MantisBT - CMake
View Issue Details
0014155CMakeModulespublic2013-05-19 06:052016-06-10 14:31
Modestas Vainius 
Clinton Stimpson 
normalminoralways
closedmoved 
amd64Debian GNU/Linuxsid
CMake 2.8.11 
 
0014155: FindQt4: define QT_IMPORTS_DIR variable even if the directory is not present on the system.
Hello,

QT_IMPORTS_DIR variable should be defined even if the directory is not present on the system. This happens when there are no plugins currently installed.

Patch here: http://anonscm.debian.org/gitweb/?p=collab-maint/cmake.git;a=blob;f=debian/patches/qt_import_dir_variable.diff;hb=HEAD [^]

See https://bugs.launchpad.net/ubuntu/+source/cmake/+bug/894805 [^] for more information.
No tags attached.
Issue History
2013-05-19 06:05Modestas VainiusNew Issue
2013-05-22 15:34Brad KingAssigned To => Stephen Kelly
2013-05-22 15:34Brad KingStatusnew => assigned
2013-05-23 11:27Stephen KellyAssigned ToStephen Kelly => Clinton Stimpson
2013-05-27 11:51Clinton StimpsonNote Added: 0033137
2013-05-27 12:36Modestas VainiusNote Added: 0033138
2016-06-10 14:28Kitware RobotNote Added: 0042283
2016-06-10 14:28Kitware RobotStatusassigned => resolved
2016-06-10 14:28Kitware RobotResolutionopen => moved
2016-06-10 14:31Kitware RobotStatusresolved => closed

Notes
(0033137)
Clinton Stimpson   
2013-05-27 11:51   
Can you please describe the use case for the expectation of finding something that doesn't exist?

See a previous discussion here:
http://www.cmake.org/pipermail/cmake/2012-January/048683.html [^]
(0033138)
Modestas Vainius   
2013-05-27 12:36   
Hello,

the variable would typically be used to install a new plugin. Since a new plugin is independent, it should not require actual presence of other unrelated plugins (even if they are typically installed by default which does not hold in case of distro splits). So you can just create imports directory if it does not exist, install a plugin to it and there is no reason why plugin wouldn't just work afterwards (given actual libraries are present obviously).

You may argue that there should be an empty imports directory present on the system then. However, I kind of don't think that cmake should make a distinction between empty and non-present directory. Especially having in mind that a native qmake returns directory path unconditionally and there isn't much point for cmake to add more logic on top of it, is there?
(0042283)
Kitware Robot   
2016-06-10 14:28   
Resolving issue as `moved`.

This issue tracker is no longer used. Further discussion of this issue may take place in the current CMake Issues page linked in the banner at the top of this page.