* superfluous download locations in oe-core's bitbake.conf
@ 2012-06-21 8:56 Robert P. J. Day
2012-06-22 13:34 ` Paul Eggleton
0 siblings, 1 reply; 6+ messages in thread
From: Robert P. J. Day @ 2012-06-21 8:56 UTC (permalink / raw)
To: OE Core mailing list
given that oe-core is defined as being a really minimal
configuration with no extraneous cruft, there seems to be a number of
unnecessary download locations in the shipped bitbake.conf, things
like:
ADOBE_MIRROR =
FREEBSD_MIRROR =
and so on. obviously, they're not doing any harm but there seems to
be little value in defining download locations that aren't used by
anything actually in oe-core.
rday
--
========================================================================
Robert P. J. Day Ottawa, Ontario, CANADA
http://crashcourse.ca
Twitter: http://twitter.com/rpjday
LinkedIn: http://ca.linkedin.com/in/rpjday
========================================================================
^ permalink raw reply [flat|nested] 6+ messages in thread
* Re: superfluous download locations in oe-core's bitbake.conf
2012-06-21 8:56 superfluous download locations in oe-core's bitbake.conf Robert P. J. Day
@ 2012-06-22 13:34 ` Paul Eggleton
2012-06-22 14:24 ` Richard Purdie
2012-06-22 15:20 ` Robert P. J. Day
0 siblings, 2 replies; 6+ messages in thread
From: Paul Eggleton @ 2012-06-22 13:34 UTC (permalink / raw)
To: Robert P. J. Day; +Cc: openembedded-core
On Thursday 21 June 2012 04:56:51 Robert P. J. Day wrote:
> given that oe-core is defined as being a really minimal
> configuration with no extraneous cruft, there seems to be a number of
> unnecessary download locations in the shipped bitbake.conf, things
> like:
>
> ADOBE_MIRROR =
> FREEBSD_MIRROR =
>
> and so on. obviously, they're not doing any harm but there seems to
> be little value in defining download locations that aren't used by
> anything actually in oe-core.
These are present for largely historical reasons - namely in OE-Classic we had
a number of recipes needing this same mirror defined and it was deemed that it
was easier to edit it in one place if it needed to change. Some of them are
even completely useless now as the sites to which they point are now gone,
e.g. HANDHELDS_CVS.
I think the way to approach this would be to just submit a patch removing the
more extraneous ones and see if anyone complains. Note that some of them that
aren't used in OE-Core might be used within meta-oe or other OE community
layers, so those might need to stay at least for the time being.
Cheers,
Paul
--
Paul Eggleton
Intel Open Source Technology Centre
^ permalink raw reply [flat|nested] 6+ messages in thread
* Re: superfluous download locations in oe-core's bitbake.conf
2012-06-22 13:34 ` Paul Eggleton
@ 2012-06-22 14:24 ` Richard Purdie
2012-06-22 14:38 ` Martin Jansa
2012-06-22 15:20 ` Robert P. J. Day
1 sibling, 1 reply; 6+ messages in thread
From: Richard Purdie @ 2012-06-22 14:24 UTC (permalink / raw)
To: Patches and discussions about the oe-core layer
On Fri, 2012-06-22 at 14:34 +0100, Paul Eggleton wrote:
> On Thursday 21 June 2012 04:56:51 Robert P. J. Day wrote:
> > given that oe-core is defined as being a really minimal
> > configuration with no extraneous cruft, there seems to be a number of
> > unnecessary download locations in the shipped bitbake.conf, things
> > like:
> >
> > ADOBE_MIRROR =
> > FREEBSD_MIRROR =
> >
> > and so on. obviously, they're not doing any harm but there seems to
> > be little value in defining download locations that aren't used by
> > anything actually in oe-core.
>
> These are present for largely historical reasons - namely in OE-Classic we had
> a number of recipes needing this same mirror defined and it was deemed that it
> was easier to edit it in one place if it needed to change. Some of them are
> even completely useless now as the sites to which they point are now gone,
> e.g. HANDHELDS_CVS.
>
> I think the way to approach this would be to just submit a patch removing the
> more extraneous ones and see if anyone complains. Note that some of them that
> aren't used in OE-Core might be used within meta-oe or other OE community
> layers, so those might need to stay at least for the time being.
Agreed, we should clean some of these up.
I would also add that we do have things in OE-Core without direct users
(at least by default), particularly things like classes but also
variables. OE-Core is a collaboration and where it makes sense to have
something defined in one location rather than in many, I'm not adverse
to that. Some of the mirror definitions do fit into that category.
Cheers,
Richard
^ permalink raw reply [flat|nested] 6+ messages in thread
* Re: superfluous download locations in oe-core's bitbake.conf
2012-06-22 14:24 ` Richard Purdie
@ 2012-06-22 14:38 ` Martin Jansa
2012-06-22 14:45 ` Paul Eggleton
0 siblings, 1 reply; 6+ messages in thread
From: Martin Jansa @ 2012-06-22 14:38 UTC (permalink / raw)
To: Patches and discussions about the oe-core layer
[-- Attachment #1: Type: text/plain, Size: 2387 bytes --]
On Fri, Jun 22, 2012 at 03:24:23PM +0100, Richard Purdie wrote:
> On Fri, 2012-06-22 at 14:34 +0100, Paul Eggleton wrote:
> > On Thursday 21 June 2012 04:56:51 Robert P. J. Day wrote:
> > > given that oe-core is defined as being a really minimal
> > > configuration with no extraneous cruft, there seems to be a number of
> > > unnecessary download locations in the shipped bitbake.conf, things
> > > like:
> > >
> > > ADOBE_MIRROR =
> > > FREEBSD_MIRROR =
> > >
> > > and so on. obviously, they're not doing any harm but there seems to
> > > be little value in defining download locations that aren't used by
> > > anything actually in oe-core.
> >
> > These are present for largely historical reasons - namely in OE-Classic we had
> > a number of recipes needing this same mirror defined and it was deemed that it
> > was easier to edit it in one place if it needed to change. Some of them are
> > even completely useless now as the sites to which they point are now gone,
> > e.g. HANDHELDS_CVS.
> >
> > I think the way to approach this would be to just submit a patch removing the
> > more extraneous ones and see if anyone complains. Note that some of them that
> > aren't used in OE-Core might be used within meta-oe or other OE community
> > layers, so those might need to stay at least for the time being.
>
> Agreed, we should clean some of these up.
>
> I would also add that we do have things in OE-Core without direct users
> (at least by default), particularly things like classes but also
> variables. OE-Core is a collaboration and where it makes sense to have
> something defined in one location rather than in many, I'm not adverse
> to that. Some of the mirror definitions do fit into that category.
Is there easy way for layer to define additional MIRRORs with global
scope for layer? e.g. putting them layer.conf?
e.g. FREESMARTPHONE_GIT is used in few recipes in meta-oe and in many in
meta-fso layer and right now is only defined but not used in oe-core.
So moving FREESMARTPHONE_GIT definition from oe-core to meta-oe will add
aditional dependency on meta-oe, but still better then duplicating it in
both layer or even in many .inc files.
I don't see any harm from ADOBE_MIRROR/FREEBSD_MIRROR in oe-core (even
when not used).
Cheers,
--
Martin 'JaMa' Jansa jabber: Martin.Jansa@gmail.com
[-- Attachment #2: Digital signature --]
[-- Type: application/pgp-signature, Size: 205 bytes --]
^ permalink raw reply [flat|nested] 6+ messages in thread
* Re: superfluous download locations in oe-core's bitbake.conf
2012-06-22 14:38 ` Martin Jansa
@ 2012-06-22 14:45 ` Paul Eggleton
0 siblings, 0 replies; 6+ messages in thread
From: Paul Eggleton @ 2012-06-22 14:45 UTC (permalink / raw)
To: Martin Jansa; +Cc: openembedded-core
On Friday 22 June 2012 16:38:12 Martin Jansa wrote:
> Is there easy way for layer to define additional MIRRORs with global
> scope for layer? e.g. putting them layer.conf?
I believe that works yes.
> e.g. FREESMARTPHONE_GIT is used in few recipes in meta-oe and in many in
> meta-fso layer and right now is only defined but not used in oe-core.
>
> So moving FREESMARTPHONE_GIT definition from oe-core to meta-oe will add
> aditional dependency on meta-oe, but still better then duplicating it in
> both layer or even in many .inc files.
I don't think we want to remove any that are going to cause that kind of pain.
If the value is still valid and more than one layer is using the variable then
it's reasonable to keep there. On the other hand, if it's invalid or only one
layer is likely to ever use it (e.g. the Opie ones that I removed earlier)
then it's a candidate for moving out, IMHO.
Cheers,
Paul
--
Paul Eggleton
Intel Open Source Technology Centre
^ permalink raw reply [flat|nested] 6+ messages in thread
* Re: superfluous download locations in oe-core's bitbake.conf
2012-06-22 13:34 ` Paul Eggleton
2012-06-22 14:24 ` Richard Purdie
@ 2012-06-22 15:20 ` Robert P. J. Day
1 sibling, 0 replies; 6+ messages in thread
From: Robert P. J. Day @ 2012-06-22 15:20 UTC (permalink / raw)
To: Paul Eggleton; +Cc: openembedded-core
On Fri, 22 Jun 2012, Paul Eggleton wrote:
> On Thursday 21 June 2012 04:56:51 Robert P. J. Day wrote:
> > given that oe-core is defined as being a really minimal
> > configuration with no extraneous cruft, there seems to be a number of
> > unnecessary download locations in the shipped bitbake.conf, things
> > like:
> >
> > ADOBE_MIRROR =
> > FREEBSD_MIRROR =
> >
> > and so on. obviously, they're not doing any harm but there seems to
> > be little value in defining download locations that aren't used by
> > anything actually in oe-core.
>
> These are present for largely historical reasons - namely in
> OE-Classic we had a number of recipes needing this same mirror
> defined and it was deemed that it was easier to edit it in one place
> if it needed to change. Some of them are even completely useless now
> as the sites to which they point are now gone, e.g. HANDHELDS_CVS.
>
> I think the way to approach this would be to just submit a patch
> removing the more extraneous ones and see if anyone complains. Note
> that some of them that aren't used in OE-Core might be used within
> meta-oe or other OE community layers, so those might need to stay at
> least for the time being.
i mentioned this bit of pedantry only because of the regular
incantation of, "does that really belong in oe-core?" so to be
consistent, if it doesn't belong, it should be removed.
if something in meta-openembedded or another layer needs a new
download location, it should provide it, one would think. i don't
think oe-core should provide variables solely on the off chance that
something else might need it some day.
rday
--
========================================================================
Robert P. J. Day Ottawa, Ontario, CANADA
http://crashcourse.ca
Twitter: http://twitter.com/rpjday
LinkedIn: http://ca.linkedin.com/in/rpjday
========================================================================
^ permalink raw reply [flat|nested] 6+ messages in thread
end of thread, other threads:[~2012-06-22 15:31 UTC | newest]
Thread overview: 6+ messages (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
2012-06-21 8:56 superfluous download locations in oe-core's bitbake.conf Robert P. J. Day
2012-06-22 13:34 ` Paul Eggleton
2012-06-22 14:24 ` Richard Purdie
2012-06-22 14:38 ` Martin Jansa
2012-06-22 14:45 ` Paul Eggleton
2012-06-22 15:20 ` Robert P. J. Day
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox