[kde-testing] and [kde-crazy] KDE4 overlays thread (part 3)

[StringCheesian]

Also, is there anyplace that shows the release schedule for the 4.2.x series? Inquiring minds want to know.

So far the 4.2.x schedule is missing dates for 4.2.1, 4.2.2, etc, but here it is anyway:
http://techbase.kde.org/Schedules/KDE4/ ... e_Schedule

[Shamus397]

Yeah, I found that too, hence the question. I guess there is no roadmap at the moment.

[KejPi]

Hi,
I have some problems with gtk-engines-qt on amd64. Currently I have 1.1-r1 but it is the same for 1.1. When I choose "use my KDE style for gtk application" the GTK applications are crashing with floating point exception. Does anybody else have the sam problem?
Thanks for any help!

[KejPi]

Hi,
I have some problems with gtk-engines-qt on amd64. Currently I have 1.1-r1 but it is the same for 1.1. When I choose "use my KDE style for gtk application" the GTK applications are crashing with floating point exception. Does anybody else have the sam problem?
Thanks for any help!

OK, I answer myself. It seems to be known bug in 1.1. It is fixed in SVN so installing live version helped.

[hunky]

4.2.61 showed up tonight in kde-testing, with the 4.3 sets. Haven't looked to see if any source is out there yet.

[coolsnowmen]

4.2.61 showed up tonight in kde-testing, with the 4.3 sets. Haven't looked to see if any source is out there yet.

It is, it looks like it is based on: svn-917530

[tampakrap]

Live users may know that those early versions of KDE 4.3 are pretty unstable and b0rked
We may not add them until .65 (but haven't decided yet)

(Hmmm, i may also add an official KDE 4.2 thread)

[tampakrap]

I suck, reavertm already added them
Have fun

[coolsnowmen]

Live users may know that those early versions of KDE 4.3 are pretty unstable and b0rked

Thanks, I was just about to ask that. I guess I'll be sticking to 4.2 for now. It's been a while since I didn't recompile kde every other week-- my computer is going to need to find something else to do on lonely sunday nights.

[gaelic]

Live users may know that those early versions of KDE 4.3 are pretty unstable and b0rked

Thanks, I was just about to ask that. I guess I'll be sticking to 4.2 for now. It's been a while since I didn't recompile kde every other week-- my computer is going to need to find something else to do on lonely sunday nights.

same problem here.

i added the x11 overlay this week ... xorg-server-9999 rulez

[aTan]

Hi, here's some error

>>> Unpacking source...
>>> Unpacking 92149-plasma-weatherforecast-0.7-kde-4.2.tar.bz2 to /var/tmp/portage/portage/kde-plasmoids/plasma-weatherforecast-0.7/work
mv: cannot stat `92149-plasma-weatherforecast-0.7-kde-4.2': No such file or directory
*
* ERROR: kde-plasmoids/plasma-weatherforecast-0.7 failed.
* Call stack:
* ebuild.sh, line 49: Called src_unpack
* environment, line 3161: Called kde4-base_src_unpack
* environment, line 2398: Called die
* The specific snippet of code:
* mv "${topleveldir}" "${P}" || die "Died while moving "${topleveldir}" to "${P}"";
* The die message:
* Died while moving "92149-plasma-weatherforecast-0.7-kde-4.2" to "plasma-weatherforecast-0.7"
*
* If you need support, post the topmost build error, and the call stack if relevant.
* A complete build log is located at '/var/log/portage/kde-plasmoids:plasma-weatherforecast-0.7:20090205-161135.log'.
* The ebuild environment file is located at '/var/tmp/portage/portage/kde-plasmoids/plasma-weatherforecast-0.7/temp/environment'.
* This ebuild is from an overlay named 'kde': '/usr/local/portage/layman/kde-testing/'

[reavertm]

Sync overlay and try again - fixed

Btw, it's rather advised to use 4.2.0 from portage (instead of unstable snapshots) and help finding bugs so that it may be marked stable in some future.
As well, please search bugzilla before reporting any new bugs - we already have over 500 KDE bugs to fix (some of them are invalid or related to old versions) and we don't really want more duplicates :)

[hunky]

Btw, it's rather advised to use 4.2.0 from portage (instead of unstable snapshots) and help finding bugs so that it may be marked stable in some future.

That's a good point. I may do that (I'm using kde-testing 4.2 right now and not particularly jazzed about redoing it all) when I finally dump 3.5.10. I'll dump that when a few remaining apps I use are ported, like k3b. Then, if I understand it correctly, I can run the portage version of 4.2 and with the kdeprefix use flag, still play around with 4.3 builds off kde-testing.?. Thanks for your work.. /jd

[titoucha]

I tested the version 4.2.61 and a lot of packages does not compile

Code: Select all

 * The following packages have failed to build or install:
 *
 *      ('ebuild', '/', 'kde-base/akonadi-4.2.61', 'merge')
 *      ('ebuild', '/', 'kde-base/kdepim-kresources-4.2.61', 'merge')
 *      ('ebuild', '/', 'kde-base/ktimetracker-4.2.61', 'merge')
 *      ('ebuild', '/', 'kde-base/kmail-4.2.61', 'merge')
 *      ('ebuild', '/', 'kde-base/korganizer-4.2.61', 'merge')
 *      ('ebuild', '/', 'kde-base/kontact-specialdates-4.2.61', 'merge')

[reavertm]

4.2.61 has been masked for a reason you know? :) Everything that is masked - is masked for a reason. Please do not report bugs related to masked packages.

[hunky]

4.2.61 has been masked for a reason you know? Everything that is masked - is masked for a reason. Please do not report bugs related to masked packages.

I'm a bit confused here. Did the person you are admonishing actually report a bug or was he just posting in the forum here? Because I can understand your quote above if he was reporting bugs. But if he was just posting his errors to this forum, then what is the point of this whole thread? I won't elaborate since perhaps he posted a bug - I hope that was the case. Otherwise...

[asturm]

Uploading a new tagged KDE release, especially the first of its kind, takes its time, hardmasked usually means "do not touch until we've sorted it out".

[hunky]

haven't we been unmasking all along in this thread? If other souls are interested in trying out the "unsupported" masked versions isn't it reasonable to have a thread discussing possible pitfalls, as the poster who was admonished did? Is this suddenly now off base?

[edit] I won't be surprised to learn I've totally missed something here - it happens all the time

[asturm]

Yes.

hardmasked means: Things aren't ready for building, ebuild upload probably hasn't finished yet, do not touch unless you know what you're doing or want to help out with meaningful backtraces and error hunting.

[hunky]

sorry for being dense here - hope you don't mind helping me out with understanding this. I'm running ~x86 on my system. I have had to unmask some of the steps up to 4.2. If I remember correctly, I had to unmask 4.96. There has been much discussion along the way of this whole kde-testing/crazy thread of people doing that, what problems they encounter, people helping people solve them.

This is in a forum called unsupported software. I'm not even sure why developers are reading it, unless to learn what problems may be cropping up. [edit] (But I'm happy they do as they have had valuable input) [/edit] But they have no obligation to help, and therefore no stake in what people are doing with "unsupported" software.

Now, we can no longer support each other if we happen to want to stick our toes into the masked pond. Even though it should mean nothing to the developers, since they should be concentrating on "supported" software problems. [edit] not only supported software problems of course - but developing or whatever it is you do. [/edit]

Am I completely missing your point?

[asturm]

Am I completely missing your point?

I think you do. There is masking by keywords, and there is (hard-)masking by package.mask. We all had to unmask KDE-4.2 during its beta state by package.keywords files in order to get it, and it was more or less unstable but for most of the people, it worked. However whenever the ebuilds were hardmasked, that meant a) some ebuilds were still missing, b) sources were still missing, or c) some other reason which prevented that version of KDE to build no matter how hard you tried. Every time though you get some people having plainly unmasked just about everything to live on the absolute KDE edge, syncing in the midst of an ebuild upload and then wondering why that stuff won't work.

I know that at some point before the 4.2.0 release 4.2.96 got hardmasked again despite it being fully functional and only ~arch keyworded before, but I don't know why (perhaps to force 4.2.0).

[hunky]

As I stated above, I'm running a ~x86 system so I'm not concerned with keyword unmasking and understand hardmasking, at least in the context here.

by the way - thanks for responding!

[edit] I guess I'm just curious why people are now being discouraged from discussing hard masked packages in an unsupported software forum.

[loftwyr]

They're not stopping you from talking about hardmasked packages, genstorm isn't correct. This is an experimental overlay and the live stuff is being discussed. If it's hardmasked, you get to work harder at making it work before you post problems.

The only concern is if you get a compilation failure, wait a day and try again. You're allowed to discuss it, just make sure you're willing to help find the solution, not just asking others for their help.

[hunky]

I fail to see what titoucha did wrong. Before him, someone said they heard the 4.2.61 builds were borked. That was enough for me to not want to touch them. Titoucha, for whatever reason, decided to give them a go - maybe out of curiousity. All he did was confirm that he couldn't compile some packages. That to me was information I could use in deciding if I, or someone more capable, wanted to play with them. It reinforced to me not to touch them. It was useful in that regard. Didn't sound like he was begging for a fix - just stating his experience.

But I'm obviously not framing my viewpoint well enough for others to grasp what I'm trying to say so I'll give it a rest. Thanks for your input. /jd

[titoucha]

I just gave my opinion, I expect no help, just wait the next release.