Camelia, the Perl 6 bug

IRC log for #darcs, 2013-07-26

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

All times shown according to UTC.

Time Nick Message
00:00 mizu_no_oto joined #darcs
00:49 dolio joined #darcs
01:13 kofno joined #darcs
02:38 kofno joined #darcs
03:18 edwardk joined #darcs
03:32 preflex joined #darcs
03:34 mizu_no_oto joined #darcs
04:09 kofno joined #darcs
04:17 mizu_no_oto joined #darcs
04:46 edwardk joined #darcs
05:14 kmels joined #darcs
06:01 lelit joined #darcs
06:14 favonia joined #darcs
07:07 alexei joined #darcs
07:40 gh_ joined #darcs
08:42 donri joined #darcs
09:29 alexei joined #darcs
09:36 mornfall jlneder: hey
09:37 jlneder here i am
09:38 jlneder i tried to compile the last state of the hashed repo, and i fail miserably
09:38 mornfall http://repos.mornfall.net/fslib/ is what you want
09:38 mornfall possibly
09:39 mornfall but the patches to port darcs over to the new codebase had been axed and rotted away
09:39 mornfall and the fslib codebase is semi-orphaned
09:40 mornfall I think the best course of action would be to fork the deprecated 0.5 branch of hashed-storage and re-absorb it into darcs
09:41 mornfall or something like that
09:41 jlneder yeah... i have been working on the 0.5.10 branch because it is the one darcs uses
09:41 mornfall anyway, I am not going to make 0.5 releases for anything else than small bugfixes
09:41 jlneder i was using this http://repos.mornfall.net/hashed-storage/0.5/
09:42 mornfall you could either port darcs to fslib (and I'd be OK with working on fslib to make it featureful enough for darcs), or fork old hashed-storage
09:42 mornfall while the former would be, in my opinion, a much better course, expect resistance from the rest of the team
09:42 jlneder which are the benefits of fslib?
09:44 mornfall I can't really remember :-) it's an evolution of hashed-storage
09:44 jlneder for me is much faster, easier and safer to use 0.5 for now. But I think it shouldn't be much of an issue to port the code later to the new 0.6 branch.
09:44 mornfall better interface, more features
09:45 mornfall whatever... you still have to fork that way
09:48 jlneder about the code. I was adding the inode info to the index, but i'm not sure is the best way. bumping the version to IHS5 i guess it should not have any problem with old index.
09:49 jlneder anyway
09:49 mornfall well, really, I don't care how you go about changing the code, as far as I am concerned, it's dead code anyway :-)
09:49 mornfall bumping index version will make darcs re-create it, yes
09:50 mornfall although you should change it DHS or something
09:52 jlneder yes, HSI5
09:53 mornfall you can actually re-use HSIx, fslib uses an 8-byte magic marker
09:54 mornfall (to avoid alignment issues on 64b architectures, I think)
10:09 raichoo joined #darcs
10:18 alexei joined #darcs
10:47 alexei joined #darcs
11:14 alexei joined #darcs
11:34 favonia joined #darcs
11:55 kofno joined #darcs
12:18 mizu_no_oto joined #darcs
12:43 edwardk joined #darcs
12:43 alexei joined #darcs
13:38 45PAA8EWZ joined #darcs
14:08 mizu_no_oto joined #darcs
14:11 alexei joined #darcs
14:14 raichoo joined #darcs
14:17 sm morning all
14:35 alexei joined #darcs
14:35 bfrank_ joined #darcs
14:45 whaletechno joined #darcs
15:10 alexei joined #darcs
15:18 kmels joined #darcs
15:23 adnap joined #darcs
15:33 adnap joined #darcs
15:42 bfrank_ when I do a pull -p, why do I get a lot of patches that don't match the pattern?
15:45 gh_ bfrank, probably because they're dependencies of those patches that do match the pattern
15:51 bfrank_ ahh, I see
15:51 bfrank_ I noticed there were some options I could pass to not show them
15:51 bfrank_ that makes sense now
15:53 bfrank_ man, darcs is so awesome
15:53 bfrank_ :)
16:15 mizu_no_oto joined #darcs
16:25 donri joined #darcs
16:48 stepcut joined #darcs
16:49 jlneder joined #darcs
17:12 lispy Heffalump: Yeah, probably. I don't really know how type families play with that stuff. So there may be some hairy corner cases?
17:12 lispy Heffalump: It's possible that I'm wrong about it helping :)
17:13 lispy Heffalump: It just seems like the closed type fams really let you get in and express things consicely at the type level.
17:13 lispy So maybe instead of making existing things cleaner it would give us more expressive power?
17:24 lelit joined #darcs
17:48 sm I'd like to show user repo counts on darcs hub front page
17:49 sm should private repo counts be kept private, or exposed ?
18:07 kmels joined #darcs
18:28 konundra joined #darcs
18:33 xymox joined #darcs
18:46 Heffalump sm: kept private
18:50 sm should the names of users with no public repos, but some private repos, be exposed ?
18:52 mizu_no_oto joined #darcs
18:53 sm I'd like to indicate the users with nothing in their account. That means also revealing users with only private repos
18:53 sm or giving up on this and not showing repo counts at all. Or not showing the full user list.
19:18 sm at http://hub.darcs.net with :8902 added, I'm showing just the public counts and not revealing who's got privates. I'll push this as the next iteration of the front page, if there are no major objections. It won't be this way forever
19:18 sm I think it's a little more interesting than what we currently show
19:34 sm another issue: the hub.darcs.net-specific code is growing, and I'd like it to be hackable by others. I think it makes sense to include much of it in the darcsden repo (as */Hub.hs, enabled by a flag).
19:38 sm I guess another approach that could work is run a separate hub.darcs.net app that talks to the same db, and fetch certain pages from it instead of darcsden - a sort of overlay
19:40 sm that would keep the generic and hub-specific code and patch history separate. But seems a bit more complicated
21:08 epta joined #darcs
21:18 kofno joined #darcs
21:29 kmels_ joined #darcs
21:30 kmels__ joined #darcs
21:34 kmels joined #darcs
21:55 alexei joined #darcs
22:06 nomeata joined #darcs
22:43 mizu_no_oto joined #darcs
22:48 Heffalump what sort of thing is hub.darcs specific
22:48 Heffalump ?
23:09 sm Heffalump: so far, the front page
23:42 bfrank_ joined #darcs
23:48 jgross_ joined #darcs

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