Openembedded Core Discussions
 help / color / mirror / Atom feed
* C.UTF-8 in glibc
@ 2023-01-01 18:32 Alex Kiernan
  2023-01-02 16:28 ` [OE-core] " Richard Purdie
  0 siblings, 1 reply; 4+ messages in thread
From: Alex Kiernan @ 2023-01-01 18:32 UTC (permalink / raw)
  To: Patches and discussions about the oe-core layer

glibc got an official C.UTF-8 in 2.35, but we don't appear to install
it by default... if I add "c" to IMAGE_LINGUAS then it gets installed,
but it feels like this ought to be default? Or even stronger and have
it separate from IMAGE_LINGUAS so even if that's set to "" you still
get it?

I tripped over it because I was trying to get the ostree ptest suite
to pass and it ended up with a time formatting failing a test because
it wasn't running in C.UTF-8.

-- 
Alex Kiernan


^ permalink raw reply	[flat|nested] 4+ messages in thread

* Re: [OE-core] C.UTF-8 in glibc
  2023-01-01 18:32 C.UTF-8 in glibc Alex Kiernan
@ 2023-01-02 16:28 ` Richard Purdie
  2023-01-02 16:40   ` Alex Kiernan
  0 siblings, 1 reply; 4+ messages in thread
From: Richard Purdie @ 2023-01-02 16:28 UTC (permalink / raw)
  To: Alex Kiernan, Patches and discussions about the oe-core layer

On Sun, 2023-01-01 at 18:32 +0000, Alex Kiernan wrote:
> glibc got an official C.UTF-8 in 2.35, but we don't appear to install
> it by default... if I add "c" to IMAGE_LINGUAS then it gets installed,
> but it feels like this ought to be default? Or even stronger and have
> it separate from IMAGE_LINGUAS so even if that's set to "" you still
> get it?
> 
> I tripped over it because I was trying to get the ostree ptest suite
> to pass and it ended up with a time formatting failing a test because
> it wasn't running in C.UTF-8.

Making it even stronger might be tricky given how our variables work
but we definitely should have that in the default locales list.

Cheers,

Richard


^ permalink raw reply	[flat|nested] 4+ messages in thread

* Re: [OE-core] C.UTF-8 in glibc
  2023-01-02 16:28 ` [OE-core] " Richard Purdie
@ 2023-01-02 16:40   ` Alex Kiernan
  2023-01-03 19:18     ` Alexandre Belloni
  0 siblings, 1 reply; 4+ messages in thread
From: Alex Kiernan @ 2023-01-02 16:40 UTC (permalink / raw)
  To: Richard Purdie; +Cc: Patches and discussions about the oe-core layer

On Mon, Jan 2, 2023 at 4:28 PM Richard Purdie
<richard.purdie@linuxfoundation.org> wrote:
>
> On Sun, 2023-01-01 at 18:32 +0000, Alex Kiernan wrote:
> > glibc got an official C.UTF-8 in 2.35, but we don't appear to install
> > it by default... if I add "c" to IMAGE_LINGUAS then it gets installed,
> > but it feels like this ought to be default? Or even stronger and have
> > it separate from IMAGE_LINGUAS so even if that's set to "" you still
> > get it?
> >
> > I tripped over it because I was trying to get the ostree ptest suite
> > to pass and it ended up with a time formatting failing a test because
> > it wasn't running in C.UTF-8.
>
> Making it even stronger might be tricky given how our variables work
> but we definitely should have that in the default locales list.
>

My original thinking on trying to make it stronger was I expect you
really want this, but given everyone's managed so far without, I guess
if they need it now they can add it back in if they're overriding
IMAGE_LINGUAS.

Just thinking about this more, we can't specify it for musl, the
obvious way would be a glibc specific append, which would then mean
that setting IMAGE_LINGUAS = "" you still get it, which doesn't
immediately feel right (plus the general wish to avoid append in
core), which I guess means it needs something with more trickery...
I'll go play.

-- 
Alex Kiernan


^ permalink raw reply	[flat|nested] 4+ messages in thread

* Re: [OE-core] C.UTF-8 in glibc
  2023-01-02 16:40   ` Alex Kiernan
@ 2023-01-03 19:18     ` Alexandre Belloni
  0 siblings, 0 replies; 4+ messages in thread
From: Alexandre Belloni @ 2023-01-03 19:18 UTC (permalink / raw)
  To: Alex Kiernan
  Cc: Richard Purdie, Patches and discussions about the oe-core layer

On 02/01/2023 16:40:20+0000, Alex Kiernan wrote:
> On Mon, Jan 2, 2023 at 4:28 PM Richard Purdie
> <richard.purdie@linuxfoundation.org> wrote:
> >
> > On Sun, 2023-01-01 at 18:32 +0000, Alex Kiernan wrote:
> > > glibc got an official C.UTF-8 in 2.35, but we don't appear to install
> > > it by default... if I add "c" to IMAGE_LINGUAS then it gets installed,
> > > but it feels like this ought to be default? Or even stronger and have
> > > it separate from IMAGE_LINGUAS so even if that's set to "" you still
> > > get it?
> > >
> > > I tripped over it because I was trying to get the ostree ptest suite
> > > to pass and it ended up with a time formatting failing a test because
> > > it wasn't running in C.UTF-8.
> >
> > Making it even stronger might be tricky given how our variables work
> > but we definitely should have that in the default locales list.
> >
> 
> My original thinking on trying to make it stronger was I expect you
> really want this, but given everyone's managed so far without, I guess
> if they need it now they can add it back in if they're overriding
> IMAGE_LINGUAS.
> 
> Just thinking about this more, we can't specify it for musl, the
> obvious way would be a glibc specific append, which would then mean
> that setting IMAGE_LINGUAS = "" you still get it, which doesn't
> immediately feel right (plus the general wish to avoid append in
> core), which I guess means it needs something with more trickery...
> I'll go play.
> 

Maybe:

IMAGE_LINGUAS_C:libc-glibc = "c"
IMAGE_LINGUAS = "${IMAGE_LINGUAS_C}"

> -- 
> Alex Kiernan

> 
> -=-=-=-=-=-=-=-=-=-=-=-
> Links: You receive all messages sent to this group.
> View/Reply Online (#175323): https://lists.openembedded.org/g/openembedded-core/message/175323
> Mute This Topic: https://lists.openembedded.org/mt/95993257/3617179
> Group Owner: openembedded-core+owner@lists.openembedded.org
> Unsubscribe: https://lists.openembedded.org/g/openembedded-core/unsub [alexandre.belloni@bootlin.com]
> -=-=-=-=-=-=-=-=-=-=-=-
> 


-- 
Alexandre Belloni, co-owner and COO, Bootlin
Embedded Linux and Kernel engineering
https://bootlin.com


^ permalink raw reply	[flat|nested] 4+ messages in thread

end of thread, other threads:[~2023-01-03 19:18 UTC | newest]

Thread overview: 4+ messages (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
2023-01-01 18:32 C.UTF-8 in glibc Alex Kiernan
2023-01-02 16:28 ` [OE-core] " Richard Purdie
2023-01-02 16:40   ` Alex Kiernan
2023-01-03 19:18     ` Alexandre Belloni

This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox