[Releasing] Meeting minutes: Release team meeting Dec 17, 2012
Linaae Hanne
Hanne.Linaae at digia.com
Mon Dec 17 18:20:15 CET 2012
Hi,
Here are the minutes from today's release team meeting:
1. Package status & 2. Bugstatus
- Latest packages: http://origin.releases.qt-project.org/digia/5.0.0/backups/2012-12-17-386/ contain fixes for the known P1 for RC2 + doc fixes.
- Packages have been extensively tested today
- Some more issues have come up. Especially QTBUG-28657 and font issue on linux
- decided to resolve what we can, but nothing standalone seem to be real showstoppers so far
3. Next steps
- build new packages with a few more fixes for tomorrow.
- test new packages tomorrow
- meet again at 4.30 PM (CET)/ 7.30 AM (PST) to make a go/nogo decision regarding whether we have packages ready for the final release.
Cheers,
Hanne
--- Meeting log ----
[16:01:00] <hanne> lars thiago steveire kkoehne_ tijensse_ mauricek treinio
iieklund akseli joaijala sahumada ZapB rosch : ping
[16:01:05] <mkalinow> hanne: pong
[16:01:17] <johanna> hanne: pong
[16:01:17] <mkalinow> andre_: ^^
[16:01:18] <rosch> pong
[16:01:23] <thiago_> hanne: pong
[16:02:34] <ZapB> hanne: pong
[16:02:41] <sahumada> pong
[16:03:08] <eskil> aavit: On windows? Since none of the examples are built in
console, I don't see any output :)
[16:04:08] <hanne> hmm - still missing lars
[16:04:30] <lars> pong... but have to leave ~16:30
[16:04:36] <hanne> lars: ok
[16:04:43] <hanne> lets get started then
[16:04:51] <hanne> Suggestion for the agenda:
[16:05:01] <hanne> 1. Package status
[16:05:01] <hanne> 2. Bug status
[16:05:02] <hanne> 3. Next steps towards release
[16:05:23] <hanne> let's start with packages.
[16:05:33] <hanne> johanna: want to give an update?
[16:06:06] <hanne> or sahumada ?
[16:06:07] <johanna> hanne: ok, seems that iieklund is not here. Well, afaik
the current packages are almost ready
[16:06:20] <johanna> there were some docs updates that were needed..?
[16:06:54] <hanne> yes, we have packages that are pretty good: 386
[16:07:00] <johanna> sahumada updated submodules this afternoon, did those
cover the doc issues?
[16:07:07] <sahumada> as far as I know .. there are only two not yet included
https://codereview.qt-project.org/42541 https://codereview.qt-
project.org/42583
[16:07:15] <sahumada> I need to cherry-pick those two
[16:07:21] <johanna> sahumada: ok
[16:07:33] <aavit> eskil: when compiled and run from designer
[16:07:39] <aavit> creator
[16:07:39] <hanne> there are some nice-to-have doc fixes that we would like in
too, so plan is to build new packages again tonight
[16:07:39] <johanna> sahumada: any estimates on when they will be there?
[16:07:51] <sahumada> johanna: passing the CI .. about 1.5 hrs
[16:07:55] <johanna> hanne: ah, ok
[16:08:09] <hanne> but essentially most fixes need should be in already in
today's packages
[16:08:29] <johanna> sahumada: ok, the we will have another build during the
night that _should_ have all needed fixes
[16:08:38] <sahumada> that's correct
[16:08:39] <hanne> they have gone through quite a bit of testing as we can see
on the releasing mailing list
[16:08:44] <tuukkat> Shall we talk the new changes on top of 386 in the
package status or bug status? I am interested in knowing what things we aim to
get on top of 386 now? Is there a list?
[16:09:01] <hanne> sahumada: ^
[16:09:13] <kkoehne_> johanna: Well, the doc fixes will only be there if
somebody updates the injection packages.
[16:09:19] <andre_> mkalinow: pong.
[16:09:26] <sahumada> tuukkat: hanne. yes, I can send a list
[16:10:02] <johanna> kkoehne_: ok, have you been discussing about that with
Iikka?
[16:10:37] <hanne> sahumada: thanks, there are only doc-fixes still to go in,
or other things too?
[16:10:39] <kkoehne_> johanna: Yes, we can do the same stunt has for the rc2,
and rc1, and have updated packages around 11:00 Oslo time tomorrow.
[16:11:06] <sahumada> hanne: doc fixes and maybe three fixes for qtwebkit.git
[16:11:15] <johanna> kkoehne_: ok
[16:11:52] <sahumada> hanne: https://codereview.qt-project.org/42127 should be
the last one for qtwebkit
[16:12:03] <johanna> sahumada: are we updating webkit with also some other
updates than just docs?
[16:12:06] <andre_> anyone with a recent windows package installed who can
pastebin 5.0.0/examples/examples.pro?
[16:12:24] <sahumada> johanna: yes https://codereview.qt-project.org/42127
[16:12:53] <johanna> sahumada: ok, hope it wont break the build at this point
:)
[16:13:31] <hanne> ok, so basically the current packages are something we
could be able to release, but we would like to still put in some doc fixes,
unless some bug reports say something else.
[16:13:48] <hanne> so let' s move to: 2. Bug status
[16:13:50] <johanna> yes
[16:14:38] <hanne> treinio is away, so i will try to give an update:
[16:14:45] <hanne> https://bugreports.qt-project.org/browse/QTBUG-27426
[16:14:58] <hanne> this shows the final release tasks
[16:15:00] <thiago_> andre_: I don't think we should have that file...
[16:15:06] <thiago_> it's installed by each package
[16:15:12] <thiago_> there's no global "examples.pro"
[16:15:55] <hanne> so it seems like we can downgrade: https://bugreports.qt-
project.org/browse/QTBUG-28414 ?
[16:15:59] <kkoehne_> thiago: Okay, I can remove it then.
[16:16:35] <ZapB> hanne: already is
[16:16:40] <jpasion> sahumada: I made some late fixes. Integrating/staged.
fixes for https://bugreports.qt-project.org/browse/QTBUG-28579
[16:16:46] <hanne> ZapB: good
[16:17:18] <sahumada> jpasion: gerrit link ?
[16:17:32] <andre_> thiago: there is a global examples.pro nowadays. and I
just noticed it even works on linux and mac, haven't try to compile on windows
[16:17:47] <kkoehne_> andre_: THere is none. It's the qtbase one
[16:17:53] <thiago> andre_: oh?
[16:17:56] <jpasion> sahumada: several: https://codereview.qt-
project.org/#dashboard,1000054
[16:18:00] <thiago> andre_: is it manually-maintained?
[16:18:18] <andre_> kkoehne_: oh.... right.
[16:18:26] <kkoehne_> We right now just do a cp -r qt5/*/examples/* examples
[16:18:31] <lars> hanne: what about the issues caroline mentioned in her test
report on clang-64?
[16:18:32] -*- andre_ is confused now
[16:18:35] <tuukkat> Not visible in the final release tasks is a bug related
to printing: https://bugreports.qt-project.org/browse/QTBUG-28657
[16:18:48] <kkoehne_> But I made sure the example.pro is from qtbase. I can
also delete it if it's better.
[16:19:15] <hanne> lars: I just noticed them, so have not investigated -
anyone else?
[16:19:21] <thiago> kkoehne_: that would be incomplete then
[16:19:35] <kkoehne_> thiago: Yes. So no examples.pro instead of an incomplete
one?
[16:19:38] <ZapB> https://bugreports.qt-project.org/browse/QTBUG-28496 looks
ot be fine for debugging applications, just not Qt itself. So I think it can
be downgraded to a P2 imho
[16:19:38] <thiago> lars: what clang-64 issue? I build on Mac with clang and
it worked for me.
[16:19:38] <tuukkat> I have commented those, see the bug status. WE can talk
what is the verdict.
[16:19:43] <thiago> kkoehne_: I'd prefer that.
[16:19:46] <kkoehne_> thiago: Done.
[16:20:38] <hanne> thiago: see report from caroline to releasing ML
[16:20:39] <tuukkat> The Mac bugs are: https://bugreports.qt-
project.org/browse/QTBUG-28496 and https://bugreports.qt-
project.org/browse/QTBUG-28657
[16:21:15] <lars> ZapB: yes, 28496 is a P2 IMO
[16:21:20] <anshaw> IMO at least QTBUG-28496 can be downgraded since its only
a package problem, as long as the app starts
[16:21:44] <tuukkat> Both are a bit annoying and certainly ones that would be
good to have fixed, but I am fine listing these as known issues.
[16:22:08] <andre_> thiago, kkoehne_: there is a 'aggregate' target in
example/examples.pro iterating through the subdirectories. when I run
qmake/make in examples e.g. ./webkitwidgets/fancybrowser/fancybrowser is build
and works. is that called "working at the toplevel"?
[16:22:23] <ZapB> tuukkat: I can't recall the last time I printed from
assistant so it's a low-pri fo rme personally
[16:22:32] <lars> tuukkat: the assistant crash is a bit worse. what about the
multimedia issues she mentioned
[16:22:34] <lars> ?
[16:22:35] <tuukkat> For https://bugreports.qt-project.org/browse/QTBUG-28657
I wonder if that is a generic printing problem, or something that can be fixed
with a workaround? It happens to me both in Asssitant and Creator for the 2nd
print.
[16:22:56] <anshaw> just did a quick test from the textedit example, its not
happening there at least
[16:22:59] <mkalinow> tuukkat: sounds generic then#
[16:23:03] <lars> tuukkat: the bug report sounds like it's a bit random. that
would point to some memory corruption
[16:23:17] <thiago> andre_: ah, I see it
[16:23:24] <thiago> kkoehne_: keep the qtbase examples.pro then
[16:23:56] <tuukkat> For the multimeadia examples some worked for me when I
tested, the ones Caroline mention did not.
[16:24:21] <kkoehne_> thiago: Okay.
[16:24:28] <lars> tuukkat: ok, so most other ones work?
[16:24:44] <lars> tuukkat: wonder whether it's a simple packaging problem that
we could fix
[16:25:04] <anshaw> re QTBUG-28657, the stack trace shows different versions
of the same symbols loaded, for the image formats for instance
[16:25:10] <tuukkat> The one I tested to play sound today worked, but I did
not run them all. Would be good to verify with real app.
[16:25:16] <anshaw> both the debug and non debug ones are loaded
[16:25:28] <kkoehne_> lars: Two of them will be fixed in
https://codereview.qt-project.org/#change,42859 . Don't know whether it'll
make it for the nighly builds, though.
[16:25:34] <anshaw> and the same for the printer support so its probably a
packaging problem
[16:26:25] <tuukkat> lars: this at least works for me Audio Output Example
[16:26:45] <kkoehne_> anshaw: Define 'packaging problem' . Were anybody able
to reproduce it with a debug_and_release build on mac?
[16:27:38] <anshaw> kkoehne_: I build from source by default and its fine for
me, but the problem is that both the debug and release plugins are loaded, the
libcocoaprintersupport one is done this way, which means its likely to be
causing a crash
[16:28:25] <anshaw> kkoehne_: since it will load both (like it did for the
cocoa platform plugin), so if both exist its going to run into trouble
[16:28:56] <johanna> anshaw: how we can fix that in packagin phase?
[16:28:58] <kkoehne_> anshaw: Okay. So a something to be fixed in qtbase?
[16:29:09] <anshaw> same way we did for the libqcocoa plugin
[16:29:36] <anshaw> it need sa proper fix long term, but we can do the same
fix for that as we did for the platform plugin I think, which unless it
changed was to not build it in debug and symlink it instead
[16:30:03] <johanna> anshaw: who is we :)? Does Iikka know this..?
[16:30:34] <thiago> btw, do we care that the fonts are completely different on
Linux?
[16:30:41] <hanne> is this something we can manage to fix for the build that
starts tonight?
[16:30:50] <anshaw> johanna: Topi and me proposed the original fix for the
qcocoa plugin, it was done for rc2, but I don't know who changed the script
[16:31:14] <hanne> thiago: is that something that recently changed?
[16:31:43] <thiago> no
[16:31:44] <johanna> anshaw: ok, I hope Iikka know someting about this, I
don't recall doing that..
[16:31:50] <hanne> johanna: is iikka still around?
[16:31:51] <thiago> Qt 5 has had that problem forever
[16:31:54] <anshaw> johanna: if he was working on it last week then yes
[16:32:00] <thiago> I'm pretty sure there are tasks about it
[16:32:06] <ZapB> yes
[16:32:08] <johanna> hanne: no
[16:32:20] <johanna> hanne: he will be in the office tomorow morning
[16:32:52] <hanne> johanna: is there anyone else who can do the fix?
[16:33:00] <lars> thiago: we're not reading the global config setting?
[16:33:35] <johanna> hanne: I don't know, not that I know..
[16:33:39] <kkoehne_> anshaw: Are you talking about https://codereview.qt-
project.org/#change,42213 ?
[16:34:10] <thiago> lars: something like that
[16:34:18] <thiago> lars: I don't know the root cause
[16:34:21] <tuukkat> hanne, johanna: Is this such a fix that only affect Mac
packaging? That shouid be possible to run during the day tomorrow, if it is
not affecting the sha1s at all.
[16:34:23] <anshaw> kkoehne_: yes
[16:34:27] <thiago> lars: but Qt Creator looks horrible to me
[16:34:35] <lars> thiago: hmmm.. it's been like that forever now.
[16:34:36] <anshaw> kkoehne_: do the same for the cocoaprintsupport and that
should solve this issue I think
[16:34:49] <anshaw> there are bound to be other issues hiding because all the
imageformats are doubled up too
[16:34:49] <kkoehne_> anshaw: Okay, but that's not a packaging fix :)
[16:34:56] <johanna> tuukkat: true, but since I don't know the details, it's
difficult to say..
[16:35:02] <lars> thiago: guess we should fix this, but do we need it for
5.0.0?
[16:35:04] <anshaw> kkoehne_: originally it was done as a workaround in the
package, but I didn't keep track of it
[16:35:11] <thiago> lars: if we want a non-horrible Qt Creator, yes
[16:35:19] <thiago> lars: that's why I am asking
[16:35:24] <thiago> do we want to block on that?
[16:35:38] <anshaw> kkoehne_: but it means it should be straight forward to
modify that patch to include the other plugin
[16:35:44] <kkoehne_> anshaw: The only fix I can think of in the packaging is
deleting one of the versions, but that would make either a debug or a release
build not print at all.
[16:35:50] <hanne> tuukkat: looking at https://codereview.qt-
project.org/#change,42213 it seems like it is need in the code?
[16:36:13] <lars> thiago: good question. unfortunately I need to run now if I
don't want to get into serious trouble at home ;-)
[16:36:23] <anshaw> kkoehne_: it loads both anyway so I don't think it would
prevent it from printing
[16:36:24] <lars> I'll be back online later tonight
[16:36:28] <tuukkat> thiago: what is the problem with Creator on Linux? I have
it (386) running on Ubuntu 12.04 and it does not look that much different than
on my Mac
[16:36:29] <lars> ~8:30
[16:36:53] <thiago> tuukkat: compare Qt 4 to Qt 5
[16:36:59] <thiago> let me create a screenshot
[16:37:21] <kkoehne_> anshaw: So you're suggesting to include only e.g.
release versions of Qt plugins in the bnaries, deleting the debug ones?
[16:37:25] <tuukkat> hanne: then it needs to be cherry picked in now to get in
tonightäs packages, or we accept it is not fixed.
[16:37:58] <hanne> tuukkat: well seems like kkoehne_ and anshaw are thinking
it ca be done in the packages
[16:38:09] <kkoehne_> hanne: We're brainstorming :)
[16:38:19] <anshaw> kkoehne_: if you do that you need to symlink it, as the
_debug ones need to exist for the debugging aspect to work, but really I think
it would be quicker to patch Qt
[16:38:29] <anshaw> kkoehne_: since its just an extra check similar to the one
in qfactoryloader.cpp
[16:38:48] <hanne> kkoehne_: yup - keep on - it might resolve in a solution :)
[16:38:49] <kkoehne_> anshaw: Okay. But that means we won't have it for
tonights packages.
[16:38:56] <anshaw> kkoehne_: indeed
[16:39:19] <anshaw> kkoehne_: but since it ended up being fixed this way for
the cocoa platform plugin I can only assume that this was deemed to be the
best way to solve it
[16:39:33] <thiago> http://wstaw.org/m/2012/12/17/plasma-desktoppj3397.png
[16:39:38] <thiago> the one on the right is Qt 5
[16:39:46] <thiago> note how the same font has different metrics
[16:39:50] <thiago> it's taller and thinner
[16:39:55] <hanne> anshaw kkoehne_ : so can we get it fixed by just changing
the packages or not?
[16:39:58] <thiago> it makes Creator close to unusable because it's harder to
read
[16:40:02] <anshaw> hanne: I would say no
[16:40:03] <eike> lars: thiago: the horrible font rendering problem is not
restricted to linux...
[16:40:10] <anshaw> hanne: in theory yes, but I think it would ultimately take
longer
[16:40:30] <anshaw> hanne: since there was a reason why we couldn't do that
for the platform plugin in the first place, I think Caroline would know more
details but I don't
[16:40:58] <hanne> iikka: welcome :)
[16:41:04] <thiago> we can say that this is a bad issue, but not a release-
stopper
[16:41:08] <iikka> hanne: hi
[16:41:17] <hanne> anshaw: can you explain to iikka ?
[16:41:28] <thiago> we just need to acknowledge it and have a plan to fix,
since people will complain
[16:41:30] <anshaw> iikka: the fix that was done to solve the cocoa platform
plugin in the Mac packages
[16:41:49] <anshaw> iikka: we need to do the same for the cocoaprintsupport
plugin as well as it is loading the release and debug ones there too
[16:41:56] <hanne> thiago: are you talking about the fonts or the printing?
[16:42:13] <anshaw> iikka: originally we were doing that by symlinking but I
believe that there was a problem with that approach hence it got fixed in the
code instead
[16:42:19] <eike> thiago: lars: tuukkat: e.g. https://bugreports.qt-
project.org/browse/QTBUG-27402
[16:42:29] <tuukkat> thiago: mine does not look like that. I emailed you
screenshots.
[16:42:38] <andre_> thiago: is that influenced by say, -style fusion on the
creator command line?
[16:42:43] <iikka> hanne:anshaw: ok, so we need a patch for that?
[16:42:44] <eike> https://bugreports.qt-project.org/browse/QTBUG-25702
[16:42:51] <eike> as well
[16:42:56] <mkalinow> thiago: lars: i thought this was all related to using a
different/unified font renderer in qt5 and hence is "desired" behavior
[16:43:02] <eike> andre_: it's the same on mac
[16:43:03] <iikka> hanne: anshaw: is that fix already done (in ci) ?
[16:43:16] <hanne> iikka: no, it isnt
[16:43:56] <andre_> eike: I get quite different results depending on whether I
use "-style fusion" on the command line, or not (on Linux).
[16:43:58] <iikka> hanne: blocker?
[16:44:04] <hanne> iikka: so either we fix it in the packaging (if possible
like now) or it is a known issue or
[16:44:10] <anshaw> iikka: it is for the cocoa platform plugin but not the
printsupport one
[16:44:13] <thiago> andre_: that is already fusion
[16:44:39] <iikka> anshaw: we can have a look tomorrow, can you pointo to jira
bug etc to get more info?
[16:45:05] <anshaw> iikka: https://bugreports.qt-
project.org/browse/QTBUG-28657
[16:45:12] <anshaw> iikka: stack trace was clear to me at least
[16:45:48] <hanne> so what is the verdict, do can we still release with these
two issues, or do we need to fix them first?
[16:45:49] <iikka> anshaw: ok
[16:46:08] <tuukkat> hanne, which 2?
[16:46:16] <anshaw> hanne: we need to fix this printing one at least because
the crash it could cause is random
[16:46:50] <hanne> tuukkat: the font issue thiago is talking about, and
QTBUG-28657
[16:47:44] <tuukkat> the font issue does not happen to me in Ubuntu 12.04
[16:48:35] <tuukkat> Maybe it is to do with how the Linux machine is set up
and can be worked around. I have quite default settings in my Ubuntu 12.04.
[16:48:36] <thiago> let me investigate the font issue further
[16:48:41] <thiago> talk to eskil for example
[16:48:57] <hanne> so sounds like we can live with the fonts, but need to fix
the printing issue
[16:49:22] <thiago> btw, that assistant crash's backtrace says
"com.trolltech.assistant"
[16:49:34] <thiago> if someone has spare cycles, hunt down where the
"com.trolltech" comes from
[16:49:35] <anshaw> someone should verify if my point about the debug library
conflicting at least
[16:49:38] <tuukkat> I do not see the font issue as a blocker, but the
printing one would be worth looking into if we can get the fix in tonight - or
if it can be fixed in packaging - during tomorrow.
[16:49:44] <anshaw> but I am quite confident it is the cause
[16:49:58] <anshaw> I can do the fix based on the same approach as the
platform one later today but not sure when exactly
[16:50:07] <thiago> but I'd say that the crash is a P2
[16:50:11] <thiago> bad, but doesn't stop the release
[16:50:35] <hanne> ok - so proposal:
[16:50:48] <mkalinow> thiago: karsten told me nobody in doc updated the
assistant doc since the nokia merge. but has been recognized to late so will
be changed later
[16:51:13] <hanne> we fix if we can tomorrow, otherwise we can still release
the packages with these as known issues
[16:51:32] <ZapB> hanne: sounds reasonable to me
[16:51:42] <thiago> tuukkat: you're showing webkit. Please compare the code
edits.
[16:51:52] <tuukkat> Sounds good. Have we been able to get the license doc
fixes in today?
[16:52:02] <tuukkat> thiago: there should be 2 screenshots
[16:52:14] <anshaw> hanne: sounds fine, I will proceed with a fix later today
and then if it gets in then great otherwise I'll live with it
[16:52:22] <thiago> tuukkat: two screenshots of the help
[16:52:25] <thiago> tuukkat: only one of them has code
[16:52:40] <hanne> tuukkat: yes, licenses fixes will be in in next package
[16:52:49] <tuukkat> thiago: will re-grab
[16:53:05] <tuukkat> font in code edit is not same as in mac, but fine for my
eye
[16:54:07] <tuukkat> hanne, sergio: so are all the things we want to get in
either in or coming through the CI with someone (Sergio) cherry-picking them
to release before the build starts?
[16:54:12] <thiago> anyway, let me follow up on this with the font guys
[16:54:23] <thiago> oh, one more thing for the release:
[16:54:31] <thiago> I've updated the qtchooser but I'm still not done
[16:54:39] <thiago> I've been having difficulty finishing it
[16:54:51] <thiago> so I cannot provide an "atomic commit" (as per ossi's
definition) for it
[16:55:03] <thiago> I will simply push the work in progress and ask people for
help
[16:55:15] <sahumada> tuukkat: all the ones I am aware of
[16:55:24] <hanne> tuukkat: yes, all the things we wanted to get in, so unless
there are more now
[16:56:21] <tuukkat> sounds good
[16:57:04] <tuukkat> so the only things for tomorrow are to: get the updated
docs in, see if we can fix mac packaging (rel/deb) and testing
[16:57:50] <tuukkat> The known issues page is not yet very much updated - some
of the things talked during today are not yet in. We need to make sure we do
not forget to mention the needed things there.
[16:58:04] <tuukkat> thiago: new grap in email
[16:58:39] <hanne> tuukkat: yes, i will make sure the known issues are updated
[17:00:02] <hanne> suggestion:
[17:00:15] <hanne> we either take today's packages with known issues, or
tomorrow's with more fixes and make these the final packages.
[17:00:25] <hanne> goal would then be to release on wednesday
[17:00:29] <thiago> tuukkat: it looks the same for you
[17:00:36] -*- ZapB votes for tomorrows unless they introduce new regressions
[17:00:54] <hanne> ZapB: agree
[17:01:03] <thiago> yeah, if tomorrow's packages are better, why would we take
worse? :-)
[17:01:14] <anshaw> if it means it comems out Wednesday over Thursday then I
am all for it
[17:01:31] <hanne> thiago: of course not, but just incase they don't build or
are worse or something else
[17:01:48] <tuukkat> thiago: then it is just me :) - the font does not look
very bad for me
[17:01:55] <hanne> so we have now moved to: 3. Next steps towards release
[17:02:15] <eskil> thiago: [font issue] Is this that Qt picks up a serif font
by default on some linux distros? I've seen that on fregl's laptop
[17:02:49] <hanne> do we need another release team meeting tomorrow?
[17:02:56] <andre_> eskil: http://wstaw.org/m/2012/12/17/plasma-
desktoppj3397.png
[17:03:20] <thiago> eskil: Creator is configured for a different font here
[17:03:28] <thiago> eskil: it's not the default. I'm using Inconsolata.
[17:03:32] <ZapB> hanne: I think we do need a go/no-go meeting tomorrow
[17:03:48] <thiago> eskil: it's a free font, you can try it. By the way, I
recommend it for coding. It's a really nice font.
[17:04:09] <eskil> thiago: What's the bug?
[17:04:11] <tuukkat> hanne: it is fine for me to have another meeting tomorrow
to decide with which packages we go.
[17:04:58] <hanne> thiago: do you want to be part of the meeting tomorrow, or
could we have it earlier without you? :)
[17:05:14] <hanne> thiago: go/nogo meeting
[17:05:38] <thiago> eskil: http://wstaw.org/m/2012/12/17/plasma-
desktoppj3397.png
[17:05:46] <thiago> eskil: left is qt4; right is qt5
[17:05:48] <eskil> thiago: Is the screenshot recent? There were some last-
minute changes after rc2 which had an impact on font selection
[17:05:59] <eskil> thiago: And is hinting turned on or off?
[17:06:02] <thiago> eskil: it's from right now
[17:06:14] <eskil> thiago: But was it like this before today? :)
[17:06:27] <eskil> thiago: Or did this regress in Qt 5.0.0 packages?
[17:06:33] <thiago> it's always been like that
[17:06:43] <thiago> the build is actually from Thursday
[17:06:44] <eskil> thiago: How always?
[17:06:52] <thiago> ever since I've tried Qt Creator with Qt 5
[17:07:26] <hanne> thiago: do you want to be part of the meeting tomorrow, or
could we have it earlier without you?
[17:07:45] <hanne> thiago: the go/nogo meeting ^
[17:07:53] <eskil> thiago: Ok, the fix I'm referring to went in on Friday
[17:07:54] <thiago> hanne: checking my calendar
[17:07:59] <thiago> eskil: rebuilding
[17:08:37] <thiago> hanne: I'd like to be part, if possible
[17:08:42] <tuukkat> For me 2 hours before would be better, I have another
meeting starting 15.00 Oslo time
[17:09:51] <hanne> ok - shall we do same time tomorrow then?
[17:10:06] <mkalinow> +1
[17:10:33] <hanne> or if possible 15 min later (would help my schedule..) ?
[17:11:01] <eskil> thiago: https://codereview.qt-project.org/#change,40003
This went in in November, so it's probably not related, right?
[17:11:04] <hanne> any objections?
[17:11:13] <andre_> hanne: not from me
[17:11:18] <eskil> thiago: It's supposed to make Qt default to Regular when
available instead of Medium
[17:11:23] <johanna> hanne: fine with me :)
[17:11:50] <thiago> eskil: it might be. The font is thinner now...
[17:12:10] <hanne> ok - let's meet again tomorrow @ 16.15 CET for a go/nogo
decision
[17:12:17] <ZapB> ack
[17:12:24] <tuukkat> Not 15:15?
[17:12:33] <thiago> eskil: no change. It still looks taller and thinner.
[17:12:41] <tuukkat> ok, will try to accommodate?
[17:12:42] <hanne> tuukkat: i cannot make it then
[17:12:52] <hanne> tuukkat: thanks
[17:13:07] <eskil> thiago: But you saw the bug before November 23rd? Because
then it should be related
[17:13:12] <hanne> anyone have anything else for the meeting?
[17:13:14] <tuukkat> Lars and I are both in anoyher meeting between 15-17 CET,
so it would be great if we can end this before it
[17:13:28] <thiago> eskil: I'm almost sure I had seen it before
[17:13:44] <hanne> tuukkat: ok - i could do later, at 17.00
[17:13:48] <thiago> eskil: but I'm retesting with the revert now
[17:13:56] <hanne> but if thiago is to join we cannot do before, it think
[17:14:07] <hanne> s/it/I
[17:14:32] <tuukkat> 16.30 should be fine, we'll try to be quick in the other
meeting
[17:14:48] <eskil> thiago: Here's a comparison the guy made for his patch:
Before: http://ubuntuone.com/30TgXSVWajLHUowlOqQEeg and After:
http://ubuntuone.com/3SpZM4PHMVpX9orDWyTbek
[17:15:01] <hanne> ok, then we meet tomorrow at 16.30 CET for go/nogo decision
[17:15:09] <hanne> any objections now? :)
[17:15:28] <thiago> 16.30 CET / 7.30 PST is fine for me
[17:15:50] <eike> tijensse_: still there?
[17:16:00] <eike> tijensse_: can I have +2 for https://codereview.qt-
project.org/43067 ? :)
[17:16:06] <eskil> thiago: I guess it would be worth checking what OpenOffice
or some Windows applications do with the Inconsolata font. Whether we've been
showing it as demibold in Qt 4.
[17:16:09] <hanne> ok - thanks all for attending the meeting - sorry it went
over time
[17:16:14] <eskil> thiago: If this turns out to be related, that is
[17:16:15] <tijensse_> eike: yes
[17:16:17] <tijensse_> eike: one moment
[17:16:23] <hanne> see you again tomorrow
[17:16:25] <hanne> bye
[17:16:31] <ZapB> bye
[17:16:34] <johanna> hanne: bye
[17:16:36] -*- eike managed to fix that for lin+win but forgot mac <facepalm/>
[17:17:03] <tuukkat> bye
[17:18:36] <thiago> eskil: will do
[17:18:46] <thiago> eskil: I need to leave now. I'll be back from the office.
[17:19:18] <eskil> thiago: Ok
[17:19:48] <eskil> thiago: I think we can probably live with this for 5.0.0
anyway
[17:19:54] <eskil> thiago: Unless there are strong objections
More information about the Releasing
mailing list