Re: lynx-dev Re: lynx should respect LANG

[Klaus Weide]

On Wed, 24 May 2000, Henry Nelson wrote:

> > or ~/.lynxrc but it is not true that there is no need to touch 
> > userdefs.h at all just "for Japanese users."
> 
> What define(s) do Japanese users _have_ to set in userdefs.h?
> 
> > - Japanese input did not work correctly with Lynx2.8.2.
> >   It is Lynx2.8.3 that one can input Japanese or delete
> >   Japanese character correctly.

You are talking about SUPPORT_MULTIBYTE_EDIT, see userdefs.h.

> You mean the built-in editor?  Well, I admit it's not perfect, but
> that is another topic.  Furthermore, it's not just Japanese.  No amount
> of munging of lynx.cfg is going to solve those problems for you.

SUPPORT_MULTIBYTE_EDIT, despite its name, probably only works right
for the two Japanese encodings.  It certainly doesn't for UTF-8.
And the comment in userdefs.h admits that it doesn't "fully support"
even the Japanese encodings ('JIS X0201 Kana').

> > - Japanese messages could not be displayed with Lynx2.8.1
> >   It is Lynx2.8.2 that uses *.po files.  And, yes, I put
> 
> You're talking about NLS here.  If you really want to do _that_ right,
> I still think the wrapper is the best way to go.  With a single binary
> you could easily have support for any number of messages catalogues,
> even offer them as choices in a menu.  With NLS you're mostly interested
> in TEXTDOMAIN and TEXTDOMAINDIR, not LANG, as I am sure you are aware.

Well, *I* am not aware of that.  Can you explain, please?  Why would
you mess with TEXTDOMAIN*, how do you do it, and why is that
preferable?

   Klaus


; To UNSUBSCRIBE: Send "unsubscribe lynx-dev" to address@hidden