Re: Unicode console patch

Guest section DW (dwguest@win.tue.nl)
Tue, 2 Feb 1999 01:59:26 +0100 (MET)


From: "H. Peter Anvin" <hpa@transmeta.com>

Well, as I said -- I'm trying to revisit all of these, especially the
currently broken handling of G0, G1, G2, G3 and the other ISO 2022
issues.

Just as a proof that the current Linux console is broken,

Well, broken - it implements a vt100-like terminal,
with some features that remind one of ISO 2022 but
are not the same.

The Linux version is documented in the kbd-FAQ.
The ISO 2022 version is documented in the charsets(4) man page.

It would have been nicer if they had been the same.
As it is, I am not quite sure whether many people need strict
ISO 2022 compliance. (And this strict compliance can be
obtained in user space by an application interpreting
the received escape sequences and doing the appropriate
corresponding Linux things.)

Linux considers 8-bit character sets to be atomic,
rather than composed of two 7-bit
character sets, which is the ISO 2022 model.

Yes.
I remember from the good old times that I had some disagreements
with Mike Jagdis about this stuff.

The question is whether one wants a console that is completely
flexible, in principle also able to handle EBCDIC or Code Page 1234,
or do we want to stipulate in advance that the console must
implement ISO 2022 ? For Linux, where many people live close
to a DOS world, the single character set model may make it easier
to interface with DOS things.

I am not really voicing an opinion now, just saying that when there
are so many standards to choose between, it is not immediately clear
that we want to pick ISO 2022, which is semi-superseded by ISO 4873
and will be made obsolete by ISO 10646.

Andries

-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.rutgers.edu
Please read the FAQ at http://www.tux.org/lkml/