Skip to content

HTTPS clone URL

Subversion checkout URL

You can clone with
or
.
Download ZIP
Branch: master
Fetching contributors…

Cannot retrieve contributors at this time

81 lines (61 sloc) 3.113 kB
Getting Started:
Building
-------------------
See README.build.txt
Using
---------------------
For the SDL variants of WRATH a script is created at WRATH root
called wrath-config, how to use that script is documented in
comments at the top of the script.
For the Qt variant, a .pri file is created for debug and release;
the debug .pri file is located in debug/QtLibConf.pri and
the release .pri file is located in release/QtLibConf.pri.
The .pri files specify paths and various compile options, they
do NOT add the necessary LIBS+= option to use the WRATH library,
your .pro file will need to do that as well.
Any source file of a project that uses WRATH must have their
very first include as "WRATHConfig.hpp". This is required
because WRATH uses STL as part of its API and the debug
build of WRATH adds _GLIBCXX_DEBUG to make use of debug
iterators. In particular, WRATH debug and WRATH release
are NOT binary compatible AND when using WRATH one
must not define _GLIBCXX_DEBUG yourself ever; instead rely
on WRATHConfig.hpp to do the right thing for you.
Documentation
-----------------------
WRATH has doxygen documentation complete with tutorials and overview text,
build the documentation with:
#cd docs && make
and find it at docs/doxy/html/index.html
Porting WRATH
-----------------------------
Most of WRATH's source code is standard vanilla C/C++ code.
The main issues of which to be aware are:
- GL version and GL header files. This is the NGL system. You will
need to hack ngl/Rules.mk for a new target to where the GL/GLES
header files are located In addition, you will need to set
correctly the GL/GLES version, in Makefile.settings,
the variable GL_VERSION
- Every GL implementation has bugs and limitations, the source
file src/WRATH/gl/WRATHGPUConfig.cpp implements various
functions to provide hints to WRATH on the GPU
- Fonts. WRATH relies on fontconfig to get a listing of fonts
from the system that it will use. For systems that do not
use font config you will need to populate the font database.
The functions of namespace WRATHFontDatabase is used to
populate the database (essentially just feed in filenames).
The fonts need to be loadable by Freetype AND must be
scalable vector fonts as well. Look to WRATHFontConfig as
an example of populating the database and hack/replace that
file as your target requires.
- WRATHReferenceCountedObject by default uses atomic increment
and decrement; one can use Mutex instead (see doxygen documentaion
on how) or change/port inc/WRATH/util/WRATHatomic.hpp to your platform
- WRATHThreadID and WRATHMutex are implemented via pthreads. When porting
to a new target you will likely want to change their implementation
accordingly.
- WRATHTime uses Unix <sys/time.h> to do its work; if not available
on your platform change it too.
- c_array and vecN make use of __gnu_cxx::__normal_iterator
to implement their iterator types. In a pinch one can simply
typedef the iterator type to the pointer type
Jump to Line
Something went wrong with that request. Please try again.