Thread: How do you manage versions of your own code ?
Hi all,
After going a bit of back and forth on different versions that I am working on I think I might as well ask how do you guys work on an open-source development project once you are disconnected ?
I mean, if I do a svn co and start working, it wouldnt allow me to check in a version since its not a distributedVCS. If I use bzr-svn, it does checkout, but I still can't check in a half-way state since it tries to dump the patch to the online SVN too which of course fails.
Apart from the current issue of me not having submit rights, how did you guys work on this when you don't have submit rights to a non-distributed VCS ?
Did you just keep storing revision patches and patch / revert on the latest checked out revision as and when you needed to switch to an older version of your code ?
(Btw, by version here I meant an older half-way-state of my own working code, not the PgAdmin version)
Thanks
Robins Tharakan
After going a bit of back and forth on different versions that I am working on I think I might as well ask how do you guys work on an open-source development project once you are disconnected ?
I mean, if I do a svn co and start working, it wouldnt allow me to check in a version since its not a distributedVCS. If I use bzr-svn, it does checkout, but I still can't check in a half-way state since it tries to dump the patch to the online SVN too which of course fails.
Apart from the current issue of me not having submit rights, how did you guys work on this when you don't have submit rights to a non-distributed VCS ?
Did you just keep storing revision patches and patch / revert on the latest checked out revision as and when you needed to switch to an older version of your code ?
(Btw, by version here I meant an older half-way-state of my own working code, not the PgAdmin version)
Thanks
Robins Tharakan
On Feb 6, 2008 3:34 AM, Robins Tharakan <tharakan@gmail.com> wrote: > Hi all, > > After going a bit of back and forth on different versions that I am working > on I think I might as well ask how do you guys work on an open-source > development project once you are disconnected ? > > I mean, if I do a svn co and start working, it wouldnt allow me to check in > a version since its not a distributedVCS. If I use bzr-svn, it does > checkout, but I still can't check in a half-way state since it tries to dump > the patch to the online SVN too which of course fails. > > Apart from the current issue of me not having submit rights, how did you > guys work on this when you don't have submit rights to a non-distributed VCS > ? > Did you just keep storing revision patches and patch / revert on the latest > checked out revision as and when you needed to switch to an older version of > your code ? > > (Btw, by version here I meant an older half-way-state of my own working > code, not the PgAdmin version) I just keep multiple copies of the tree around if I'm working on more than one large project at a time, and take periodic diff's against SVN if I think I need to. I've never really found the need for a distributed scm with pgAdmin. /D
Dave Page wrote: > On Feb 6, 2008 3:34 AM, Robins Tharakan <tharakan@gmail.com> wrote: >> Hi all, >> >> After going a bit of back and forth on different versions that I am working >> on I think I might as well ask how do you guys work on an open-source >> development project once you are disconnected ? >> >> I mean, if I do a svn co and start working, it wouldnt allow me to check in >> a version since its not a distributedVCS. If I use bzr-svn, it does >> checkout, but I still can't check in a half-way state since it tries to dump >> the patch to the online SVN too which of course fails. >> >> Apart from the current issue of me not having submit rights, how did you >> guys work on this when you don't have submit rights to a non-distributed VCS >> ? >> Did you just keep storing revision patches and patch / revert on the latest >> checked out revision as and when you needed to switch to an older version of >> your code ? >> >> (Btw, by version here I meant an older half-way-state of my own working >> code, not the PgAdmin version) > > I just keep multiple copies of the tree around if I'm working on more > than one large project at a time, and take periodic diff's against SVN > if I think I need to. I've never really found the need for a > distributed scm with pgAdmin. > I don't work with so many patches to feel the need of something better than subversion. I'm usually working with only one patch. -- Guillaume. http://www.postgresqlfr.org http://dalibo.com
Hi, 2008/2/6, Robins Tharakan <tharakan@gmail.com>: > After going a bit of back and forth on different versions that I am working > on I think I might as well ask how do you guys work on an open-source > development project once you are disconnected ? We use Mercurial ( http://www.selenic.com/mercurial/wiki/ ). Works great. New versions from the main project we import to separate subtrees and carry our changes over with a diff from the older subtree. I wonder whether anyone has a better method for that. Looking forward to provide the cursored querying I promised last year but that we are still working on, -- Kalle Hallivuori +358-41-5053073 http://korpiq.iki.fi/
Robins Tharakan wrote: > After going a bit of back and forth on different versions that I am working > on I think I might as well ask how do you guys work on an open-source > development project once you are disconnected ? > > I mean, if I do a svn co and start working, it wouldnt allow me to check in > a version since its not a distributedVCS. If I use bzr-svn, it does > checkout, but I still can't check in a half-way state since it tries to dump > the patch to the online SVN too which of course fails. > > Apart from the current issue of me not having submit rights, how did you > guys work on this when you don't have submit rights to a non-distributed VCS > ? > Did you just keep storing revision patches and patch / revert on the latest > checked out revision as and when you needed to switch to an older version of > your code ? I don't do much pgAdmin development, but with Postgres, I keep an rsync'd copy of the whole CVS repository on my laptop, and work against that. Much more convenient, you can do diffs, checkouts, view histories, all offline and quickly. -- Heikki Linnakangas EnterpriseDB http://www.enterprisedb.com
On Wed, Feb 06, 2008 at 09:04:46AM +0530, Robins Tharakan wrote: > Hi all, > > After going a bit of back and forth on different versions that I am working > on I think I might as well ask how do you guys work on an open-source > development project once you are disconnected ? > > I mean, if I do a svn co and start working, it wouldnt allow me to check in > a version since its not a distributedVCS. If I use bzr-svn, it does > checkout, but I still can't check in a half-way state since it tries to dump > the patch to the online SVN too which of course fails. > > Apart from the current issue of me not having submit rights, how did you > guys work on this when you don't have submit rights to a non-distributed VCS > ? > Did you just keep storing revision patches and patch / revert on the latest > checked out revision as and when you needed to switch to an older version of > your code ? > > (Btw, by version here I meant an older half-way-state of my own working > code, not the PgAdmin version) I'm never offline long enough for that to matter. Since you can do a diff without being online, that's usually enough. If I do work on multiple patches at the same time that aren't entirely isolated, I check out multiple copies. //Magnus