[gentoo-user] stable java virtuals require unstable java packages

[gentoo-user] stable java virtuals require unstable java packages

[Alexander Kapshuk]

[-- Attachment #1: Type: text/plain, Size: 1533 bytes --]

Is it normal for a stable virtual to require an unstable package?

Running 'emerge -auND' generates the following output on my system:

Calculating dependencies  .... ....... done!
[ebuild  NS   ~] dev-java/icedtea-bin-7.2.5.3 [6.1.13.5] USE="X alsa cups
nsplugin -cjk -doc -examples (-selinux) -source -webstart"
[ebuild  NS    ] virtual/jdk-1.7.0 [1.6.0-r2]
[ebuild  NS    ] virtual/jre-1.7.0 [1.6.0-r1]
[ebuild     U ~] dev-java/icedtea-web-1.5.1-r1 [1.4.2] USE="icedtea7%*
-tagsoup%"

The following keyword changes are necessary to proceed:
 (see "package.accept_keywords" in the portage(5) man page for more details)
# required by virtual/jdk-1.7.0::gentoo
# required by virtual/jre-1.7.0::gentoo
# required by sys-libs/db-4.8.30-r2::gentoo[java]
# required by dev-libs/redland-1.0.16::gentoo[berkdb]
# required by app-office/libreoffice-4.4.1.2::gentoo
# required by @selected
# required by @world (argument)
=dev-java/icedtea-bin-7.2.5.3 ~x86
# required by dev-java/icedtea-bin-7.2.5.3::gentoo[nsplugin]
# required by virtual/jdk-1.7.0::gentoo
# required by virtual/jre-1.7.0::gentoo
# required by sys-libs/db-4.8.30-r2::gentoo[java]
# required by dev-libs/redland-1.0.16::gentoo[berkdb]
# required by app-office/libreoffice-4.4.1.2::gentoo
# required by @selected
# required by @world (argument)
=dev-java/icedtea-web-1.5.1-r1 ~x86

/usr/portage/virtual/jre/jre-1.7.0.ebuild:12
        =virtual/jdk-1.7.0*
/usr/portage/virtual/jdk/jdk-1.7.0.ebuild:12,13
        =dev-java/icedtea-bin-7*
        =dev-java/icedtea-7*

Thanks.

[-- Attachment #2: Type: text/html, Size: 1768 bytes --]

Re: [gentoo-user] stable java virtuals require unstable java packages

[Alan McKinnon]

On 21/04/2015 07:28, Alexander Kapshuk wrote:
> Is it normal for a stable virtual to require an unstable package?


No, that is definitely not how virtuals should work.

It's either a bug in virtual/jdk-1.7 ebuild or more likely, stable
request for icedtead-bin-7 is lagging behind.

I know the java team is badly understaffed, lots of mentions on -dev, so
maybe log a stable request but against icedtea-bin-7



> 
> Running 'emerge -auND' generates the following output on my system:
> 
> Calculating dependencies  .... ....... done!
> [ebuild  NS   ~] dev-java/icedtea-bin-7.2.5.3 [6.1.13.5] USE="X alsa
> cups nsplugin -cjk -doc -examples (-selinux) -source -webstart"
> [ebuild  NS    ] virtual/jdk-1.7.0 [1.6.0-r2]
> [ebuild  NS    ] virtual/jre-1.7.0 [1.6.0-r1]
> [ebuild     U ~] dev-java/icedtea-web-1.5.1-r1 [1.4.2] USE="icedtea7%*
> -tagsoup%"
> 
> The following keyword changes are necessary to proceed:
>  (see "package.accept_keywords" in the portage(5) man page for more details)
> # required by virtual/jdk-1.7.0::gentoo
> # required by virtual/jre-1.7.0::gentoo
> # required by sys-libs/db-4.8.30-r2::gentoo[java]
> # required by dev-libs/redland-1.0.16::gentoo[berkdb]
> # required by app-office/libreoffice-4.4.1.2::gentoo
> # required by @selected
> # required by @world (argument)
> =dev-java/icedtea-bin-7.2.5.3 ~x86
> # required by dev-java/icedtea-bin-7.2.5.3::gentoo[nsplugin]
> # required by virtual/jdk-1.7.0::gentoo
> # required by virtual/jre-1.7.0::gentoo
> # required by sys-libs/db-4.8.30-r2::gentoo[java]
> # required by dev-libs/redland-1.0.16::gentoo[berkdb]
> # required by app-office/libreoffice-4.4.1.2::gentoo
> # required by @selected
> # required by @world (argument)
> =dev-java/icedtea-web-1.5.1-r1 ~x86
> 
> /usr/portage/virtual/jre/jre-1.7.0.ebuild:12
>         =virtual/jdk-1.7.0*
> /usr/portage/virtual/jdk/jdk-1.7.0.ebuild:12,13
>         =dev-java/icedtea-bin-7*
>         =dev-java/icedtea-7*
> 
> Thanks.
> 


-- 
Alan McKinnon
alan.mckinnon@gmail.com

Re: [gentoo-user] stable java virtuals require unstable java packages

[Florian Gamböck]

Am 21.04.2015 um 07:42 schrieb Alan McKinnon:
> It's either a bug in virtual/jdk-1.7 ebuild or more likely, stable 
> request for icedtead-bin-7 is lagging behind.

You are right, the stable request is still going on: 
https://bugs.gentoo.org/show_bug.cgi?id=546902

Obviously the virtual got stableized a bit too fast.

Re: [gentoo-user] stable java virtuals require unstable java packages

[gottlieb]

On Tue, Apr 21 2015, Florian Gamböck wrote:

> Am 21.04.2015 um 07:42 schrieb Alan McKinnon:
>> It's either a bug in virtual/jdk-1.7 ebuild or more likely, stable
>> request for icedtead-bin-7 is lagging behind.
>
> You are right, the stable request is still going on:
> https://bugs.gentoo.org/show_bug.cgi?id=546902
>
> Obviously the virtual got stableized a bit too fast.

I was going to ask about this too.  I noticed that the virtual/jdk-1.7
actually gives a choice of jdk's.  One of them
   dev-java/oracle-jdk-bin
does have a stable 1.7 version (indeed also a stable 1.8).

I am not anxious to switch oracle-*.

allan

Re: [gentoo-user] stable java virtuals require unstable java packages

[Andreas K. Huettel]

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA512

Am Dienstag, 21. April 2015, 15:37:34 schrieb gottlieb@nyu.edu:
> On Tue, Apr 21 2015, Florian Gamböck wrote:
> 
> > Am 21.04.2015 um 07:42 schrieb Alan McKinnon:
> >> It's either a bug in virtual/jdk-1.7 ebuild or more likely, stable
> >> request for icedtead-bin-7 is lagging behind.
> >
> > You are right, the stable request is still going on:
> > https://bugs.gentoo.org/show_bug.cgi?id=546902
> >
> > Obviously the virtual got stableized a bit too fast.
> 
> I was going to ask about this too.  I noticed that the virtual/jdk-1.7
> actually gives a choice of jdk's.  One of them
>    dev-java/oracle-jdk-bin
> does have a stable 1.7 version (indeed also a stable 1.8).
> 
> I am not anxious to switch oracle-*.
> 
> allan
> 
> 

This is a typical problem with virtuals. 

A virtual can (by definition) be stable as soon as one of its providers is stable. Which one does not matter, and stabilization can be pretty much automatic and be done by any dev. 

So, after oracle-* went stable the virtual went stable too. Now portage is trying to upgrade it and can only resolve that by either keywording icedtea-bin or switching to oracle. Which solution it ends up with differs between portage versions, I suspect... 

This should be improved, obviously, but it's not a trivial issue.

My personal advice (without knowing the packages in detail)- if you dont want to switch to oracle, just accept the keywording suggestion from portage and give the suggested ~arch icedtea-bin a try. Don't worry too much about mixing stable and testing (as long as it's only for a few packages). 

- -- 

Andreas K. Huettel
Gentoo Linux developer 
dilfridge@gentoo.org
http://www.akhuettel.de/

-----BEGIN PGP SIGNATURE-----
Version: GnuPG v2.0
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=x6Fh
-----END PGP SIGNATURE-----

Re: [gentoo-user] stable java virtuals require unstable java packages

[Alexander Kapshuk]

[-- Attachment #1: Type: text/plain, Size: 2212 bytes --]

On Tue, Apr 21, 2015 at 8:42 AM, Alan McKinnon <alan.mckinnon@gmail.com>
wrote:

> On 21/04/2015 07:28, Alexander Kapshuk wrote:
> > Is it normal for a stable virtual to require an unstable package?
>
>
> No, that is definitely not how virtuals should work.
>
> It's either a bug in virtual/jdk-1.7 ebuild or more likely, stable
> request for icedtead-bin-7 is lagging behind.
>
> I know the java team is badly understaffed, lots of mentions on -dev, so
> maybe log a stable request but against icedtea-bin-7
>
>
>
> >
> > Running 'emerge -auND' generates the following output on my system:
> >
> > Calculating dependencies  .... ....... done!
> > [ebuild  NS   ~] dev-java/icedtea-bin-7.2.5.3 [6.1.13.5] USE="X alsa
> > cups nsplugin -cjk -doc -examples (-selinux) -source -webstart"
> > [ebuild  NS    ] virtual/jdk-1.7.0 [1.6.0-r2]
> > [ebuild  NS    ] virtual/jre-1.7.0 [1.6.0-r1]
> > [ebuild     U ~] dev-java/icedtea-web-1.5.1-r1 [1.4.2] USE="icedtea7%*
> > -tagsoup%"
> >
> > The following keyword changes are necessary to proceed:
> >  (see "package.accept_keywords" in the portage(5) man page for more
> details)
> > # required by virtual/jdk-1.7.0::gentoo
> > # required by virtual/jre-1.7.0::gentoo
> > # required by sys-libs/db-4.8.30-r2::gentoo[java]
> > # required by dev-libs/redland-1.0.16::gentoo[berkdb]
> > # required by app-office/libreoffice-4.4.1.2::gentoo
> > # required by @selected
> > # required by @world (argument)
> > =dev-java/icedtea-bin-7.2.5.3 ~x86
> > # required by dev-java/icedtea-bin-7.2.5.3::gentoo[nsplugin]
> > # required by virtual/jdk-1.7.0::gentoo
> > # required by virtual/jre-1.7.0::gentoo
> > # required by sys-libs/db-4.8.30-r2::gentoo[java]
> > # required by dev-libs/redland-1.0.16::gentoo[berkdb]
> > # required by app-office/libreoffice-4.4.1.2::gentoo
> > # required by @selected
> > # required by @world (argument)
> > =dev-java/icedtea-web-1.5.1-r1 ~x86
> >
> > /usr/portage/virtual/jre/jre-1.7.0.ebuild:12
> >         =virtual/jdk-1.7.0*
> > /usr/portage/virtual/jdk/jdk-1.7.0.ebuild:12,13
> >         =dev-java/icedtea-bin-7*
> >         =dev-java/icedtea-7*
> >
> > Thanks.
> >
>
>
> --
> Alan McKinnon
> alan.mckinnon@gmail.com
>
>
>
Understood. Thanks.

[-- Attachment #2: Type: text/html, Size: 3096 bytes --]

Re: [gentoo-user] stable java virtuals require unstable java packages

[Alan McKinnon]

On 21/04/2015 20:54, Alexander Kapshuk wrote:
> On Tue, Apr 21, 2015 at 8:42 AM, Alan McKinnon <alan.mckinnon@gmail.com
> <mailto:alan.mckinnon@gmail.com>> wrote:
> 
>     On 21/04/2015 07:28, Alexander Kapshuk wrote:
>     > Is it normal for a stable virtual to require an unstable package?
> 
> 
>     No, that is definitely not how virtuals should work.
> 
>     It's either a bug in virtual/jdk-1.7 ebuild or more likely, stable
>     request for icedtead-bin-7 is lagging behind.
> 
>     I know the java team is badly understaffed, lots of mentions on -dev, so
>     maybe log a stable request but against icedtea-bin-7
> 
> 
> 
>     >
>     > Running 'emerge -auND' generates the following output on my system:
>     >
>     > Calculating dependencies  .... ....... done!
>     > [ebuild  NS   ~] dev-java/icedtea-bin-7.2.5.3 [6.1.13.5] USE="X alsa
>     > cups nsplugin -cjk -doc -examples (-selinux) -source -webstart"
>     > [ebuild  NS    ] virtual/jdk-1.7.0 [1.6.0-r2]
>     > [ebuild  NS    ] virtual/jre-1.7.0 [1.6.0-r1]
>     > [ebuild     U ~] dev-java/icedtea-web-1.5.1-r1 [1.4.2] USE="icedtea7%*
>     > -tagsoup%"
>     >
>     > The following keyword changes are necessary to proceed:
>     >  (see "package.accept_keywords" in the portage(5) man page for
>     more details)
>     > # required by virtual/jdk-1.7.0::gentoo
>     > # required by virtual/jre-1.7.0::gentoo
>     > # required by sys-libs/db-4.8.30-r2::gentoo[java]
>     > # required by dev-libs/redland-1.0.16::gentoo[berkdb]
>     > # required by app-office/libreoffice-4.4.1.2::gentoo
>     > # required by @selected
>     > # required by @world (argument)
>     > =dev-java/icedtea-bin-7.2.5.3 ~x86
>     > # required by dev-java/icedtea-bin-7.2.5.3::gentoo[nsplugin]
>     > # required by virtual/jdk-1.7.0::gentoo
>     > # required by virtual/jre-1.7.0::gentoo
>     > # required by sys-libs/db-4.8.30-r2::gentoo[java]
>     > # required by dev-libs/redland-1.0.16::gentoo[berkdb]
>     > # required by app-office/libreoffice-4.4.1.2::gentoo
>     > # required by @selected
>     > # required by @world (argument)
>     > =dev-java/icedtea-web-1.5.1-r1 ~x86
>     >
>     > /usr/portage/virtual/jre/jre-1.7.0.ebuild:12
>     >         =virtual/jdk-1.7.0*
>     > /usr/portage/virtual/jdk/jdk-1.7.0.ebuild:12,13
>     >         =dev-java/icedtea-bin-7*
>     >         =dev-java/icedtea-7*
>     >
>     > Thanks.
>     >
> 
> 
>     --
>     Alan McKinnon
>     alan.mckinnon@gmail.com <mailto:alan.mckinnon@gmail.com>
> 
> 
> 
> Understood. Thanks.

Turns out the virtual is working as designed - see Andreas's post above

I recall now a discussion on -dev about this ages ago, and a consensus
emerged then to keep things as they currently are (changing it requires
much effort and has all manner of effects on the tree). The actual rule is:

A virtual can (by definition) be stable as soon as one of its providers
is stable.



-- 
Alan McKinnon
alan.mckinnon@gmail.com

Re: [gentoo-user] stable java virtuals require unstable java packages

[Dale]

Alan McKinnon wrote:
> Turns out the virtual is working as designed - see Andreas's post
> above I recall now a discussion on -dev about this ages ago, and a
> consensus emerged then to keep things as they currently are (changing
> it requires much effort and has all manner of effects on the tree).
> The actual rule is: A virtual can (by definition) be stable as soon as
> one of its providers is stable. 

So if we really don't want one of the other packages that satisfies what
the virtual needs, we need to mask the others locally?  

Great.  :/

Dale

:-)  :-) 

Re: [gentoo-user] stable java virtuals require unstable java packages

[hydra]

[-- Attachment #1: Type: text/plain, Size: 713 bytes --]

On Wed, Apr 22, 2015 at 8:09 AM, Dale <rdalek1967@gmail.com> wrote:

> Alan McKinnon wrote:
> > Turns out the virtual is working as designed - see Andreas's post
> > above I recall now a discussion on -dev about this ages ago, and a
> > consensus emerged then to keep things as they currently are (changing
> > it requires much effort and has all manner of effects on the tree).
> > The actual rule is: A virtual can (by definition) be stable as soon as
> > one of its providers is stable.
>
> So if we really don't want one of the other packages that satisfies what
> the virtual needs, we need to mask the others locally?
>
> Great.  :/
>
> Dale
>
> :-)  :-)
>
>

https://bugs.gentoo.org/show_bug.cgi?id=546902

[-- Attachment #2: Type: text/html, Size: 1255 bytes --]

Re: [gentoo-user] stable java virtuals require unstable java packages

[Alan McKinnon]

On 22/04/2015 08:09, Dale wrote:
> Alan McKinnon wrote:
>> Turns out the virtual is working as designed - see Andreas's post
>> above I recall now a discussion on -dev about this ages ago, and a
>> consensus emerged then to keep things as they currently are (changing
>> it requires much effort and has all manner of effects on the tree).
>> The actual rule is: A virtual can (by definition) be stable as soon as
>> one of its providers is stable. 
> 
> So if we really don't want one of the other packages that satisfies what
> the virtual needs, we need to mask the others locally?  
> 
> Great.  :/
> 
> Dale
> 
> :-)  :-) 
> 


Not totally. Pick which package you want and emerge it, portage knows
you have something that satisfies the virtual and will be happy with it.

If you don't use the main provider that's first in the list, like
Alexander has here, then portage gets wordy when the provider is not yet
stabilized. Take note, keyword it if you need to, and move along with
the rest of your life.

No need to mask all the other providers


-- 
Alan McKinnon
alan.mckinnon@gmail.com

Re: [gentoo-user] stable java virtuals require unstable java packages

[Dale]

Alan McKinnon wrote:
> On 22/04/2015 08:09, Dale wrote:
>> Alan McKinnon wrote:
>>> Turns out the virtual is working as designed - see Andreas's post
>>> above I recall now a discussion on -dev about this ages ago, and a
>>> consensus emerged then to keep things as they currently are (changing
>>> it requires much effort and has all manner of effects on the tree).
>>> The actual rule is: A virtual can (by definition) be stable as soon as
>>> one of its providers is stable. 
>> So if we really don't want one of the other packages that satisfies what
>> the virtual needs, we need to mask the others locally?  
>>
>> Great.  :/
>>
>> Dale
>>
>> :-)  :-) 
>>
>
> Not totally. Pick which package you want and emerge it, portage knows
> you have something that satisfies the virtual and will be happy with it.
>
> If you don't use the main provider that's first in the list, like
> Alexander has here, then portage gets wordy when the provider is not yet
> stabilized. Take note, keyword it if you need to, and move along with
> the rest of your life.
>
> No need to mask all the other providers
>
>


Whew!!  Had me worried for a minute there. 

Dale

:-)  :-) 

Re: [gentoo-user] stable java virtuals require unstable java packages

[R0b0t1]

I've been using 1.7 for some time now, I wouldn't have too many qualms using it.

Re: [gentoo-user] stable java virtuals require unstable java packages

[Dale]

R0b0t1 wrote:
> I've been using 1.7 for some time now, I wouldn't have too many qualms using it.
>
>

Well, I was wondering about other situations not this one.  There are
quite a few virtuals in the tree.  One recent change was the
libav/ffmpeg.  I'm sure there are others to.

Dale

:-)  :-) 

Re: [gentoo-user] stable java virtuals require unstable java packages

[Andreas K. Huettel]

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA512

Am Mittwoch, 22. April 2015, 10:18:06 schrieb Dale:
> R0b0t1 wrote:
> > I've been using 1.7 for some time now, I wouldn't have too many qualms using it.
> >
> >
> 
> Well, I was wondering about other situations not this one.  There are
> quite a few virtuals in the tree.  One recent change was the
> libav/ffmpeg.  I'm sure there are others to.
> 
> Dale
> 
> :-)  :-) 
> 

The recent changes to the libav/ffmpeg handling were done (also) to fix this particular problem there. 

Before, the virtual pulled in one solution, but it was "up to portage" to decide which one. So sometimes the decision flipped to another solution (e.g. because of delays in stabling), and (since ffmpeg and libav block each other) things went awry.

Now, the virtual and the packages get an explicit instruction via the libav useflag: "I want *this* solution."

- -- 

Andreas K. Huettel
Gentoo Linux developer 
dilfridge@gentoo.org
http://www.akhuettel.de/

-----BEGIN PGP SIGNATURE-----
Version: GnuPG v2.0
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=I3wu
-----END PGP SIGNATURE-----

Re: [gentoo-user] stable java virtuals require unstable java packages

[Alexander Kapshuk]

[-- Attachment #1: Type: text/plain, Size: 3267 bytes --]

On Wed, Apr 22, 2015 at 8:41 AM, Alan McKinnon <alan.mckinnon@gmail.com>
wrote:

> On 21/04/2015 20:54, Alexander Kapshuk wrote:
> > On Tue, Apr 21, 2015 at 8:42 AM, Alan McKinnon <alan.mckinnon@gmail.com
> > <mailto:alan.mckinnon@gmail.com>> wrote:
> >
> >     On 21/04/2015 07:28, Alexander Kapshuk wrote:
> >     > Is it normal for a stable virtual to require an unstable package?
> >
> >
> >     No, that is definitely not how virtuals should work.
> >
> >     It's either a bug in virtual/jdk-1.7 ebuild or more likely, stable
> >     request for icedtead-bin-7 is lagging behind.
> >
> >     I know the java team is badly understaffed, lots of mentions on
> -dev, so
> >     maybe log a stable request but against icedtea-bin-7
> >
> >
> >
> >     >
> >     > Running 'emerge -auND' generates the following output on my system:
> >     >
> >     > Calculating dependencies  .... ....... done!
> >     > [ebuild  NS   ~] dev-java/icedtea-bin-7.2.5.3 [6.1.13.5] USE="X
> alsa
> >     > cups nsplugin -cjk -doc -examples (-selinux) -source -webstart"
> >     > [ebuild  NS    ] virtual/jdk-1.7.0 [1.6.0-r2]
> >     > [ebuild  NS    ] virtual/jre-1.7.0 [1.6.0-r1]
> >     > [ebuild     U ~] dev-java/icedtea-web-1.5.1-r1 [1.4.2]
> USE="icedtea7%*
> >     > -tagsoup%"
> >     >
> >     > The following keyword changes are necessary to proceed:
> >     >  (see "package.accept_keywords" in the portage(5) man page for
> >     more details)
> >     > # required by virtual/jdk-1.7.0::gentoo
> >     > # required by virtual/jre-1.7.0::gentoo
> >     > # required by sys-libs/db-4.8.30-r2::gentoo[java]
> >     > # required by dev-libs/redland-1.0.16::gentoo[berkdb]
> >     > # required by app-office/libreoffice-4.4.1.2::gentoo
> >     > # required by @selected
> >     > # required by @world (argument)
> >     > =dev-java/icedtea-bin-7.2.5.3 ~x86
> >     > # required by dev-java/icedtea-bin-7.2.5.3::gentoo[nsplugin]
> >     > # required by virtual/jdk-1.7.0::gentoo
> >     > # required by virtual/jre-1.7.0::gentoo
> >     > # required by sys-libs/db-4.8.30-r2::gentoo[java]
> >     > # required by dev-libs/redland-1.0.16::gentoo[berkdb]
> >     > # required by app-office/libreoffice-4.4.1.2::gentoo
> >     > # required by @selected
> >     > # required by @world (argument)
> >     > =dev-java/icedtea-web-1.5.1-r1 ~x86
> >     >
> >     > /usr/portage/virtual/jre/jre-1.7.0.ebuild:12
> >     >         =virtual/jdk-1.7.0*
> >     > /usr/portage/virtual/jdk/jdk-1.7.0.ebuild:12,13
> >     >         =dev-java/icedtea-bin-7*
> >     >         =dev-java/icedtea-7*
> >     >
> >     > Thanks.
> >     >
> >
> >
> >     --
> >     Alan McKinnon
> >     alan.mckinnon@gmail.com <mailto:alan.mckinnon@gmail.com>
> >
> >
> >
> > Understood. Thanks.
>
> Turns out the virtual is working as designed - see Andreas's post above
>
> I recall now a discussion on -dev about this ages ago, and a consensus
> emerged then to keep things as they currently are (changing it requires
> much effort and has all manner of effects on the tree). The actual rule is:
>
> A virtual can (by definition) be stable as soon as one of its providers
> is stable.
>
>
>
> --
> Alan McKinnon
> alan.mckinnon@gmail.com
>
>
>
I ended up doing as suggested by Andreas and yourself.

Thanks.

[-- Attachment #2: Type: text/html, Size: 4843 bytes --]