339 
340 /*
341  * Callback function.
342  * Add a space-entry to the value space.
343  */
344 static void fun_addcut(const char *s, void *data, int num)
345 {
346  chrmaptab tab = (chrmaptab) data;
347  tab->input = set_map_string(tab->input, tab->nmem, s, strlen(s),
348  (char*) CHR_CUT, 0);
349 }
350 
351 /*
352  * Create a string containing the mapped characters provided.
353  */
354 static void fun_mkstring(const char *s, void *data, int num)
355 {
356  chrwork *arg = (chrwork *) data;
357  const char **res, *p = s;
358 
359  res = chr_map_input(arg->map, &s, strlen(s), 0);
360  if (*res == (char*) CHR_UNKNOWN)
361  yaz_log(YLOG_WARN, "Map: '%s' has no mapping", p);
362  strncat(arg->string, *res, CHR_MAXSTR - strlen(arg->string));
363  arg->string[CHR_MAXSTR] = '\0';
364 }
365 
366 /*
367  * Create an unmodified string (scan_string handler).
368  */
369 static void fun_add_equivalent_string(const char *s, void *data, int num)
370 {
371  chr_equiv_work *arg = (chr_equiv_work *) data;
372 
373  if (arg->no_eq == CHR_MAXEQUIV)
374  return;
375  arg->eq[arg->no_eq++] = nmem_strdup(arg->nmem, s);
376 }
377 
378 /*
379  * Add a map to the string contained in the argument.
380  */
381 static void fun_add_map(const char *s, void *data, int num)
382 {
383  chrwork *arg = (chrwork *) data;
384 
385  assert(arg->map->input);
386  yaz_log(YLOG_DEBUG, "set map %.*s", (int) strlen(s), s);
387  set_map_string(arg->map->input, arg->map->nmem, s, strlen(s), arg->string,
388  0);
389  for (s = arg->string; *s; s++)
390  yaz_log(YLOG_DEBUG, " %3d", (unsigned char) *s);
391 }
392 
393 static int scan_to_utf8(yaz_iconv_t t, ucs4_t *from, size_t inlen,
394  char *outbuf, size_t outbytesleft)
395 {
396  size_t inbytesleft = inlen * sizeof(ucs4_t);
397  char *inbuf = (char*) from;
398  size_t ret;
399 
400  if (t == 0)
401  *outbuf++ = *from; /* ISO-8859-1 is OK here */
402  else
403  {
404  ret = yaz_iconv(t, &inbuf, &inbytesleft, &outbuf, &outbytesleft);
405  if (ret != (size_t) (-1))
406  ret = yaz_iconv(t, 0, 0, &outbuf, &outbytesleft);
407 
408 
409  if (ret == (size_t) (-1))
410  {
411  yaz_log(YLOG_LOG, "from: %2X %2X %2X %2X",
412  from[0], from[1], from[2], from[3]);
413  yaz_log(YLOG_WARN|YLOG_ERRNO, "bad unicode sequence");
414  return -1;
415  }
416  }
417  *outbuf = '\0';
418  return 0;
419 }
420 
421 static int scan_string(char *s_native,
422  yaz_iconv_t t_unicode, yaz_iconv_t t_utf8,
423  void (*fun)(const char *c, void *data, int num),
424  void *data, int *num)
425 {
426  char str[1024];
427 
428  ucs4_t arg[512];
429  ucs4_t arg_prim[512];
430  ucs4_t *s = arg;
431  ucs4_t c, begin, end;
432  size_t i;
433 
434  if (t_unicode != 0)
435  {
436  char *outbuf = (char *) arg;
437  char *inbuf = s_native;
438  size_t outbytesleft = sizeof(arg)-4;
439  size_t inbytesleft = strlen(s_native);
440  size_t ret;
441  ret = yaz_iconv(t_unicode, &inbuf, &inbytesleft,
442  &outbuf, &outbytesleft);
443  if (ret != (size_t)(-1))
444  ret = yaz_iconv(t_unicode, 0, 0, &outbuf, &outbytesleft);
445 
446  if (ret == (size_t)(-1))
447  return -1;
448  i = (outbuf - (char*) arg)/sizeof(ucs4_t);
449  }
450  else
451  {
452  for (i = 0; s_native[i]; i++)
453  arg[i] = s_native[i] & 255; /* ISO-8859-1 conversion */
454  }
455  arg[i] = 0; /* terminate */
456  if (s[0] == 0xfeff || s[0] == 0xfeff) /* skip byte Order Mark */
457  s++;
458  while (*s)
459  {
460  switch (*s)
461  {
462  case '{':
463  s++;
464  begin = zebra_prim_w(&s);
465  if (*s != '-')
466  {
467  yaz_log(YLOG_FATAL, "Bad range in char-map");
468  return -1;
469  }
470  s++;
471  end = zebra_prim_w(&s);
472  if (end <= begin)
473  {
474  yaz_log(YLOG_FATAL, "Bad range in char-map");
475  return -1;
476  }
477  s++;
478  for (c = begin; c <= end; c++)
479  {
480  if (scan_to_utf8(t_utf8, &c, 1, str, sizeof(str)-1))
481  return -1;
482  (*fun)(str, data, num ? (*num)++ : 0);
483  }
484  break;
485  case '(':
486  ++s;
487  i = 0;
488  while (*s != ')' || s[-1] == '\\')
489  {
490  if (*s == '\0')
491  {
492  yaz_log(YLOG_FATAL, "Missing ) in charmap");
493  return -1;
494  }
495  arg_prim[i++] = zebra_prim_w(&s);
496  }
497  arg_prim[i] = 0;
498  if (scan_to_utf8(t_utf8, arg_prim, zebra_ucs4_strlen(arg_prim), str, sizeof(str)-1))
499  return -1;
500  (*fun)(str, data, num ? (*num)++ : 0);
501  s++;
502  break;
503  default:
504  c = zebra_prim_w(&s);
505  if (scan_to_utf8(t_utf8, &c, 1, str, sizeof(str)-1))
506  return -1;
507  (*fun)(str, data, num ? (*num)++ : 0);
508  }
509  }
510  return 0;
511 }
512 
513 chrmaptab chrmaptab_create(const char *tabpath, const char *name,
514  const char *tabroot)
515 {
516  FILE *f;
517  char line[512], *argv[50];
518  chrmaptab res;
519  int lineno = 0;
520  int no_directives = 0;
521  int errors = 0;
522  int argc, num = (int) *CHR_BASE, i;
523  NMEM nmem;
524  yaz_iconv_t t_unicode = 0;
525  yaz_iconv_t t_utf8 = 0;
526  unsigned endian = 31;
527  const char *ucs4_native = "UCS-4";
528 
529  yaz_log(YLOG_DEBUG, "maptab %s open", name);
530  if (!(f = yaz_fopen(tabpath, name, "r", tabroot)))
531  {
532  yaz_log(YLOG_WARN|YLOG_ERRNO, "%s", name);
533  return 0;
534  }
535 
536  if (*(char*) &endian == 31) /* little endian? */
537  ucs4_native = "UCS-4LE";
538 
539  t_utf8 = yaz_iconv_open("UTF-8", ucs4_native);
540 
541  nmem = nmem_create();
542  res = (chrmaptab) nmem_malloc(nmem, sizeof(*res));
543  res->nmem = nmem;
544  res->input = (chr_t_entry *) nmem_malloc(res->nmem, sizeof(*res->input));
545  res->input->target = (unsigned char **)
546  nmem_malloc(res->nmem, sizeof(*res->input->target) * 2);
547  res->input->target[0] = (unsigned char*) CHR_UNKNOWN;
548  res->input->target[1] = 0;
549  res->input->children = (chr_t_entry **)
550  nmem_malloc(res->nmem, sizeof(res->input) * 256);
551  for (i = 0; i < 256; i++)
552  {
553  res->input->children[i] = (chr_t_entry *)
554  nmem_malloc(res->nmem, sizeof(*res->input));
555  res->input->children[i]->children = 0;
556  res->input->children[i]->target = (unsigned char **)
557  nmem_malloc(res->nmem, 2 * sizeof(unsigned char *));
558  res->input->children[i]->target[1] = 0;
559  res->input->children[i]->target[0] = (unsigned char*) CHR_UNKNOWN;
560  }
561  res->q_input = (chr_t_entry *)
562  nmem_malloc(res->nmem, sizeof(*res->q_input));
563  res->q_input->target = 0;
564  res->q_input->children = 0;
565 
566  for (i = *CHR_BASE; i < 256; i++)
567  res->output[i] = 0;
568  res->output[(int) *CHR_SPACE] = (unsigned char *) " ";
569  res->output[(int) *CHR_UNKNOWN] = (unsigned char*) "@";
570  res->base_uppercase = 0;
571 
572  while (!errors && (argc = readconf_line(f, &lineno, line, 512, argv, 50)))
573  {
574  no_directives++;
575  if (!yaz_matchstr(argv[0], "lowercase"))
576  {
577  if (argc != 2)
578  {
579  yaz_log(YLOG_FATAL, "Syntax error in charmap");
580  ++errors;
581  }
582  if (scan_string(argv[1], t_unicode, t_utf8, fun_addentry,
583  res, &num) < 0)
584  {
585  yaz_log(YLOG_FATAL, "Bad value-set specification");
586  ++errors;
587  }
588  res->base_uppercase = num;
589  res->output[(int) *CHR_SPACE + num] = (unsigned char *) " ";
590  res->output[(int) *CHR_UNKNOWN + num] = (unsigned char*) "@";
591  num = (int) *CHR_BASE;
592  }
593  else if (!yaz_matchstr(argv[0], "uppercase"))
594  {
595  if (!res->base_uppercase)
596  {
597  yaz_log(YLOG_FATAL, "Uppercase directive with no lowercase set");
598  ++errors;
599  }
600  if (argc != 2)
601  {
602  yaz_log(YLOG_FATAL, "Missing arg for uppercase directive");
603  ++errors;
604  }
605  if (scan_string(argv[1], t_unicode, t_utf8, fun_addentry,
606  res, &num) < 0)
607  {
608  yaz_log(YLOG_FATAL, "Bad value-set specification");
609  ++errors;
610  }
611  }
612  else if (!yaz_matchstr(argv[0], "space"))
613  {
614  if (argc != 2)
615  {
616  yaz_log(YLOG_FATAL, "Syntax error in charmap for space");
617  ++errors;
618  }
619  if (scan_string(argv[1], t_unicode, t_utf8,
620  fun_addspace, res, 0) < 0)
621  {
622  yaz_log(YLOG_FATAL, "Bad space specification");
623  ++errors;
624  }
625  }
626  else if (!yaz_matchstr(argv[0], "cut"))
627  {
628  if (argc != 2)
629  {
630  yaz_log(YLOG_FATAL, "Syntax error in charmap for cut");
631  ++errors;
632  }
633  if (scan_string(argv[1], t_unicode, t_utf8,
634  fun_addcut, res, 0) < 0)
635  {
636  yaz_log(YLOG_FATAL, "Bad cut specification");
637  ++errors;
638  }
639  }
640  else if (!yaz_matchstr(argv[0], "map"))
641  {
642  chrwork buf;
643 
644  if (argc != 3)
645  {
646  yaz_log(YLOG_FATAL, "charmap directive map requires 2 args");
647  ++errors;
648  }
649  buf.map = res;
650  buf.string[0] = '\0';
651  if (scan_string(argv[2], t_unicode, t_utf8,
652  fun_mkstring, &buf, 0) < 0)
653  {
654  yaz_log(YLOG_FATAL, "Bad map target");
655  ++errors;
656  }
657  if (scan_string(argv[1], t_unicode, t_utf8,
658  fun_add_map, &buf, 0) < 0)
659  {
660  yaz_log(YLOG_FATAL, "Bad map source");
661  ++errors;
662  }
663  }
664  else if (!yaz_matchstr(argv[0], "equivalent"))
665  {
666  chr_equiv_work w;
667 
668  if (argc != 2)
669  {
670  yaz_log(YLOG_FATAL, "equivalent requires 1 argument");
671  ++errors;
672  }
673  w.nmem = res->nmem;
674  w.no_eq = 0;
675  if (scan_string(argv[1], t_unicode, t_utf8,
676  fun_add_equivalent_string, &w, 0) < 0)
677  {
678  yaz_log(YLOG_FATAL, "equivalent: invalid string");
679  ++errors;
680  }
681  else if (w.no_eq == 0)
682  {
683  yaz_log(YLOG_FATAL, "equivalent: no strings");
684  ++errors;
685  }
686  else
687  {
688  char *result_str;
689  int i, slen = 5;
690 
691  /* determine length of regular expression */
692  for (i = 0; i<w.no_eq; i++)
693  slen += strlen(w.eq[i]) + 1;
694  result_str = nmem_malloc(res->nmem, slen + 5);
695 
696  /* build the regular expression */
697  *result_str = '\0';
698  slen = 0;
699  for (i = 0; i<w.no_eq; i++)
700  {
701  result_str[slen++] = i ? '|' : '(';
702  strcpy(result_str + slen, w.eq[i]);
703  slen += strlen(w.eq[i]);
704  }
705  result_str[slen++] = ')';
706  result_str[slen] = '\0';
707 
708  /* each eq will map to this regular expression */
709  for (i = 0; i<w.no_eq; i++)
710  {
711  set_map_string(res->q_input, res->nmem,
712  w.eq[i], strlen(w.eq[i]),
713  result_str, 0);
714  }
715  }
716  }
717  else if (!yaz_matchstr(argv[0], "encoding"))
718  {
719  if (t_unicode != 0)
720  yaz_iconv_close(t_unicode);
721  t_unicode = yaz_iconv_open(ucs4_native, argv[1]);
722  }
723  else
724  {
725  yaz_log(YLOG_WARN, "Syntax error at '%s' in %s", line, name);
726  errors++;
727  }
728  }
729  yaz_fclose(f);
730  if (no_directives == 0)
731  {
732  yaz_log(YLOG_WARN, "No directives in '%s'", name);
733  errors++;
734  }
735  if (errors)
736  {
737  chrmaptab_destroy(res);
738  res = 0;
739  }
740  yaz_log(YLOG_DEBUG, "maptab %s num=%d close %d errors", name, num, errors);
741  if (t_utf8 != 0)
742  yaz_iconv_close(t_utf8);
743  if (t_unicode != 0)
744  yaz_iconv_close(t_unicode);
745  return res;
746 }
747 
749 {
750  if (tab)
751  nmem_destroy(tab->nmem);
752 }
753 
754 
755 /*
756  * Local variables:
757  * c-basic-offset: 4
758  * c-file-style: "Stroustrup"
759  * indent-tabs-mode: nil
760  * End:
761  * vim: shiftwidth=4 tabstop=8 expandtab
762  */
763 
const char ** chr_map_q_input(chrmaptab maptab, const char **from, int len, int first)
Definition: charmap.c:207
const char * chr_map_output(chrmaptab maptab, const char **from, int len)
Definition: charmap.c:221
const char ** chr_map_input_x(chrmaptab maptab, const char **from, int *len, int first)
Definition: charmap.c:184
unsigned ucs4_t
Definition: charmap.c:37
const char * CHR_CUT
Definition: charmap.c:50
static int zebra_ucs4_strlen(ucs4_t *s)
Definition: charmap.c:231
int base_uppercase
Definition: charmap.c:58
struct chrwork chrwork
chr_t_entry * q_input
Definition: charmap.c:56
const char * CHR_BASE
Definition: charmap.c:51
#define CHR_MAXSTR
Definition: charmap.c:43
const char ** chr_map_input(chrmaptab maptab, const char **from, int len, int first)
Definition: charmap.c:194
static void fun_addcut(const char *s, void *data, int num)
Definition: charmap.c:344
char string[CHR_MAXSTR+1]
Definition: charmap.c:77
const char * CHR_SPACE
Definition: charmap.c:49
static chr_t_entry * find_entry_x(chr_t_entry *t, const char **from, int *len, int first)
Definition: charmap.c:141
static void * begin(struct zebra_register *reg, void *class_handle, RSET rset, NMEM nmem, TERMID *terms, int numterms)
Definition: rank1.c:107
static int scan_to_utf8(yaz_iconv_t t, ucs4_t *from, size_t inlen, char *outbuf, size_t outbytesleft)
Definition: charmap.c:393
struct chrmaptab_info * chrmaptab
Definition: charmap.h:38
static void fun_addentry(const char *s, void *data, int num)
Definition: charmap.c:318
const char * CHR_UNKNOWN
Definition: charmap.c:48
static void fun_add_equivalent_string(const char *s, void *data, int num)
Definition: charmap.c:369
unsigned char * output[256]
Definition: charmap.c:57
static void fun_add_map(const char *s, void *data, int num)
Definition: charmap.c:381
void chrmaptab_destroy(chrmaptab tab)
Definition: charmap.c:748
Definition: charmap.c:65
static void fun_mkstring(const char *s, void *data, int num)
Definition: charmap.c:354
const unsigned char CHR_FIELD_BEGIN
Definition: charmap.c:46
unsigned char ** target
Definition: charmap.c:68
chrmaptab chrmaptab_create(const char *tabpath, const char *name, const char *tabroot)
Definition: charmap.c:513
chr_t_entry * input
Definition: charmap.c:55
static void end(struct zebra_register *reg, void *set_handle)
Definition: rank1.c:156
chr_t_entry ** children
Definition: charmap.c:67
static int scan_string(char *s_native, yaz_iconv_t t_unicode, yaz_iconv_t t_utf8, void(*fun)(const char *c, void *data, int num), void *data, int *num)
Definition: charmap.c:421
ucs4_t zebra_prim_w(ucs4_t **s)
Definition: charmap.c:239
char * eq[CHR_MAXEQUIV]
Definition: charmap.c:87
static void fun_addspace(const char *s, void *data, int num)
Definition: charmap.c:333
chrmaptab map
Definition: charmap.c:76
#define CHR_MAXEQUIV
Definition: charmap.c:44
static chr_t_entry * set_map_string(chr_t_entry *root, NMEM nmem, const char *from, int len, char *to, const char *from_0)
Definition: charmap.c:92