NEW VERSION OF MULE STUFF: Very close to working on all platforms

Mostly I've done a bunch of cleanup work. I also fixed the various bugs pointed out by Fabrice, Daniel Pittman, and others, although not yet Matej's, which are harder. It definitely should work on Windows w/Mule [probably without as well], and Cygwin w/o Mule. Probably on other platforms as well. One problem I've noticed so far is with Cygwin w/Mule: you get a crash at startup in parse-unicode-translation-table. I don't quite know what the problem is and haven't been able to debug it yet, as the debugger keeps locking up. MS Win does the same code, though, without problems, so it almost looks like either a GCC bug or a bug in sscanf[]. Anyway, I'd love it if people on other platforms could test with and without Mule and see how [or whether] it works. Also, pdump might be working; I'm not sure. It compiles but I haven't had much chance to test the result. Another thing to note is that coding-system support is now ON by default on all platforms: but on Unix non-Mule platforms, all defaults are set to `binary', so that auto-decoding is disabled and essentially you should get behavior just like when not compiling with file-coding. Pretty soon I'm going to disable the option of compiling without file-coding. Please also test this and see how it comes out. Once I finally lick all of these configuration issues and such, I'll be tackling the following: -- some annoying bugs pointed out by Matej, e.g. not writing out the FFFE at the beginning of Unicode files. -- improving the detection scheme, both so that you get better results and it's more flexible, and especially importantly so that you don't get data loss! This is currently a BIG problem. -- The dreaded "Russian C-x" problem. some days ago i did a bunch of work getting the underlying stuff up to snuff, and I'm not so far from finishing this. Perhaps at that point XEmacs/Mule will really be usable in Windows! ben