[Twisted-Python] Moving sandbox out of the trunk.
If no one yells at me during the next 24 hours, tomorrow I will run svn commit -m 'Move sandbox outside of trunk.' \ svn+ssh://tv@cvs.twistedmatrix.com/svn/Twisted/trunk/sandbox \ svn+ssh://tv@cvs.twistedmatrix.com/svn/Twisted/sandbox This is your only chance to prevent it. I want it gone.
On Sun, 2004-12-26 at 15:42 +0200, Tommi Virtanen wrote:
If no one yells at me during the next 24 hours, tomorrow I will run
svn commit -m 'Move sandbox outside of trunk.' \ svn+ssh://tv@cvs.twistedmatrix.com/svn/Twisted/trunk/sandbox \ svn+ssh://tv@cvs.twistedmatrix.com/svn/Twisted/sandbox
Consider this a yell. What is your motivation for doing this? I'm not saying I disagree, necessarily, but I'd like some discussion first.
Itamar Shtull-Trauring wrote:
If no one yells at me during the next 24 hours, tomorrow I will run
Consider this a yell. What is your motivation for doing this? I'm not saying I disagree, necessarily, but I'd like some discussion first.
I think other people's sandbox changes just clutter up "svn st -u" and "svn log", sandbox isn't really part of the twisted project (should not get included in released source; should not get tagged; should not get branched), and say 80% of the people co'ing trunk have no interest in the sandbox (and when they have, it's easy to co when really wanted). What's the _harm_ in moving it outside of trunk? Nevow has its sandbox outside the trunk, and I really prefer it that way.
-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 I agree with the change, but I'm one of the people who will be negatively impacted. Probably, everyone doing serious work in the sandbox will be in the same boat with me. Published in mailing list discussions or wikis or webpages, one occasionally mentions one's new project which is not yet ready for Twisted, and one occasionally gets email from people trying it out. Changing the location of sandbox will mean confused emails from people trying out that experimental code. That said, I'm okay with this. C Tommi Virtanen wrote: | Itamar Shtull-Trauring wrote: | |>> If no one yells at me during the next 24 hours, |>> tomorrow I will run |>> |> |> Consider this a yell. What is your motivation for doing this? I'm not |> saying I disagree, necessarily, but I'd like some discussion first. |> |> | I think other people's sandbox changes just clutter up "svn st -u" and | "svn log", sandbox isn't really part of the twisted project (should not | get included in released source; should not get tagged; should not | get branched), and say 80% of the people co'ing trunk have no interest | in the sandbox (and when they have, it's easy to co when really wanted). | | What's the _harm_ in moving it outside of trunk? Nevow has its sandbox | outside the trunk, and I really prefer it that way. | | | _______________________________________________ | Twisted-Python mailing list | Twisted-Python@twistedmatrix.com | http://twistedmatrix.com/cgi-bin/mailman/listinfo/twisted-python -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.2.5 (MingW32) Comment: Using GnuPG with Thunderbird - http://enigmail.mozdev.org iD8DBQFBzxXD3A5SrXAiHQcRAimHAJ45aZmqW5VgvmugEzixPsyHSVssYwCeJEkZ 8ewBuaThiRlVbG6eeUjKRuA= =7mSJ -----END PGP SIGNATURE-----
Tommi Virtanen wrote:
I think other people's sandbox changes just clutter up "svn st -u" and "svn log", sandbox isn't really part of the twisted project (should not get included in released source; should not get tagged; should not get branched), and say 80% of the people co'ing trunk have no interest in the sandbox (and when they have, it's easy to co when really wanted).
Also, sandbox changes trigger buildbot, adding noise to the waterfall.
-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 Actually, they're not supposed to. If they are triggering the buildbot, it's a recent regression. C Tommi Virtanen wrote: | Tommi Virtanen wrote: | |> I think other people's sandbox changes just clutter up "svn st -u" and |> "svn log", sandbox isn't really part of the twisted project (should not |> get included in released source; should not get tagged; should not |> get branched), and say 80% of the people co'ing trunk have no interest |> in the sandbox (and when they have, it's easy to co when really wanted). | | | Also, sandbox changes trigger buildbot, adding noise to the waterfall. | | _______________________________________________ | Twisted-Python mailing list | Twisted-Python@twistedmatrix.com | http://twistedmatrix.com/cgi-bin/mailman/listinfo/twisted-python -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.2.5 (MingW32) Comment: Using GnuPG with Thunderbird - http://enigmail.mozdev.org iD8DBQFB0D3A3A5SrXAiHQcRAr9OAJ956hUcVDUMfOKx7XCkcZngm2yvJACfQi6B gSvpXP1wpOA34mvT/KxhY9U= =v78m -----END PGP SIGNATURE-----
On Sun, 2004-12-26 at 11:49 -0800, Cory Dodt wrote:
That said, I'm okay with this.
I won't be as negatively impacted, but I agree with this sentiment. I believe the commit should go ahead - I hope to make branches a more integral part of the development process soon, so anything that makes it easier to check them out is worthwhile. So, +1
Glyph Lefkowitz wrote:
On Sun, 2004-12-26 at 11:49 -0800, Cory Dodt wrote:
That said, I'm okay with this.
I won't be as negatively impacted, but I agree with this sentiment.
I believe the commit should go ahead - I hope to make branches a more integral part of the development process soon, so anything that makes it easier to check them out is worthwhile.
So, +1
You two seem to be happy and itamar only wanted discussion :) r12804 moved sandbox out of the trunk. To get your hands on the new sandbox, use something like svn co svn://cvs.twistedmatrix.com/svn/Twisted/sandbox svn co svn+ssh://cvs.twistedmatrix.com/svn/Twisted/sandbox
FYI, this may cause problems with existing checkouts such that the
only (obvious?) fix is to checkout a new copy. At least it did with
mine.
After a failed attempt to 'svn up' (which said some stuff about not
deleting the current directory since it was changed), svn stat told me
"! .", which means ". is not what I think it is". It was totally
incomprehensible to me, so I just checked out again.
On Tue, 28 Dec 2004 01:58:41 +0200, Tommi Virtanen
Glyph Lefkowitz wrote:
On Sun, 2004-12-26 at 11:49 -0800, Cory Dodt wrote:
That said, I'm okay with this.
I won't be as negatively impacted, but I agree with this sentiment.
I believe the commit should go ahead - I hope to make branches a more integral part of the development process soon, so anything that makes it easier to check them out is worthwhile.
So, +1
You two seem to be happy and itamar only wanted discussion :)
r12804 moved sandbox out of the trunk.
To get your hands on the new sandbox, use something like
svn co svn://cvs.twistedmatrix.com/svn/Twisted/sandbox svn co svn+ssh://cvs.twistedmatrix.com/svn/Twisted/sandbox
_______________________________________________ Twisted-Python mailing list Twisted-Python@twistedmatrix.com http://twistedmatrix.com/cgi-bin/mailman/listinfo/twisted-python
-- Twisted | Christopher Armstrong: International Man of Twistery Radix | -- http://radix.twistedmatrix.com | Release Manager, Twisted Project \\\V/// | -- http://twistedmatrix.com |o O| | Founding Member, Hobart Hacking Society w----v----w-+ -- http://hackingsociety.org/chapters/hash
Christopher Armstrong wrote:
FYI, this may cause problems with existing checkouts such that the only (obvious?) fix is to checkout a new copy. At least it did with mine.
After a failed attempt to 'svn up' (which said some stuff about not deleting the current directory since it was changed), svn stat told me "! .", which means ". is not what I think it is". It was totally incomprehensible to me, so I just checked out again.
You had non-versioned stuff in sandbox/ that was left around for you to clean up manually. See what's left and finally rm -rf sandbox. That happens whenever svn up sees repository has removed a file, but there's a "conflict" caused by local files hanging around. Just as a counterpoint, my sandbox directory was clean and svn up removed it happily.
participants (6)
-
Christopher Armstrong -
Cory Dodt -
Glyph Lefkowitz -
Itamar Shtull-Trauring -
Jp Calderone -
Tommi Virtanen