// Count number of utf8 characters needed to convert multi-byte characters from src
FXint FXTextCodec::mb2utflen(const FXchar* src,FXint nsrc) const {
register FXint nr,len=0;
FXwchar w;
if(src && 0<nsrc){
do{
nr=mb2wc(w,src,nsrc);
if(nr<=0) return nr;
src+=nr;
nsrc-=nr;
len+=utflen(w);
}
while(0<nsrc);
}
return len;
}
// Convert multi-byte characters from src to utf8 characters at dst
FXint FXTextCodec::mb2utf(FXchar* dst,FXint ndst,const FXchar* src,FXint nsrc) const {
register FXint nr,nw,len=0;
FXwchar w;
if(dst && src && 0<nsrc){
do{
nr=mb2wc(w,src,nsrc);
if(nr<=0) return nr;
src+=nr;
nsrc-=nr;
nw=wc2utf(dst,ndst,w);
if(nw<=0) return nw;
len+=nw;
dst+=nw;
ndst-=nw;
}
while(0<nsrc);
}
return len;
}
int
main(int argc, char **argv)
{ dtd_parser *p = NULL;
char *s;
int xml = FALSE;
int output = TRUE;
int nodefs = FALSE; /* include defaulted attributes */
setlocale(LC_CTYPE, "");
s = strchr(argv[0], '/');
program = s == NULL ? argv[0] : s + 1;
if (streq(program, "xml"))
xml = TRUE;
shift;
while (argc > 0 && argv[0][0] == '-')
{ if (streq(argv[0], "-xml"))
{ xml = TRUE;
} else if (streq(argv[0], "-s"))
{ output = FALSE;
} else if (streq(argv[0], "-nodefs"))
{ nodefs = TRUE;
} else if (streq(argv[0], "-style"))
{ style_messages = TRUE;
} else
{ usage();
}
shift;
}
if (argc > 0)
{ char *slash = strchr(argv[0], '/');
char *dot = strchr(argv[0], '.');
char *ext = dot == 0 || (slash != 0 && slash > dot) ? "." : dot;
if (strcaseeq(ext, ".dtd"))
{ char doctype[256];
strncpy(doctype, argv[0], ext - argv[0]);
doctype[ext - argv[0]] = '\0';
p = new_dtd_parser(new_dtd(mb2wc(doctype)));
load_dtd_from_file(p, mb2wc(argv[0]));
shift;
} else if (strcaseeq(ext, ".html") || strcaseeq(ext, ".htm"))
{ p = new_dtd_parser(new_dtd((ichar const *) "html"));
load_dtd_from_file(p, L"html.dtd");
} else if (xml || strcaseeq(ext, ".xml"))
{ dtd *dtd = new_dtd(no_dtd);
set_dialect_dtd(dtd, DL_XML);
p = new_dtd_parser(dtd);
} else
{ p = new_dtd_parser(new_dtd(no_dtd));
}
} else
{ p = new_dtd_parser(new_dtd(no_dtd));
}
if (nodefs)
p->flags |= SGML_PARSER_NODEFS;
switch (argc)
{ case 1:
{ set_functions(p, output);
sgml_process_file(p, mb2wc(argv[0]), 0);
free_dtd_parser(p);
if (output && nerrors == 0)
printf("C\n");
return 0;
}
case 0:
{ set_functions(p, output);
set_file_dtd_parser(p, IN_FILE, L"stdin");
set_mode_dtd_parser(p, DM_DATA);
sgml_process_stream(p, stdin, 0);
free_dtd_parser(p);
if (output && nerrors == 0 && nwarnings == 0)
printf("C\n");
return 0;
}
default:
{ usage();
return EXIT_FAILURE;
}
}
}
static void
cutfields(Cut_t* cut, Sfio_t* fdin, Sfio_t* fdout)
{
register unsigned char *sp = cut->space;
register unsigned char *cp;
register unsigned char *wp;
register int c, nfields;
register const int *lp = cut->list;
register unsigned char *copy;
register int nodelim, empty, inword=0;
register unsigned char *ep;
unsigned char *bp, *first;
int lastchar;
wchar_t w;
Sfio_t *fdtmp = 0;
long offset = 0;
unsigned char mb[8];
/* process each buffer */
while ((bp = (unsigned char*)sfreserve(fdin, SF_UNBOUND, -1)) && (c = sfvalue(fdin)) > 0)
{
cp = bp;
ep = cp + --c;
if((lastchar = cp[c]) != cut->eob)
*ep = cut->eob;
/* process each line in the buffer */
while (cp <= ep)
{
first = cp;
if (!inword)
{
nodelim = empty = 1;
copy = cp;
if (nfields = *(lp = cut->list))
copy = 0;
else
nfields = *++lp;
}
else if (copy)
copy = cp;
inword = 0;
do
{
/* skip over non-delimiter characters */
if (cut->mb)
for (;;)
{
switch (c = sp[*(unsigned char*)cp++])
{
case 0:
continue;
case SP_WIDE:
wp = --cp;
while ((c = mb2wc(w, cp, ep - cp)) <= 0)
{
/* mb char possibly spanning buffer boundary -- fun stuff */
if ((ep - cp) < mbmax())
{
int i;
int j;
int k;
if (lastchar != cut->eob)
{
*ep = lastchar;
if ((c = mb2wc(w, cp, ep - cp)) > 0)
break;
}
if (copy)
{
empty = 0;
if ((c = cp - copy) > 0 && sfwrite(fdout, (char*)copy, c) < 0)
goto failed;
}
for (i = 0; i <= (ep - cp); i++)
mb[i] = cp[i];
if (!(bp = (unsigned char*)sfreserve(fdin, SF_UNBOUND, -1)) || (c = sfvalue(fdin)) <= 0)
goto failed;
cp = bp;
ep = cp + --c;
if ((lastchar = cp[c]) != cut->eob)
*ep = cut->eob;
j = i;
k = 0;
while (j < mbmax())
mb[j++] = cp[k++];
if ((c = mb2wc(w, (char*)mb, j)) <= 0)
{
c = i;
w = 0;
}
first = bp = cp += c - i;
if (copy)
{
copy = bp;
if (w == cut->ldelim.chr)
lastchar = cut->ldelim.chr;
else if (w != cut->wdelim.chr)
{
empty = 0;
if (sfwrite(fdout, (char*)mb, c) < 0)
goto failed;
}
}
c = 0;
}
else
{
w = *cp;
c = 1;
}
break;
}
cp += c;
c = w;
if (c == cut->wdelim.chr)
{
c = SP_WORD;
break;
}
if (c == cut->ldelim.chr)
{
c = SP_LINE;
break;
}
continue;
default:
wp = cp - 1;
break;
}
break;
}
else
{
while (!(c = sp[*cp++]));
wp = cp - 1;
}
/* check for end-of-line */
if (c == SP_LINE)
{
if (cp <= ep)
break;
if (lastchar == cut->ldelim.chr)
break;
/* restore cut->last character */
if (lastchar != cut->eob)
*ep = lastchar;
inword++;
if (!sp[lastchar])
break;
}
nodelim = 0;
if (--nfields > 0)
continue;
nfields = *++lp;
if (copy)
{
empty = 0;
if ((c = wp - copy) > 0 && sfwrite(fdout, (char*)copy, c) < 0)
goto failed;
copy = 0;
}
else
/* set to delimiter unless the first field */
copy = empty ? cp : wp;
} while (!inword);
if (!inword)
{
if (!copy)
{
if (nodelim)
{
if (!cut->sflag)
{
if (offset)
{
sfseek(fdtmp,(Sfoff_t)0,SEEK_SET);
sfmove(fdtmp,fdout,offset,-1);
}
copy = first;
}
}
else
sfputc(fdout,'\n');
}
if (offset)
sfseek(fdtmp,offset=0,SEEK_SET);
}
if (copy && (c=cp-copy)>0 && (!nodelim || !cut->sflag) && sfwrite(fdout,(char*)copy,c)< 0)
goto failed;
}
/* see whether to save in tmp file */
if(inword && nodelim && !cut->sflag && (c=cp-first)>0)
{
/* copy line to tmpfile in case no fields */
if(!fdtmp)
fdtmp = sftmp(BLOCK);
sfwrite(fdtmp,(char*)first,c);
offset +=c;
}
}
failed:
if(fdtmp)
sfclose(fdtmp);
}
String * LibraryItem::getContainer()
{
String *result = new String(mb2wc(containerPath.c_str()));
return result;
}
String * LibraryItem::getTitle()
{
String *result = new String(mb2wc(title.c_str()));
return result;
}
String * LibraryItem::getFilename()
{
String *result = new String(mb2wc(localPath.c_str()));
return result;
}
String * LibraryItem::getAlbum()
{
String *result = new String(mb2wc(album.c_str()));
return result;
}
String * LibraryItem::getArtist()
{
String *result = new String(mb2wc(artist.c_str()));
return result;
}
int wc_count(Wc_t *wp, Sfio_t *fd, const char* file)
{
register char* type = wp->type;
register unsigned char* cp;
register Sfoff_t nbytes;
register Sfoff_t nchars;
register Sfoff_t nwords;
register Sfoff_t nlines;
register Sfoff_t eline = -1;
register Sfoff_t longest = 0;
register ssize_t c;
register unsigned char* endbuff;
register int lasttype = WC_SP;
unsigned int lastchar;
ssize_t n;
ssize_t o;
unsigned char* buff;
wchar_t x;
unsigned char side[32];
sfset(fd,SF_WRITE,1);
nlines = nwords = nchars = nbytes = 0;
wp->longest = 0;
if (wp->mb < 0 && (wp->mode & (WC_MBYTE|WC_WORDS)))
{
cp = buff = endbuff = 0;
for (;;)
{
if (cp >= endbuff || (n = mb2wc(x, cp, endbuff-cp)) < 0)
{
if ((o = endbuff-cp) < sizeof(side))
{
if (buff)
{
if (o)
memcpy(side, cp, o);
mbinit();
}
else
o = 0;
cp = side + o;
if (!(buff = (unsigned char*)sfreserve(fd, SF_UNBOUND, 0)) || (n = sfvalue(fd)) <= 0)
{
if ((nchars - longest) > wp->longest)
wp->longest = nchars - longest;
break;
}
nbytes += n;
if ((c = sizeof(side) - o) > n)
c = n;
if (c)
memcpy(cp, buff, c);
endbuff = buff + n;
cp = side;
x = mbchar(cp);
if ((cp-side) < o)
{
cp = buff;
nchars += (cp-side) - 1;
}
else
cp = buff + (cp-side) - o;
}
else
{
cp++;
x = -1;
}
if (x == -1 && eline != nlines && !(wp->mode & WC_QUIET))
eline = invalid(file, nlines);
}
else
cp += n ? n : 1;
if (x == '\n')
{
if ((nchars - longest) > wp->longest)
wp->longest = nchars - longest;
longest = nchars + 1;
nlines++;
lasttype = 1;
}
else if (iswspace(x))
lasttype = 1;
else if (lasttype)
{
lasttype = 0;
nwords++;
}
nchars++;
}
if (!(wp->mode & WC_MBYTE))
nchars = nbytes;
}
else if (!wp->mb && !(wp->mode & WC_LONGEST) || wp->mb > 0 && !(wp->mode & (WC_MBYTE|WC_WORDS|WC_LONGEST)))
{
if (!(wp->mode & (WC_MBYTE|WC_WORDS|WC_LONGEST)))
{
while ((cp = (unsigned char*)sfreserve(fd, SF_UNBOUND, 0)) && (c = sfvalue(fd)) > 0)
{
nchars += c;
endbuff = cp + c;
if (*--endbuff == '\n')
nlines++;
else
*endbuff = '\n';
for (;;)
if (*cp++ == '\n')
{
if (cp > endbuff)
break;
nlines++;
}
}
}
else
{
while ((cp = buff = (unsigned char*)sfreserve(fd, SF_UNBOUND, 0)) && (c = sfvalue(fd)) > 0)
{
nchars += c;
/* check to see whether first character terminates word */
if (c==1)
{
if (eol(lasttype))
nlines++;
if ((c = type[*cp]) && !lasttype)
nwords++;
lasttype = c;
continue;
}
if (!lasttype && type[*cp])
nwords++;
lastchar = cp[--c];
*(endbuff = cp+c) = '\n';
c = lasttype;
/* process each buffer */
for (;;)
{
/* process spaces and new-lines */
do
{
if (eol(c))
for (;;)
{
/* check for end of buffer */
if (cp > endbuff)
goto beob;
nlines++;
if (*cp != '\n')
break;
cp++;
}
} while (c = type[*cp++]);
/* skip over word characters */
while (!(c = type[*cp++]));
nwords++;
}
beob:
if ((cp -= 2) >= buff)
c = type[*cp];
else
c = lasttype;
lasttype = type[lastchar];
/* see if was in word */
if (!c && !lasttype)
nwords--;
}
if (eol(lasttype))
nlines++;
else if (!lasttype)
nwords++;
}
}
else
{
int lineoff=0;
int skip=0;
int adjust=0;
int state=0;
int oldc;
int xspace;
int wasspace = 1;
unsigned char* start;
lastchar = 0;
start = (endbuff = side) + 1;
xspace = iswspace(0xa0) || iswspace(0x85);
while ((cp = buff = (unsigned char*)sfreserve(fd, SF_UNBOUND, 0)) && (c = sfvalue(fd)) > 0)
{
nbytes += c;
nchars += c;
start = cp-lineoff;
/* check to see whether first character terminates word */
if(c==1)
{
if(eol(lasttype))
nlines++;
if((c = type[*cp]) && !lasttype)
nwords++;
lasttype = c;
endbuff = start;
continue;
}
lastchar = cp[--c];
endbuff = cp+c;
cp[c] = '\n';
if(mbc(lasttype))
{
c = lasttype;
goto mbyte;
}
if(!lasttype && spc(type[*cp]))
nwords++;
c = lasttype;
/* process each buffer */
for (;;)
{
/* process spaces and new-lines */
spaces:
do
{
if (eol(c))
{
/* check for end of buffer */
if (cp > endbuff)
goto eob;
if(wp->mode&WC_LONGEST)
{
if((cp-start)-adjust > longest)
longest = (cp-start)-adjust-1;
start = cp;
}
nlines++;
nchars -= adjust;
adjust = 0;
}
} while (spc(c = type[*cp++]));
wasspace=1;
if(mbc(c))
{
mbyte:
do
{
if(c&WC_ERR)
goto err;
if(skip && (c&7))
break;
if(!skip)
{
if(!(c&7))
{
skip=1;
break;
}
skip = (c&7);
adjust += skip;
state = 0;
if(skip==2 && (cp[-1]&0xc)==0 && (state=(cp[-1]&0x3)))
oldc = *cp;
else if(xspace && cp[-1]==0xc2)
{
state = 8;
oldc = *cp;
}
}
else
{
skip--;
if(state && (state=chkstate(state,oldc)))
{
if(state==10)
{
if(!wasspace)
nwords++;
wasspace = 1;
state=0;
goto spaces;
}
oldc = *cp;
}
}
} while (mbc(c = type[*cp++]));
wasspace = 0;
if(skip)
{
if(eol(c) && (cp > endbuff))
goto eob;
err:
skip = 0;
state = 0;
if(eline!=nlines && !(wp->mode & WC_QUIET))
eline = invalid(file, nlines);
while(mbc(c) && ((c|WC_ERR) || (c&7)==0))
c=type[*cp++];
if(eol(c) && (cp > endbuff))
{
c = WC_MB|WC_ERR;
goto eob;
}
if(mbc(c))
goto mbyte;
else if(c&WC_SP)
goto spaces;
}
if(spc(c))
{
nwords++;
continue;
}
}
/* skip over word characters */
while(!(c = type[*cp++]));
if(mbc(c))
goto mbyte;
nwords++;
}
eob:
lineoff = cp-start;
if((cp -= 2) >= buff)
c = type[*cp];
else
c = lasttype;
lasttype = type[lastchar];
/* see if was in word */
if(!c && !lasttype)
nwords--;
}
if ((wp->mode&WC_LONGEST) && ((endbuff + 1 - start) - adjust - (lastchar == '\n')) > longest)
longest = (endbuff + 1 - start) - adjust - (lastchar == '\n');
wp->longest = longest;
if (eol(lasttype))
nlines++;
else if (!lasttype)
nwords++;
if (wp->mode & WC_MBYTE)
nchars -= adjust;
else
nchars = nbytes;
}
wp->chars = nchars;
wp->words = nwords;
wp->lines = nlines;
return 0;
}
const void* __declspec(dllexport) strncvt (const void* str, int strLen, unsigned int cpFrom, unsigned int cpTo, int* len) {
if (!str) return NULL;
if (strLen<0) strLen = (cpFrom>=1200&&cpFrom<=1203? wcslen(str) : strlen(str));
if (cpFrom==cpTo) {
if (len) *len = strLen;
return str;
}
const void* from = NULL, *result = NULL;
BOOL alloced = FALSE;
switch(cpFrom){
case CP_UTF16_LE:
from=str;
if (len) *len = strLen;
break;
case CP_UTF16_LE_BOM:
from = ((const wchar_t*)str) +1;
strLen--;
if (len) *len = strLen;
break;
case CP_UTF16_BE_BOM:
from = ((const wchar_t*)str) +1;
strLen--;
case CP_UTF16_BE :
if (!from) from = str;
from = wcsdup(from);
unicodeSwitchEndianess(from, 1+wcslen(from));
if (len) *len = strLen;
alloced=TRUE;
break;
case CP_UTF8_BOM :
from = ((const char*)str) +3;
cpFrom = CP_UTF8;
strLen -= 3;
default :
if (!from) from=str;
from = mb2wc(from, strLen, cpFrom, &strLen);
if (len) *len = strLen;
alloced = TRUE;
break;
}
if (!from) return NULL;
switch(cpTo){
case CP_UTF16_LE:
result = from;
((wchar_t*)result)[strLen]=0;
alloced = FALSE;
break;
case CP_UTF16_LE_BOM :
if (alloced) {
result = from;
memmove( ((wchar_t*)result)+1, ((wchar_t*)result), sizeof(wchar_t)*(1+strLen));
}
else {
result = malloc(sizeof(wchar_t) * (4+strLen));
memcpy( ((wchar_t*)result)+1, from, sizeof(wchar_t)*(1+strLen));
}
(*((wchar_t*)result)) = 0xFEFF;
((wchar_t*)result)[++strLen] = 0;
if (len) *len = strLen;
alloced = FALSE;
break;
case CP_UTF16_BE :
result = (alloced? from : wcsdup(from));
unicodeSwitchEndianess(result, 3+wcslen(result));
alloced = FALSE;
((wchar_t*)result)[strLen] = 0;
if (len) *len = strLen;
break;
case CP_UTF16_BE_BOM :
if (alloced) {
result = from;
memmove( ((wchar_t*)result)+1, ((wchar_t*)result), sizeof(wchar_t)*(1+strLen));
}
else {
result = malloc(sizeof(wchar_t) * (4+strLen));
memcpy(((wchar_t*)result)+1, from, sizeof(wchar_t)*(1+strLen));
}
(*(((wchar_t*)result))) = 0xFEFF;
((wchar_t*)result)[++strLen] = 0;
unicodeSwitchEndianess(result, 3+strLen);
if (len) *len = strLen;
alloced = FALSE;
break;
case CP_UTF8_BOM :
result = wc2mb(from, strLen, CP_UTF8, &strLen);
memmove( ((const char*)result)+3, ((const char*)result), 1+strLen);
memcpy(result, "\xEF\xBB\xBF", 3);
if (len) (*len) = strLen+3;
break;
default :
result = wc2mb(from, strLen, cpTo, &strLen);
if (len) *len = strLen;
break;
}
if (alloced) free(from);
return result;
}
const wchar_t* fromEncoding (const char* str, int encoding) {
if (!str) return NULL;
if (strptr) free(strptr);
strptr = mb2wc(str, encoding);
return strptr;
}
