Re: [AD] COLORCONV_KEEP_TRANS in datafiles + cleanup

["Eric Botcazou" <ebotcazou@xxxxxxxxxx>]

> Oh, heh, I guess the new if() is quite self-explanatory, looking at the
> flag which is tested and seeing the next line..

Ok, but a comment helps when reading the patch (and explaining its
rationale).

> The reason is simple: datafiles in memory store 15bit-bitmaps as 15bit,
> and dat2s stores the memory format of datafiles, and not the on-disk
> format :)

Ha! Of course...

> Well, I was surprised by it myself, because in fact, the EOL marker (which
> again is 0xf83f for 15bit RLE sprites) does seem to be placed into the
> data area of RLE sprites, in line 729 (after applying my patch). But the
> 15bit RLE drawer obviously is smart enough to draw the sprites properly
> nevertheless. So there's a hidden feature: With RLE sprite, you can
> actually display the magenta color on the screen with Allegro :)

Ok, thanks for this detailed explanation.

--
Eric Botcazou
ebotcazou@xxxxxxxxxx