Re: [screen-devel] [PATCH] Make RC_OTHER always choose some other

[Micah Cowan]

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

Sadrul Habib Chowdhury wrote:
> * Jeroen Roovers had this to say on [18 Sep 2008, 18:52:11 +0200]:
> [snip]
>> Sometimes when you do RC_OTHER, you get presented with "This IS window (n)."
>> That's rather stupid as you didn't even ask for window n in the first place,
>> but simply _some_ other window. -JeR
>>
>> --- process.c.orig   2008-07-23 08:36:05.000000000 +0200
>> +++ process.c        2008-09-18 18:21:19.000000000 +0200
>> @@ -1731,7 +1731,7 @@
>>        /* FALLTHROUGH */
>>      case RC_OTHER:
>>        if (MoreWindows())
>> -    SwitchWindow(display && D_other ? D_other->w_number : NextWindow());
>> +    SwitchWindow(display && (D_other != fore) ? D_other->w_number : 
>> NextWindow());
>>        break;
>>      case RC_META:
>>        if (user->u_Esc == -1)
> 
> The patch, as it is, looks good. Thanks!
> 
> But, perhaps a more appropriate fix would be to make sure that
> display->d_other is updated appropriately? Do you have a screenrc file
> that can be used to easily reproduce the bug?

I know I've run across it a few times. Don't have a reproduction example
at the moment.

Also, I'm wondering whether it's still important to check whether
D_other is null, as in the - line. I applied the patch and couldn't seem
to get it to crash, but I'd have to look deeper at the code to know.
OTOH, keeping the check against null isn't going to _hurt_ anything...

- --
Micah J. Cowan
Programmer, musician, typesetting enthusiast, gamer.
GNU Maintainer: wget, screen, teseq
http://micah.cowan.name/
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.6 (GNU/Linux)
Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org

iD8DBQFI01jL7M8hyUobTrERAhxzAJ94rlG1KsDNY3pB+zMVjKplvO4ZEACeIRw8
FLgNNI+z/qlILZoSFag+jPM=
=I6Hn
-----END PGP SIGNATURE-----