Perl 6 - the future is here, just unevenly distributed

IRC log for #darcs, 2014-10-07

| Channels | #darcs index | Today | | Search | Google Search | Plain-Text | summary

All times shown according to UTC.

Time Nick Message
00:55 mizu_no_oto joined #darcs
01:37 mizu_no_oto joined #darcs
01:46 ilbot3 joined #darcs
01:46 Topic for #darcs is now http://darcs.net/ | logs: http://irclog.perlgeek.de/darcs/ | darcs 2.8.5 is out http://darcs.net/Releases/2.8
02:34 siel joined #darcs
06:25 lelit joined #darcs
06:50 c74d joined #darcs
07:11 c74d joined #darcs
07:11 lelix joined #darcs
07:15 lambdabot joined #darcs
07:56 gal_bolle joined #darcs
08:05 mornfall http get in 2.8.4 appears to be broken :\ (built with ghc 7.6.3)
08:05 mornfall darcs: /var/tmp/mainline_2/_darcs/patches/0000000442-f19c0c4654fb31c3a2abdd8da13b258185366ad9adbb50492e40812b9312cbf9-new_c2d2ef30f877a07298d39872c4794cc8: getSymbolicLinkStatus: does not exist (No such file or directory)
08:06 mornfall and, well, --no-http-pipelining seems to fix it
09:43 byorgey_ joined #darcs
10:28 lelit joined #darcs
10:57 stulli joined #darcs
11:07 stulli Is there a reason we have so many upper bounds in the darcsden.cabal? I was able to build it with just hsp<0.8 and random <1.1 (well, and bumping a webdriver dependency)
11:09 mizu_no_oto joined #darcs
11:54 stulli Why is there a darcs-beta package which is older than the current release? Shouldn't that be updated/deleted?
12:36 stulli joined #darcs
13:03 mizu_no_oto joined #darcs
14:22 gh joined #darcs
14:23 gh__ hi
14:40 mizu_no_oto joined #darcs
14:47 gh__ Heffalump, +1 for a sqlite-backed darcsden :)
14:55 gh__ mornfall, can you try darcs 2.8.5? it was released on purpose to fix that bug.
14:56 mornfall heh, I just corrupted an sqlite db :-)
14:56 mornfall but whatever
14:56 mornfall gh__: I'm using packages
14:57 mornfall I'm trying a server-side fix
14:57 gh__ oh :(
14:58 mornfall I guess darcs packages aren't a priority these days...
14:58 mornfall and specifically, this is not hitting me personally but someone else (probably not in a position to upgrade darcs either)
14:59 gh__ we need a recipe to generate static executables
14:59 mornfall let's see if tweaking nginx can prevent the problem from happening
15:00 mornfall gh__: I don't think that's really an answer
15:03 gh__ it would be just an alternative
15:03 mornfall on a multi-user, multi-computer network, static binaries sound like a nightmare
15:03 mornfall unless your management system somehow takes care of them
15:04 gh__ alright, I was thinking about simpler use cases
15:08 mornfall hmm, setting keepalive_requests to 1 server side makes get really slow but the bug is still there :| darn
15:42 stepkut joined #darcs
16:42 aristid joined #darcs
16:54 wavis joined #darcs
17:05 alexei___ joined #darcs
17:38 mizu_no_oto joined #darcs
17:53 Heffalump mornfall: is the resulting repo actually broken?
17:54 Heffalump gh: I'd prefer flat files really
17:54 Heffalump and push as much info into the repos themselves
17:54 stulli joined #darcs
17:55 mornfall Heffalump: I think it can be used but is missing patches
17:56 mornfall but well, we will just document the problem and hope for the best
17:56 mornfall eventually .5 will hit distros I guess
18:21 Heffalump mornfall: I wouldn't actually have expected the bug that was fixed in .5 to show up with a GHC 7.6 build
18:21 Heffalump there is a different problem where darcs prints out spurious messages about files it couldn't delete, but where the repo is actually fine
19:27 alexei___ joined #darcs
19:45 lelit joined #darcs
19:50 amgarchIn9 joined #darcs
20:57 mizu_no_oto joined #darcs
23:29 mizu_no_oto joined #darcs

| Channels | #darcs index | Today | | Search | Google Search | Plain-Text | summary