Age | Commit message (Collapse) | Author |
|
AmigaOS4 core devs tell that the special directory "SObjs" (that holds
the shared object files) should only be used by the end user (games,
apps, etc.) but not by projects that are to be build.
This patch fixes it.
svn-id: r51269
|
|
svn-id: r51057
|
|
* Added a yes/no variable _unix to configure, which controls when
-DUNIX is added to DEFINES
* Enable SEQ MIDI via _seq_midi by default on UNIX type systems,
except for those which override that.
* Switch SEQ MIDI code to check #define USE_SEQ_MIDI
(alternatively, we could compile it only conditionally...)
svn-id: r51055
|
|
* Remove unused add_to_config_mk_if_no()
* Add two new functions: define_in_config_h_if_yes() and
define_in_config_if_yes(). These make it much more convenient
to #define something in config.h, or in both config.h and config.mk
simultaneously.
* Convert configure code to use the two new functions, were appropriate.
* Add preliminary USE_SEQ_MIDI support, as discussed on scummvm-devel.
This is incomplete as it does not actually detect anything, it just
allows turning off _seq support.
svn-id: r51054
|
|
svn-id: r50982
|
|
This removes the final private library - we should be completely above
board now.
svn-id: r50677
|
|
Conversion of the old build system is incomplete. Binaries created
using this new build system support most likely will not run!
More work is needed, but the work done so far should be enough
to get basic NDS support on our buildbot working.
svn-id: r50672
|
|
svn-id: r50670
|
|
arm-*-linux-androideabi, since this is what the upcoming gcc-4.6
change uses.
Work around Android 2.0 and newer dynamic linker bug by explicitly
linking plugins against the core libscummvm.so.
Mark Android packages as preferring to be on external storage.
Move -lm from LDFLAGS to LIBS.
svn-id: r50665
|
|
svn-id: r50582
|
|
svn-id: r50556
|
|
In recent discussions on -devel it turned out, that this
feature is rather superfluous and instead we should rather
implement a proper error reporting in our GUI.
I also removed the dependency on iconv along with this.
svn-id: r50335
|
|
--enable-release was specified on configure run.
svn-id: r50246
|
|
USE_TRANSLATION is set but USE_TERMCONV is not.
svn-id: r49941
|
|
svn-id: r49903
|
|
This removes them from the old scripts in /backends/platform/gp2x/build/ and puts them in the right place.
svn-id: r49892
|
|
(ENABLE_)TRANSLATION, DETECTLANG and TERMCONV.
svn-id: r49885
|
|
svn-id: r49884
|
|
svn-id: r49882
|
|
with buildbot.
This removes them from the old scripts in /backends/platform/gp2xwiz/build/ and puts them in the right place as make targets.
svn-id: r49877
|
|
svn-id: r49868
|
|
Based on patch #2903830: "Updated Translation Prototype" by alexbevi
which in turn is based on patch #1739965 by jvprat.
Currently it builds all translations right into ScummVM. Once the
feature will be accepted more widely, i.e. more translations will
pop up, it will be trivial to move translation strings to external
file.
Finished translation: Russian
Unfinished translation: Hungarian
Things which are nice to do:
- Language code -> language mapping for more user friendness
- Specifying fonts to be used with language
- Updating of interface language without restart. It will require
moving of much code to reflowLayout() methods for each dialog
The .po files must be in single byte encodings. I.e. no support
for Unicode.
svn-id: r49759
|
|
svn-id: r49504
|
|
svn-id: r49449
|
|
svn-id: r49167
|
|
Configure reflects that now.
svn-id: r48996
|
|
With these changes, we are a bit closer to how autoconf works, the
de-facto standard for configure scripts. The changes make it easier
for packagers to modify where ScummVM install things (e.g. the
Maemo port might benefit from this). It is now also trivial to
override certain paths when invoking make, e.g. you can now do
make install prefix=/opt
and also datadir, mandir etc. will reflect the change.
svn-id: r48989
|
|
svn-id: r48980
|
|
configure.
svn-id: r48977
|
|
svn-id: r48969
|
|
svn-id: r48968
|
|
svn-id: r48803
|
|
capitalization
svn-id: r48770
|
|
svn-id: r48688
|
|
Since no porter complained about these during the last release,
this should be safe. In any case it would be trivial to put them
back, but that is only acceptable if an explanatory comment is
included.
svn-id: r48676
|
|
svn-id: r48672
|
|
svn-id: r48671
|
|
svn-id: r48669
|
|
svn-id: r48664
|
|
svn-id: r48632
|
|
svn-id: r48619
|
|
compile, the test code
svn-id: r48603
|
|
svn-id: r48602
|
|
and adapt calling code accordingly
svn-id: r48601
|
|
*require* a non-trivial LDFLAGS & CXXFLAGS; applying same fix to the memalign test code
svn-id: r48600
|
|
svn-id: r48599
|
|
svn-id: r48597
|
|
svn-id: r48596
|
|
svn-id: r48595
|
|
svn-id: r48594
|