htmlfmt: use uhtml for character set conversion

diff --git a/sys/src/cmd/htmlfmt/dat.h b/sys/src/cmd/htmlfmt/dat.h
index f3b05605cf207771204ca43ce7d11ebbbc73c8b6..b23b2125ae0420f00af195e43d1453eab3ba90c0 100644 (file)
--- a/sys/src/cmd/htmlfmt/dat.h
+++ b/sys/src/cmd/htmlfmt/dat.h
@@ -28,12 +28,10 @@ struct URLwin
 extern char*   url;
 extern int             aflag;
 extern int             width;
-extern int             defcharset;
 
 extern char*   loadhtml(int);
 
 extern char*   readfile(char*, char*, int*);
-extern int     charset(char*);
 extern void*   emalloc(ulong);
 extern char*   estrdup(char*);
 extern char*   estrstrdup(char*, char*);

diff --git a/sys/src/cmd/htmlfmt/html.c b/sys/src/cmd/htmlfmt/html.c
index fc3ea4915ac4dfdce8472b3a76e83319a5dbf941..5f9ea86bed8c345a9ab03d896db430fb64752609 100644 (file)
--- a/sys/src/cmd/htmlfmt/html.c
+++ b/sys/src/cmd/htmlfmt/html.c
@@ -285,40 +285,13 @@ rerender(URLwin *u)
        free(t);
 }
 
-/*
- * Somewhat of a hack.  Not a full parse, just looks for strings in the beginning
- * of the document (cistrstr only looks at first somewhat bytes).
- */
-int
-charset(char *s)
-{
-       char *meta, *emeta, *charset;
-
-       if(defcharset == 0)
-               defcharset = ISO_8859_1;
-       meta = cistrstr(s, "<meta");
-       if(meta == nil)
-               return defcharset;
-       for(emeta=meta; *emeta!='>' && *emeta!='\0'; emeta++)
-               ;
-       charset = cistrstr(s, "charset=");
-       if(charset == nil)
-               return defcharset;
-       charset += 8;
-       if(*charset == '"')
-               charset++;
-       if(cistrncmp(charset, "utf-8", 5) || cistrncmp(charset, "utf8", 4))
-               return UTF_8;
-       return defcharset;
-}
-
 void
 rendertext(URLwin *u, Bytes *b)
 {
        Rune *rurl;
 
-       rurl = toStr((uchar*)u->url, strlen(u->url), ISO_8859_1);
-       u->items = parsehtml(b->b, b->n, rurl, u->type, charset((char*)b->b), &u->docinfo);
+       rurl = toStr((uchar*)u->url, strlen(u->url), UTF_8);
+       u->items = parsehtml(b->b, b->n, rurl, u->type, UTF_8, &u->docinfo);
 //     free(rurl);
 
        rerender(u);

diff --git a/sys/src/cmd/htmlfmt/main.c b/sys/src/cmd/htmlfmt/main.c
index f85bbb484b73d6e44c2affaa1c0dfb63ce7c1ff4..871c658c96ad35035a350efa6f21dd3408d1d919 100644 (file)
--- a/sys/src/cmd/htmlfmt/main.c
+++ b/sys/src/cmd/htmlfmt/main.c
@@ -8,7 +8,34 @@
 char *url = "";
 int aflag;
 int width = 70;
-int defcharset;
+char *defcharset = "latin1";
+
+int
+uhtml(int fd)
+{
+       int p[2];
+
+       if(pipe(p) < 0)
+               return fd;
+       switch(fork()){
+       case -1:
+               break;
+       case 0:
+               dup(fd, 0);
+               dup(p[1], 1);
+               close(p[1]);
+               close(p[0]);
+               execl("/bin/uhtml", "uhtml", "-c", defcharset, nil);
+               execl("/bin/cat", "cat", nil);
+               exits("exec");
+       default:
+               dup(p[0], fd);
+               break;
+       }
+       close(p[0]);
+       close(p[1]);
+       return fd;
+}
 
 void
 usage(void)
@@ -21,7 +48,7 @@ void
 main(int argc, char *argv[])
 {
        int i, fd;
-       char *p, *err, *file;
+       char *err, *file;
        char errbuf[ERRMAX];
 
        ARGBEGIN{
@@ -29,9 +56,7 @@ main(int argc, char *argv[])
                aflag++;
                break;
        case 'c':
-               p = smprint("<meta charset=\"%s\">", EARGF(usage()));
-               defcharset = charset(p);
-               free(p);
+               defcharset = EARGF(usage());
                break;
        case 'l': case 'w':
                err = EARGF(usage());
@@ -50,7 +75,7 @@ main(int argc, char *argv[])
        err = nil;
        file = "<stdin>";
        if(argc == 0)
-               err = loadhtml(0);
+               err = loadhtml(uhtml(0));
        else
                for(i=0; err==nil && i<argc; i++){
                        file = argv[i];
@@ -60,7 +85,7 @@ main(int argc, char *argv[])
                                err = errbuf;
                                break;
                        }
-                       err = loadhtml(fd);
+                       err = loadhtml(uhtml(fd));
                        close(fd);
                        if(err)
                                break;