atlas0fd00m

this is really doing well. the primary problems on i386 right now have to do with multifunction codeblocks and codeblocks overlapping for libc (there's a thing where jcc over 1...

yay, merged in master and commented out setLevel(DEBUG) line in switchcase.py, and we're passing :)

@rakuy0 , [210408-switchcase_020045ca.zip](https://github.com/vivisect/vivisect/files/6282945/210408-switchcase_020045ca.zip) one of the "nicer" long annoying graphs for switchcase. currently takes 45 seconds to get a working path to the switchcase. from chown:0x20045ca

@williballenthin @rakuy0 @invisig0th @mr-tz @uxmal @Ana06 @bat-serjo @upside2 @chompie-vertex @vEpiphyte @evil-e i'd like your input on something. we seem to have ended up with two distinct pathing methods to carve...

TODO: improve performance TODO: cleanup unittests TODO: code review TODO: merge this sucker, it's long in the tooth.

comparing against the vendor's own disassembly, we're identical aside from reg-range operands (affecting a couple instructions, and which won't matter unless we're trying to run the output through an assembler.

hmmm... actually, i now recall this was also intended to allow for the resolution of C++ demangled names with `(stuff)`. may need a little more refining.

hmmm.... gotta rework these.... ``` vw.setMeta('FileBytes', v_parsers.compressBytes(byts))``` since that overwrites that metadata for each file loaded. filemeta, perhaps? or is it time to move to loading the file as a...

TODO: ensure analysis modules can be "friendly" when run multiple times, or restrict to only new files TODO: wait until larger release since this is a significant change.

todo: * make sure, during loading (ie. matching of exports to imports), the Call Graph is updated appropriately