|
Bugzilla – Full Text Bug Listing |
| Summary: | yast-gtk crashes clicking on basic settings tab of print queue edit dialog | ||
|---|---|---|---|
| Product: | [openSUSE] openSUSE 11.0 | Reporter: | JP Rosevear <jpr> |
| Component: | YaST2 | Assignee: | Forgotten User h13THG8RK1 <forgotten_h13THG8RK1> |
| Status: | RESOLVED DUPLICATE | QA Contact: | Jiri Srain <jsrain> |
| Severity: | Critical | ||
| Priority: | P1 - Urgent | CC: | alberto.passalacqua, coolo, forgotten_79ta3g7mmD, forgotten_h13THG8RK1, lars.vogdt, obelich |
| Version: | Final | ||
| Target Milestone: | --- | ||
| Hardware: | Other | ||
| OS: | Other | ||
| Whiteboard: | gnome-update | ||
| Found By: | --- | Services Priority: | |
| Business Priority: | Blocker: | --- | |
| Marketing QA Status: | --- | IT Deployment: | --- |
| Attachments: |
i586 test rpm
x86_64 test rpm |
||
|
Description
JP Rosevear
2008-06-20 18:02:45 UTC
Yep, that would be bug 394309. I thought we had repackaged -gtk after that though. Fixed the packaging error. You already packaged it though, right? I added there some slight change to the radio/check buttons behavior that would be cool to ship... In case it hasn't been shipped yet... A patch would be very welcome soon :-) Regards, Alberto Ok, something very odd happened here - the changelog in the package is: 2008-06-02 Michael Meeks <michael.meeks@novell.com> * src/YGUI.cc (YGApplication::makeScreenShot), * src/ygtkwizard.c (ygtk_help_dialog_init), * src/ygtkwizard.c (sync_window_title): tag more missing strings for translation in the next version bug #395803# 2008-06-02 Ricardo Cruz <rpmcruz@alunos.dcc.fc.up.pt> * src/YGUI.cc: fixed bug 395484. Wrong check that assumed error was set, crashing... * src/YGUI.cc: fixed bug 395484. Wrong check that assumed error was set, crashing... which does *not* match the trunk changelogs for 2008-06-02. Perhaps there was a branch at the time that caused the problem? We have a branch for 11.0 together as do all the yast modules and as we had before. I haven't touched the branch; I'm doing my stuff in the trunk. Doesn't seem to make much sense to maintain a branch that will just be a copy of that of trunk: yast2-gtk being a layer between yast2 and gtk only needs to keep up with yast2-libui. If some stuff is added to yast2-libui, it's much more manageable to add #ifdefs checks on the version and share the rest of the code. Of course, we had to eventually create a trunk/10_3 because yast2-ui API was re-written; this is unlikely to happen now though. Perhaps the branch didn't have the fix, but it didn't cause it. This is a much older bug that comes from 10.3 and before: the combo box was the first widget I did, and it seems gtk_combo_box_get_active_text() only works for GtkComboBoxEntry (unless the GtkComboBox is created by some specialized construct). It returns NULL otherwise, crashing at the std::string constructor. It got triggered only now because YCP developers tend to use `CurrentItem for ordinary ComboBox because it returns the current item as an integer, unlike `Value which returns a string, usually being used for the editable ComboBoxes. Created attachment 223608 [details]
i586 test rpm
Test rpms attached
Created attachment 223609 [details]
x86_64 test rpm
Anja, swamp id for this and bug 402316 please. When will the fix appear in updates? A. This patch is not in the test update repo yet even, but its been taken in autobuild - Anja? *** Bug 404865 has been marked as a duplicate of this bug. *** *** Bug 406012 has been marked as a duplicate of this bug. *** There is talk for an update at bug 394309 as well, which is about the exact same thing. We probably want to set one of them as the update of the other. Doesn't make much sense to test the same thing twice. «We probably want to set one of them as the (update) of the other.» I meant: ^ duplicate *** This bug has been marked as a duplicate of bug 394309 *** |