My guess, from studying this bug a couple of times, is that both of your patches together triggered some latent bug. Not an uncommon way for difficult bugs to appear, since this means that the bug may be anywhere in the code and not just in the lines changed.
Basically, any key you map in the workspace context becomes unmapped in every other context.
3.6 + my patches work, but none of the 3.7 alphas I've tried do, including alpha5.
I do not know why, or what patch caused this, but it seems that it's an interaction between my workspace patch (#1) and later changes. I do
1) Someone else takes ownership of the bug, finds it and fixes it. 2) We start from 3.6, apply patches until it manifests, and then back out the patch that causes the problem until it can be resolved. 3) Back out the workspace patch.
4) For the time being, consider this a "Known bug" and accept that this (excuse me for saying so) minor patch does not work exactly as expected. As far as I know, this bug does not cause ctwm to crash - does it?
Yes, I've looked at alpha5, it's still broken.
In the sense that the bug is still there.
I did make an effort myself, both to find out what was wrong myself,
Keep at it.
fairly serious bug at the start of 3.7 that nobody wants to take
How serious is this? How many ctwm users use the workspace manager context for keymapping?
Don't get me wrong here, I understand that this feature is important to you. However, you have yourself stated that 3.6 + this patch works fine. So as I see it, either you use that combination or you fix the bug.
patches for, my suggestion is that we effectively back out that entire blob. Dan, I don't see why you think this is such an unreasonable thing to suggest, given the circumstances, nor why you feel the need to be abusive about it.
I assume there was a substancial amount of work involved in getting from 3.6 to 3.7a4. I _know_ the amount of work needed to get from 3.7a4 to 3.7a5. Your suggestion means that we throw all of this away because we have a known bug in the project. Apart from this I have gotten only positive feedback on a5. Most people who tried it said nothing at all - and being a software developer I consider this positive feedback.
Disrespecting other people's work is about the most abusive thing I know. Which is why I "feel the need" to bite back. Actually, my first intuition told me to just unsubscribe from this list and leave ctwm to bleed.
Now, can we please stop wasting even more time on this matter. My counter-suggestion is that we move current to 3.7b1. As far as I know we
have pretty much taken care of any other a5 issues (such as Rudy's bug). We should of course document the bug discussed above in the appropriate places (manpage and readme?).
I'm sure we'll get a lot more issues reported if we try to move to a stable release in a near future, many are prone not to use neither alpha nor beta versions. And this time we should propably issue a few bugfix releases (3.7.1 etc) instead of leaving bunches of patches dangling in void - I suspect this is what really created this mess.
Richard, you said you had a few more tickets you wanted to look at before a beta-release? I think I found some minor problem with my late f.changesize patch, but I think we can live with that too in a beta (I'll report this when I have the time and possiblity to reproduce it).
Any others against moving to 3.7b1?
Regards,
//\\ /"\ dL - Dan Lilliehorn \ / ASCII ribbon campaign http://www.dL.nu/ X against HTML email [EMAIL PROTECTED] / \
