[darcs-devel] [issue501] failure to push into new (empty) repo with 1.0.9

2007-07-23 Thread Zooko
New submission from Zooko [EMAIL PROTECTED]: I created a new directory on a remote server named hanford, and ran darcs init therein. (hanford has darcs 1.0.8.) Then I ran darcs push -a from a different computer to try to push stuff into that repo on hanford. The different computer is

[darcs-devel] [issue503] Problem with get

2007-07-23 Thread Augustsson, Lennart
New submission from Augustsson, Lennart [EMAIL PROTECTED]: $ darcs get --partial http://darcs.haskell.org/ghc This is the GHC darcs repository (HEAD branch) For more information, visit the GHC developer wiki at http://hackage.haskell.org/trac/ghc ** Copying patch 210 of

Re: [darcs-devel] Pristine cache on a single file

2007-07-23 Thread David Roundy
On Sun, Jul 15, 2007 at 08:33:32PM +0200, Salvatore Insalaco wrote: Today I had a strange idea: why not using sqlite (an embedded database) for data storage? I tried a bit today, even with very large (100s of MB) binary files, and the time overhead seems very low from the first tests I made

Re: [darcs-devel] Pristine cache on a single file

2007-07-23 Thread zooko
Today I had a strange idea: why not using sqlite (an embedded database) for data storage? I think this is quite reasonable. That's what monotone has been doing for years. If it didn't work well, they'd have stopped doing it by now. http://monotone.ca Regards, Zooko

[darcs-devel] Use System.Directory.copyFile for file copying

2007-07-23 Thread Kevin Quick
This is a resubmission, splitting the original patch into two components: one that changes all internal file copies in darcs to use cloneFile, and the second (more controversial) that changes cloneFile to use System.Directory.copyFile. These patches replace the previous submission. -KQ Sun Jul

[darcs-devel] issue308 update: Shorter summary/separate test

2007-07-23 Thread Kevin Quick
Resend of fix for issue308. This submission shortens the summary name of the patch and also splits the test for issue 308 out to a separate patch as requested (note that there is an update to the selftests in the main patch, since the main patch causes the tests to break without this change).

[darcs-devel] Re: [issue405] darcs send -o test.dpatch --repodir=somerepo leaves output file in repository directory

2007-07-23 Thread Tommy Pettersson
[CC-ing darcs-devel] I'm just brain storming a little here. We have absolute paths, relative paths, and repo-root-relative paths. There is no file system way to express a repo-root-relative path. What if darcs set the environment variable $rrr to the repo root, be it ../../some/where,

Re: [darcs-devel] darcs patch: fjsda

2007-07-23 Thread Zachary P. Landau
On Mon, Jul 23, 2007 at 08:58:35PM -0400, Zachary P. Landau wrote: Sat Jul 21 19:38:50 EDT 2007 Zachary P. Landau [EMAIL PROTECTED] * fjsda Whoops, feel free to ignore that. My finger 'slipped' while trying to use the -o option. -- Zachary P. Landau [EMAIL PROTECTED] GPG: gpg --recv-key

[darcs-devel] [issue405] darcs send -o test.dpatch --repodir=somerepo leaves output file in repository directory

2007-07-23 Thread Eric Kow
Eric Kow [EMAIL PROTECTED] added the comment: Hey, moving fast on this! :-) On Tue, Jul 24, 2007 at 01:41:43 -, Zachary P. Landau wrote: A lot of talk for what seems like a small patch. Can someone take a look at this and see if it makes sense to them? It seems to me that all of the