Camelia, the Perl 6 bug

IRC log for #darcs, 2013-06-27

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

All times shown according to UTC.

Time Nick Message
00:42 lispy Did you folks see the reddit post about darcs?
00:43 lispy http://www.reddit.com/r/haskell/comments/1h4rak/d​arcs_source_code_is_some_of_the_cleanest_haskell/
01:08 mizu_no_oto joined #darcs
01:46 mizu_no_oto joined #darcs
01:51 lambdabot joined #darcs
01:56 intripoon_ joined #darcs
02:35 konundra joined #darcs
03:30 preflex joined #darcs
05:10 Heffalump sm: http://irclog.perlgeek.de/​darcs/2012-08-03#i_5872448 is the most relevant conversation I can find
06:46 lelit joined #darcs
07:17 raichoo joined #darcs
07:35 stepcut joined #darcs
07:36 alexei___ joined #darcs
07:37 xymox_ joined #darcs
07:37 preflex joined #darcs
07:37 dolio joined #darcs
07:37 teratorn joined #darcs
07:37 sm joined #darcs
07:37 mal`` joined #darcs
07:37 idnar joined #darcs
07:37 haasn joined #darcs
07:37 byorgey joined #darcs
07:37 dixie joined #darcs
07:37 Heffalump joined #darcs
07:45 preflex joined #darcs
08:11 idris joined #darcs
08:13 donri joined #darcs
08:35 whaletechno joined #darcs
08:52 Moggle joined #darcs
09:15 mulander joined #darcs
09:21 edwardk joined #darcs
09:23 donri joined #darcs
09:58 gh_ joined #darcs
11:15 owst joined #darcs
11:22 epta joined #darcs
12:36 mizu_no_oto joined #darcs
12:45 gh_ joined #darcs
12:53 mornfall joined #darcs
12:57 konundra joined #darcs
13:20 mizu_no_oto joined #darcs
13:38 uniquenick joined #darcs
13:45 favonia joined #darcs
14:01 edwardk joined #darcs
14:04 sm yup that's it
14:15 felipe joined #darcs
14:21 idnar joined #darcs
14:23 mizu_no_oto joined #darcs
14:33 * sm browses recent darcs work thanks to the reddit thread
14:34 sm "Suffix extraction used to be worst case quadratic complexity, which is now reduced to worst case linear." looks nice
14:38 * sm tries to understand the fix,  http://hub.darcs.net/darcs/darcs-s​creened/patch/20130228151120-ae621
14:58 sm I think.. the function adds recent patches which are in the repo but not the patch index ("new"), to the index (and ignores patches in the index but not in the repo ("old")). It used to filter all index patches by (`elem` new), which would be INDEXPATCHES * NEWPATCHES, now instead it relies on the new patches being at the end, so the complexity is just that of finding the common prefix which is linear
14:59 sm so... the old way was O(M*N) ? and that's called quadratic ?
15:00 mtp no, quadratic would be O(n^2)
15:00 mtp i always hated doing complexity analysis :/
15:01 sm ah right.. so I see why he called it that
15:01 sm I need to understand it a bit better. Now that there's a cheatsheet :) http://bigocheatsheet.com/
15:03 dolio joined #darcs
15:05 sm incidentally darcs is currently 31k lines of haskell and 1k of C, plus 10k lines of shell for the tests
15:06 sm (#haskell was discussing what are the large haskell code bases other than ghc. The only larger one mentioned was Agda at 50k)
15:58 donri joined #darcs
16:11 raichoo joined #darcs
16:45 edwardk joined #darcs
17:02 whaletechno joined #darcs
17:14 alexei___ joined #darcs
17:22 dolio joined #darcs
17:23 konundra joined #darcs
17:29 onetom joined #darcs
17:51 alexei___ joined #darcs
17:58 lelit joined #darcs
18:12 edwardk joined #darcs
18:44 idris joined #darcs
18:47 gh_ joined #darcs
18:52 mizu_no_oto joined #darcs
18:55 edwardk joined #darcs
19:01 alexei___ joined #darcs
19:02 Moggle_ joined #darcs
19:04 carter joined #darcs
19:24 konundra joined #darcs
19:25 alexei___ joined #darcs
19:25 dolio joined #darcs
20:22 edwardk joined #darcs
20:39 donri joined #darcs
20:49 preflex joined #darcs
20:59 mizu_no_oto joined #darcs
22:09 carter joined #darcs
22:22 dolio joined #darcs
22:40 mizu_no_oto joined #darcs
23:30 edwardk joined #darcs
23:32 mizu_no_oto joined #darcs
23:48 edwardk joined #darcs

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