Re: [PATCH] Fix console utf8 composing

["Alexander E. Patrakov" <patrakov@ums.usu.ru>]

Jan Engelhardt wrote:

>>Differences between our versions are described below.

>What do you think we should do? I have not given this fix any thought, 
>because it applied fine for long time (+/- fuzz), so I cannot comment on 
>anything in your version being better or not.

If you really don't care about characters beyond 0xffff (but see below),
please consider applying this on top of your patch:

--- linux-2.6.15-rc6/drivers/char/consolemap.c	2005-12-25 
10:02:11.000000000 +0500
+++ linux-2.6.15-rc6.my/drivers/char/consolemap.c	2005-12-25 
10:01:22.000000000 +0500
@@ -216,6 +216,7 @@
  	u16 **p1, *p2;
  	u16 *q;
  	
+	if (!p) return;
  	q = p->inverse_trans_unicode;
  	if (!q) {
  		q = p->inverse_trans_unicode =
@@ -289,6 +290,7 @@
  		p = (struct uni_pagedir *)*vc_cons[i].d->vc_uni_pagedir_loc;
  		if (p && p != q) {
  			set_inverse_transl(vc_cons[i].d, p, USER_MAP);
+			set_inverse_trans_unicode(vc_cons[i].d, p);
  			q = p;
  		}
  	}

>>My version of to_utf8() takes uint as a second argument and handles values
>>beyonf 0xffff.
>
>I doubt that there is reason to support characters beyond 0xffff.
>CJK is within 0xffff, besides that console fonts just do not have the 
>capacity to support CJK in a meaningful way.

Using console fonts is not the only way to display characters. You
forgot about jfbterm.

-- 
Alexander E. Patrakov