On Thursday 12 September 2013 16:14:36 Jan Kundrát wrote:
> On Thursday, 12 September 2013 15:52:38 CEST, Pali Rohár wrote:
> > First I changed all find_package with trojita_find_package
> > which handle also feature summary cmake module and second
> > there is for example qt dbus and qt test libra
On Monday, 16 September 2013 18:39:45 CEST, Pali Rohár wrote:
What is bad on using here also my wrapper trojita_find_package?
Qt5 upstream recommends [1] using qt5_use_modules for specifying which of
the many Qt5 modules are needed by each target. Before this macro is
available, though, one h
On Thursday, 12 September 2013 15:52:38 CEST, Pali Rohár wrote:
First I changed all find_package with trojita_find_package which
handle also feature summary cmake module and second there is for
example qt dbus and qt test libraries which are not required
always (only if dbus and test options ar
On Thursday 12 September 2013 10:43:05 Jan Kundrát wrote:
> On Thursday, 12 September 2013 09:37:43 CEST, Pali Rohár wrote:
> > I copied file CMakeDependentOption.cmake from system and I
> > rewritten it. If you want I can include it in git, but you
> > will see only that file was replaced.
>
> If
On Thursday, 12 September 2013 09:37:43 CEST, Pali Rohár wrote:
I copied file CMakeDependentOption.cmake from system and I
rewritten it. If you want I can include it in git, but you will
see only that file was replaced.
If it's a complete rewrite, there should be no copyright header pointing t
On Wednesday 11 September 2013 18:34:52 Jan Kundrát wrote:
> On Wednesday, 11 September 2013 17:45:02 CEST, Pali Rohár
wrote:
> > First version of code is in my pali-gsoc branch.
>
> Looks like you based your code on something from cmake's git.
> Could you please make a separate commit that adds
On Wednesday, 11 September 2013 17:45:02 CEST, Pali Rohár wrote:
First version of code is in my pali-gsoc branch.
Looks like you based your code on something from cmake's git. Could you
please make a separate commit that adds the original file (possibly under
the new name, not the original on
On Tuesday, 13 August 2013 08:30:54 CEST, Pali Rohár wrote:
I do not like idea to include compile output path into binaries.
This sounds like that something is bad if some random path
(=compile path) must be present in application executable.
I asked for opinion, I got one -- thanks. I happen
On Monday 12 August 2013 18:25:40 Jan Kundrát wrote:
> On Sunday, 11 August 2013 12:45:23 CEST, Pali Rohár wrote:
> > So do you think that this is real problem?
>
> Maybe not really, after all. However, what about adding a code
> which checks whether the qApp->applicationDir() matches the
> cmake'
On Sunday, 11 August 2013 12:45:23 CEST, Pali Rohár wrote:
So do you think that this is real problem?
Maybe not really, after all. However, what about adding a code which checks
whether the qApp->applicationDir() matches the cmake's binary output path
and if that is the case, it disabled the
On Sunday 11 August 2013 09:40:15 Jan Kundrát wrote:
> On Sunday, 11 August 2013 08:54:58 CEST, Pali Rohár wrote:
> > I think "trojita_plugin_" prefix is needed because plugins
> > should be usable also from application directory (when
> > trojita is not installed to system path). And it is better
On Sunday, 11 August 2013 08:54:58 CEST, Pali Rohár wrote:
I think "trojita_plugin_" prefix is needed because plugins should
be usable also from application directory (when trojita is not
installed to system path). And it is better to tell qpluginloader
to try open only files which could be plu
On Saturday 10 August 2013 17:10:00 Jan Kundrát wrote:
> Hi, a couple of comments:
>
> - The tests should be built by default. That's the whole point
> of them -- whenever someone compiles Trojita (and
> *especially* when someone makes any change), the tests must
> be run.
>
OK.
> - The build s
On Saturday 10 August 2013 17:10:00 Jan Kundrát wrote:
>
> - The plugin directory must not be something overly generic
> like /usr/lib. Please use configure.cmake.in (you might have
> to rebase on top of current master for that) and use
> something like
> "@CMAKE_INSTALL_PREFIX@/libexec/trojita" a
Hi, a couple of comments:
- The tests should be built by default. That's the whole point of them --
whenever someone compiles Trojita (and *especially* when someone makes any
change), the tests must be run.
- The build should abort when someone attempts to build with KDE and Qt5
support at t
15 matches
Mail list logo
