At 10:55 AM 10/30/01 -0500, Les Schaffer wrote: You really need to talk to BillP. He is the only one qualified to fix GTK bugs. andy

On Thu, 13 Dec 2001 00:43:54 -0500, wmperry(a)gnu.org (William M. Perry) said: Just for the record ... I build a GTK build every couple of months to see if the problem I reported to you a long time ago, an extra frame appearing and a message in my terminal Gtk-CRITICAL **: file gtkwidget.c: line 3722 (gtk_widget_get_parent_window): assertion `widget->parent != NULL' failed. has gotten better. It hasn't. This is on FreeBSD-current with both gnome and GTK installed. None of my other gnome or GTK apps do this. I know that's not much help and I don't expect you to fix what you can't reproduce ... I just don't know what I can tell you that will be more helpful. -- Michael D. Harnois bilocational bivocational Pastor, Redeemer Lutheran Church Washburn, Iowa 1L, UST School of Law Minneapolis, Minnesota "Believe those who are seeking the truth. Doubt those who find it." -- Andre Gide

Michael Harnois <mharnois(a)cpinternet.com&gt; writes: Kick ass! Thank you for the repro recipe. I just sent out a patch to the list (and it is checked into CVS). Cool... I vaguely remember this but thought it had been fixed. Could someone put this on the bug list ＠atrian.net? Hrm - could you send me your config.log? -bp -- Ceterum censeo vi esse delendam

my original set of messages were related to exact same problems encountered on 21.5.3 builds. so therefore don't respond to bug reports???? just to be clear here: i am not complaining about there being bugs in the gtk code, not complaining about perhaps needing to fix them myself, not complaining about gtk being experimental. am complaining about lack of response especially since i had the intent to pitch in and help out. i'd be happy with just a pointer like, oh, go look in the blah-blah code section, and i would be off and running. nope. same problems. see my earlier reports. um, i thought thats why i posted to this list: thinking there might be like minded people wanting to sort out the gtk stuff. if noone responded ON THE LIST, why would they respond to offlist requests??? is the xemacs-beta list essentially now a closed list where only select coders are collaborating on code? if so, i will unsub so as to not disturn your efforts further. ====== anywho, anyone interested in putting heads together and ironing out some of the gtk bugs offlist??? the two problems i see: 1.) when the buffer tab is set to top, left, bottom, i get a: Gtk-CRITICAL **: file gtkwidget.c: line 3722 (gtk_widget_get_parent_window): assertion `widget->parent != NULL' failed. and a seperate frame is opended for the tab instead. interestingly, when "right" location is chosen, this does NOT happen, though the buffer tab displays incorrectly (basically, it doesnt display anything, just takes up space). 2.) redisplay after progress bar completion is hosed. see the pic at http://folks.astrian.net/godzilla/xemacs-redisplay.jpg and more complete report at: http://list-archive.xemacs.org/xemacs-beta/200109/msg00125.html les schaffer

Les> so therefore don't respond to bug reports???? Yes, exactly. It's in the priority queue at low priority, what more can we guarantee? We are trying to turn 21.4 into something that allows us to say "There's no point in 21.1 any more, 21.4 is all that and much more." It isn't there yet, and until it is, _I_ have _no_ time for GTK, and I'm not going to ask people who pay attention to my requests to work on it, either. Sorry, that's the way it has to be. Les> just to be clear here: i am not complaining about there being Les> bugs in the gtk code, not complaining about perhaps needing Les> to fix them myself, not complaining about gtk being Les> experimental. Les> am complaining about lack of response especially since i had Les> the intent to pitch in and help out. i'd be happy with just a Les> pointer like, oh, go look in the blah-blah code section, and Les> i would be off and running. The GTK port was basically implemented by one guy, Bill Perry, in a few months on contract to a company which doesn't exist anymore. If he doesn't respond, there are no other experts yet. Me, I read all bug reports. If they don't ring a bell and aren't relevant to 21.4 support, they go into my round.tuit file at nice 10 (which now has over 400 messages in it since April), and I say a quick prayer that the expert will read them and answer. But I had no pointers to give, except to Bill, which I did, and a short essay on how you might go about gathering support for your priorities. Les> um, i thought thats why i posted to this list: thinking there Les> might be like minded people wanting to sort out the gtk Les> stuff. if noone responded ON THE LIST, why would they respond Les> to offlist requests??? Well, for starters the subject was a little off-putting. Flamebait, even. Many people would prefer not to <AOL> on that. Or maybe they may not be reading the list due to general lack of progress on GTK and unwillingness to take initiative themselves. Or maybe they read it, but didn't have a clue to offer. But they might respond to personal mail if someone took that initiative to actually cull the archives for people who had posted multiple times or otherwise showed pretty strong interest. Yes, personal communication actually works. It's especially important when there's something you think needs doing but the powers that be seem uninterested. I'm well aware that the review board's, and especially my, attitude has a chilling effect on some of these projects. That's part and parcel of the responsibility of maintainership. We don't like that effect, but we can't avoid it if we're going to concentrate effort, including other people's, on getting 21.4 perfected. Eg, really, I'd rather _you_ work on the font-lock problems. If you don't have interest, by all means, work on GTK. I have to make those priorities public. That's what we call "leadership." In the face of that, to get people to respond honestly, you may have to go off-line. That's how 21.4 itself happened, as well as other things over the past year. Man, I _been_ where you are, I did what I suggested that you do, it bloody well worked. That's why I recommended it to you. Les> is the xemacs-beta list essentially now a closed list where Les> only select coders are collaborating on code? if so, i will Les> unsub so as to not disturn your efforts further. Not at all. Did anyone say "that's OT?" Of course not. I do read everything on the list. I just don't respond when all I have to post is "I don't know." There's nothing wrong with not responding if you have nothing to contribute to the thread. Please note, at this point you're no longer asking for collaboration. You're asking for people to drop whatever it is they're already doing and work on what bugs you. (That is the unavoidable inference from the phrasing of the subject.) _There's nothing wrong with that_, and xemacs-beta is _exactly_ the place to start. I wish more people would do it. Hey, you've already gotten at least one positive response. All I was doing is to point out that we have explicitly said "this is cool, we'd love it if someone _else_ would pick it up and run with it, but it is not part of what we can encourage devoting effort to supporting (_yet_)." That's what "experimental" means in practice. There are probably a dozen features like that in 21.4. -- Institute of Policy and Planning Sciences http://turnbull.sk.tsukuba.ac.jp University of Tsukuba Tennodai 1-1-1 Tsukuba 305-8573 JAPAN Don't ask how you can "do" free software business; ask what your business can "do for" free software.