Re: [PATCH] RFC/RFA fix exec-charset problems - take 2

[Mark Mitchell <mark at codesourcery dot com>]

Eric Christopher wrote:

> Here's a second take at the exec-charset problems.
>
> I've got a set of testcases to go with these, but I'm not sure what to
> do with them since they require that exec-charset translate to something
> like say IBM1047 and iconv doesn't necessarily exist on all platforms so
> I don't want to add things that'll just fail on some hosts. However, if
> this is an ok solution then I can do that. :)
>
> Mark: If this is, again, not acceptable for 3.4 then I'll submit a patch
> to turn off the option on the branch as it isn't useful without it.
>
> Thoughts? OK?
The C++ changes look OK to me, although I haven't tried to figure out if 
the switching is happening in the right places, exactly.  I'll take your 
word for it.

I don't thin we want tests that will fail on some hosts .  You should 
either make it so that they are skipped or XFAIL'd on those hosts, or 
you should not check them in.  You might be able to rip off some of the 
dg-require machinery to create a dg-require-iconv thingy that would 
handle that.

I don't see that this is compelling for 3.4.  Is this a regression?  

Otherwise, let's hold off until 3.5.

Thanks,

--
Mark Mitchell
CodeSourcery, LLC
(916) 791-8304
mark@codesourcery.com