| Summary: | Commit 0f149f44 broke Konqueror's session management | ||
|---|---|---|---|
| Product: | [Applications] konqueror | Reporter: | pat_h | 
| Component: | session restore | Assignee: | Eduardo Robles Elvira <edulix> | 
| Status: | RESOLVED FIXED | ||
| Severity: | major | CC: | aacid, bugs.kde.org, cfeck, faure, gwierzchowski, kde, kde, lueck | 
| Priority: | NOR | ||
| Version First Reported In: | 4.14.1 | ||
| Target Milestone: | --- | ||
| Platform: | Debian testing | ||
| OS: | Linux | ||
| URL: | https://bugs.debian.org/cgi-bin/bugreport.cgi?bug=763281 | ||
| Latest Commit: | http://commits.kde.org/kde-baseapps/dcc45ea70b0b408c8017c6f423b978fe6ca6163a | Version Fixed/Implemented In: | 4.14.3 | 
| Sentry Crash Report: | |||
| 
        
          Description
        
        
          pat_h
        
        
        
        
          2014-10-05 12:30:07 UTC
        
       Adding committer. confirmed with 14.0 + 14.1 from distribution (kubuntu 14.04 + ppa) and with kde 4.14 compiled from sources. I just noted, that it's not explicit in the bug report's description: The patch already exists, it's simply reverting 0f149f44, which is a very small commit (it does not have to be reverted completely, but it's obvious when looking at it). *** Bug 339968 has been marked as a duplicate of this bug. *** Is there anything more I can do not being a KDE developer? I mean this is a bug generating quite a lot of extra work for some users, the bug itself is confirmed, the fix is known, it works for me now for a month, but it does not get fixed in KDE? Is there something more I am supposed to do or is this the normal timeframe for fixing a bug in KDE? This is a serous question as I do not know the details of KDE's workflow. The Konqueror maintainer did not have the time yet to check this issue. If the fix is obvious to you, please suggest a patch (here or at reviewboard.kde.org). What is so "obviously" wrong about commit 0f149f44696 ? Please explain the patch you have in mind, to fix this without reintroducing queryExit()? Ah - does it work better if you swap the tests?
--- a/konqueror/src/konqmainwindow.cpp
+++ b/konqueror/src/konqmainwindow.cpp
@@ -5190,7 +5190,7 @@ void KonqMainWindow::closeEvent( QCloseEvent *e )
           QApplication::sendEvent( (*it)->part()->widget(), e );
   }
   KParts::MainWindow::closeEvent( e );
-  if (stayPreloaded() && !kapp->sessionSaving()) {
+  if (!kapp->sessionSaving() && stayPreloaded()) {
       e->ignore();
       hide();
   }
We don't want to call stayPreloaded() when closing the session.
Feel free to drop me a private email if this fix works for you; I am too swamped by bugzilla.(In reply to David Faure from comment #8) > Ah - does it work better if you swap the tests? > No session management ist still broken as before the swap I just tested it, and the patch fixes the issue for me... Git commit dcc45ea70b0b408c8017c6f423b978fe6ca6163a by David Faure. Committed on 11/11/2014 at 11:17. Pushed by dfaure into branch 'KDE/4.14'. Unbreak session management: don't clear this window (for reusing) while saving. FIXED-IN: 4.14.4 (maybe this can be added to 4.14.3?) CCMAIL: release-team@kde.org M +1 -1 konqueror/src/konqmainwindow.cpp http://commits.kde.org/kde-baseapps/dcc45ea70b0b408c8017c6f423b978fe6ca6163a Oh, 14.12 is already branched, so right now, the commit is lost in a dead branch (4.14.3 is the last release from the KDE/4.14 branch). Reopening until it is clear if we can get it into 4.14.3, but I had the impression it is released today. Otherwise, the commit needs to be in master or 14.12 branch. |