On Thu, Dec 08, 2005 at 11:20:43AM -0000, Simon Marlow wrote:
Certainly performance of the --partial tree seems good enough, though I don't like that I can't see the history for individual files. I can't
Just to clarify, that limitation exists only when using --partial. One thing we could do is put a tar.bz2 up periodically that contains a full copy of the full history, which should be useful for core ghc hackers such as yourself. It should be faster than downloading the 20,000 patches individually. OTOH, the pain of a get without --partial only has to be endured once per person, and there are probably very few people that care about the full history of things dating back before their own involvement. (You'll have the full history, on a per-file basis, of things starting from the date of the last snapshot and moving forward when using --partial.) And then there are also advantages to consider: with CVS, you can't get history of things *at all* unless you have a live Internet connection, and even then it doesn't preserve things such as renames. With Darcs, once you have a local repo, you can look at changelogs all you want without ever having to hit the network.
get browsing to work using Trac: with the full darcs repository it takes
I have never worked much with these web front-ends. My understanding is that Trac is probably not the most efficient front-end to darcs, as it tries to put things in a more svn-like model. I wonder if one of the other frontends might be a better performer? (But this is a question more for a darcs list, I guess.) -- John