Gentoo Websites Logo
Go to: Gentoo Home Documentation Forums Lists Bugs Planet Store Wiki Get Gentoo!
Bug 582800 - =kde-apps/kate-4.14.3 does not build against PyQt4 snaphot version.
Summary: =kde-apps/kate-4.14.3 does not build against PyQt4 snaphot version.
Status: RESOLVED FIXED
Alias: None
Product: Gentoo Linux
Classification: Unclassified
Component: [OLD] KDE (show other bugs)
Hardware: All Linux
: Normal normal (vote)
Assignee: Qt Bug Alias
URL:
Whiteboard:
Keywords:
Depends on:
Blocks:
 
Reported: 2016-05-12 03:47 UTC by Oleh
Modified: 2016-06-12 11:11 UTC (History)
1 user (show)

See Also:
Package list:
Runtime testing required: ---


Attachments
build.log (build.log,756.42 KB, text/plain)
2016-05-12 03:48 UTC, Oleh
Details
sipAPIkate.h (sipAPIkate.h,488.39 KB, text/x-chdr)
2016-06-02 03:21 UTC, Oleh
Details
sipkatepart0.cpp (sipkatepart0.cpp,349.33 KB, text/x-c++src)
2016-06-02 03:22 UTC, Oleh
Details

Note You need to log in before you can comment on or make changes to this bug.
Description Oleh 2016-05-12 03:47:56 UTC
as in summary.

Reproducible: Always

Steps to Reproduce:
1. make sure PyQt4-4.12_pre1605051544 installed
2. attempt to merge =kate-4.14.3
3. notice a build failure
4. downgrade to PyQt4-4.11.4
5. notice a successful build.
Comment 1 Oleh 2016-05-12 03:48:35 UTC
Created attachment 434028 [details]
build.log
Comment 2 Michael Palimaka (kensington) gentoo-dev 2016-05-12 13:57:22 UTC
CCing qt in case there's something obvious, but I'm not sure there's much we can do beyond restricting the dep - kate:4 is no longer developed and kate:5 doesn't ship python stuff.
Comment 3 Davide Pesavento gentoo-dev 2016-05-12 14:22:42 UTC
This _could_ be an unintended API break in PyQt4, it should be reported upstream immediately. Oleg, can you notify them please?

(btw, what's up with the build.log? it's so messed up by those random characters that it becomes almost unreadable)
Comment 4 Oleh 2016-05-12 16:00:46 UTC
not sure whats wrong with build.log. i seeing it normally. i don't know where to report this upstream
Comment 5 Davide Pesavento gentoo-dev 2016-05-12 16:20:10 UTC
(In reply to Oleg from comment #4)
> not sure whats wrong with build.log. i seeing it normally.

I guess it's the escape sequences for colored output, used by both cmake and gcc.

> i don't know
> where to report this upstream

equery meta -U dev-python/PyQt4

You can write either to the ML (requires registration), or to Phil directly. Thanks a lot.
Comment 6 Oleh 2016-05-12 17:20:14 UTC
maybe you, guys, better contact him as he probably know qt team (of Gentoo) better?
Comment 7 Johannes Huber (RETIRED) gentoo-dev 2016-05-30 09:48:41 UTC
kde-apps/kate:5 is stable since some days.

commit 5b5e4f1a6f77fac08296c2346f0a647be5ca4358
Author: Mikle Kolyada <zlogene@gentoo.org>
Date:   Tue May 24 14:34:56 2016 +0300

    kde-apps/kate: amd64 stable wrt bug #579992
    
    Package-Manager: portage-2.2.28
Comment 8 Davide Pesavento gentoo-dev 2016-05-30 10:17:20 UTC
As I said above, this could be an unintentional API break in PyQt4. So even if this is no longer relevant for kate:4, it could still be a PyQt4 bug. I had forgotten to report it upstream.
Comment 9 Oleh 2016-05-30 13:03:46 UTC
(In reply to Johannes Huber from comment #7)
> kde-apps/kate:5 is stable since some days.
> 
> commit 5b5e4f1a6f77fac08296c2346f0a647be5ca4358
> Author: Mikle Kolyada <zlogene@gentoo.org>
> Date:   Tue May 24 14:34:56 2016 +0300
> 
>     kde-apps/kate: amd64 stable wrt bug #579992
>     
>     Package-Manager: portage-2.2.28

Sorry, this is not how things fixed :)
Comment 10 Davide Pesavento gentoo-dev 2016-05-30 19:37:17 UTC
https://www.riverbankcomputing.com/pipermail/pyqt/2016-May/037579.html

waiting for feedback...
Comment 11 Johannes Huber (RETIRED) gentoo-dev 2016-05-31 15:00:14 UTC
>[addons/kate/pate/sip/CMakeFiles/python_module_PyKate4_kate.dir/kate/sipkatepart0.o] Error 1


By the way: in the past we had a lot of build failures (mostly marked as invalid) regarding PyQt4 when sip was not rebuilded....
Comment 12 Davide Pesavento gentoo-dev 2016-05-31 21:42:10 UTC
(In reply to Johannes Huber from comment #11)
> By the way: in the past we had a lot of build failures (mostly marked as
> invalid) regarding PyQt4 when sip was not rebuilded....

IIUC, you're referring to the sip ABI breaks that happened from time to time... those shouldn't be a problem anymore, now that sip has a proper subslot and its reverse deps use slot operators. Besides, sip ABI has been pretty stable lately. So I don't think that's the cause here.
Comment 13 Johannes Huber (RETIRED) gentoo-dev 2016-06-01 07:43:58 UTC
(In reply to Davide Pesavento from comment #12)
> (In reply to Johannes Huber from comment #11)
> > By the way: in the past we had a lot of build failures (mostly marked as
> > invalid) regarding PyQt4 when sip was not rebuilded....
> 
> IIUC, you're referring to the sip ABI breaks that happened from time to
> time... those shouldn't be a problem anymore, now that sip has a proper
> subslot and its reverse deps use slot operators. Besides, sip ABI has been
> pretty stable lately. So I don't think that's the cause here.

When i look in PyQt4 and sip ebuilds its the other way arround. When sip ABI changes PyQt4 is rebuilded, but not when PyQt4 is updated sip is not rebuilded. The compile error indicates that sip needs to be rebuilded.
Comment 14 Davide Pesavento gentoo-dev 2016-06-01 09:29:42 UTC
(In reply to Johannes Huber from comment #13)
> When i look in PyQt4 and sip ebuilds its the other way arround. When sip ABI
> changes PyQt4 is rebuilded,

Correct.

> but not when PyQt4 is updated sip is not
> rebuilded. The compile error indicates that sip needs to be rebuilded.

Why would that be needed? sip is completely independent from PyQt4.
Comment 15 Johannes Huber (RETIRED) gentoo-dev 2016-06-01 11:27:20 UTC
(In reply to Davide Pesavento from comment #14)
> (In reply to Johannes Huber from comment #13)
> > When i look in PyQt4 and sip ebuilds its the other way arround. When sip ABI
> > changes PyQt4 is rebuilded,
> 
> Correct.
> 
> > but not when PyQt4 is updated sip is not
> > rebuilded. The compile error indicates that sip needs to be rebuilded.
> 
> Why would that be needed? sip is completely independent from PyQt4.

I am not that familiar with python stuff to give you a valuable answer ;), but i tried to reproduced the build failure. I have downgraded to kate-4.14.3 successfully with PyQt4-4.12_pre1605051544 installed. PyQt4, pykde4 and sip wasn't installed before -> it is a) missing sip rebuild or b) missing pykde4 rebuild. Both can't be automated at the moment as PyQt4 has no subslot and it is likely not worth it to invest here more time as kde-apps/kate:4 will be tree cleaned soon.
Comment 16 Oleh 2016-06-01 12:20:01 UTC
just in case (to reduce guesses :)) this is build from fresh attempt, i.e emerge kate:4, which is pulling sip/pyqt4, which are never installed before. I.e fresh install, so this is likely some API breaks happening somewhere.
Comment 17 Davide Pesavento gentoo-dev 2016-06-01 23:49:36 UTC
Oleg, can you attach the C++ bindings file generated by sip? (I think it's called sipkatepart0.cpp)
Comment 18 Oleh 2016-06-02 03:21:48 UTC
Created attachment 436126 [details]
sipAPIkate.h
Comment 19 Oleh 2016-06-02 03:22:25 UTC
Created attachment 436128 [details]
sipkatepart0.cpp
Comment 20 Davide Pesavento gentoo-dev 2016-06-11 11:35:47 UTC
According to upstream, this should be fixed in the latest PyQt4 snapshot, which I just pushed to the tree as PyQt4-4.12_pre1606101416. Please test and report back.
Comment 21 Oleh 2016-06-12 10:45:49 UTC
newer PyQt4 snapshot fixes this, indeed. Thx
Comment 22 Davide Pesavento gentoo-dev 2016-06-12 11:11:20 UTC
Thanks