These are the things that are the patches I'm applying:
*Modify getPropertyNames() so that it returns mixed case variable names (parser.CaseInsensitiveHashMap.java:73)
Changes to CaseInsensitiveHashMap will probably break using saved files from Character Tool in Map Tool. These will probably break the drag and drop from Init Tool also. So I'd like a chance to fix these problems before it is final. Will that be possible?
Ouch! Thanks for pointing that out! I don't have either of those checked out so there were no obvious dependencies for me to notice when I did the References check...
I can leave the existing code as-is and simply add getPropertyNamesRaw()
or something similar that preserves the case-sensitivity. Then the MTscript language can use that instead, perhaps even adding the corresponding MTscript macro function as well? (I don't like bloating the MTscript API unnecessarily, but I'm not sure it's a big deal.)
Does that work for you?
*Assets in RPToks are still being written out as MIME64 encoded text instead of as binary images (maptool.util.PersistenceUtil.java:453)
*Fix for RPToks ballooning in size by not merging old assets from file into new file (unless requested?) (rplib.lib.io.PackedFile.java:222-251)
Are these two what you're referring to for the dnd functionality?
I'll do some more extensive testing of the saveAssets() and loadAssets() in the PersistenceUtil class, including backward compatibility of dnd data.
For the duplicate assets, I can't see how that would affect CT or IT. Basically, when MT writes out an RPTOK, it preserves the original assets and then writes any new ones. Over time the RPTOK becomes very large and yet the assets inside are not being referenced elsewhere in the RPTOK. Do you see a problem with CT or IT in that regard?
Ooh, I haven't looked or thought about that in weeks - you may want to take a careful look at it before committing it to the final. It wasn't in a "final" stage at last look.
Oh, okay. Thanks for the heads-up.