[gentoo-dev] eclass caching fix. was More unhappy news for overlay-utilizing devs :(

[gentoo-dev] eclass caching fix. was More unhappy news for overlay-utilizing devs :(

[Brian Harring]

On Fri, Feb 18, 2005 at 11:09:30PM +0900, Jason Stubbs wrote:
> Please, PLEASE, next time you see strange behaviour like this, speak loudly. 
> Most bugs are known and classified, even if only in individual portage dev's 
> minds, but unknown bugs are terrible - especially when they effect committing 
> to the main tree. *PLEASE* ask/notify!
...

Um. yeah, seriously, if people know of bugs, but don't report it, they need a swift kick. >:)

The portage devs -didn't- know the cache 'staleness' detection was broken for overlays + eclasses... so a bug 
way back whenever people noticed it would've been *quite* useful.  

Seriously, we would *much* rather have an extra 100 bugs of portage oddities that wind up as invalid bugs, then not know...

So... stepping off the stool and getting back technical stuff, a fix is available at 
http://dev.gentoo.org/~ferringb/eclass-validation-fix.patch
It's also InCvs.

Devs should use this patch... won't break anything either, so no excuse in not using it :)
~brian
--
gentoo-dev@gentoo.org mailing list

[gentoo-dev] Re: [gentoo-core] eclass caching fix. was More unhappy news for overlay-utilizing devs :(

[Donnie Berkholz]

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

Brian Harring wrote:
> So... stepping off the stool and getting back technical stuff, a fix is available at 
> http://dev.gentoo.org/~ferringb/eclass-validation-fix.patch
> It's also InCvs.
> 
> Devs should use this patch... won't break anything either, so no excuse in not using it :)

If we should use it, it would be helpful if we didn't have to keep track
of where it was (i.e., apply it to released portage as a patch).
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.0 (GNU/Linux)

iD8DBQFCGq6NXVaO67S1rtsRAiQdAJ9x/iWqaEy+mq8LxXm6JF9U7wjKPACcDqDb
I5KZJLTT/KtSBVmMl7gfRFU=
=aHni
-----END PGP SIGNATURE-----
--
gentoo-dev@gentoo.org mailing list

Re: [gentoo-dev] Re: [gentoo-core] eclass caching fix. was More unhappy news for overlay-utilizing devs :(

[Brian Harring]

On Mon, Feb 21, 2005 at 08:01:17PM -0800, Donnie Berkholz wrote:
> If we should use it, it would be helpful if we didn't have to keep track
> of where it was (i.e., apply it to released portage as a patch).
E'yep.  :)

Until 2.0.51.16 is stabled, pushing patches into the tree isn't an option though.  Pushing out a 2.0.51-r16 isn't 
really viable.

You actually just touched on the reason why portage is jumping from .51-rN to .51.N; so we can use the -rN version 
component to push out patches while releases are being put through the testing process...
~brian

--
gentoo-dev@gentoo.org mailing list

Re: [gentoo-dev] Re: [gentoo-core] eclass caching fix. was More unhappy news for overlay-utilizing devs :(

[Donnie Berkholz]

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

Brian Harring wrote:
> On Mon, Feb 21, 2005 at 08:01:17PM -0800, Donnie Berkholz wrote:
> 
>>If we should use it, it would be helpful if we didn't have to keep track
>>of where it was (i.e., apply it to released portage as a patch).
> 
> E'yep.  :)
> 
> Until 2.0.51.16 is stabled, pushing patches into the tree isn't an option though.  Pushing out a 2.0.51-r16 isn't 
> really viable.
> 
> You actually just touched on the reason why portage is jumping from .51-rN to .51.N; so we can use the -rN version 
> component to push out patches while releases are being put through the testing process...

No! Don't tell me portage is finally using a sane numbering scheme!
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.0 (GNU/Linux)

iD8DBQFCGrIbXVaO67S1rtsRAhzFAKCRDD6lhIUBrx3BpWqqcSJtbKA3lgCdHXvi
sPXWa+aWCMGZvy24rJ4Ejlk=
=1NcF
-----END PGP SIGNATURE-----
--
gentoo-dev@gentoo.org mailing list

Re: [gentoo-dev] Re: [gentoo-core] eclass caching fix. was More unhappy news for overlay-utilizing devs :(

[Spider]

[-- Attachment #1: Type: text/plain, Size: 426 bytes --]

On Mon, 2005-02-21 at 20:16 -0800, Donnie Berkholz wrote:
> 
> No! Don't tell me portage is finally using a sane numbering scheme!

*Hands on ears singing the smurf-theme loudly*
LA LA lala LA LAAAAA!

..... I won't believe it I don't believe it, I can't believe it until
its staable.....



-- 
begin  .signature
Tortured users / Laughing in pain
See Microsoft KB Article Q265230 for more information.
end


[-- Attachment #2: This is a digitally signed message part --]
[-- Type: application/pgp-signature, Size: 189 bytes --]