Gentoo Forums
Gentoo Forums
Gentoo Forums
Quick Search: in
Updating ebuilds without incrementing the revision number
View unanswered posts
View posts from last 24 hours

 
Reply to topic    Gentoo Forums Forum Index Portage & Programming
View previous topic :: View next topic  
Author Message
drheld
n00b
n00b


Joined: 18 Mar 2005
Posts: 2

PostPosted: Fri Mar 18, 2005 7:49 am    Post subject: Updating ebuilds without incrementing the revision number Reply with quote

Am I the only one who finds it annoying when ebuilds are updated without having their revision number bumped?

If an ebuild is released, regardless of how broken it is, when it is, I think it would be better if the revision number is incremented when it is fixed. I'm not sure why it sometimes isn't, as I can't see the harm in incrementing it. On the other hand, it can be misleading for the exact same ebuild to change.

Perhaps there should be some sort of policy on this?
Back to top
View user's profile Send private message
Earthwings
Administrator
Administrator


Joined: 14 Apr 2003
Posts: 7751
Location: Karlsruhe, Germany

PostPosted: Fri Mar 18, 2005 7:57 am    Post subject: Reply with quote

There is a policy for that already.
http://www.gentoo.org/proj/en/devrel/handbook/handbook.xml?part=3&chap=1#doc_chap1 wrote:
Versioning and revision bumps
Package revision numbers should be incremented by Gentoo Linux developers when the ebuild has changed to the point where users would want to upgrade. Typically, this is the case when fixes are made to an ebuild that affect the resultant installed files, but the ebuild uses the same source tarball as the previous release. If you make an internal, stylistic change to the ebuild that does not change any of the installed files, then there is no need to bump the revision number. Likewise, if you fix a compilation problem in the ebuild that was affecting some users, there is no need to bump the revision number, since those for whom it worked perfectly would see no benefit in installing a new revision, and those who experienced the problem do not have the package installed (since compilation failed) and thus have no need for the new revision number to force an upgrade. A revision bump is also not necessary if a minority of users will be affected and the package has a nontrivial average compilation time; use your best judgement in these circumstances.
Back to top
View user's profile Send private message
Element Dave
Tux's lil' helper
Tux's lil' helper


Joined: 10 Nov 2006
Posts: 82

PostPosted: Sat Mar 23, 2013 9:12 pm    Post subject: Reply with quote

Eight years later.... and the policy is still fucking INSANE. The original poster is spot on with. A revision occurs when something changes, regardless of whether a random developer feels the change is sufficient that users should "want" to upgrade.

Quote:
Likewise, if you fix a compilation problem in the ebuild that was affecting some users, there is no need to bump the revision number, since those for whom it worked perfectly would see no benefit in installing a new revision, and those who experienced the problem do not have the package installed (since compilation failed) and thus have no need for the new revision number to force an upgrade.


Of course, if package-x.y.z failed to compile for a subset of users, they could reasonably assume that attempting to remerge it would also fail, whereas a new revision might contain a fix. Who on Earth does software development like this?

Please excuse the language, but the level of insanity with which the portage tree is maintained has really gotten on my nerves.
Back to top
View user's profile Send private message
NeddySeagoon
Administrator
Administrator


Joined: 05 Jul 2003
Posts: 32091
Location: 56N 3W

PostPosted: Sat Mar 23, 2013 9:38 pm    Post subject: Reply with quote

I can see both points of view.

Do you really want to rebuild packages that have a keyword revision for another arch?
I don't. If a packages gets ~sparc or ~arm keywords, I really don't want to rebuild to get identical binaries on my ~amd64 and ~x86 systems.

Its just as insane to force revision bumps for everything.

The policy is a reasonable middle ground.
_________________
Regards,

NeddySeagoon

Computer users fall into two groups:-
those that do backups
those that have never had a hard drive fail.
Back to top
View user's profile Send private message
leo.the_zoo
Tux's lil' helper
Tux's lil' helper


Joined: 04 Jul 2005
Posts: 148
Location: Poland

PostPosted: Sat Mar 23, 2013 10:11 pm    Post subject: Reply with quote

@Element Dave
I understand your point to certain extent but doing a version bump for every little change would force, say, 90% of users, to waste resources on unnecessary recompilation just to get the same binaries. If you are among 10% of those for whom the ebuild did not work as expected and the fix would only address this issue, version bump is not necessary since Portage will be trying to upgrade every single time you run emerge world -DuNv... unless you mask the failing ebuild. Portage could address this issue by logging these failures somewhere and then attempt to update with the same version only upon change of ebuild's checksums. But the current policy can still be considered a decent compromise.
Back to top
View user's profile Send private message
phajdan.jr
Developer
Developer


Joined: 23 Mar 2006
Posts: 1767
Location: Poland

PostPosted: Sat Mar 23, 2013 10:19 pm    Post subject: Re: Updating ebuilds without incrementing the revision numbe Reply with quote

drheld wrote:
If an ebuild is released, regardless of how broken it is, when it is, I think it would be better if the revision number is incremented when it is fixed. I'm not sure why it sometimes isn't, as I can't see the harm in incrementing it. On the other hand, it can be misleading for the exact same ebuild to change.


drheld, could you give an example?

One example where it's actually not a good idea to revision bump is when fixing compile failures. There is no need to trigger an update on systems where it compiled successfully, and on systems where it failed a revision bump doesn't change anything. Add to that a package that takes quite some computing power to compile (e.g. www-client/chromium or libreoffice) and I hope you can see the problem.
_________________
http://phajdan-jr.blogspot.com/
Back to top
View user's profile Send private message
leo.the_zoo
Tux's lil' helper
Tux's lil' helper


Joined: 04 Jul 2005
Posts: 148
Location: Poland

PostPosted: Sat Mar 23, 2013 11:02 pm    Post subject: Reply with quote

I'm afraid drheld won't give you any example since he started this thread back in 2005 (and does not seem to drop by since then). Element Dave resurrected it after a long time, so perhaps he's the one to answer since he supports the views of drheld. :)
Back to top
View user's profile Send private message
Display posts from previous:   
Reply to topic    Gentoo Forums Forum Index Portage & Programming All times are GMT
Page 1 of 1

 
Jump to:  
You cannot post new topics in this forum
You cannot reply to topics in this forum
You cannot edit your posts in this forum
You cannot delete your posts in this forum
You cannot vote in polls in this forum