[ Thread Index |
Date Index
| More lists.liballeg.org/allegro-developers Archives
]
On Sun April 19 2009, Evert Glebbeek wrote:
> On 19 Apr 2009, at 11:52 , Milan Mimica wrote:
> > Just wanted to point out that in current SVN, most (but not all)
> > examples that
> > load a file fail at it.
> >
> > $ ./ex_stream_file Rear_Center.wav
> > Voice created.
> > Mixer created.
> > Assert failed at line 76 of /home/mimix/Documents/4.9/src/file.c
> > Aborted
> >
> > $ ./ex_gldepth
> > $ cat allegro.log | grep Error
> > Error loading `data/fixed_font.tga'
> >
> >
> > Examples that use mysha seem to work though.
>
> It's probably related to the assertions I'm getting:
>
> $ ./ex_bitmap
> Assert failed at line 13 of /Users/eglebbk/Program/Allegro5/4.9/src/
> file.c
> Abort trap
>
> This assertion triggers when al_fopen() is called while
> "al_get_new_file_interface()" returns NULL. I'm not sure what that
> means, but presumably it means that al_fopen() is called before
> Allegro has been initialised. Relevant backtrace from gdb:
>
> #5 0x000fd22e in al_fopen (path=0x153b6c "allegro5.cfg",
> mode=0x152b36 "r") at /Users/eglebbk/Program/Allegro5/4.9/src/file.c:13
> #6 0x000e4a40 in al_load_config_file (filename=0x153b6c
> "allegro5.cfg") at /Users/eglebbk/Program/Allegro5/4.9/src/config.c:343
> #7 0x001145e8 in read_allegro_cfg () at /Users/eglebbk/Program/
> Allegro5/4.9/src/system_new.c:134
>
> I think Thomas' code had some logic in place to make the stdio driver
> work as default even before Allegro was initialised, but I'm not sure.
> I'm also not sure what change exactly broke this. Peter?
Yup. Allegro needs the api itself before allegro itself initializes, so it
better damn well work. The main vtable is initialized at compile time to the
stdio driver to make things just work.
> Evert
>
--
Thomas Fjellstrom
tfjellstrom@xxxxxxxxxx