[Comerrers] Welcome and proposal

[Love]

Jeffrey Hutzelman <jhutz@cmu.edu> writes:

> Actually, I think the right answer is that there is a separate domain for
> each error table, derived algorithmically from the name of the error
> package.  The com_err library can then use dgettext() or a similar
> interface to look up translations in the appropriate domain.

So you mean that the domain for the kerberos5 package name will be
something like this, where number2name is the reverse of name2number() in
compile_et, awk/perl-foo

        "et_" | number2name(error_code)

Guess that will work. None will have a clue that message catalog the file
is for, but then I guess you should use you package system for that.

So then a function that takes a domain needs to be passed in (dgettext
instead of gettext) to com_err.

Still I would think I would be nice if the error message could be prefixed
with package that is generating the error so you don't need to duplicate
this information in the error string. But to be able to do this, we need to
break api/abi, and I don't that is a good idea.

Love