So, hi, I have a question. Say one would want to convert a darcs repo to bzr. Say, hypothetically, the tailor repo. One would do:
% darcs get http://darcs.arstecnica.it/tailor /tmp/darcs-repo And then convert it to bzr with a config file like: [tailor] source = darcs:tailor target = bzr:tailor root-directory = /tmp/tailor [darcs:tailor] repository = /tmp/darcs-repo [bzr:tailor] # empty When tailor finishes, /tmp/tailor is both a copy of the darcs tailor repository, and a bzr version of it. If one wants to keep this up to date, one can: % cd /tmp/darcs-repo && darcs pull -a % tailor --config /tmp/tailor.config But I wanted to keep only one copy of the darcs repo, so I expected that if I changed (after the initial migration is finished) the [darcs:tailor] stanza above to: [darcs:tailor] repository = http://darcs.arstecnica.it/tailor I could just rm -rf /tmp/darcs-repo, and have tailor update /tmp/tailor directly from arstecnica.it. The result is that it works, but it's excruciatingly slow. Those who know their way inside darcs already know why, but it took me a bit to figure out: seems like, whilst the pulled-from-the-web repo in /tmp/darcs-repo has a optimized _darcs/inventory, going only from the last tag onwards, pulls are fast. But the _darcs/inventory file in the repo created by tailor in /tmp/tailor is not a short one, but a full one, thus making the pulling very slow. What I did was an extra step: % cp /tmp/darcs-repo/_darcs/inventory /tmp/tailor/_darcs % cp /tmp/darcs-repo/_darcs/inventories/* /tmp/tailor/_darcs/inventories With this, it works the way I want, but I can't help wonder if tailor could do something so that the resulting repo in /tmp/tailor is useable right away. Thanks. -- Adeodato Simó dato at net.com.org.es Debian Developer adeodato at debian.org Listening to: LFO - Mokeylips _______________________________________________ Tailor mailing list [email protected] http://lists.zooko.com/mailman/listinfo/tailor
