There is already a framework-phonon module-set with the right entry for
phonon (phonon/phonon instead of phonon, which will build all
phonon-related modules).
Log said "make: *** No rule to make target `Experimental'. Stop."
Script said:
Use of uninitialized value $numTests in numeric gt (>) at /d/kde/inst/kde4.8/bin/kdesrc-build line 2420, <$logf> line 2.
Now it says:
Some tests failed for konsole, consult [...]/log/2011-12-27-06/konsole/test-results.log for info
If we create the full source path the rest of the script will get
confused and think there's existing source there already (since I've
been too lazy to code a check if the source dir is empty or not).
Although I don't think this was the exact cause of the
kde-build-metadata problems (but instead the moved SCM type setting),
this does at least have the effect of moving an update under the 'Script
started processing' message, which is probably where it should be.
Having the proper SCM type ensures that the &filter subroutine in
expandXMLModules actually sets the repository value for our
kde-build-metadata module.
This unfortunately isn't really the best way to do it (in retrospect
this should be handled in ensure_projects_xml_present since we want the
metadata whenever we'd be downloading the XML anyways). Plus it really
probably isn't necessary to subclass KDEProjectUpdate.
However, the perfect is the enemy of the good and this seems to be
working for me now and I will be away on holidays for a week or so, so
best to get it in now, imperfect as it is.
Let me know if this doesn't work, also keep in mind the
'kde-build-metadata' repo is open for business if there's other modules
that should be ignored! ;)
BUG:288378
At some point this will mean ignoring modules that are active but which
we shouldn't do anything with, and maybe even real dependency resolution
someday.
Note that despite the comment, forcing kde-build-metadata to only be in
the build phase doesn't seem to work yet.
CCBUG:288378
This is kind of a workaround for the "kdev-www is in the default
kdevelop build but doesn't have a build system fiasco". This really
needs support from the build software repository to be fixed finally,
and that requires me to design and actually implement such support. :(
CCBUG:288378
This is mostly an excuse to get rid of the last usage of
module_uses_cmake, and therefore get rid of module_uses_cmake
completely.
If I were designing this from scratch it would probably be an
integration point for the build system where the build system object is
passed the context and told to make the environment ready for its
module...
This actually is already present, but is checked for by looking for
cmake buildsystems directly, whereas this determination should really
belong to the build system (now that such an entity is defined).
Note that it's not supported with l10n from what I understand, even
though it uses CMake... though this is just me porting the old behavior,
it may work fine now.
Apparently Perl has included (since version 5.0, no less) a module that
handles almost the exact case split_quoted_on_whitespace exists for.
The impetus for this is that the '${MODULE}' syntax for dest-dir is
apparently broken due to problems with the mixture of braces and quotes
in split_quoted_on_whitespace.
This change does not break the test suite (but does add a relevant test)
and does seem to fix the issue with strings with braces.
CCMAIL:rakuco@FreeBSD.org
This commit should make the 'prefix' option expand this variable to the
module's name, so one can easily install each module into a different
directory.
Reviewed-by: mpyne
Turns out I removed the auto-svn-switching feature by mistake, noted by
dfaure. Too bad it took so long for me to dig through the git history,
this turned out to be much easier to find than I'd thought it would be.
CCMAIL:faure@kde.org
Prior to this, safe_make was handling a lot of different things that
unnecessarily cluttered it up (e.g. determining whether it was safe to
use make-options or not based solely on the args parameter, determining
whether to use make-install-prefix, etc.)
Now the actual callers of safe_make (where it is already known whether
or not to use make-options or make-install-prefix) pass that kind of
information in.
Even with an increase in code at the calling sites and the more
extensive comments, the simplifications in safe_make allow for an
overall decrease in code size.
2 changes related to handling --pretend + no-source-dir.
1. svnInfo:
If we can't change to the source dir before running svn info that should
be an error, so treat it as such.
Except, in pretend mode that's OK... so just check for that and return
some nonsense to keep things going.
2. Qt configure:
Again if there's no source dir due to pretending, just move on.
In a previous commit I made a "large refactoring" that involved using
mixin classes to decide at runtime what build system or scm handler to
use for a given module.
In retrospect, using what is essentially a static class member to handle
a per-object decision was not my most impressive moment ever... this is
easily fixed by having specific scm() and buildSystem() members for each
Module instance, but does involve yet-more modification of code in lots
of places. :-/
Should be mostly the same as far as operation goes, but in this commit I
introduce a build system mixin similar to the scm handler mixin. There's
still more code that should be moved but we're getting closer to having
a simple build process where all the scm-specific stuff doesn't clutter
the main code with the equivalent of #ifdefs everywhere.
Add specific update classes (including a separate one for l10n support,
which probably doesn't work still...)
Still need to go further to split out this into easy-to-implement blocks
but the design is looking better.
$! should be treated like C's errno, which you can basically assume has
its value reset between libc or system calls. So, save the return value
before doing too much other work (to include I/O).
Now you can specify the private key to be passed to your SSH agent if
you forgot to run that before running kdesrc-build.
Proof-of-concept patch graciously provided by Ralf Jung.
BUG:284071
FIXED-IN:1.14.2
Michael Pyne
Eike Hein
Raphael Kubo da Costa
David Faure
Andre Woebbeking