[OE-core] [PATCH 2/2] vim: Update to 8.1.1198

[richard.purdie at linuxfoundation.org]

On Tue, 2019-04-23 at 18:01 -0400, Tom Rini wrote:
> On Tue, Apr 23, 2019 at 04:02:49PM +0100, Richard Purdie wrote:
> > On Sun, 2019-04-21 at 13:57 -0400, Tom Rini wrote:
> > > This brings us to the current version of vim.
> > > 
> > > Signed-off-by: Tom Rini <trini at konsulko.com>
> > > ---
> > >  meta/recipes-support/vim/{vim-tiny_8.1.1017.bb => vim-
> > > tiny_8.1.1198.bb} | 0
> > >  meta/recipes-support/vim/{vim_8.1.1017.bb =>
> > > vim_8.1.1198.bb}           | 2 +-
> > >  2 files changed, 1 insertion(+), 1 deletion(-)
> > >  rename meta/recipes-support/vim/{vim-tiny_8.1.1017.bb => vim-
> > > tiny_8.1.1198.bb} (100%)
> > >  rename meta/recipes-support/vim/{vim_8.1.1017.bb => vim_8.1.1198.bb}
> > > (98%)
> > 
> > https://autobuilder.yoctoproject.org/typhoon/#/builders/40/builds/527
> 
> Does the current version even work in this case?  Digging down into
> https://autobuilder.yoctoproject.org/typhoon/#/builders/40/builds/527/steps/7/logs/step1b
> there's a certain level of "how does that even work today then?" that
> springs to mind.
> 
> I need a few more details to know how to reproduce this locally as I
> don't see whatever must be in site.conf/etc to turn regular poky into
> no-x11.  The errors are also around iconv stuff so maybe nls is also
> removed here?  Thanks!

perhaps simpler, qemux86-world also failed:

https://autobuilder.yoctoproject.org/typhoon/#/builders/52/builds/515

with:

| /home/pokybuild/yocto-worker/qemux86-world/build/build/tmp/work/i586-
poky-linux/vim/8.1.1198-r0/recipe-sysroot-native/usr/bin/msgfmt: Cannot
convert from "ISO-8859-2" to "UTF-8". msgfmt relies on iconv(), and
iconv() does not support this conversion.
| make[2]: *** [gvim.desktop] Error 1

So perhaps there is host contamination and this works one some build
machine setups but not more minimal ones?

The build configuration hasn't changed, just the new version upgrade
and the other patch. I suppose its possible it depends which of our
workers its built upon but that seems unlikely for two world failures
(no-x11 is also a world failure).

Cheers,

Richard