| 1 |
> > Having a live tree requires people to be perfect. People are not |
| 2 |
> > perfect and requiring it is ridiculous. I love having commits in my |
| 3 |
> > local tree within the hour, but having a stable and unstable branch |
| 4 |
> > makes a lot of sense. |
| 5 |
> |
| 6 |
> Does it? How does having a stable and unstable branch differ from |
| 7 |
> having stable and unstable keywords? |
| 8 |
|
| 9 |
Agreed. That doesn't make sense. |
| 10 |
|
| 11 |
Subversion has what is known as pre-commit hooks. |
| 12 |
Using those it's very easy to: |
| 13 |
* prevent (most or some) committers from designating ebuilds as stable |
| 14 |
* allow committers to designate ebuilds as stable under a certain path only |
| 15 |
* strictly limit a commiters access to a part of the tree |
| 16 |
|
| 17 |
Slightly harder, but could be done too: |
| 18 |
* deny commits if it breaks ebuild dependencies |
| 19 |
|
| 20 |
If you want central control of what's happening in the repository, |
| 21 |
Subversion seems like the way to go. |
| 22 |
|
| 23 |
> SVN requires at least 2x the tree size for storage on the local machine, |
| 24 |
|
| 25 |
This is a feature, not a bug. |
| 26 |
It allows you to keep operations local, fx. do a diff without being |
| 27 |
online with the repository. |
| 28 |
|
| 29 |
> checkouts take something akin to an order of magnitude longer than CVS. |
| 30 |
|
| 31 |
In the past Subversion performance was sub-par. I haven't |
| 32 |
systematically tested performance, but I would expect it to be much |
| 33 |
better now. |
| 34 |
|
| 35 |
Performance is gradually improved, see fx. r18867, r18944 and r19420: |
| 36 |
http://svn.collab.net/viewvc/svn?view=rev&revision=18867 |
| 37 |
http://svn.collab.net/viewvc/svn?view=rev&revision=18944 |
| 38 |
http://svn.collab.net/viewvc/svn?view=rev&revision=19420 |
| 39 |
|
| 40 |
GIT might perform better, since Linus specifically emphasized |
| 41 |
non-O(n^2) performance in the design. But being a decentralized tool, |
| 42 |
I'm not sure how well it fits the needs here. |
| 43 |
|
| 44 |
> The former is annoying, but liveable, but |
| 45 |
> the latter is a deal-breaker. |
| 46 |
|
| 47 |
I don't think so. You really rarely do a complete checkout. |
| 48 |
|
| 49 |
> - No changeset/merge tracking |
| 50 |
|
| 51 |
Solutions exist, including svnmerge and svk. |
| 52 |
"Official" solution actively worked on at the moment, check out the |
| 53 |
svn dev list. |
| 54 |
|
| 55 |
A benefit of Subversion that I personally like is that FSFS |
| 56 |
repositories are extremely easy to rsync. |
| 57 |
|
| 58 |
-- |
| 59 |
gentoo-dev@g.o mailing list |
Archives