grfcodec 0.9.7 test version

[Patchman]

Since alpha 57 will allow including new sound samples (although they won't be able to be used yet), here is a version of grfcodec to accomplish this.

In order to include sound samples (or any file, really), you need action 11 and the following kind of entries after it:

Code: Select all

   16 ** sounds/horn1.wav
   17 ** sounds/horn2.wav
   18 ** sounds/horn3.wav

The maximum possible size of these files is 65532 minus the length of the filename (without the directory; the name is stored so that after decoding, the extracted file will have the right name).

I'd appreciate if the interested parties could test this for a bit, before I make an official release of 0.9.7. Thanks.

[eis_os]

Since it's A LOT easier to understand large pseudo code blocks with comments I added support for comments in pseudo sprite blocks.

[Oracle]

Thanks for that comments feature, Oskar - it will be really useful when I next code stations!
Josef, I might get around to trying sound encoding briefly in the next few days. Do you just want me to check that the sounds encode/decode properly and that the patch doesn't mind them in GRF files?

[Patchman]

Josef, I might get around to trying sound encoding briefly in the next few days. Do you just want me to check that the sounds encode/decode properly and that the patch doesn't mind them in GRF files?

Well, mainly that, and now that action 11 accepts them properly and rejects non-binary files. Also that I didn't accidentally break anything else

[Patchman]

Just a heads-up, I've released grfcodec 0.9.7 at the usual place.

Changes since Oskar's version:

• Fixed bug with grfcodec occasionally trying to decode first sprite as binary include
• fixed incorrect error message while encoding if .nfo file could not be read (now it reports "Cannot read %s" instead of Creating sprites/: File exists")
• made code -Wall clean
• remove myalloc stuff

[OzTrans]

I'm just going to suggest it here ...

Would it be possible to allow those string codes (e.g. \94) found in ttdpttxt.txt to be placed in plain text strings in action-4 and then have them converted to hex by GRFCodec. That would make it simpler to define those F8xx texts in a .grf. At the same time eliminating the problem with imbedded null-bytes (\00).

allow this ...

Code: Select all

5 * 175     04 48 9F 01 00 F8 
          "\94\80\n\90  Bribe the local authority to increase your rating, at the risk "
          "of a severe penalty if caught.\n  Cost: \7f" 00

instead of having to do it this way ...

Code: Select all

5 * 175     04 48 9F 01 00 F8 
          94 80 0D 90 "  Bribe the local authority to increase your rating, at the risk "
          "of a severe penalty if caught." 0D "  Cost: " 7F 00

[Patchman]

This would not solve the problem of null bytes though.

Still, it shouldn't be too hard to copy the mkpttxt code for the backslash handling to grfcodec, but I doubt I'll find the time to work on grfcodec in the near future.