Version: svn-775407 (using Devel) Installed from: Compiled sources Compiler: i686-pc-linux-gnu-4.1. CFLAGS="-march=k8 -O2 -pipe -fomit-frame-pointer" OS: Linux compile Error in containment.cpp at line 121. Reproduction: every compile time Solution(?): initialize the variables
Upgrade to qt-4.4
*** Bug 157909 has been marked as a duplicate of this bug. ***
the proper place to report compilation errors if you think they exist is panel-devel@kde.org. you may also wish to visit http://developer.kde.org/~dirk/dashboard/ to see what it says about the state of the build. if it's green and you're not finding any love, then ask either on irc in #plasma or on the panel-devel mailing list. as a reference point: it's very unusual for a compilation issue to persist more than an hour or two at the *most*.
I didn't realize there were two places to report bugs. I will mark that in page in my bookmarks. Thanks sincerly for the tips. On 16 Feb 2008 05:27:25 -0000, Aaron J. Seigo <aseigo@kde.org> wrote: [bugs.kde.org quoted mail] I didn't realize there were two places to report bugs.<br><br>I will mark that in page in my bookmarks. Thanks sincerly for the tips.<br><br><div class="gmail_quote">On 16 Feb 2008 05:27:25 -0000, Aaron J. Seigo <<a href="mailto:aseigo@kde.org">aseigo@kde.org</a>> wrote:<br> <blockquote class="gmail_quote" style="border-left: 1px solid rgb(204, 204, 204); margin: 0pt 0pt 0pt 0.8ex; padding-left: 1ex;"><div class="Ih2E3d">------- You are receiving this mail because: -------<br>You are on the CC list for the bug, or are watching someone who is.<br> <br><a href="http://bugs.kde.org/show_bug.cgi?id=157888" target="_blank">http://bugs.kde.org/show_bug.cgi?id=157888</a><br><br><br><br><br></div>------- Additional Comments From aseigo kde org 2008-02-16 06:27 -------<br> the proper place to report compilation errors if you think they exist is panel-devel kde org<br><br>you may also wish to visit <a href="http://developer.kde.org/%7Edirk/dashboard/" target="_blank">http://developer.kde.org/~dirk/dashboard/</a> to see what it says about the state of the build. if it's green and you're not finding any love, then ask either on irc in #plasma or on the panel-devel mailing list.<br> <br>as a reference point: it's very unusual for a compilation issue to persist more than an hour or two at the *most*.<br></blockquote></div><br>
Re: comment #4 I do hope that you were being sarcastic, because that is the way I took it. There are NOT two places to report bugs. However, I would suggest that compilation errors should be reported to the kde-devel mailing list first and only file a bug if that yields no resolution and if it appears to be something more than a simple coding error. If you report compile issues to the kde-devel mailing list, expect that you will receive the same complaint -- that you should post it to a specific list. Well, just ignore them. You cannot be expected to subscribe to EVERY mailing list just to report compile problems. Note that the "dashboard" reports: kdebase-workspace r776070 FAILED:
James, you might be right about the part to report compile errors to kde-devel, but thats no reason to re-open the bugreport. trunk/ requires qt4.4 now.
Re: comment #6 SAY WHAT!!!!! If, in fact, the problem here is that Qt-4.4 is required, then there is still a bug here. The build system must check the version of the Qt-4 that it finds and report an error if Qt-4.4 isn't available. That will actually fix this bug. Which is a valid bug although I would have reported the problem to kde-devel rather than opening a bug. But the bug exists and it should remain open till it is properly fixed.
If you want to keep a bugreport open for "kdelibs depends on qt4.4 but doesn't check for qt minimum version > 4.3.0" then open a bugreport for that. However this bugreport is invalid.
It is unfortunate that Andreas Pakulat fails to understand the issues. We shouldn't look for excuses to close bugs, we should fix them. While I am not sure that the error message: [ 45%] Building CXX object workspace/libs/plasma/CMakeFiles/plasma.dir/containment.o /usr/SVN/trunk/KDE/kdebase/workspace/libs/plasma/containment.cpp: In member function 'virtual void Plasma::Containment::init()': /usr/SVN/trunk/KDE/kdebase/workspace/libs/plasma/containment.cpp:121: error: 'NoCache' was not declared in this scope /usr/SVN/trunk/KDE/kdebase/workspace/libs/plasma/containment.cpp:121: error: 'setCacheMode' was not declared in this scope make[2]: *** [workspace/libs/plasma/CMakeFiles/plasma.dir/containment.o] Error 1 make[1]: *** [workspace/libs/plasma/CMakeFiles/plasma.dir/all] Error 2 make: *** [all] Error 2 Is due to the fact that the reporter didn't have Qt-4.4 installed (actually a pre-release of it since it has not been released). Consulting the Qt dox for 4.3 seems to indicate that it isn't (or shouldn't be) the problem. The point is that if Qt-4.4 is required, this error would NOT occur if the CMake scripts checked for a Qt version >= to 4.4.0. Therefore, that is the problem -- the problems is with the CMake scripts. This would actually fix this bug. The user could no longer get this mysterious error message but rather would get a clear error message stating that Qt-4.4.0 was needed. To put this generally, if we can determine the problem from a reporter's bug report then we should fix it. We shouldn't use the fact that the reporter's report isn't exactly correct as a lame excuse to close the bug. This is both unprofessional and nonproductive (and it sucks). Also note that the README is wrong. For KDELibs, it says: If configure claims Qt cannot be found, have a look at ftp://ftp.trolltech.com and download the latest Qt 4.x.x release, at least Qt 4.3.0. Quality management is seriously lacking.
I'm not failing to understand the issue, I'm trying to get proper bugtracking. And the problem that kdelibs doesn't check for Qt4.4 (which is actually by purpose to give people some time to update their qt-copy) is different from this one, hence it needs to be adressed in a separate bugreport. BTW: This is a development version of KDE you're complaining about, you can't expect it to compile all the time and there are certainly more important things to do than fixing a README in this early stage of 4.1 development. People who track trunk should actually be aware of the fact that things might not compile all the time.
"People who track trunk should actually be aware of the fact that things might not compile all the time." I realize that. I did wait a couple of days before reporting this because of that. The snapshot qt 4.4 did fix the problem. The problem is the error I was given would in no way leave me to think this was a qt error. I assumed the problem was with plasma. It would be nice for cmake to spit out the error that the qt version was too low like it does for xine and other libs. I did remove my build and try again, so my dependencies were re-checked. Either way, thanks for the solution. Like I mentioned, that resolved the problem. "People who track trunk should actually be aware of the fact that things might not compile all the time."<br><br>I realize that. I did wait a couple of days before reporting this because of that.<br><br>The snapshot qt 4.4 did fix the problem. The problem is the error I was given would in no way leave me to think this was a qt error. I assumed the problem was with plasma.<br> <br>It would be nice for cmake to spit out the error that the qt version was too low like it does for xine and other libs. I did remove my build and try again, so my dependencies were re-checked.<br><br>Either way, thanks for the solution. Like I mentioned, that resolved the problem.<br>
Re: Comment #11 After discussion on the kde-devel list, this bug has actually been fixed. If you try to compile KDE TRUNK with Qt < 4.4 you now get an error message when you try to build KDELibs. Due to quirks in CMake, this will not appear if you try to update other modules without first updating KDELibs. This is not a bug in CMake, but something that everyone should be aware of: CMake files are installed by KDELibs which are used when building other modules so you should always update KDELibs first.