If you have been following the ##crawl-dev IRC channel, you may already be aware that we recently (this time last week) moved our code base from SourceForge.net to Gitorious.org. This move comes mainly from some minor frustrations with SourceForge, most of which relating to their recent reliability problems. SourceForge have been a excellent host, but as we’ve recently been butting up against its limitations, it seemed like a good opportunity to move.

You can find the main page for the Crawl code on Gitorious at: <http://gitorious.org/crawl>.

From there, you can see all of the contained repository, begin watching any of them, and see all of the changes represented as a stream; from the repositories, you can clone and begin tracking and pushing changes to these clones.

Therefore, if you’ve been following Crawl development via Git on SourceForge, you’ll need to make some changes in order to switch to tracking Gitorious instead.

For newer versions of Git, you can execute the following command to update your “origin” remote — replacing “origin” with whichever name you use for Crawl’s main remote:

git remote set-url origin git://gitorious.org/crawl/crawl.git

For all other versions of Git, including newer ones, you can use the following command:

git config remote.origin.url git://gitorious.org/crawl/crawl.git

If you have commit access to Gitorious, you should replace git://gitorious.org/crawl/crawl.git with ssh://git@gitorious.org:crawl/crawl.git in all of the above statements.

Our suite of submodules (specifically including Lua, Luajit, sqlite, and so-on) currently still point to SourceForge, however I intend to properly update them to point to the relevant repository on Gitorious.

So, how does this impact Crawl’s external or contributed development procedure?

For simple patches: Currently, our process for submitting patches remains the same: you can still upload them as an attachment to an issue on our Mantis tracker, and these patches will be applied and merged back into the upstream code base as per our previous records.

For larger changes, it is now possible to register for Gitorious and clone our repository there. We currently have no procedures for dealing with merge requests from Gitorious, however, I would recommend that you still open a ticket on Mantis, and provide a link to the merge request on Gitorious along with the ticket.

As usual, if you have any issues, please leave a comment here, or alternately, drop by ##crawl-dev and hopefully someone there can help. Happy crawling!