--- Log opened Sun Feb 01 00:00:14 2009
01:06 < mae_> Saizan_: syb-with-class version bump on hackage? whats the status? :)
01:44 < mae_> omg
01:44 < mae_> cocaine badger
01:44 < mae_> so funny
01:44 < mae_> http://cocainebadger.ytmnd.com/
11:36 < rovar> hola
11:37 < aluink> hey! first sign of life in a long time
11:37 < rovar> happy to help
11:39 < aluink> yeah, i'm a n00b haskeller and happstacker, so i'm just hear to glean off conversations, and talk about working on docs(call i can handle right now) ;)
11:40 < rovar> well it looks like we're in the same boat :)
11:40 < aluink> cool!
11:40 < rovar> I'm working on a HAML parser in Haskell as a learning exercise
11:42 < koeien> okay. it's good to see that also beginning Haskellers are interested!
11:43 < aluink> yeah, i'm working through RWH right now, in ch06 now
11:43 < aluink> though i've looked into c14 already (Monads) :P
11:43 < koeien> i bought the book, i liked the chapter on profiling best
11:43 < koeien> also i thought the STM chapter was great
11:44 < rovar> don't tell me the ending. I ordered it, should be arriving tomorrow
11:44 < koeien> :)
11:44 < aluink> my housemate bought hte book, i'm kinda tight on cash right now, so i'm just doing the online stuff right now
11:45 < aluink> STM?
11:46 < koeien> software transactional memory
11:46 < aluink> ahh, cool
11:46 < rovar> one of the reasons i decided to pick up Haskell, actually
11:46 < aluink> STM, now is that what Happstack does?
11:46 < rovar> professionally, I spend 90% of my time on complex distributed systems.
11:47 < rovar> I don't know if MACID uses STM. Seems like it would
11:48 < aluink> man, that's the killer with getting into Happstack, a lot of MLA (ManyLetterAcronyms) i need to learn and understand
11:49 < koeien> ACID comes from database theory
11:49 < rovar> or chemistry
11:49 < koeien> ;)
11:49 < aluink> Bronsted or Lewis?
11:50 < rovar> ACID stands for Atomic, Consisten, Isolated, Durable
11:50 < Saizan> yes, MACID is implemented on top of STM, locally
11:50 < aluink> yeah, learning what MACID means and does is on my TODO
11:52 < rovar> at this point i'll be happy with figuring out what Parsec.tokenPrimEx does
12:14 < chessguy> hi ya'all
12:14 < koeien> hello
12:15 < chessguy> how goes?
12:15 < rovar> ello
12:15 < koeien> buildbot still offline ;) i hope to get it online this evening
12:16 < aluink> tail . tail "othello"
12:16 < wchogg> koeien : not a problem!  No one has comitted in a while anyway.
12:16 < chessguy> > tail . tail "othello"
12:16 < chessguy> aww
12:17 < chessguy> anyway, i think you mean tail . tail $ "hello"'
12:17 < aluink> ahh, yes
12:17 < aluink> check #happs, i was trying it out there
12:17 < chessguy> or (tail . tail) "hello"
12:17 < aluink> right
13:19 < rovar> i like the last version better
13:40 < aluink> foo "othello" where foo = tail . tail
13:43 < aluink> (tails "othello")!!2
13:45 < aluink> foo "othello" where foo (a:b:xs) = xs
14:01 < chessguy> aluink, partial functions are evil
14:02 < chessguy> though i guess tail . tail is too
14:02 < rovar> why are partial functions evil?
14:03 < rovar> i thought the (tail . tail) "othello"  was the most elegant
14:03 < chessguy> rovar, well, essentially because it has a potential side effect
14:06 < aluink> explain the potential for side effects
14:10 < mae_> Saizan: syb-with-class?
14:25 < gwern> aluink: exceptions - head [] throws an exception
14:25 < gwern> which can kill the program, which is certainly nowhere in the type [a] -> a
14:30 < koeien> however, all functions of type [a] -> a are partial
14:34 < gwern> true, which is why safeHead goes a -> [a] -> a
14:43 < mae_> gwern: howdy
14:54 < Saizan> mae_: i'm finishing reading the SYB papers, it's been a full weekend :)
15:08 < gwern> mae_: 'lo
15:13 < mae_> Saizan: okey :)
15:15 < wchogg> mae_ : Is there anything else I can do to help with the 0.1 release?  It looks like everything tagged 0.1 is covered by someone now.
15:16 < mae_> wchogg: documentation patches
15:16 < mae_> for haddock
15:16 < mae_> we need to get all the exported ones haddock'd
15:16 < rovar> hello
15:17 < mae_> wchogg: all doesn't have to be done by 0.1 but it can't hurt to start now as it is harmless (in terms of program behavior)
15:17 < wchogg> mae_ : Sure, that makes sense.
15:31 < rovar> <-- Rick Richardson
15:31 < rovar> btw
15:45 < mae_> rovar: hi rick :)
15:45 < mae_> mr happstack, the bunny with the coke habit..
15:45 < rovar> i like to think that he does it recreationally, and doesn't actually have a habit
15:46 < wchogg> It's just a hobby?
15:53 < mae_> pushed up some changes
15:53 < mae_> to fix some errors
15:53 < mae_> err warnings
15:54 < mae_> where is the build bot :\
15:54 < koeien> dead
15:54 < mae_> !
15:54 < koeien> i'm integrating the test script
15:54 < mae_> ah ok
15:55 < koeien> it should be online again soon
16:01 < mae_> um
16:03 < thomashartman1> hey! so, is anybody that answered to happstutorial maintainership thread around?
16:03 < mae_> HAppS.Data.HList doesn't seem to be imported internally anywhere, or used in any of the happstack packages
16:03 < mae_> anyone against the idea of removing HAppS.Data.HList since gwern has made http://hackage.haskell.org/cgi-bin/hackage-scripts/package/HList
16:03 < gwern> mae_: I'd first make sure that HList is usable
16:03 < gwern> (I've long been suspicious of hlist)
16:04 < mae_> gwern: HList is an export from HAppS-Data
16:04 < mae_> its not used internally anywhere
16:05 < mae_> there might be a slight chance that an end-user is using it for something but somehow i doubt it
16:06 < gwern> this is indepedent of whether anyone uses it; if the happs hlist works better than thecabalized hlist, it obviously shouldn't be scrapped
16:07 < mae_> i see.
16:07 < mae_> well it requires overlapping instances to compile
16:08 < gwern> so?
16:08 < mae_> so its causing lots of warnings
16:09 < mae_> and we're not using it for anything internally.. yet..
16:10 < mae_> i will keep it around per your suggestion for now.
16:11 < gwern> put it up on the bugtracker - 'todo: evaluate our hlist and see whether it works better/worse/the-same-as regular hlist'
16:16 < mae_> http://code.google.com/p/happstack/issues/detail?id=34
16:17 < wchogg> thomashartman1 : Oops, well I'm around 15 minutes later if you still want to talk.
16:21 < gwern> oh, we're up to 34 issues now? that's good
16:26 < stepcut>  :)
16:26 < thomashartman1> wchogg: there?
16:27 < thomashartman1> irc question: is there a client that will audibly ping when someone addresses you as user: with a colon like?
16:27 < chessguy> thomashartman1, most clients can do that, i think
16:27 < wchogg> thomashartma1 : here
16:27 < thomashartman1> hey creighton
16:27 < stepcut> thomashartman1: you can probably 'configure' erc to do that easily
16:27 < thomashartman1> chesssguy: i use erc (emacs)
16:28 < thomashartman1> k, I'll look into it.
16:28 < chessguy> ah, i use emacs, but not as an IRC client
16:28 < thomashartman1> chessguy: what do you use? I need something works on linux
16:28 < thomashartman1> chessguy: who're you again? I remember you emailed me...
16:29 < chessguy> i use xchat on windows and colloquy on my mac. not sure if either of those have linux versions off the top of my head
16:29 < chessguy> Andrew Wagner
16:29 < thomashartman1> anyways, first item of business: anybody here that's ready at this point in time to become the official happstut maintainer?
16:29 < wchogg> Yes.
16:29 < thomashartman1> I could hack out an interim hackage release with the maintainer field changed to kick things off.
16:29 < thomashartman1> cool :)
16:29 < thomashartman1> anyone else?
16:29 < stepcut> thomashartman1: this might have something useful: http://www.emacswiki.org/emacs/ErcSound
16:29 < thomashartman1> thx stepcut
16:30 < thomashartman1> wchogg: I'll do  a hackage release and sign it over to you.
16:30 < stepcut> thomashartman1: I think you use EmasSpeak sot hat it even reads the message out loud to you ;)
16:31 < chessguy_work> sorry, i got booted
16:31 < thomashartman1> stepcut: no thanks, don't fancy robo-voices :)
16:31 < chessguy_work> thomashartman1, i'd like to hear more details on what is needed for happstut
16:31 < thomashartman1> I have a text todo list under darcs.
16:31 < thomashartman1> http://patch-tag.com/publicrepos/happs-tutorial/todo
16:31 < thomashartman1> it's probably pretty impressionistic / confusing to an outsider
16:32 < thomashartman1> I guess first item of business would be to get this cleaned up and the items moved to the happstack issue tracker.
16:32 < thomashartman1> mae: ok with that?
16:32 < wchogg> thomashartman1 : It's not that confusing, don't worry.
16:32 < thomashartman1> everybody seemed to be against a separate googlegroup, so I figure that would extend to the issue tracker as well. or?
16:32 < stepcut> I wonder if there is a mechanical turk, text-to-speech API
16:32 < chessguy> wchogg, were you wanting to take over happstut?
16:32 < thomashartman1> stepcut: interesting idea
16:33 < wchogg> chessguy : Yes, I also had e-mailed Thomas about it.
16:33 < thomashartman1> happstut folk: wchogg piped up first so he has dibs for maintainer, and it's up to him if it gets shared with anyone else, I guess.
16:34 < thomashartman1> very glamorous job :)
16:34 < chessguy> good enough, just let me know if you want any help
16:34 < thomashartman1> chessguy: probably, there's a lot of todos.
16:34 < stepcut> thomashartman1: its more glamorous than the guy who follows the police horses around at parades and sweeps up after them
16:35 < wchogg> chessguy : I figured it'd be a team effort.
16:35 < chessguy> wchogg, sure, but someone should spearhead it
16:36 < wchogg> mae_ : ping?
16:36 < mae_> pong
16:36 < thomashartman1> wchogg: do you agree that the todos should move to the happstack todo list?
16:36 < wchogg> mae_ : happstutorial issues list?  Merge with happstack or make a new google project?
16:37 < mae_> thomashartman1: yes i'm ok with that, happstutorial is adjunct to happstack since it is basically an example
16:37 < wchogg> thomashartman1 : I could go either way.  If they're merged, I'd want to just make sure we use a separate tag for them.
16:37 < mae_> but doesn't really relate to any milestones..
16:37 < chessguy> i agree with a separate tag, but not a separaate project
16:37 < wchogg> Okay, same project separate tag it is then.
16:38 < mae_> right
16:38 < thomashartman1> wchogg: suggest you transfer as much of the text todos to happstack issues list as you can, anything confusing/muddled remaining I then do myself.
16:38 < mae_> in the future once its cleaned up we can probably put that in contrib (maybe?)
16:38 < thomashartman1> there's also some private email related to happstut issues/errors that I haven't put into the todo.
16:39 < thomashartman1> I'll put that in a separate file called todos-misc.txt or something and push to the repo.
16:39 < gwern> I once used emacspeak. after the incredible difficulty of getting it set up and working on linux, I discovered that it was incredibly annoying
16:39 < wchogg> mae_ : that might be a reasonable idea
16:39 < gwern> cool idea but not worthwhile for anyone with half-working vision - eyes are just so fast, I discovered
16:39 < wchogg> thomashartman1 : Right, I'll go ahead and start transfering the issues to the happstack list
16:39 < chessguy> gwern, haha
16:40 < mae_> wchogg: how many issues?
16:40 < chessguy> gwern, you should try living with someone who  uses a fast-speaking screen-reader :)
16:41 < wchogg> mae_ : I'm guessing at least 10-15?  I haven't split up Thomas's notes in my head to bite sized items.
16:41 < gwern> chessguy: sounds painful
16:42 < mae_> ok I added a default label to issue tracker
16:42 < mae_> 'HappstackTutorial'
16:42 < chessguy> gwern, yeah, but she's cute, so it's worth it :)
16:42 < thomashartman1> my stepbrothers' annoying father toook me for a ride in his new sports car one day and tried to show off the speak recognizing built-in cellular phone. "stepdad: call my office" "phone: request-not-understood". he kept trying and getting negged. it was hilarious.
16:42 < mae_> so please label those accordingly
16:42 < wchogg> mae_ : Thanks
16:42 < mae_> wchogg: np
16:42 < thomashartman1> mae: good idea
16:43 < chessguy> thomashartman1, we took a GPS when we went to see my wife's dad in texas. he argued with it the whole time
16:43 < chessguy> anyway, i'm off to a super bowl party
16:43 < chessguy> go steelers!
16:44 < wchogg> Oh yeah...that's today.
16:44 < wchogg> I keep forgetting what days are what.
16:44 < gwern> today?
16:45 < stepcut> the superbowl is just a warm up for the real party tomorrow
16:45 < gwern> what's tomorrow?
16:45 < thomashartman1> wchogg: suggest you create your own public repo called happstack-tutorial, transfer existing happs-tutorial over to that, and add me as a project member. then I'll delete happs-tutorial repo.
16:45 < stepcut> gwern: groundhog day!
16:45 < wchogg> thomashartman1 : Sounds good to me
16:46 < thomashartman1> and old happs-tutorial record at code.haskell.org
16:46 < wchogg> stepcut : Groundhog day parties?
16:47 < mae_> thomashartman1: yeah its big enough to where it probably belongs in its own repo ;)
16:47 < thomashartman1> but I'll hold off on doing so until after we do the hext hackage release, with addresses updated at http://happstutorial.com/tutorial/run-tutorial-locally
16:47 < stepcut> wchogg: of course, groundhog shaped cookies, watching groundhog day the movie...
16:47 < thomashartman1> adn within the cabal file of course.
16:48 < stepcut> why else would they sell this, http://www.thecookiecuttershop.com/cabin/cabin16.shtml
16:48 < wchogg> stepcut : everyone all spazzing out at their shadows & crawling under the couch.
16:48 < wchogg> stepcut : I could see how that could be a fun party
16:49 < stepcut> that's what all my parties are like
16:49 < wchogg> stepcut : ...you're a better man than I
16:55 < wchogg> buildbot, you live!
16:55 < koeien> yes. for now the messages go to me only since i don't want to annoy you
16:55 < mae_> buildbot build me some packages!
16:55 < mae_> hehe
16:56 < mae_> koeien: is 6.8.3 still building?
16:56 < mae_> koeien: i submitted some patches
16:56 < koeien> no. it is broken, it just told me
16:56 < koeien> unrecognized flag
16:56 < wchogg> koeien : I wasn't annoyed by the messages, honestly.
16:56 < koeien> ghc-6.8.3: unrecognised flags: -fno-warn-deprecated-flags
16:58 < wchogg> thomashartman1 : What did you have a dependency on Crypto for in happstutorial that wasn't in the HAppS.Crypto ?
17:01 < mae_> koeien: heh
17:01 < mae_> that was me :\
17:02 < koeien> the results are now stored in a *cough* mysql database
17:02  * koeien runs and evades stones thrown at him
17:02 < wchogg> *throws koeien seeking smart-stones*
17:03 < mae_> koeien: oh lord.
17:03 < mae_> heh
17:03 < koeien> makes it easy to insert data from another process
17:04 < stepcut> don't you have a REST API for that ;)
17:05 < koeien> SQL is a kludge. i made a typing mistake and spent half an hour tracing the bug
17:06 < mae_> koeien: just pushed a rollback patch, let me know if 6.8.3 builds
17:08 < stepcut> koeien: exactly.
17:08 < mae_> koeien: i would like to in the future, make macid also easy to interact with
17:09 < koeien> yeah i have to push log files, they're pretty big
17:09 < mae_> with rest
17:09 < mae_> then also some command line tools which use the rest interface
17:09 < thomashartman1> wchogg: still around?
17:10 < thomashartman1> wchogg: i'll email ya
17:10 < stepcut> is HList getting moved from HAppS-Data? There are some HList unit tests that go with it
17:10 < mae_> stepcut: i was debating getting rid of it
17:10 < mae_> since gwern has an hlist cabal package
17:10 < mae_> stepcut: but it will stay for now
17:10 < stepcut> ok
17:11 < stepcut> I'll hook up the tests then
17:11 < mae_> stepcut: you talking about the tests folder?
17:11 < mae_> with the stdout
17:11 < stepcut> yeah
17:12 < koeien> stepcut: i parse the HUnit output
17:12 < wchogg> thomashartman1 : Sorry, yes, I'm here.
17:12 < stepcut> koeien: in buildbot?
17:12 < thomashartman1> wchogg: there's no more crypto dep.
17:12 < koeien> stepcut: it would be very much appreciated if we could standardize on HUnit
17:12 < koeien> stepcut: yes
17:12 < wchogg> thomashartman1 : Ah, okay.
17:12 < stepcut> koeien: well,that is my plan
17:12 < thomashartman1> wchogg: happstut depended on pbkdf2 for login for a while.
17:12 < thomashartman1> I was going to brag that it had resolved the whole "is the login really secure" issue that was discussed on haskell cafe a while ago.
17:12 < koeien> stepcut: ok. i count the total number of testcases failed + total number of cases done
17:13 < stepcut> koeien: also, the tests will be exported as a library function, so you could write your own wrapper that calls them, if that made things any easier
17:13 < koeien> i have a small script for now. i build twice, once with -f tests and once without
17:13 < wchogg> thomashartman1 : But you cut out the pbkdf2 dependency in the darcs head?
17:13 < thomashartman1> however, pbkdf2 introduced a dep on crypto, and crypto (despite promises from maintainer dominic) still doesn't compile cleanly on ghc 6.10.1
17:13 < thomashartman1> wchogg: either there or in the last hackage release.
17:14 < thomashartman1> can't remembe.
17:14 < thomashartman1> the pbkdf2 stuff is actually kind of neat but not worth the packaging headache.
17:14 < thomashartman1> good for a blog post sometime if I can ever get it woring cleanly ootb.
17:14 < thomashartman1> patch-tag actually *does* use pbkdf2 fwiw :)
17:15 < wchogg> Right, because you're not worried about it being packaged up.
17:15 < thomashartman1> pretty much.
17:15 < h_buildbot> Build for ghc-6.8.3 failed. Check http://buildbot.happstack.com/ for details.
17:16 < koeien> mae: this in happstack-data. the flag is still not recognized. but this could also be the buildbot screwing up
17:16 < wchogg> thomashartman1 : So how much of the current todo list in darcs head is relevant still?  I was assuming it all was current.
17:16 < thomashartman1> this pbkdf2 stuff is really rather tending toward paranoia, and there might be disadvantages towards using it as well: eg, might be able to dos the server by repeated logins (checking the login is expensive)
17:16 < thomashartman1> wchogg: might be some cruft in there.
17:16 < wchogg> thomashartman1 : Alright, well if there ends up being irrelevant tickets then I'll just close them when I find them.
17:16 < wchogg> Not a big deal
17:17 < thomashartman1> wchogg: have you already started creating items in the happstack bug tracker?
17:17 < wchogg> thomashartman1 : Only a couple
17:17 < wchogg> but yes
17:17 < mae_> koeien: i don't see any logs for the 1st ?
17:17 < thomashartman1> wchogg: let's be efficient. I'll go through the text file, clean up, and push.
17:17 < thomashartman1> then you take from there.
17:17 < h_buildbot> Build for ghc-6.10.1 OK. Test suite ran. 0/1 test cases failed. Check http://buildbot.happstack.com/ for details.
17:17 < koeien> mae_: the logs are not being created in text files :(
17:18 < wchogg> thomashartman1 : Sure, sounds fine.
17:18 < thomashartman1> ok, i'll get to that and ping ya when done.
17:18 < wchogg> Thanks, I appreciate it.
17:19 < stepcut> h_buildbot: how could the test fail!!
17:19  * stepcut checks
17:19 < koeien> stepcut: 0/1 failed
17:19 < koeien> stepcut: so all succeeded
17:19 < stepcut> ah
17:19 < mae_> koeien: ok another patch, let me know
17:19 < koeien> maybe i'm too negative... ;)
17:19 < stepcut> :)
17:20 < stepcut> I read it as: (0 of 1 passed), test cases failed.
17:21 < mae_> koeien: flying blind without the build bot : ( i was depending on it already heh :)
17:21 < h_buildbot> Build for ghc-6.8.3 failed. Check http://buildbot.happstack.com/ for details.
17:21 < koeien> mae_: in fact the buildbot is building 6.8.3 now again for you
17:21 < koeien> hmm
17:22 < koeien> now it's happstack-contrib
17:22 < koeien> same error message
17:22 < koeien> did this work in the past, or did you add the flags to remove warnings?
17:23 < mae_> i did add the flag
17:23 < mae_> but didn't realize it would break 6.8.3
17:23 < stepcut> the HAppS/test/test file boggles my mind. Is that supposed to be machine readable? Or just notes on what a person should do?
17:23 < mae_> koeien: pushed up another patch
17:24 < h_buildbot> Build for ghc-6.10.1 OK. Test suite ran. 0/1 test cases failed. Check http://buildbot.happstack.com/ for details.
17:24 < koeien> mae_: the buildbot is in automagic mode, so we will hear about it in a few minutes
17:25 < mae_> koeien: just an aesthetic tweak but how about 'All tests passed.' if 0/N.
17:25 < mae_> err rather if none failed
17:25 < koeien> no problemo
17:28 < h_buildbot> Build for ghc-6.8.3 OK. Test suite ran, all tests passed! Check http://buildbot.happstack.com/ for details.
17:28 < mae_> thank goodnes you are back buildbot
17:28 < mae_> i missed you
17:30 < stepcut> I am adding the test stuff to happstack-data now, and adding the top-level portion as well which will run all the tests for happstack-util and happstack-data
17:30 < h_buildbot> Build for ghc-6.10.1 OK. Test suite ran, all tests passed! Check http://buildbot.happstack.com/ for details.
17:33 < koeien> stepcut: I already have a script for that in place
17:33 < stepcut> koeien: a script for what?
17:33 < koeien> stepcut: i only need to know which executables to run
17:33 < koeien> stepcut: for building & installing the testsuite
17:33 < stepcut> koeien: ok
17:34 < stepcut> you will only need to run one executable
17:34 < koeien> perfect
17:35 < stepcut> though, if some of the modules fail to build, you would still be able to run the testsuites for the modules that passed if you wanted
17:35 < koeien> stepcut: i don't use exit codes, only the HUnit output. it should be no problem if you execute more than one HUnit tests, they should be added up
17:35 < koeien> stepcut: if the build fails, at this point i'm not going to execute the testsuite. I don't really see the point of it atm
17:36 < stepcut> well, if HAppS-State fails to build, and then someone introduces a bug in HAppS-Data, it could report that bug which is unrelated to HAppS-State. Probably overkill at the moment though.
17:36 < koeien> possible. but it's not that large a project
17:37 < stepcut> wait, we are replacing apache *AND* mysql, and it's *not that large of a project* :p
17:37 < stepcut> (and memcached)
17:37 < koeien> i'm counting loc ;)
17:38 < wchogg> Right.  Small in size, not in grandeur.
17:39 < koeien> at the moment i think it's overkill to run a subset of the testsuite if the thing FTBFS. `not compiling' is a pretty important bug ;)
17:40 < koeien> a higher priority feature is getting the logs to be displayed on the web page
17:41 < stepcut> :)
17:53 < Saizan> mae_: hackage's hlist (aside from being a poor cabalization of oleg's work) is much more complex than the subset implemented in happs, though i don't think the latter has seen much use either
17:53 < Saizan> it looks related to dealing with xml
17:55 < stepcut> heh, happstack-data has one of my favorite kinds of tests: namely testing that certain code *does* generate compiler errors
17:57 < stepcut> it also appears that some of the Xml* tests are failing
17:59 < Saizan> am i right that we have an .HList in -Server too?
18:07 < wchogg> I don't think so
18:09 < wchogg> Ah, it looks like Mr. Elder removed it last week
18:13 < mae_> yep
18:13 < mae_> but it is still in HAppS-Data
18:14 < mae_> there was one in Data and one in Server
18:14 < mae_> o.O
18:18 < Saizan> they are different, i think
18:19 < Saizan> "HList" is a vague concept :)
18:24 < mae_> homogenous type list :)
18:25 < Saizan> heterogeneous list, actually :P
18:28 < mae_> err yeah
18:28 < mae_> thats what i meant
18:28 < mae_> haha
18:33 < gwern> Saizan: what was all that bad of my cabalization?
18:34 < Saizan> gwern: i had to expose many other modules before i could actually use it, and you didn't respect the multiple choices available for labels
18:35 < Saizan> well, the two are related
18:35 < Saizan> but it's been a while
18:35 < gwern> ah. I thought Data.HList re-exported all the rest
18:35 < Saizan> you can't re-export everything
18:35 < Saizan> from one place
18:36 < gwern> all the useful rest, I guess I meant
18:36 < Saizan> you're supposed to pick one of the possible strategies
18:36 < Saizan> have you ever done something with it?
18:36 < koeien> http://buildbot.happstack.com/ is somewhat operational. could you take a look at it?
18:36 < gwern> no
18:40 < koeien> mae_: you can now look at the exact error message ;)
18:48 < koeien> anyway, i'm heading to bed. let me know what you think about the new page
18:53 < stepcut> we should probably update the Tested-with: field in the .cabal files
19:23 < stepcut> koeien: I pushed updates for the test suite
19:24 < stepcut> koeien: there is a new script, bin/build-install-test-all.sh, which builds everything with the testsuite enabled. Then you just run happstack-tests afterwards.
19:24 < stepcut> E-Z.
19:25 < stepcut> and, there are only 15 errors/failures :)
19:28 < h_buildbot> Build for ghc-6.8.3 OK. Test suite ran, all tests passed! Check http://buildbot.happstack.com/ for details.
19:29 < stepcut> bah. it's not running the full test suite yet :)
19:31 < h_buildbot> Build for ghc-6.10.1 OK. Test suite ran, all tests passed! Check http://buildbot.happstack.com/ for details.
19:36  * stepcut stomps around
20:00 < thetallguy> dude, it's the superbowl
20:01  * thetallguy thinks stepcut is unreasonable to expect any bug fixes until fifteen minutes after the last commercial
20:56 < h_buildbot> Build for ghc-6.8.3 failed. Check http://buildbot.happstack.com/ for details.
20:57 < h_buildbot> Build for ghc-6.10.1 failed. Check http://buildbot.happstack.com/ for details.
21:02 < Saizan> uhm, the page isn't updated
21:03 < Saizan> koeien: you need to "cabal update" on the build-bots
23:54 < rovar> h_buildbot, you're alive!
--- Log closed Mon Feb 02 00:00:18 2009