Re: vidcontrol(1) complains about Bad magic, in base/head, amd64, sc console, r268165

On Thu, 3 Jul 2014 17:50+0200, Trond Endrestøl wrote:

> On Thu, 3 Jul 2014 12:48+0300, Aleksandr Rybalko wrote:
> 
> > On Thu, 3 Jul 2014 08:31:45 +0200 (CEST)
> > Trond Endrestøl <Trond.Endrestol_at_fagskolen.gjovik.no> wrote:
> > 
> > > On Thu, 3 Jul 2014 08:21+0200, Trond Endrestøl wrote:
> > > 
> > > > On Wed, 2 Jul 2014 17:37-0400, Ed Maste wrote:
> > > > 
> > > > > On 2 July 2014 17:09, Trond Endrestøl
> > > > > <Trond.Endrestol_at_fagskolen.gjovik.no> wrote:
> > > > > > On Wed, 2 Jul 2014 16:43-0400, Ed Maste wrote:
> > > > > >
> > > > > >> On 2 July 2014 14:51, Trond Endrestøl
> > > > > >> <Trond.Endrestol_at_fagskolen.gjovik.no> wrote:
> > > > > >> > Hi,
> > > > > >> >
> > > > > >> > Is it just me or is there something wrong with vidcontrol(1) in
> > > > > >> > base/head, amd64, sc console, r268165?
> > > > > >>
> > > > > >> Should be fixed in r268175.
> > > > > >
> > > > > > Looks good, thanks.
> > > > > 
> > > > > Thanks for the report, and sorry for the trouble.
> > > > 
> > > > No trouble at all, I follow base/head (and stable/{8,9,10}) on various 
> > > > VMs at home only to know what's ahead. ;-)
> > > > 
> > > > Since neither kbdcontrol(1) nor I mind using the old syscons keymap
> > > > file norwegian.iso.kbd, wouldn't it be nice if kbdcontrol(1), while in
> > > > vt(4) mode, would search for keymaps in /usr/share/syscons/keymaps
> > > > after searching for them in /usr/share/vt/keymaps?
> > > > 
> > > > E.g.:
> > > > 
> > > > Index: usr.sbin/kbdcontrol/kbdcontrol.c
> > > > ===================================================================
> > > > --- usr.sbin/kbdcontrol/kbdcontrol.c    (revision 268203)
> > > > +++ usr.sbin/kbdcontrol/kbdcontrol.c    (working copy)
> > > > _at__at_ -804,7 +804,7 _at__at_
> > > >         char    *postfix[] = {blank, dotkbd, NULL};
> > > > 
> > > >         if (is_vt4())
> > > > -               prefix[2] = vt_keymap_path;
> > > > +               prefix[1] = vt_keymap_path;
> > > >         cp = getenv("KEYMAP_PATH");
> > > >         if (cp != NULL)
> > > >                 asprintf(&(prefix[0]), "%s/", cp);
> > > 
> > > Or maybe this patch is even better, as it leaves one instance of blank 
> > > in the array when KEYMAP_PATH is set in the environment, at prefix[1], 
> > > and sadly add a redundant blank at prefix[2] when KEYMAP_PATH is not 
> > > set in the environment.
> > > 
> > > Index: usr.sbin/kbdcontrol/kbdcontrol.c
> > > ===================================================================
> > > --- usr.sbin/kbdcontrol/kbdcontrol.c    (revision 268203)
> > > +++ usr.sbin/kbdcontrol/kbdcontrol.c    (working copy)
> > > _at__at_ -800,7 +800,7 _at__at_
> > >         char    *name, *cp;
> > >         char    blank[] = "", keymap_path[] = KEYMAP_PATH;
> > >         char    vt_keymap_path[] = VT_KEYMAP_PATH, dotkbd[] = ".kbd";
> > > -       char    *prefix[]  = {blank, blank, keymap_path, NULL};
> > > +       char    *prefix[]  = {blank, blank, blank, keymap_path, NULL};
> > >         char    *postfix[] = {blank, dotkbd, NULL};
> > > 
> > >         if (is_vt4())
> > > 
> > > For now I could just stick to using an absolute pathname for keymap= 
> > > in /etc/rc.conf.
> > 
> > Hi Trond,
> > 
> > It is not so good idea to fallback to syscons keymaps, because vt(4)
> > works with Unicode only char codes. So fallback will make input with
> > non-English characters - unreadable.
> > 
> > Instead of that fallback you can convert keymaps you can verify by
> > follow instructions in [1], then please check it and send it to list,
> > so me or someone else will commit it. 
> > 
> > Thank you for reports!
> > 
> > 1.
> > http://raybsd.blogspot.com/2013/10/newcons-international-keyboard-input.html
> 
> I tried to follow the instructions, but I honestly don't see any 
> difference.
> 
> I downloaded uakbd2ukbd.pl.gz, decompressed the file, installed 
> converters/p5-Text-Iconv, copied /usr/share/syscons/norwegian.iso.kbd 
> to cwd, and ran:
> 
> ./uakbd2ukbd.pl norwegian.iso.kbd ISO-8859-1 > norwegian.utf8.kbd
> 
> Running diff -u norwegian.* shows absolutely no difference.
> 
> Is it pilot error on my part?

Yes, perhaps it was pilot error.

Running:

./uakbd2ukbd.pl norwegian.iso.kbd ISO-8859-15 > norwegian.utf8-15.kbd

produced differences when compared to norwegian.iso.kbd.

The keyboard checked out on all three occasions, with the original 
norwegian.iso.kbd, with the norwegian.utf8.kbd, and with the 
norwegian.utf8-15.kbd.

-- 
+-------------------------------+------------------------------------+
| Vennlig hilsen,               | Best regards,                      |
| Trond Endrestøl,              | Trond Endrestøl,                   |
| IT-ansvarlig,                 | System administrator,              |
| Fagskolen Innlandet,          | Gjøvik Technical College, Norway,  |
| tlf. mob.   952 62 567,       | Cellular...: +47 952 62 567,       |
| sentralbord 61 14 54 00.      | Switchboard: +47 61 14 54 00.      |
+-------------------------------+------------------------------------+