xfnw/plan9fox
1
0
Fork 0
Code Issues Pull requests Projects Releases Wiki Activity

uhtml: check if document is valid utf8 even with charset specified

Browse source 
often, documents specify charsets but are really utf-8 encoded.
we now try to decode as utf-8 and only if that fails assume
the charset specified in the document.
This commit is contained in:
cinap_lenrek 2015-05-28 16:37:55 +02:00
parent e601e1605b
commit 3d1e12363d
1 changed files with 59 additions and 59 deletions
Show stats Download patch file Download diff file Expand all files Collapse all files

30
sys/src/cmd/uhtml.c
View file

@ -49,7 +49,7 @@ void
main(int argc, char *argv[])
{
int n, q, pfd[2], pflag = 0;
char *arg[4], *s, *e, *p, *g, *a, t;
char *arg[4], *s, *g, *e, *p, *a, t;
Rune r;
ARGBEGIN {
@ -69,34 +69,34 @@ main(int argc, char *argv[])
sysfatal("open: %r");
}
nbuf = 0;
p = buf;
g = buf;
while(nbuf < sizeof(buf)-1){
if((n = read(0, buf + nbuf, sizeof(buf)-1-nbuf)) <= 0)
break;
nbuf += n;
buf[nbuf] = 0;
if(nbuf == n){
if(memcmp(p, "\xEF\xBB\xBF", 3)==0){
}
p = buf;
if(nbuf >= 3 && memcmp(p, "\xEF\xBB\xBF", 3)==0){
p += 3;
nbuf -= 3;
cset = "utf";
goto Found;
}
if(memcmp(p, "\xFE\xFF", 2) == 0){
if(nbuf >= 2 && memcmp(p, "\xFE\xFF", 2) == 0){
p += 2;
nbuf -= 2;
cset = "unicode-be";
goto Found;
}
if(memcmp(p, "\xFF\xFE", 2) == 0){
if(nbuf >= 2 && memcmp(p, "\xFF\xFE", 2) == 0){
p += 2;
nbuf -= 2;
cset = "unicode-le";
goto Found;
}
}
s = g;
s = p;
do {
if((s = strchr(s, '<')) == nil)
break;
@ -122,26 +122,26 @@ main(int argc, char *argv[])
}
t = *e;
*e = 0;
if((a = attr(g, "encoding")) || (a = attr(g, "charset"))){
*e = t;
if((a = attr(g, "encoding")) != nil || (a = attr(g, "charset")) != nil){
cset = a;
goto Found;
*e = t;
break;
}
*e = t;
s = ++e;
} while(t);
}
if(cset)
goto Found;
s = p;
while(s+UTFmax < p+nbuf){
s += chartorune(&r, s);
if(r == Runeerror){
if(cset == nil)
cset = "latin1";
goto Found;
}
}
cset = "utf";
Found:
if(pflag){
print("%s\n", cset);