tech-userlevel archive

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index][Old Index]

Re: curses vs non-ASCII

>> "each octet is one character" semantic I want for this application.
> It's a long time since I tought the Perl Curses module to correctly deal wit$

I'm not sure how usefully describe which flavour of curses I'm using
except "the one that comes with 5.2".  Is there something I can look at
that will tell me?

Looking at addbytes.c (which is where all the add-string functions seem
to end up), I see nothing at all to prevent throwing the octet string
at mbrtowc, I think it was, except building without HAVE_WCHAR.

I've since done some build attempts, and this is pretty ridiculous.
First I tried to build with CITRUS=no, to make the library backing the
i18n stuff go away entirely.  That failed - something failed to link (I
forget what).  Then I tried to build with RUNE=no, to make libc's
multibyte support stub itself out despite citrus's presence.  Something
else failed to link (again, I forget what).  Then I tried
DISABLE_WCHAR=yes, to make libcurses not use the wide character stuff.
Then vi failed to link.  Now I'm trying DISABLE_WCHAR=yes (for
libcurses) and USE_WIDECHAR=no (for vi).  If I can't get this to work
within at most a few more iterations I think I'll go back to CITRUS=no
(because that's really what I want here, to make all wide-character
goop go away) and wade in with a code machete on anything that gets
upset by that.

It really seems to me it shouldn't be this hard to use a single-byte
character set like 8859-1 or KOI-8 with curses.  (It's still possible
all I need to do is set something in the environment, but I have no
idea what; the documentation is a maze of pointers without the content
I need and my attempts to USTL foundered on the maze of pointers in the
code.  I probably could dig through with debuggers and figure the code
out, but if I have to use a debugger and the source tree to figure it
out, something is pretty badly broken, even if only the doc.  And the
list hasn't helped any, so apparently nobody who's on the list and
interested in saying anything has managed to make it work either.  If I
thought it would produce anything but "closed - EOL, wontfix" I'd file
a PR.)

/~\ The ASCII				  Mouse
\ / Ribbon Campaign
 X  Against HTML
/ \ Email!	     7D C8 61 52 5D E7 2D 39  4E F1 31 3E E8 B3 27 4B

Home | Main Index | Thread Index | Old Index