Openembedded Core Discussions
 help / color / mirror / Atom feed
From: Richard Purdie <richard.purdie@linuxfoundation.org>
To: Otavio Salvador <otavio@ossystems.com.br>
Cc: openembedded-core <openembedded-core@lists.openembedded.org>
Subject: Re: [PATCH] allarch: Allow class to be included but overridden
Date: Mon, 26 Nov 2012 11:22:10 +0000	[thread overview]
Message-ID: <1353928930.21863.57.camel@ted> (raw)
In-Reply-To: <CAP9ODKpjfvi-xZPG3+wJGVzv7n704dsSYDs-mcKrbhQQ5EvbAw@mail.gmail.com>

On Mon, 2012-11-26 at 09:19 -0200, Otavio Salvador wrote:
> On Mon, Nov 26, 2012 at 9:17 AM, Richard Purdie
> <richard.purdie@linuxfoundation.org> wrote:
> > On Mon, 2012-11-26 at 09:08 -0200, Otavio Salvador wrote:
> >> On Sun, Nov 25, 2012 at 6:23 PM, Richard Purdie
> >> <richard.purdie@linuxfoundation.org> wrote:
> >> > We have cases where we'd like to inherit this class by default but allow
> >> > special cases to override it. This change makes the code of the class
> >> > conditional on PACKAGE_ARCH remaining set to "all", allowing it to be
> >> > overridden. packagegroup usage is one case this is desirable.
> >> >
> >> > Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
> >> > ---
> >>
> >> Richard, I fail to see the usecase of packagroup. Can you clarify?
> >
> > packagegroup.bbclass previously set PACKAGE_ARCH = "all" which was
> > confusing other code. Anything using the all package arch should really
> > use allarch so we are consistent about how we set the various variables.
> >
> > Even though packagegroup.bbclass now uses allarch, we need specific
> > packagegroup packages to be able to override the class e.g. marking
> > themselves machine specific. This is why we need to make this change.
> 
> My only concern is if we could make allarch to raise an exception if
> you inherit it and set package arch to another value?

We're deciding that we're supporting that exact usecase though?

> The only bad effect about it is we won't be able to change something
> in a bbappend which makes the package arch specific that way.

Why not? This should be perfectly usable from a .bbappend?

Cheers,

Richard






      reply	other threads:[~2012-11-26 11:36 UTC|newest]

Thread overview: 5+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2012-11-25 20:23 [PATCH] allarch: Allow class to be included but overridden Richard Purdie
2012-11-26 11:08 ` Otavio Salvador
2012-11-26 11:17   ` Richard Purdie
2012-11-26 11:19     ` Otavio Salvador
2012-11-26 11:22       ` Richard Purdie [this message]

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=1353928930.21863.57.camel@ted \
    --to=richard.purdie@linuxfoundation.org \
    --cc=openembedded-core@lists.openembedded.org \
    --cc=otavio@ossystems.com.br \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox