Re: gtk_text_buffer_insert() problem character code transfer

[tomas tuxteam de]

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

On Fri, Feb 02, 2007 at 05:03:20PM +0800, æå wrote:
> hi everyone,
>
> I have a GtkTextView to display the data in the buffer  *buffer.
>
> The buffer i received from a socket  consists of English and Chinese
> Characters.
>
>  How to display this Characters one by one using the API :
> gtk_text_buffer_insert()?

I think I don't understand quite what you mean with "one by one".
Typically you'd add as much as you got from the socket -- like so:

  /* at the beginning */
  GtkTextBuffer *tb = gtk_text_buffer_new(NULL); /* or create a tag table */


  /* everytime you get a load full of chars,
     typically in a callback function */
  int len;
  char buf[1024];
  if( (len=read(socket, buf, sizeof(buf))) > 0 ) {
    GtkTextIter it;
    gtk_text_buffer_get_end_iter(tb, &it); /* position it at end of text */
    gtk_text_buffer_insert(tb, &it, buf, len); /* insert at end */
  }

CAVEATS:
 * this is totally untested code.
 * you'll probably want a more intelligent function than read() to fetch
   a bufferfull: the input will come in utf-8 (I assume), and read()
   might brutally split multi-byte sequences; I'd expect
   gtk_text_buffer_insert() to get upset about this: not what you want.
   See g_io_channel_read_chars() & friends for a more robust soultion.
   You'll find there also functions to use different encodings should
   your input be other than utf-8.

HTH
- -- tomÃs
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.1 (GNU/Linux)

iD8DBQFFxHpuBcgs9XrR2kYRAlVNAJ98ALAMDC8XfPxEosj5yjsj/njggwCfUV6d
k8/IkimivncOo3o5H8uoDik=
=t9J7
-----END PGP SIGNATURE-----