Camelia, the Perl 6 bug

IRC log for #darcs, 2013-10-20

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

All times shown according to UTC.

Time Nick Message
00:43 mtp joined #darcs
00:43 mtp joined #darcs
00:54 alexei_ joined #darcs
01:30 intripoon joined #darcs
02:10 alexei_ joined #darcs
02:17 dixie joined #darcs
02:36 alexei_ joined #darcs
02:44 alexei___ joined #darcs
03:09 mizu_no_oto joined #darcs
03:40 alexei___ joined #darcs
03:42 preflex joined #darcs
04:26 favonia joined #darcs
04:59 mizu_no_oto joined #darcs
05:11 mizu_no_oto joined #darcs
05:34 frozencemetery joined #darcs
05:35 frozencemetery joined #darcs
05:44 favonia joined #darcs
07:37 schlaftier joined #darcs
09:19 lelit joined #darcs
10:31 donri joined #darcs
11:11 alexei___ joined #darcs
11:53 lelix joined #darcs
11:57 nomeata joined #darcs
14:40 alexei___ joined #darcs
14:45 gal_bolle joined #darcs
15:05 mizu_no_oto joined #darcs
15:18 kowey joined #darcs
15:19 kowey hallo!
15:19 kowey it's HCAR time
15:20 mornfall afternoon kowey
15:23 kowey hope gh_ is having a good time at the GSoC mentor summit
15:24 kowey I wonder if there's something that would make darcs get of the xmonad repo more than 8 seconds for somebody else
15:24 kowey like some geographical thing?
15:25 kowey (sorry using #darcs as my brain - first task is to dig up and read the last HCAR report; second task is trawl through the mailing lists)
15:33 kowey whatever happened with Sebastian Fischer's work with Factis?
15:36 kowey our last HCAR said 4 things: darcs 2.8.4, we're working on rebase, factis, and hub
15:40 kowey wow, lots of nice looking work on making darcs record smoother: less whatsnew -l stack space kaboomy, automatic detection of mvs…
15:40 kowey patience diff
15:53 kowey ah-hah!
15:54 kowey enabling packs requires that I log in to hub.darcs.net, click on my repo and go to settings and click on Build Packs
15:54 kowey (I somehow assumed it'd be magically going on behind the scenes)
15:55 kowey so packed Agda and packed xmonad would be nice to see
16:04 dolio Haskell could really use some better tools for formatting usage/help output to fit nice in consoles.
16:04 dolio I think ever command args library I've used is terrible at that.
16:05 dolio And, while it's nice that the --super-long-options-in-some-actions are now put on their own line, darcs help get still doesn't fit in the usual 80 characters.
16:08 mizu_no_oto joined #darcs
16:10 kowey mornfall, is that something you worked on at some point?
16:11 kowey I forget if cmdlib handles this reflow stuff
16:11 kowey it's one of the oldest bugs in our tracker!
16:11 kowey how do I still remember that?
16:11 mornfall don't know :-)
16:12 mornfall probably not
16:12 mornfall although could be fixed I suppose :)
16:12 mornfall it just passes things to getopt
16:12 mornfall iirc anyway
16:12 dolio Someone rewrote it to be a little smarter than that.
16:12 mornfall I didn't touch Haskell for a long while, sadly.
16:13 mornfall dolio: cmdlib? I'd probably know about that.
16:13 dolio It used to put '--no-look-for-adds, --don't-look-for-adds' on the same line.
16:13 dolio No, the code in darcs for formatting help.
16:13 mornfall I see, nothing to do with me, that.
16:13 gal_bolle joined #darcs
16:13 dolio Now it puts each of those on a separate line.
16:14 kowey little details like this are so nice to get right
16:14 kowey that extra tiny bit of thoughtfulness
16:14 kowey ah!
16:14 dolio Yeah.
16:14 mornfall Eric! :-P
16:14 dolio I do know it's tough in Haskell, because I've looked for libraries that handle this automatically for my own stuff, and never find anything. :)
16:15 kowey oh btw the link to that tweet
16:15 kowey was really yet-another-example for me of somebody who says “darcs is slow! nice in theory, sucky in practice!”
16:15 kowey when what they (understandably because they can't be expected to know the ins and outs of darcs) really mean is “boy, when I do darcs get, it sure takes a long time”
16:16 kowey sometimes I wonder if it's just something to do with code.haskell.org specifically
16:16 kowey but in any case, it really feels like all this perception about darcs performance is about this one command
16:16 lelit joined #darcs
16:16 * kowey hangs his head for commenting about this stuff without really getting stuck in himself
16:16 mornfall kowey: don't worry it's not that easy
16:32 kowey I push draft! http://darcs.net/hcar/Darcs.tex
16:32 kowey ok, so some small-circulation feedback would be great
16:32 kowey after that I'll send this out on darcs-users for more comments
16:34 kowey kind of funny, this has injected a lot of darcs-enthusiasm into my bloodstream (state lately has been a bit agh-too-much-going-on-can't-think-about-darcs)
16:34 kowey will probably revert to said state though, but it's nice to feel it again
16:34 * kowey makes soup
16:35 adnap joined #darcs
16:43 dolio Perhaps a change to the formatting of darcs' help would be useful? For instance, git just delegates to man pages, as far as I can see.
16:44 dolio That might not be acceptable for darcs, but man pages usually seem to be written in a way that doesn't require laying out multiple columns.
16:44 dolio Which makes a lot of things easier.
16:45 dolio Mercurial actually formats multiple column output somehow.
16:45 dolio (And bazaar has the same problems as darcs. :))
16:49 kowey there has been some consolidation lately
16:49 kowey the whole help/manual/man page latex-markdown thing
16:56 sm dolio: did you try cmdargs ? I've not noticed any layout problems with it
16:56 dolio Don't remember.
16:56 sm welcome back kowey!
16:58 kowey heya, sm!
16:58 kowey I wish I were back longer instead of just randomly resurfacing like this :-)
16:58 sm us too. Life treating you well I hope ? Still using darcs ?
16:59 kowey definitely still using darcs (well, actually am using git for some work-work, which is giving me loads of nuance)
17:00 kowey both more reasons to love darcs (git ui has lots of crazy things, like how add -p doesn't work on new files, how if you cherry pick with add -p and then you say git commit <file>, it doesn't do what you might think)
17:00 dolio sm: Is there a way to use cmdargs where I don't have to worry about what magic is going on with unsafePerformIO now?
17:01 kowey and also some things to be jealous about (now I know a good bit more about why everybody loves git rebase -i so much! being able to consolidate a patch by changing it to “fixup”, ah! so nice)
17:01 sm dolio: sure, see the docs
17:01 kowey how's things? have gotten back into tracking things with hledger over here
17:01 kowey still the only finance-type software i think i understand
17:02 sm great.. I've been blogging about it
17:02 sm all fine here. I've also made friends with git at last.. using both now
17:03 kowey it's been a nice read; hledger/ledger cooperation sounds like a great model to follow, and encouraging muhaha-y to see John becoming a Haskell nut
17:37 sm kowey: I think you just tweeted something about ledger ? I've not used twitter on this phone for a long time, but it popped up as a notification.. yet now I can't find it
17:38 sm maybe I'm subscribed to a search for #ledger ?
17:39 kowey ah yes, @dysinger posted a tweet saying it was neat, and I made a not-quite-accurate-in-hindsight claim that you were on twitter: https://twitter.com/dysinge​r/status/391978626456899584
17:39 kowey so you probably a notification because I used your twitter handle
17:39 sm ahh, makes sense!
17:40 sm sorry, a bit off topic
17:44 kowey perhaps more on topic: any comments on the recent HCAR update?
17:44 kowey I purposely put bsrkaditya's project second so I could segue into a hub.darcs.net update if there was more to say outside of GSoC
17:45 kowey http://darcs.net/hcar/Darcs.tex to avoid backscroll
17:47 sm urgh work. Thanks for doing it kowey
17:48 kowey :-) i was procrastinating on it, but trying to explain what people did in terms of how it would make darcs users' lives better made feel suddenly more enthusiastic about darcs
17:48 sm http://joyful.com/blog/2013-09-26-da​rcsden-darcs-hub-gsoc-complete.html was my wrap-up for the darcs hub gsoc. Unfortunately it might not be the complete feature list
17:49 sm but it links to bsrk's full list
17:49 sm latest hub user/repo counts are on the front page
17:49 kowey I should link to your post, which is quite a bit more fleshed-out and user-centric
17:50 sm not too much else going on with hub just now.. I'm happy to say it's running itself at the moment
17:50 kowey oh, I'll mention the Agda repo as a good candidate for packs tests
17:50 kowey ok, so I'll link your post when talking about Aditya's work then
17:51 sm and your expectation that hub would do packing automatically was a totally reasonable one
17:51 sm (btw)
17:51 kowey sort of an assumption I didn't question, partly due to darcs telling me there was a problem fetching packs
17:51 kowey but it could also be one of those tradeoff things
17:51 sm I don't believe pack getting is working yet
17:51 kowey like maybe packs not good for small repos
17:52 kowey ah!
19:17 mizu_no_oto joined #darcs
20:13 schlaftier joined #darcs
20:22 stepcut joined #darcs
20:26 alexei___ joined #darcs
22:09 arpunk joined #darcs
23:00 dolio joined #darcs

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