Re: [Groff] grops and Unicode, CVS test.

[Wartan Hachaturow]

On 5/16/05, Werner LEMBERG <address@hidden> wrote:

> The CVS groff version currently supports Cyrillic for html and utf8
> also.  utf8 should be just fine, html should be fine too, but you get
..
> Please test the CVS and report problems.

So here I am, with 20050518 CVS snapshot.

try.ms file contains a ".PP" followed by a "Hello, world!" in Russian, KOI8-R
encoding.

Now, I suppose this:
iconv -fKOI8-R -tUTF-8 ./try.ms | groff -Tutf8 -mms | iconv -fUTF-8 -tKOI8-R
should work as expected, right? (As far as I understand, devutf8 is essentially
a one-to-one mapper when it comes to characters?)
This doesn't work:
<standard input>:2: warning: invalid input character code 159
<standard input>:2: warning: invalid input character code 128
<standard input>:2: warning: invalid input character code 130
<standard input>:2: warning: invalid input character code 128
     iconv: illegal input sequence at position 11

The output, when saved, actually doesn't look like proper cyrillic UTF-8 when I 
`cat` it in uxterm.

-- 
Regards, Wartan.