Created attachment 1510 [details]
perl programme to demonstrate  the problem

Created attachment 1511 [details]
Snapshot of the output produced by the demo in attachment 1510 [details] (cauldron)

The version number at the right of the snapshot is the text obtained from Gtk2::VERSION.

It can be a bug relating to the new theme (oxygen) ?

CC: (none) => olav, thierry.vignaud

tv uploaded a new gtk+2.0 without oxygen (version gtk+2.0-2.24.10-2.mga2)

IMO, relying on Oxygen just means additional bugs. Remember problems with Evolution as well.

(In reply to comment #3)
> It can be a bug relating to the new theme (oxygen) ?

Indeed as proven by testing several engines. eg:
GTK2_RC_FILES=/usr/share/themes/oxygen-gtk/gtk-2.0/gtkrc perl /entry.pl&
GTK2_RC_FILES=/usr/share/themes/Ia\ Ora/gtk-2.0/gtkrc perl /entry.pl&

(In reply to comment #4)
> tv uploaded a new gtk+2.0 without oxygen (version gtk+2.0-2.24.10-2.mga2)

That only means we won't pull the gtk+2 oxygen them engine anymore by default.
You'll still have engine installed for older installs.

BTW, please open a bug report against oxygen-gtk3 for your evolution issue.

CC: (none) => hugo
Assignee: bugsquad => juan.baptiste
Source RPM: gtk+2.0-2.24.10-1.mga2 => oxygen-gtk

I don't use Evolution. I know because I read such a bugreport upstream. Even think I saw the same in a Mageia bugreport, not sure though. I'll try and see if I can find it.

(it's the bug 2627)

Definitly an oxygen-gtk bug, which I'll try to fix.
Now concerning using oxygen or not as default mageia theme, I guess not using it would lower my motivation to fixing the bugs related to it directly from mageia bugzilla. (since there is enough already on bugs.kde.org)

I understand your position Hugo but also understand ours, we are at less than a month of version freeze (March 7) and there still are many important bugs with oxygen-gtk that need to be solved to be able to use oxygen-gtk as a default.

But also IMO, taking the decision to not use oxygen-gtk as default now is a very precipitated one. We should wait a little more, at least until one week before version freeze and see how all of this bugs have evolved. We need to have into consideration the good will of Hugo, the upstream developer, willing to work with us to fix the bugs, so there's a pretty good chance of getting them fixed for mga 2, right ? :)

Well:

1) usage of oxygen on new install has been only disabled for gtk+2, not for gtk+3.

2) since gtk+2 metacity crashes when one try to log in with gnome classic for quite some times now, I don't see how it can be seen as precipitated.

mmm. Quite some times means February 4th (according to bug 4408). So 10 days ago ?

Also note that I have been unable to reproduce the bug, and have attempted to send 3 patches nonetheless.
Anyway, the decision is yours to take ... I was just mentionning my motivation to help.

Also: I don't see the point of using oxygen-gtk3 and not oxygen-gtk2 (my undestanding is that most gtk apps around are still gtk2. Notably the most used one).

(and to some extend, same as using oxygen-qt, and not oxygen-gtk).

I've seen the bug for longuer but yes I could have reported it earlier.
There're similar issues with XFCE opened for more time.

But since you cannot fix it and since I see it on several machines I don't see why we should let default install crashes like this.

(and cannot reproduce).

But anyway, fine with me.
Bye.

@Thierry,

Why can't we wait a little longer until taking the decision of not making oxygen-gtk2 ? there are four weeks left before version freeze, there are still time left to work on these bugs. Can we wait until for example, March 5?

This can be changed at any time.
But for now why should we get bug reports about newly installed system not working at all?

Because we need users testing what is being fixed and report any other issues that could arise ?

> Because we need users testing what is being fixed and report any other issues
> that could arise ?

No hurry from the user's side. I think the yardstick for when to take your decision is that the "boxed apps" which come with Gtk have enough time for getting user-tested and for new bugs then being discovered to have a chance to get fixed.

Since Beta is around the corner, I just wanted to make sure that this problem is on some todo list.

But I agree with Thierry with respect to bugs that throw segfaults: installed systems must be in some kind of satisfactory working state - otherwise people will stop doing the testing they ought to do.

(In reply to comment #19)
> But I agree with Thierry with respect to bugs that throw segfaults: installed
> systems must be in some kind of satisfactory working state - otherwise people
> will stop doing the testing they ought to do.

And I agree too, what I say is that we can still wait two or three weeks more before removing oxygen-gtk as the default theme. Removing it now will reduce testing of it, waiting a little more won't hurt anybody and will help to find and fix remaining issues faster. If in say, three weeks there still are major issues like segfaults, then the right thing to do will be to make another theme the default, but we can wait a little longer before doing that :)

Fixed in oxygen-gtk branch 1.2 and master.
Branch 1.2 will be tagged to 1.2.1 in about 4 days.

Will there be a new release of the gtk+3 branch too?

Keywords: (none) => PATCH, UPSTREAM
Summary: Gtk2 does not correctly display entry widgets => oxygen-gtk2 does not correctly display entry widgets
Severity: normal => minor

yes 
1.0.1

Please test new oxygen rpm.

CC: (none) => dmorganec

Works great, many thanks - and sorry for the delay

So closing. Thanks all.

Status: NEW => RESOLVED
Resolution: (none) => FIXED