#8458 closed patch (fixed)
Possible Kyra fix
| Reported by: | eriktorbjorn | Owned by: | cyxx |
|---|---|---|---|
| Priority: | normal | Component: | Engine: Kyra |
| Version: | Keywords: | ||
| Cc: | Game: | Kyrandia 1 |
Description
Lately, the Kyrandia 1 intro has been crashing for me in drawShape() because assert(shapeData) failed.
This is because when seq_setPanPages() is called to create the hand shapes, the 'data' pointer points to the following data in my copy of the game:
03 00 10 00 00 00 59 0a 00 00 44 14 00 00 3a 1e
"03 00" is the number of shapes. The offset is read by READ_LE_UINT16(data + 2 + shape * 2), so the offsets are "10 00", "00 00", and "59 0a". And that's the problem: The second offset is 0, so that pointer will be NULL.
Now, it looks to me as if the offsets are actually 32-bit quantities, i.e. the offsets are "10 00 00 00", "59 0a 00 00" and "44 14 00 00", in which case the attached patch should fix it. But I don't know if that's really true even for my version of Kyrandia 1, let alone any other version.
Ticket imported from: #1280643. Ticket imported from: patches/563.
Attachments (1)
Change History (5)
by , 19 years ago
comment:1 by , 19 years ago
| Owner: | set to |
|---|
comment:2 by , 19 years ago
I just verified, in my talkie version the writing.cps file has 32 bits for each offset (or maybe 2 extra bytes for something else...). So, I suppose it's just another difference in the datafiles between cdrom and floppy versions (until now, I only tested my code on the latter).
I just committed some code to handle that in CVS, Kallak hands should be displayed now.
comment:3 by , 19 years ago
| Resolution: | → fixed |
|---|---|
| Status: | new → closed |
comment:4 by , 6 years ago
| Component: | → Engine: Kyra |
|---|---|
| Game: | → Kyrandia 1 |
Patch against a September 2 CVS snapshot