Sebastian Harl <[EMAIL PROTECTED]> wrote Mon, Jun 04, 2007:
> Hi,
>
> On Mon, Jun 04, 2007 at 11:57:12AM +0200, Xavier Bestel wrote:
> > On Mon, 2007-06-04 at 08:22 +0200, Jonas Fonseca wrote:
> > > Note [...] that a newer tig version is
> > > available (0.7) and that I am not able to reproduce this behavior on my
> > > system with a newer version.
> > >
> > > Would it be possible if you could check if newer versions has this
> > > problem?
> >
> > Debian has only 0.5 available. I'll try to have a look at compiling it
> > myself, if time permits.
This should not be necessary.
> I already talked about this issue with Jonas a while ago - I think it is fixed
> in 0.6 already even though neither of us can explain why ;-)
Yes, I just now stumbled upon our past discussion. :)
Trying to find the explaination for the cause of the bug I found that
the fix seems to be commit cf4d82e60e23c1a6a457c2df0b9b4c8afef40bef:
Hardwire ERR to mean REQ_NONE in the main loop
It changes handling of ERR returned by wgetch() for non-blocking reads.
Before this patch the "request" variable in the main loop would remain
unchanged, after it is set to the no-op request identifier REQ_NONE.
> 0.7 will be available in unstable in a short while.
Cool!
> PS: Please no not send replies to any bug reports to [EMAIL PROTECTED]
> Replying to <bug number>@bugs.debian.org is all you need.
Sorry, I just hit the reply-all ('g') key. Perhaps I need to
tell Mutt that this is a "mailing list".
--
Jonas Fonseca
--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
