Bug 216213 - [patch] "Size & Orientation" asks me to apply changes, but I did not change anything
Summary: [patch] "Size & Orientation" asks me to apply changes, but I did not change a...
Status: RESOLVED FIXED
Alias: None
Product: systemsettings
Classification: Applications
Component: kcm_randr (show other bugs)
Version: unspecified
Platform: unspecified Linux
: NOR normal
Target Milestone: ---
Assignee: Gustavo Pichorim Boiko
URL:
Keywords:
: 217822 219470 220031 222050 226542 226869 (view as bug list)
Depends on:
Blocks:
 
Reported: 2009-11-26 05:43 UTC by Christoph Feck
Modified: 2010-02-15 04:08 UTC (History)
10 users (show)

See Also:
Latest Commit:
Version Fixed In:

Attachments
Proposed patch (1.50 KB, patch)
2009-12-25 17:01 UTC, Dario Andres 		Details
View All Add an attachment

Note You need to log in before you can comment on or make changes to this bug.
Description Christoph Feck 2009-11-26 05:43:43 UTC 
Version:           1.0 (using 4.3.77 (KDE 4.3.77 (KDE 4.4 >= 20091123)), compiled sources)
Compiler:          gcc
OS:                Linux (i686) release 2.6.31.5-0.1-desktop

Just visiting module Display/Size & Orientation, and going back. I did not modify anything, but it still asks me to apply changes.
Comment 1 Christoph Feck 2009-12-08 06:26:46 UTC 
*** Bug 217822 has been marked as a duplicate of this bug. ***
Comment 2 Dario Andres 2009-12-20 23:28:53 UTC 
*** Bug 219470 has been marked as a duplicate of this bug. ***
Comment 3 BRULE Herman 2009-12-22 02:19:07 UTC 
Same with the beta2
Comment 4 Dario Andres 2009-12-25 14:17:44 UTC 
*** Bug 220031 has been marked as a duplicate of this bug. ***
Comment 5 Dario Andres 2009-12-25 17:01:13 UTC 
Created attachment 39333 [details]
Proposed patch

Outputconfig constructor is connecting the combobox changed signals to the setDirty function prior to loading the screens data (which will programmatically change the comboboxes, forcing a "unsaved" state in KCM, even when loading)

Proposed patch: load the screen config prior to connection the signals.
And, in the delayed load, disconnect the signals, and later reconnect them at the end.

I don't know all the possible workflows so this should be tested.
Comment 6 Christoph Feck 2010-01-10 13:28:00 UTC 
*** Bug 222050 has been marked as a duplicate of this bug. ***
Comment 7 Christoph Feck 2010-01-26 05:53:34 UTC 
Please apply the patch to trunk, it can be backported to 4.4 branch when no regressions are visible.
Comment 8 Dario Andres 2010-01-26 12:10:53 UTC 
SVN commit 1080502 by darioandres:

- Do not mark the KCM as modified during the first load
(going to backport to 4.4 if no regressions are visible)

CCBUG: 216213


 M  +8 -1      outputconfig.cpp  


WebSVN link: http://websvn.kde.org/?view=rev&revision=1080502
Comment 9 Dario Andres 2010-02-12 15:08:05 UTC 
*** Bug 226542 has been marked as a duplicate of this bug. ***
Comment 10 m.wege 2010-02-12 17:34:17 UTC 
seems like the backport was forgotten. have there been regressions? otherwise it would be nice to see it in 4.4.1.
Comment 11 BRULE Herman 2010-02-12 19:15:23 UTC 
Yes, bug is here in kde 4.4.0
Comment 12 Dario Andres 2010-02-12 19:20:11 UTC 
Indeed, it was too risky and late to backport an untested patch to 4.4.0
So far, no more errors were reported, so I think that the fix could be backported...
Comment 13 Dario Andres 2010-02-12 20:17:08 UTC 
SVN commit 1089244 by darioandres:

Backport to 4.4 of:
SVN commit 1080502 by darioandres:

- Do not mark the KCM as modified during the first load
(going to backport to 4.4 if no regressions are visible)

BUG: 216213


 M  +8 -1      outputconfig.cpp  


WebSVN link: http://websvn.kde.org/?view=rev&revision=1089244
Comment 14 Dario Andres 2010-02-12 20:23:47 UTC 
Ha, I forgot to remove the message about "going to backport", the fix is already on 4.4 branch, to be released with KDE SC 4.4.1 :)
Comment 15 Dario Andres 2010-02-15 04:08:58 UTC 
*** Bug 226869 has been marked as a duplicate of this bug. ***