[gentoo-dev] New USE_EXPAND: CLAWS_MAIL_PLUGINS

[gentoo-dev] New USE_EXPAND: CLAWS_MAIL_PLUGINS

[Christian Faulhammer]

[-- Attachment #1: Type: text/plain, Size: 640 bytes --]

Hello everybody,

the upcoming Claws Mail release will have no separation between
internal plugins (stuff that is built on mail-client/claws-mail with
USE="crypt bogofilter") and external ones (all packages
mail-client/claws-mail-*) anymore.
All are now integrated, which would mean 21 additional USE flags for
Claws or alternatively a new set of CLAWS_MAIL_PLUGINS which includes
the old and new internal plugins.

Any objections against the addition?

Wishes

V-Li

-- 
Christian Faulhammer, Gentoo Lisp project
<URL:http://www.gentoo.org/proj/en/lisp/>, #gentoo-lisp on FreeNode

<URL:http://gentoo.faulhammer.org/>

[-- Attachment #2: signature.asc --]
[-- Type: application/pgp-signature, Size: 198 bytes --]

Re: [gentoo-dev] New USE_EXPAND: CLAWS_MAIL_PLUGINS

[Gilles Dartiguelongue]

Le dimanche 21 avril 2013 à 23:38 +0200, Christian Faulhammer a écrit :
> Hello everybody,
> 
> the upcoming Claws Mail release will have no separation between
> internal plugins (stuff that is built on mail-client/claws-mail with
> USE="crypt bogofilter") and external ones (all packages
> mail-client/claws-mail-*) anymore.
> All are now integrated, which would mean 21 additional USE flags for
> Claws or alternatively a new set of CLAWS_MAIL_PLUGINS which includes
> the old and new internal plugins.
> 
> Any objections against the addition?

We had a discussion about this not so long ago:

Subject: [gentoo-dev] Add HEXCHAT_PLUGINS to USE_EXPAND
Date: Thu, 21 Mar 2013 14:09:54 +0100
Message-ID: <514B06A2.8000803@politeia.in>

To me, the conclusion was that we should be careful about adding
IUSE_EXPAND variables as we might hit system limits sooner than we
expect.

Most probably, we should avoid using it when we cannot add IUSE
dependency on reverse dependencies ?

Other than this technical point, I have no objection.

-- 
Gilles Dartiguelongue <eva@gentoo.org>
Gentoo

Re: [gentoo-dev] New USE_EXPAND: CLAWS_MAIL_PLUGINS

[René Neumann]

Am 21.04.2013 23:38, schrieb Christian Faulhammer:
> Hello everybody,
> 
> the upcoming Claws Mail release will have no separation between
> internal plugins (stuff that is built on mail-client/claws-mail with
> USE="crypt bogofilter") and external ones (all packages
> mail-client/claws-mail-*) anymore.
> All are now integrated, which would mean 21 additional USE flags for
> Claws or alternatively a new set of CLAWS_MAIL_PLUGINS which includes
> the old and new internal plugins.
> 
> Any objections against the addition?
> 
> Wishes
> 
> V-Li
> 

As more and more packages seem to (mis)use USE_EXPAND: Can we get the
possibility to set this directly in package.use? Having to write
'claws_mail_plugins_foo' does not help readability, and setting it in
make.conf is also not the right way (as is package.env).

And could anyone explain to me, why having 21 additional use-flags is
bad? I mean, you have them now too, the only 'advantage' being a nicer
display on 'emerge -v'.

- René

Re: [gentoo-dev] New USE_EXPAND: CLAWS_MAIL_PLUGINS

[Zac Medico]

On 04/24/2013 02:51 AM, René Neumann wrote:
> As more and more packages seem to (mis)use USE_EXPAND: Can we get the
> possibility to set this directly in package.use? Having to write
> 'claws_mail_plugins_foo' does not help readability, and setting it in
> make.conf is also not the right way (as is package.env).

How about if we use a share a single PLUGINS USE_EXPAND among multiple
unrelated packages? That way, you're package.use settings will be less
verbose (plugins_foo versus claws_mail_plugins_foo).

> And could anyone explain to me, why having 21 additional use-flags is
> bad? I mean, you have them now too, the only 'advantage' being a nicer
> display on 'emerge -v'.

Using a generic PLUGINS USE_EXPAND will also have the advantage of
making the emerge -v display less verbose.
-- 
Thanks,
Zac

Re: [gentoo-dev] New USE_EXPAND: CLAWS_MAIL_PLUGINS

[Michał Górny]

[-- Attachment #1: Type: text/plain, Size: 1480 bytes --]

On Sun, 21 Apr 2013 23:38:44 +0200
Christian Faulhammer <fauli@gentoo.org> wrote:

> the upcoming Claws Mail release will have no separation between
> internal plugins (stuff that is built on mail-client/claws-mail with
> USE="crypt bogofilter") and external ones (all packages
> mail-client/claws-mail-*) anymore.
> All are now integrated, which would mean 21 additional USE flags for
> Claws or alternatively a new set of CLAWS_MAIL_PLUGINS which includes
> the old and new internal plugins.
> 
> Any objections against the addition?

Wouldn't it be better to use the feature-intended USE flags and common
sense to provide the users with best possible experience? Use proper
USE flags for features which having switchable benefits user, and just
enable others unconditionally (with USE=-minimal, maybe).

This is, for example, how ekg2 ebuild does. It means that users need
not to be aware of exact plugin structure, and the global USE flags
help them in getting the features they expect. Since plugins can be
enabled and disabled dynamically from within the program, they still
have the full flexibility.

For example, I'd expect USE=gpg to bring me the GPG encryption support,
or USE=html to bring HTML message viewing support. Then we have flags
to further narrow down the engine choice, like USE=webkit. Use them
wisely and it will be much more helpful than claws_mail_plugin_gpg
and claws_mail_plugin_fancy.

-- 
Best regards,
Michał Górny

[-- Attachment #2: signature.asc --]
[-- Type: application/pgp-signature, Size: 966 bytes --]

Re: [gentoo-dev] New USE_EXPAND: CLAWS_MAIL_PLUGINS

[René Neumann]

Am 24.04.2013 19:15, schrieb Zac Medico:
> On 04/24/2013 02:51 AM, René Neumann wrote:
>> As more and more packages seem to (mis)use USE_EXPAND: Can we get the
>> possibility to set this directly in package.use? Having to write
>> 'claws_mail_plugins_foo' does not help readability, and setting it in
>> make.conf is also not the right way (as is package.env).
> 
> How about if we use a share a single PLUGINS USE_EXPAND among multiple
> unrelated packages? That way, you're package.use settings will be less
> verbose (plugins_foo versus claws_mail_plugins_foo).
> 
>> And could anyone explain to me, why having 21 additional use-flags is
>> bad? I mean, you have them now too, the only 'advantage' being a nicer
>> display on 'emerge -v'.
> 
> Using a generic PLUGINS USE_EXPAND will also have the advantage of
> making the emerge -v display less verbose.
> 

Yes, this sounds like a good compromise. I'd still go with Michał if I
had to choose ... but I'm not the maintainer :).

[gentoo-dev] Re: New USE_EXPAND: CLAWS_MAIL_PLUGINS

[Christian Faulhammer]

[-- Attachment #1: Type: text/plain, Size: 950 bytes --]

Hi,

Christian Faulhammer <fauli@gentoo.org>:
> the upcoming Claws Mail release will have no separation between
> internal plugins (stuff that is built on mail-client/claws-mail with
> USE="crypt bogofilter") and external ones (all packages
> mail-client/claws-mail-*) anymore.
> All are now integrated, which would mean 21 additional USE flags for
> Claws or alternatively a new set of CLAWS_MAIL_PLUGINS which includes
> the old and new internal plugins.
> 
> Any objections against the addition?

 Considering all this I will likely drop the idea of IUSE_EXPAND and go
with USE flags.  All plugins that need no special dependencies will
just be built (as they have to be loaded in the interface -> so no
cruft) , the rest will be configurable by normal USE flags.

V-Li

-- 
Christian Faulhammer, Gentoo Lisp project
<URL:http://www.gentoo.org/proj/en/lisp/>, #gentoo-lisp on FreeNode

<URL:http://gentoo.faulhammer.org/>

[-- Attachment #2: signature.asc --]
[-- Type: application/pgp-signature, Size: 198 bytes --]

[gentoo-dev] USE_EXPAND is not an IUSE replacement [was: New USE_EXPAND: CLAWS_MAIL_PLUGINS]

[René Neumann]

Am 24.04.2013 11:51, schrieb René Neumann:
> As more and more packages seem to (mis)use USE_EXPAND: Can we get the
> possibility to set this directly in package.use? Having to write
> 'claws_mail_plugins_foo' does not help readability, and setting it in
> make.conf is also not the right way (as is package.env).

I just noted, that USE_EXPAND variables are even worse than I thought...

NGINX_MODULES_HTTP="-access" emerge -av nginx
Invalid '-' operator in non-incremental variable 'NGINX_MODULES_HTTP':
'-access

In other words: They do not even work from commandline (in a stacking
manner).

So, what exactly are the use-cases again, and why are normal use-flags
replaced by them? I already see a future where packages carry a
"FOOBAR_CONFIGURE_OPTIONS" use-expand ... when IUSE is not hip enough
any more.

- René

P.S.: Using USE_EXPAND for PYTHON_TARGETS or LINGUAS or VIDEO_CARDS is
of course good, as I (usually) set this system-wide.

Re: [gentoo-dev] USE_EXPAND is not an IUSE replacement [was: New USE_EXPAND: CLAWS_MAIL_PLUGINS]

[Zac Medico]

On 05/03/2013 05:59 AM, René Neumann wrote:
> Am 24.04.2013 11:51, schrieb René Neumann:
>> As more and more packages seem to (mis)use USE_EXPAND: Can we get the
>> possibility to set this directly in package.use? Having to write
>> 'claws_mail_plugins_foo' does not help readability, and setting it in
>> make.conf is also not the right way (as is package.env).
> 
> I just noted, that USE_EXPAND variables are even worse than I thought...
> 
> NGINX_MODULES_HTTP="-access" emerge -av nginx
> Invalid '-' operator in non-incremental variable 'NGINX_MODULES_HTTP':
> '-access
> 
> In other words: They do not even work from commandline (in a stacking
> manner).

The long-standing portage behavior is for a make.conf setting like
VIDEO_CARDS="foo" is to discard any VIDEO_CARDS settings from the
profile, which is a non-incremental behavior. If we make all USE_EXPAND
variables behave more like real incremental variables, then users will
have to use VIDEO_CARDS="-* foo" to get the equivalent of the old
behavior. Is it worth changing?

Note that PMS says USE_EXPAND variables are supposed to be treated as
incremental. Also note that portage implements pseudo-increment behavior
for settings of USE_EXPAND variables in profile make.defaults, but it
only works for positive values (which are stacked incrementally only
after they have been translated to the corresponding USE flags).
-- 
Thanks,
Zac

Re: [gentoo-dev] USE_EXPAND is not an IUSE replacement [was: New USE_EXPAND: CLAWS_MAIL_PLUGINS]

[Ciaran McCreesh]

[-- Attachment #1: Type: text/plain, Size: 235 bytes --]

On Fri, 03 May 2013 13:20:15 -0700
Zac Medico <zmedico@gentoo.org> wrote:
> Note that PMS says USE_EXPAND variables are supposed to be treated as
> incremental.

PMS says nothing about user configuration.

-- 
Ciaran McCreesh

[-- Attachment #2: signature.asc --]
[-- Type: application/pgp-signature, Size: 198 bytes --]

Re: [gentoo-dev] USE_EXPAND is not an IUSE replacement [was: New USE_EXPAND: CLAWS_MAIL_PLUGINS]

[René Neumann]

Am 03.05.2013 22:20, schrieb Zac Medico:
> Is it worth changing?

Nope. What's worth changing is the excessive use of USE_EXPAND for no
reason (your described usecase makes sense for reasonable USE_EXPAND
stuff like VIDEO_CARDS). But seems like I'm the only one concerned by
this, so I should probably rest my case and switch to silent sobbing
instead ;-).

- René

Re: [gentoo-dev] USE_EXPAND is not an IUSE replacement [was: New USE_EXPAND: CLAWS_MAIL_PLUGINS]

[Mike Gilbert]

On Fri, May 3, 2013 at 5:08 PM, René Neumann <lists@necoro.eu> wrote:
> Am 03.05.2013 22:20, schrieb Zac Medico:
>> Is it worth changing?
>
> Nope. What's worth changing is the excessive use of USE_EXPAND for no
> reason (your described usecase makes sense for reasonable USE_EXPAND
> stuff like VIDEO_CARDS). But seems like I'm the only one concerned by
> this, so I should probably rest my case and switch to silent sobbing
> instead ;-).
>
> - René
>

Well, I do agree that they are overused. For example, GRUB_PLATFORMS
is used by only one package.

I would convert it back into a normal set of use flags, but that would
mean any users of sys-boot/grub:2 would have to update their
configuration. I don't think it is worth it at this point.

Re: [gentoo-dev] USE_EXPAND is not an IUSE replacement [was: New USE_EXPAND: CLAWS_MAIL_PLUGINS]

[Rick "Zero_Chaos" Farina]

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

On 05/03/2013 04:20 PM, Zac Medico wrote:
> On 05/03/2013 05:59 AM, Ren￩ Neumann wrote:
>> Am 24.04.2013 11:51, schrieb Ren￩ Neumann:
>>> As more and more packages seem to (mis)use USE_EXPAND: Can we get the
>>> possibility to set this directly in package.use? Having to write
>>> 'claws_mail_plugins_foo' does not help readability, and setting it in
>>> make.conf is also not the right way (as is package.env).
>>
>> I just noted, that USE_EXPAND variables are even worse than I thought...
>>
>> NGINX_MODULES_HTTP="-access" emerge -av nginx
>> Invalid '-' operator in non-incremental variable 'NGINX_MODULES_HTTP':
>> '-access
>>
>> In other words: They do not even work from commandline (in a stacking
>> manner).
> 
> The long-standing portage behavior is for a make.conf setting like
> VIDEO_CARDS="foo" is to discard any VIDEO_CARDS settings from the
> profile, which is a non-incremental behavior. If we make all USE_EXPAND
> variables behave more like real incremental variables, then users will
> have to use VIDEO_CARDS="-* foo" to get the equivalent of the old
> behavior. Is it worth changing?

Honestly some days I'd really want to say YES.  But with a cool head I
can very firmly say no.  Changing would cause pretty significant
breakage for a lot of users (and likely even our own profiles).
> 
> Note that PMS says USE_EXPAND variables are supposed to be treated as
> incremental. Also note that portage implements pseudo-increment behavior
> for settings of USE_EXPAND variables in profile make.defaults, but it
> only works for positive values (which are stacked incrementally only
> after they have been translated to the corresponding USE flags).
> 
That is a bit of an issue honestly, but a rare time when I'd say we
should make PMS match portage behavior.

Remember you can always USE=-use_expand_whatever (makes for some
interesting tests I must say).

- -Zero
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v2.0.19 (GNU/Linux)
Comment: Using GnuPG with Thunderbird - http://www.enigmail.net/
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=CAeL
-----END PGP SIGNATURE-----

Re: [gentoo-dev] USE_EXPAND is not an IUSE replacement [was: New USE_EXPAND: CLAWS_MAIL_PLUGINS]

[Rick "Zero_Chaos" Farina]

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

On 05/03/2013 05:28 PM, Mike Gilbert wrote:
> On Fri, May 3, 2013 at 5:08 PM, René Neumann <lists@necoro.eu> wrote:
>> Am 03.05.2013 22:20, schrieb Zac Medico:
>>> Is it worth changing?
>>
>> Nope. What's worth changing is the excessive use of USE_EXPAND for no
>> reason (your described usecase makes sense for reasonable USE_EXPAND
>> stuff like VIDEO_CARDS). But seems like I'm the only one concerned by
>> this, so I should probably rest my case and switch to silent sobbing
>> instead ;-).
>>
>> - René
>>
> 
> Well, I do agree that they are overused. For example, GRUB_PLATFORMS
> is used by only one package.
> 
> I would convert it back into a normal set of use flags, but that would
> mean any users of sys-boot/grub:2 would have to update their
> configuration. I don't think it is worth it at this point.
> 

Considering grub:2 isn't really used by the majority of users I feel
compelled to note that you could convert now and just pick a sane set of
use defaults.  I know it's been a while since EAPI2 was introduced but I
do love that little feature...  :-)

Personally I don't care as I have GRUB_PLATFORMS set in my profile
already ;-)

- -Zero
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v2.0.19 (GNU/Linux)
Comment: Using GnuPG with Thunderbird - http://www.enigmail.net/
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=oHZS
-----END PGP SIGNATURE-----

Re: [gentoo-dev] USE_EXPAND is not an IUSE replacement [was: New USE_EXPAND: CLAWS_MAIL_PLUGINS]

[Dustin C. Hatch]

On 5/3/2013 16:08, René Neumann wrote:
> Am 03.05.2013 22:20, schrieb Zac Medico:
>> Is it worth changing?
>
> Nope. What's worth changing is the excessive use of USE_EXPAND for no
> reason (your described usecase makes sense for reasonable USE_EXPAND
> stuff like VIDEO_CARDS). But seems like I'm the only one concerned by
> this, so I should probably rest my case and switch to silent sobbing
> instead ;-).
>
> - René
>
I definitely agree with you. USE_EXPAND combined with nearly everything 
on-by-default (ENLIGHTENMENT_PLUGINS, for example) really annoys me. 
There are two ways to "turn off" USE_EXPADed flags: explicitly set 
everything but what you don't want in make.conf, or use the 
fully-qualified flag in make.conf:USE or package.use, which defeats the 
purpose entirely. Both of these are overly verbose, in my opinion. I 
don't know how the over-utilization of USE_EXPAND got started, but I 
would very much like to see it go away.

-- 
♫Dustin
http://dustin.hatch.name/

Re: [gentoo-dev] USE_EXPAND is not an IUSE replacement [was: New USE_EXPAND: CLAWS_MAIL_PLUGINS]

[vivo75]

Il 04/05/2013 08:11, Dustin C. Hatch ha scritto:
> On 5/3/2013 16:08, René Neumann wrote:
>> Am 03.05.2013 22:20, schrieb Zac Medico:
>>> Is it worth changing?
>>
>> Nope. What's worth changing is the excessive use of USE_EXPAND for no
>> reason (your described usecase makes sense for reasonable USE_EXPAND
>> stuff like VIDEO_CARDS). But seems like I'm the only one concerned by
>> this, so I should probably rest my case and switch to silent sobbing
>> instead ;-).
>>
>> - René
>>
> I definitely agree with you. USE_EXPAND combined with nearly
> everything on-by-default (ENLIGHTENMENT_PLUGINS, for example) really
> annoys me. There are two ways to "turn off" USE_EXPADed flags:
> explicitly set everything but what you don't want in make.conf, or use
> the fully-qualified flag in make.conf:USE or package.use, which
> defeats the purpose entirely. Both of these are overly verbose, in my
> opinion. I don't know how the over-utilization of USE_EXPAND got
> started, but I would very much like to see it go away.
>
Actually this would not be that bad if USE_EXPAND where immutable (fix
it once and be ok) but they are not and as such suffer the same problem
as USE="-*"

Re: [gentoo-dev] USE_EXPAND is not an IUSE replacement [was: New USE_EXPAND: CLAWS_MAIL_PLUGINS]

[Zac Medico]

On 05/03/2013 11:11 PM, Dustin C. Hatch wrote:
> On 5/3/2013 16:08, René Neumann wrote:
>> Am 03.05.2013 22:20, schrieb Zac Medico:
>>> Is it worth changing?
>>
>> Nope. What's worth changing is the excessive use of USE_EXPAND for no
>> reason (your described usecase makes sense for reasonable USE_EXPAND
>> stuff like VIDEO_CARDS). But seems like I'm the only one concerned by
>> this, so I should probably rest my case and switch to silent sobbing
>> instead ;-).
>>
>> - René
>>
> I definitely agree with you. USE_EXPAND combined with nearly everything
> on-by-default (ENLIGHTENMENT_PLUGINS, for example) really annoys me.
> There are two ways to "turn off" USE_EXPADed flags: explicitly set
> everything but what you don't want in make.conf, or use the
> fully-qualified flag in make.conf:USE or package.use, which defeats the
> purpose entirely.

FWIW, portage supports both positive and negative wildcard settings for
USE_EXPAND flags like USE="video_cards_* -enlightenment_plugins_*".

> Both of these are overly verbose, in my opinion. I
> don't know how the over-utilization of USE_EXPAND got started, but I
> would very much like to see it go away.
-- 
Thanks,
Zac

Re: [gentoo-dev] USE_EXPAND is not an IUSE replacement [was: New USE_EXPAND: CLAWS_MAIL_PLUGINS]

[Ciaran McCreesh]

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

On Fri, 03 May 2013 22:58:57 -0400
"Rick \"Zero_Chaos\" Farina" <zerochaos@gentoo.org> wrote:
> > Note that PMS says USE_EXPAND variables are supposed to be treated
> > as incremental. Also note that portage implements pseudo-increment
> > behavior for settings of USE_EXPAND variables in profile
> > make.defaults, but it only works for positive values (which are
> > stacked incrementally only after they have been translated to the
> > corresponding USE flags).
> > 
> That is a bit of an issue honestly, but a rare time when I'd say we
> should make PMS match portage behavior.

Uhm, PMS governs how this works for profiles, not for user
configuration or any crazy stuff you do on the command line. Portage is
not required to use exactly the same code for both...

- -- 
Ciaran McCreesh
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v2.0.19 (GNU/Linux)

iEYEARECAAYFAlGFKboACgkQ96zL6DUtXhHoEgCg27zs7DoujKY4jXkx4ZXv2j6X
Qu4AoNrGtxLJbiNpCkwyd9J36kWGMf2S
=MfPG
-----END PGP SIGNATURE-----

Re: [gentoo-dev] USE_EXPAND is not an IUSE replacement [was: New USE_EXPAND: CLAWS_MAIL_PLUGINS]

[Rick "Zero_Chaos" Farina]

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

On 05/04/2013 11:31 AM, Ciaran McCreesh wrote:
> On Fri, 03 May 2013 22:58:57 -0400
> "Rick \"Zero_Chaos\" Farina" <zerochaos@gentoo.org> wrote:
>>> Note that PMS says USE_EXPAND variables are supposed to be treated
>>> as incremental. Also note that portage implements pseudo-increment
>>> behavior for settings of USE_EXPAND variables in profile
>>> make.defaults, but it only works for positive values (which are
>>> stacked incrementally only after they have been translated to the
>>> corresponding USE flags).
>>>
>> That is a bit of an issue honestly, but a rare time when I'd say we
>> should make PMS match portage behavior.
> 
> Uhm, PMS governs how this works for profiles, not for user
> configuration or any crazy stuff you do on the command line. Portage is
> not required to use exactly the same code for both...
> 
> 
Thanks for clearing that up Ciaran.  That certainly makes a difference.

- -Zero
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v2.0.19 (GNU/Linux)
Comment: Using GnuPG with Thunderbird - http://www.enigmail.net/

iQIcBAEBAgAGBQJRhTk1AAoJEKXdFCfdEflK2o4QAILB7VhsnktsW7KQC5Y7F/0q
fyq1K9q5lg9Fz/2WWZ+yVsjE6XSzLAPjes6kJzez0wZ0BUaDuI7eQOefFFqZ5Lwq
P0CdqXdnou3TfaSWmC8s3vVSDX1xV/O2y+9R/zd4YaVFa7vLzVAkFpNNRtzgwGcm
o8w5L+Lso1odhLzhsz2Wql5F1L8mNPpV4bAa27XnENrx/BEkRuEJHoj90ZpqPTIv
zxMircMWYSnfauhUV4KSe6ieOeUeY7rOKVPMgoqTF0hYR0Oox7Vg5HZx77yy4uQw
fkczatmzIp6L/I/zQws7Gi9bgh9wsfwd/jscfH5+8bFTH914YPbs2BejC9HKrnUQ
uNN2/jCwlrchlWKWkwxTW/ay2gufsymR8btR3099zupwjCgEpgRDYCkEGS5hqN0V
It5Nm7QWHrtBAZJRg/U4Pe/zV8fwYCWKEgreXEZzB5lT+vFkgn7kf/cJkmpOBwvG
QR1LAlvnWaRNV0fbaoZydcjDPVZc/yBs1h3J+YOKvKZtF+dkH5/uHbjJPydgtpl4
qG+r2KdtV9aOTbJuidTh97gzAOOkEEbLoKIyqwHvrEzzb93fawVj1IaowUwHBlTq
Yutbcbw2W4po5tg1nDy416sZRD2pAgsut47vnRVMUhPlkZn4PBYFYkOVzYAOPwMu
PFR2Za+sLwsMYnOshi5W
=BkWW
-----END PGP SIGNATURE-----

Re: [gentoo-dev] USE_EXPAND is not an IUSE replacement [was: New USE_EXPAND: CLAWS_MAIL_PLUGINS]

[William Hubbs]

[-- Attachment #1: Type: text/plain, Size: 1576 bytes --]

On Fri, May 03, 2013 at 11:03:41PM -0400, Rick "Zero_Chaos" Farina wrote:
> -----BEGIN PGP SIGNED MESSAGE-----
> Hash: SHA1
> 
> On 05/03/2013 05:28 PM, Mike Gilbert wrote:
> > On Fri, May 3, 2013 at 5:08 PM, René Neumann <lists@necoro.eu> wrote:
> >> Am 03.05.2013 22:20, schrieb Zac Medico:
> >>> Is it worth changing?
> >>
> >> Nope. What's worth changing is the excessive use of USE_EXPAND for no
> >> reason (your described usecase makes sense for reasonable USE_EXPAND
> >> stuff like VIDEO_CARDS). But seems like I'm the only one concerned by
> >> this, so I should probably rest my case and switch to silent sobbing
> >> instead ;-).
> >>
> >> - René
> >>
> > 
> > Well, I do agree that they are overused. For example, GRUB_PLATFORMS
> > is used by only one package.
> > 
> > I would convert it back into a normal set of use flags, but that would
> > mean any users of sys-boot/grub:2 would have to update their
> > configuration. I don't think it is worth it at this point.
> > 
> 
> Considering grub:2 isn't really used by the majority of users I feel
> compelled to note that you could convert now and just pick a sane set of
> use defaults.  I know it's been a while since EAPI2 was introduced but I
> do love that little feature...  :-)

I would also like to chime in here and point out that all grub:2 users
are ~arch users. If we were going to break stable it would be more of a
shake-up, but since it isn't happening that way, I would say go for it.
~arch users are expected to be able to handle things like this.

William

[-- Attachment #2: Digital signature --]
[-- Type: application/pgp-signature, Size: 198 bytes --]

Re: [gentoo-dev] New USE_EXPAND: CLAWS_MAIL_PLUGINS

[Zac Medico]

On 04/24/2013 12:48 PM, René Neumann wrote:
> Am 24.04.2013 19:15, schrieb Zac Medico:
>> On 04/24/2013 02:51 AM, René Neumann wrote:
>>> As more and more packages seem to (mis)use USE_EXPAND: Can we get the
>>> possibility to set this directly in package.use? Having to write
>>> 'claws_mail_plugins_foo' does not help readability, and setting it in
>>> make.conf is also not the right way (as is package.env).
>>
>> How about if we use a share a single PLUGINS USE_EXPAND among multiple
>> unrelated packages? That way, you're package.use settings will be less
>> verbose (plugins_foo versus claws_mail_plugins_foo).
>>
>>> And could anyone explain to me, why having 21 additional use-flags is
>>> bad? I mean, you have them now too, the only 'advantage' being a nicer
>>> display on 'emerge -v'.
>>
>> Using a generic PLUGINS USE_EXPAND will also have the advantage of
>> making the emerge -v display less verbose.
>>
>
> Yes, this sounds like a good compromise. I'd still go with Michał if I
> had to choose ... but I'm not the maintainer :).

We've got a feature request bug here for a package.use USE_EXPAND syntax 
extension:

  https://bugs.gentoo.org/show_bug.cgi?id=471776
-- 
Thanks,
Zac