-/*
- * Copyright (C) 1996-2002, Index Data
- * All rights reserved.
- * Sebastian Hammer, Adam Dickmeiss
- *
- * $Id: charmap.c,v 1.20 2002-04-04 20:50:37 adam Exp $
- *
- */
+/* $Id: charmap.c,v 1.26 2002-08-28 19:52:29 adam Exp $
+ Copyright (C) 1995,1996,1997,1998,1999,2000,2001,2002
+ Index Data Aps
+
+This file is part of the Zebra server.
+
+Zebra is free software; you can redistribute it and/or modify it under
+the terms of the GNU General Public License as published by the Free
+Software Foundation; either version 2, or (at your option) any later
+version.
+
+Zebra is distributed in the hope that it will be useful, but WITHOUT ANY
+WARRANTY; without even the implied warranty of MERCHANTABILITY or
+FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License
+for more details.
+
+You should have received a copy of the GNU General Public License
+along with Zebra; see the file LICENSE.zebra. If not, write to the
+Free Software Foundation, 59 Temple Place - Suite 330, Boston, MA
+02111-1307, USA.
+*/
+
+
/*
* Support module to handle character-conversions into and out of the
#include <string.h>
#include <assert.h>
+typedef unsigned ucs4_t;
+
#include <yaz/yaz-util.h>
#include <charmap.h>
+
#define CHR_MAXSTR 1024
#define CHR_MAXEQUIV 32
unsigned char zebra_prim(char **s)
{
unsigned char c;
- unsigned int i;
+ unsigned int i = 0;
yaz_log (LOG_DEBUG, "prim %.3s", *s);
if (**s == '\\')
{
+ (*s)++;
+ c = **s;
+ switch (c)
+ {
+ case '\\': c = '\\'; (*s)++; break;
+ case 'r': c = '\r'; (*s)++; break;
+ case 'n': c = '\n'; (*s)++; break;
+ case 't': c = '\t'; (*s)++; break;
+ case 's': c = ' '; (*s)++; break;
+ case 'x': sscanf(*s, "x%2x", &i); c = i; *s += 3; break;
+ case '0':
+ case '1':
+ case '2':
+ case '3':
+ case '4':
+ case '5':
+ case '6':
+ case '7':
+ case '8':
+ case '9':
+ sscanf(*s, "%3o", &i);
+ c = i;
+ *s += 3;
+ break;
+ default:
+ (*s)++;
+ }
+ }
+ else
+ {
+ c = **s;
+ ++(*s);
+ }
+ return c;
+}
+
+ucs4_t zebra_prim_w(ucs4_t **s)
+{
+ ucs4_t c;
+ ucs4_t i = 0;
+ char fmtstr[8];
+
+ yaz_log (LOG_DEBUG, "prim %.3s", (char *) *s);
+ if (**s == '\\')
+ {
(*s)++;
c = **s;
switch (c)
case 'n': c = '\n'; (*s)++; break;
case 't': c = '\t'; (*s)++; break;
case 's': c = ' '; (*s)++; break;
- case 'x': sscanf(*s, "x%2x", &i); c = i; *s += 3; break;
+ case 'x':
+ fmtstr[0] = (*s)[0];
+ fmtstr[1] = (*s)[1];
+ fmtstr[2] = (*s)[2];
+ fmtstr[3] = 0;
+ sscanf(fmtstr, "x%2x", &i);
+ c = i;
+ *s += 3; break;
case '0':
case '1':
case '2':
case '7':
case '8':
case '9':
- sscanf(*s, "%3o", &i);
+ fmtstr[0] = (*s)[0];
+ fmtstr[1] = (*s)[1];
+ fmtstr[2] = (*s)[2];
+ fmtstr[3] = 0;
+ sscanf(fmtstr, "%3o", &i);
c = i;
*s += 3;
break;
logf (LOG_DEBUG, " %3d", (unsigned char) *s);
}
+static int scan_to_utf8 (yaz_iconv_t t, ucs4_t *from, size_t inlen,
+ char *outbuf, size_t outbytesleft)
+{
+ size_t inbytesleft = inlen * sizeof(ucs4_t);
+ char *inbuf = (char*) from;
+ size_t ret;
+
+ if (t == 0)
+ *outbuf++ = *from; /* ISO-8859-1 is OK here */
+ else
+ {
+ ret = yaz_iconv (t, &inbuf, &inbytesleft, &outbuf, &outbytesleft);
+ if (ret == (size_t) (-1))
+ {
+ yaz_log (LOG_WARN|LOG_ERRNO, "bad unicode sequence");
+ return -1;
+ }
+ }
+ *outbuf = '\0';
+ return 0;
+}
-static int scan_string(char *s,
+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)
{
- unsigned char c, str[1024], begin, end, *p;
-
+ char str[1024];
+
+ ucs4_t arg[512];
+ ucs4_t *s0, *s = arg;
+ ucs4_t c, begin, end;
+ size_t i;
+
+ if (t_unicode != 0)
+ {
+ char *outbuf = (char *) arg;
+ char *inbuf = s_native;
+ size_t outbytesleft = sizeof(arg)-4;
+ size_t inbytesleft = strlen(s_native);
+ size_t ret;
+ ret = yaz_iconv(t_unicode, &inbuf, &inbytesleft,
+ &outbuf, &outbytesleft);
+ if (ret == (size_t)(-1))
+ return -1;
+ i = (outbuf - (char*) arg)/sizeof(ucs4_t);
+ }
+ else
+ {
+ for (i = 0; s_native[i]; i++)
+ arg[i] = s_native[i] & 255; /* ISO-8859-1 conversion */
+ }
+ arg[i] = 0; /* terminate */
+ if (s[0] == 0xfeff || s[0] == 0xfeff) /* skip byte Order Mark */
+ s++;
while (*s)
{
switch (*s)
{
case '{':
s++;
- begin = zebra_prim(&s);
+ begin = zebra_prim_w(&s);
if (*s != '-')
{
logf(LOG_FATAL, "Bad range in char-map");
return -1;
}
s++;
- end = zebra_prim(&s);
+ end = zebra_prim_w(&s);
if (end <= begin)
{
logf(LOG_FATAL, "Bad range in char-map");
s++;
for (c = begin; c <= end; c++)
{
- str[0] = c; str[1] = '\0';
- (*fun)((char *) str, data, num ? (*num)++ : 0);
+ if (scan_to_utf8 (t_utf8, &c, 1, str, sizeof(str)-1))
+ return -1;
+ (*fun)(str, data, num ? (*num)++ : 0);
}
break;
case '[': s++; abort(); break;
case '(':
- p = (unsigned char*) ++s;
- /* Find the end-marker, ignoring escapes */
- do
- {
- if (!(p = (unsigned char*) strchr((char*) p, ')')))
- {
- logf(LOG_FATAL, "Missing ')' in string");
- return -1;
- }
- }
- while (*(p - 1) == '\\');
- *p = 0;
- (*fun)(s, data, num ? (*num)++ : 0);
- s = (char*) p + 1;
+ ++s;
+ s0 = s;
+ while (*s != ')' || s[-1] == '\\')
+ s++;
+ *s = 0;
+ if (scan_to_utf8 (t_utf8, s0, s - s0, str, sizeof(str)-1))
+ return -1;
+ (*fun)(str, data, num ? (*num)++ : 0);
+ s++;
break;
default:
- c = zebra_prim(&s);
- str[0] = c; str[1] = '\0';
- (*fun)((char *) str, data, num ? (*num)++ : 0);
+ c = zebra_prim_w(&s);
+ if (scan_to_utf8 (t_utf8, &c, 1, str, sizeof(str)-1))
+ return -1;
+ (*fun)(str, data, num ? (*num)++ : 0);
}
}
return 0;
int errors = 0;
int argc, num = (int) *CHR_BASE, i;
NMEM nmem;
+ yaz_iconv_t t_unicode = 0;
+ yaz_iconv_t t_utf8 = 0;
+ unsigned endian = 31;
+ const char *ucs4_native = "UCS-4";
+
+ if (*(char*) &endian == 31) /* little endian? */
+ ucs4_native = "UCS-4LE";
+ t_utf8 = yaz_iconv_open ("UTF-8", ucs4_native);
logf (LOG_DEBUG, "maptab %s open", name);
- if (!(f = yaz_path_fopen_base(tabpath, name, "r", tabroot)))
+ if (!(f = yaz_fopen(tabpath, name, "r", tabroot)))
{
logf(LOG_WARN|LOG_ERRNO, "%s", name);
return 0;
logf(LOG_FATAL, "Syntax error in charmap");
++errors;
}
- if (scan_string(argv[1], fun_addentry, res, &num) < 0)
+ if (scan_string(argv[1], t_unicode, t_utf8, fun_addentry,
+ res, &num) < 0)
{
logf(LOG_FATAL, "Bad value-set specification");
++errors;
logf(LOG_FATAL, "Missing arg for uppercase directive");
++errors;
}
- if (scan_string(argv[1], fun_addentry, res, &num) < 0)
+ if (scan_string(argv[1], t_unicode, t_utf8, fun_addentry,
+ res, &num) < 0)
{
logf(LOG_FATAL, "Bad value-set specification");
++errors;
logf(LOG_FATAL, "Syntax error in charmap");
++errors;
}
- if (scan_string(argv[1], fun_addspace, res, 0) < 0)
+ if (scan_string(argv[1], t_unicode, t_utf8,
+ fun_addspace, res, 0) < 0)
{
logf(LOG_FATAL, "Bad space specification");
++errors;
}
buf.map = res;
buf.string[0] = '\0';
- if (scan_string(argv[2], fun_mkstring, &buf, 0) < 0)
+ if (scan_string(argv[2], t_unicode, t_utf8,
+ fun_mkstring, &buf, 0) < 0)
{
logf(LOG_FATAL, "Bad map target");
++errors;
}
- if (scan_string(argv[1], fun_add_map, &buf, 0) < 0)
+ if (scan_string(argv[1], t_unicode, t_utf8,
+ fun_add_map, &buf, 0) < 0)
{
logf(LOG_FATAL, "Bad map source");
++errors;
}
buf.map = res;
buf.string[0] = '\0';
- if (scan_string(argv[2], fun_mkstring, &buf, 0) < 0)
+ if (scan_string(argv[2], t_unicode, t_utf8,
+ fun_mkstring, &buf, 0) < 0)
{
logf(LOG_FATAL, "Bad qmap target");
++errors;
}
- if (scan_string(argv[1], fun_add_qmap, &buf, 0) < 0)
+ if (scan_string(argv[1], t_unicode, t_utf8,
+ fun_add_qmap, &buf, 0) < 0)
{
logf(LOG_FATAL, "Bad qmap source");
++errors;
}
}
+ else if (!yaz_matchstr(argv[0], "encoding"))
+ {
+ if (t_unicode != 0)
+ yaz_iconv_close (t_unicode);
+ t_unicode = yaz_iconv_open (ucs4_native, argv[1]);
+ }
else
{
logf(LOG_WARN, "Syntax error at '%s' in %s", line, name);
}
- fclose(f);
+ yaz_fclose(f);
if (errors)
{
chrmaptab_destroy(res);
res = 0;
}
logf (LOG_DEBUG, "maptab %s close %d errors", name, errors);
+ if (t_utf8 != 0)
+ yaz_iconv_close(t_utf8);
+ if (t_unicode != 0)
+ yaz_iconv_close(t_unicode);
return res;
}