Feature addition and bugfix to SCI resource compression tool

[agfor]

The first commit fixes incorrect offset adjustment for Freddy Pharkas resources with incorrect sizes.

The second commit adds support for compressing KQ5 and Jones in the Fast Lane audio resources.

[lordhoto]

New should throw an exception in case allocation fails, so this is probably not needed here.

The old code didn't seem to think about that, but I guess it's not a reason to add it again ;-). Same goes for lines 307+308 and 328+329.

[agfor]

I made it consistent, the old code checked some allocations and not others. I'll remove them.

[lordhoto]

That's a good thing :-).. But in this case we just do not need it, thus we can drop it at this occasion too ;-).

[lordhoto]

Wouldn't it be simpler to use the "fake" entry as proper end of file indicator?

[agfor]

The intention was to make sure not to reach the end of the file. I'll change it.

[lordhoto]

Any reason why you do not want to reach the end of the file? It seems simpler to take advantage of the marker in the resource map. Since that would be pretty much just a simple while loop.

[agfor]

There seem to be a lot of malformed files that the tool compensates for in various ways. I was being cautious in case there was no fake entry at the end in some file. It's easy enough to put it back if that turns out to be the case; for now I've switched it as you suggested.

[lordhoto]

Please use && instead of and.

[agfor]

Too much Python. I'll fix this and the "or".

[lordhoto]

Please use || instead of or.

Also please use decimal numbers for file sizes instead of hexadecimal ones.

[agfor]

This was again to be consistent with the rest of the file, specifically lines 113 & 115. Would you like me to change all of them to decimal?

[lordhoto]

I am afraid I am not really familiar with this tool. I just wanted to give some feedback, since it seems no one else bothered to do so, so I didn't know it's using this in other occasions too. Personally I think decimal numbers are better for file sizes, since it's easier to compare it to your files. So it might be a good idea to change these at the lines you mentioned too.

[lordhoto]

Why is this a warning? Shouldn't it rather be a debug output?

[agfor]

There were other "warning()"s in the file, and no debugs, so I just went with consistency, although the others are actually places where the original files are malformed. I'll change it.

[agfor]

I forgot (it's been a while since I wrote this) -- the is no "debug()", only "warning()" and "print()". "warning" just prepends "WARNING" to the string. I'll change it to "print()".

[lordhoto]

Yes, using print is probably what you want.

[lordhoto]

Warning seems odd here too, it should probably use something just for status output like print. Also I think we automatically add new lines.

[agfor]

Consistency again - the other "warning()"s and "print()"s have newlines. I'll change it to "print()". Do you want me to remove all the newlines or leave them?

[lordhoto]

I checked up again, it seems our tools code doesn't add newlines automatically. I remember it differently. So sorry about that, just keep the new lines.

[bluegr]

Thanks for updating the tool :) The changes you added should help compressing those huge raw audio files in KQ5 and Jones.

After lordhoto's comments and suggestions (thanks!), this looks to be in a good shape to be merged. Unfortunately, I can't test the changes, as I'm on vacations right now :( So, unless someone else can test this, I can test and merge it in about 10 days, once I get back.

[agfor]

Ten days sounds great to me; the patch sat on the tracker for 3 months so far :). If you're trying to make the ultimate ScummVM USB stick, the 50mb saved on KQ5 counts! Jones? Well, it wasn't any extra work, anyway.

[agfor]

Sorry, misclick.

[bluegr]

Great work :) Works beautifully, thanks!