rfivet
/
uemacs
mirror of https://github.com/rfivet/uemacs.git
1
0
Fork 0
Code Releases Activity

&left and &mid handling of UTF-8 encoded characters.

This commit is contained in:
Renaud 2017-05-16 12:13:12 +08:00
parent 918a310601
commit 4873021e37
2 changed files with 46 additions and 10 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

48
eval.c
View File

@ -412,8 +412,22 @@ static char *gtfun( char *fname) {
retstr = result ;
}
break ;
case UFLEFT | DYNAMIC:
sz = atoi( arg2) ;
case UFLEFT | DYNAMIC: {
int sz1, i ;
sz1 = strlen( arg1) ;
sz = 0 ;
for( i = atoi( arg2) ; i > 0 ; i -= 1) {
unicode_t c ;
int bytc ;
bytc = utf8_to_unicode( arg1, sz, sz1, &c) ;
if( bytc == 0)
break ;
else
sz += bytc ;
}
if( sz >= ressize) {
free( result) ;
result = malloc( sz + 1) ;
@ -423,6 +437,7 @@ static char *gtfun( char *fname) {
strncpy( result, arg1, sz) ;
result[ sz] = 0 ;
retstr = result ;
}
break ;
case UFRIGHT | DYNAMIC:
sz = atoi( arg2) ;
@ -434,17 +449,40 @@ static char *gtfun( char *fname) {
retstr = strcpy( result, &arg1[ strlen( arg1) - sz]) ;
break ;
case UFMID | TRINAMIC:
sz = atoi( arg3) ;
case UFMID | TRINAMIC: {
int sz1, start, i, bytc ;
unicode_t c ;
sz1 = strlen( arg1) ;
start = 0 ;
for( i = atoi( arg2) - 1 ; i > 0 ; i -= 1) {
bytc = utf8_to_unicode( arg1, start, sz1, &c) ;
if( bytc == 0)
break ;
else
start += bytc ;
}
sz = start ;
for( i = atoi( arg3) ; i > 0 ; i -= 1) {
bytc = utf8_to_unicode( arg1, sz, sz1, &c) ;
if( bytc == 0)
break ;
else
sz += bytc ;
}
sz -= start ;
if( sz >= ressize) {
free( result) ;
result = malloc( sz + 1) ;
ressize = sz + 1 ;
}
strncpy( result, &arg1[ atoi( arg2) - 1], sz) ;
strncpy( result, &arg1[ start], sz) ;
result[ sz] = 0 ;
retstr = result ;
}
break ;
case UFNOT | MONAMIC:
retstr = ltos( stol( arg1) == FALSE) ;

8
utf8.c
View File

@ -17,17 +17,15 @@
* NOTE 2! This does *not* verify things like minimality. So overlong forms
* are happily accepted and decoded, as are the various "invalid values".
*/
unsigned utf8_to_unicode(char *line, unsigned index, unsigned len, unicode_t *res)
{
unsigned utf8_to_unicode( char *line, unsigned index, unsigned len,
unicode_t *res) {
unicode_t value ;
unsigned c ;
unsigned bytes, mask, i;
assert( index < len) ;
#ifdef NDEBUG
if( index >= len)
return 0 ;
#endif
*res = c = line[ index] & 0xFFU ;
/*