Hi Jean-Michel,

BESSOT Jean-Michel wrote on Sun, Sep 15, 2019 at 03:44:59PM +0200:

> I did take a screenshot of the uniname output, it is more talkative to 
> see a theta in place of an eacute .
> 
> The picture: https://tof.cx/image/qql3Fithe picture:

Thanks.  Let me summarize.  That output proves that

 1. Your file utf8.txt is correctly UTF-8 encoded:
    the columns "encoded as" and "name" printed by uniname(1) prove that.

 2. Your xterm(1) is configured for multibyte character handling:
    the fact that one single-width glyph each is shown for two- and
    three- byte sequences proves that.

 3. Not 100% sure, but your xterm(1) does seem to be configured for
    UTF-8: for any other encoding, it would seem unlikely that all
    the two- and three-byte dequences shown would map to single
    characters at the same time.

The most likely explanation seems to me: in some place other than
~/.Xdefaults, you may have manually configured xterm(1) or the font
system to use some weird non-Unicode-compatible font that has Greek
characters in the glyph range A0-FF.  It doesn't look like a font
intended for use with ISO-8859-7 though; maybe a font intended for
mathematical typesetting?

There doesn't seem to be a bug because this is certainly not the
default behaviour; you must have configured *something*, and whatever
it is, it doesn't look like a good idea.

So i think bugs@ is no longer the right venue for this question.

I never configured fonts, so i can't help with that.  If you fail
to find out what you did, post a short summary of what we found so
far (including the link to tof.cx) to misc@ and ask there which
files you should look at.  There may be people around on that list
who do not read bugs@.

Yours,
  Ingo

Reply via email to