Re: iconv()

[Carey Evans <c.evans@xxxxxxxxxxxx>]

"Jason M. Felice" <jasonf@Baldwingroup.COM> writes:

> On Wed, Jan 05, 2000 at 11:18:28PM +1300, Carey Evans wrote:

[...]

> > At least with the glibc 2.1 iconv(), error handling is very poor.  If
> > it encounters a character it can't translate, it errors out without
> > any way to recover.
> 
> Hmm, is that a design or implementation issue?

The specification doesn't seem to be very clear on it.  IBM have a
whole lot of extra flags you can pass to the OS/400 iconv_open() to
say what to do.  The glibc docs seem to say that things will get
better, too.

> The aforementioned libiconv claims to map characters to similar-looking
> characters when it can't map them directly.

This sounds good.  Maybe libiconv with added EBCDIC could be a
prerequisite for tn5250, falling back to a static mapping, until glibc
or the system's libc catches up.

> Think think think think.... there should be no way to determine whether the
> current console font is iso-8859-1 or some other, huh?  Except for the
> locale, if they have that set, maybe?  Humm...

Well, XFree86 4.0 xterm will completely support UTF-8 output.
However, the Curses wide-character functions specify wchar_t
parameters, and there doesn't seem to be any way to find out what they
are.

As long as only single-byte output is assumed, then getting the
locale's character set and feeding that to iconv() should be good
enough.  Hmm, is there a function call to find that out?

-- 
         Carey Evans  http://home.clear.net.nz/pages/c.evans/

CONFIG_IPL_RDR
  Select this option if you want to IPL the image from a real card reader.
+---
| This is the LINUX5250 Mailing List!
| To submit a new message, send your mail to LINUX5250@midrange.com.
| To subscribe to this list send email to LINUX5250-SUB@midrange.com.
| To unsubscribe from this list send email to LINUX5250-UNSUB@midrange.com.
| Questions should be directed to the list owner/operator: david@midrange.com
+---