[MPlayer-cvslog] CVS: main/libmpdemux ebml.c,1.4,1.5

[Reimar Döffinger]

Hi,
> long double itself is not portable!!! the IEEE spec only specifies lower 
> bounds on the precission, i doubt that the x86 long double can be read by any 
> non x86 system with the union hack, though i didnt try ...

I'm for banning floating point, it's just too ugly *g*. Well, actually
it's just annonying that they decided to allow it in matroska (I think we
have that problem really nowhere else).

> nowhere, the format and bit ordering the cpu uses internally for its floating 
> point variables has no effect at all on the av_*2*() code, it would even work 
> if sizeof(float) != 4

Ok, saw that after reading it more closely. But the input if the int2*
function is in host endianness, and normal bswap will work, right?
Time to fix a few places in MPlayer then I guess...
Anybody volunteering to fix ebml.c, bswap.h and
libmpdemux/nuppelvideo.h?
Since internal libavcodec still is not mandatory I guess that will also
mean copying those funtions :-(.

Greetings,
Reimar Döffinger