DOS Japanese crap

[tgies]

Does anyone have any idea how to get cmd.exe in Windows XP to correctly handle programs that want to output text in a Japanese character encoding?

I'm trying to use this 6502 assembler and it's giving me messages which I can't read and can't even determine the encoding of because it's munged them so badly. It's kind of annoying.

I'm aware this is

Comments

[disastrouscode]

have you tried piping it to a text file and then opening it with a browser/advanced text editor with encoding options?

[tgies]

Yes, that's the first thing I did, but it still winds up getting munged. I've actually done a lot of work programming internationalization stuff, so I can usually identify encoding at a glance by looking at the binary or ASCII-garbage representation of the text, but this looks almost random and has a lot of features I can't identify and it's just completely ridiculous. If anyone knows what the hell DOS does to the internal representation of the encoding that screws it up so badly, that would also help me a lot.

[njyoder]

You checked if it's Unicode or Shift-JIS? I'm assuming you have all the relevant fonts installed, and if not, read daev's comment. I'm not sure if this would work, but the unix 'file' utility (also available with cygwin and possibly GNU-Win32) may be able to identify the encoding.

Do you have a link to the software? I may be able to identify it myself.

[tgies]

I tried file, yeah. Examining it more closely, it turns out it's Shift-JIS shifted three bits to the left and with the rightmost three bits replaced with 101. So yeah, it's getting totally munged somewhere along the line. Software works fine on a Japanese MS-DOS machine, so I want to blame Windows for this.

[autodidacticphd]

yeah, i'd blame the concept of regioncoding and MS for participating. though, if it is DOS old, it may have just been a matter of convinience to make the system opperate differently at the time.