00:32:41 <TimeCat> I have a question about dependencies and happstack-5.0.1
00:34:01 <stepkut> TimeCat: oh ?
00:34:09 <stepkut> TimeCat: ask away!
00:35:15 <TimeCat> I am trying to go through the tutorial.happstack.com site, but on cabal install I get a message saying that tutorial requires mtl >= && < while happstack-0.5.1 requires mtl ==
00:37:15 <gienah> hack the cabal file, bump the dependencies, fix the breakages is my philosophy :-)
00:37:59 <gienah> sometimes bumping mtl can be somewhat challenging
00:38:23 <TimeCat> gienah: yeah that was my first thought but it still seemed to fail, if you want to see the output let me know
00:38:59 <stepkut> ah.. tutorial is maintained by someone else (thomas hartman). it has not really been updated for 0.5.1. Ultimately it is supposed to be replaced by the happstack crash course
00:39:26 <stepkut> though, if someone wants to take over authorship of it and bring it up-to-date, there is cool too :)
00:39:48 <stepkut> I can provide some guidance on the build issues, but I do not have time to keep it up-to-date myself
00:39:57 <TimeCat> stepkut: That is kind of what I figured might be the case with it. In the mean time I have been going through the crash course, but thought I would give that a look too
00:40:51 <TimeCat> stepkut: Despite my name I don't think I have enough time or experience to do anything with that tutorial, though I might let you know in the future
00:41:23 <TimeCat> as in maintain, I do have enough spare time to follow it (of course)
00:41:29 <stepkut> well, if by fixing the tutorial you would gain the experience ;)
00:42:35 <stepkut> I think john meacham decided to learn haskell by writing a haskell compiler :p
00:43:33 <stepkut> TimeCat: anyway, the source for the tutorial is available via darcs. So if you fix the build errors, you can send a patch for that, which would be useful
00:43:42 <TimeCat> stepkut: I took a breif look around it, and if I manage to fix it I will send a patch, but I am certianly no meacham.
00:44:01 <stepkut> I'm no meacham either :)
00:44:13 <stepkut> but, a patch would be awesome!
00:47:07 <TimeCat> stepkut: i'll see what I can do, and thanks again
02:24:27 <mafs> Hmm...happstack server really slow under OS X for some reason
02:25:47 <mafs> 7 second response time hosted on localhost, `main = simpleHTTP nullConf $ ok "A"`. run via runhaskell
02:34:33 <stepcut> hmmm
02:34:51 <stepcut> usually it is faster
02:35:00 <stepcut> I have not tested since I made some recent changes though
02:35:17 <stepcut> I should test again.. that is in the cards for the next week anyway
02:35:28 <mafs> alright, just wanted to bring it to your attention at least
02:35:35 <stepcut> yeah
02:35:37 <mafs> I'm probably going to hop back into linux for some real development anyway
02:35:56 <stepcut> heh
02:36:08 <mafs> hard to work efficiently without a tiling window manager :)
02:36:28 <stepcut> i use xmonad on OS X.. but only for X apps...
02:36:50 <mafs> yeah, wish it worked for all. best I've found so far is divvy, but it's mainly manual management
07:10:57 <stepkut> the docs for happstack 0.5.1 are now online, http://www.happstack.com/docs/0.5.1/
07:11:19 <stepkut> time to start sending patches with improvements to Happstack.Server.* ;)
07:12:20 <mafs> sweet
07:12:27 <mafs> which reminds me, was going to tell you earlier: the docs *rock*
07:12:38 <mafs> I was reading through the haddock documentation a bit earlier, extremely well written and helpful
07:13:51 <stepkut> cool!
07:14:53 <stepkut> I have been trying to make sure we have 100% coverage for the user portion of Happstack.Server.*, and that there are plenty of examples
07:15:50 <stepkut> Happstack.Server.Internal.* is not as critical, since most people should never need to directly import anything from there
07:16:27 <stepkut> also trying to make sure that the haddock docs refer to sections of the crash course when appropriate
07:17:41 <mafs> ooh yeah, good idea
07:23:02 <stepkut> the haddock-all.sh script is annoying difficult to get right
07:23:17 <stepkut> but I got it working enough to generate some docs I can upload to the site :)
07:23:19 <stepkut> now it is time for bed
07:31:46 <mafs> g'night
11:36:21 <Entroacceptor> stepkut: do I need haddock-all if I have documentation: True in my cabal/config?
15:33:01 <mafs> stepkut: bodypolicy asks for a directory to store files--but it doesn't store *any* data to disk you'd want to access that's been PUT or POSTed, right, or does it?
15:45:46 <mafs> Nevermind, got it :)
18:24:11 <mafs> Hmm, time to actually start writing the oauth stuff it seems
18:37:41 <stepkut> :-/
18:37:43 <stepkut> good luck!
18:37:49 <mafs> thanks
18:37:59 <mafs> I was postponing it as long as possible, using plaintext passwords for the prototyping portion
18:38:07 <mafs> but it'll make my life a lot easier once it's done
18:38:14 <stepkut> yeah
18:50:11 <mafs> Hmm, I need to see about a way to pass something from path to withData (or lookup a path segment in the fromData definition in instance FromData..)
18:51:52 <stepkut> I don't generally use FromData
18:51:59 <stepkut> s/generally/ever/
18:52:31 <mafs> Really?
18:52:46 <mafs> Any reason why not?
18:54:04 <stepkut> just seems like extra work with no real gain
18:54:16 <stepkut> and makes it hard to parameterize things with extra values
18:54:48 <mafs> so what tactic do you take then when trying to get useful data out of querystring info?
18:55:19 <stepkut> I just use the look functions directly in my code
18:55:45 <mafs> That'll probably end up being a lot easier, plus it'll let me keep my current URL styling which I'm fond of
18:55:48 <stepkut> like this, http://happstack.com/docs/crashcourse/RqData.html#rqdata
18:56:24 <mafs> aha, thanks
18:57:13 <stepkut> I think FromData is a left over remnant from a style of development that was going to use a lot of generics and stuff
19:04:00 <mafs> \o/ thanks
19:04:08 <mafs> Just switched the code to use look directly, much nicer
19:04:40 <stepkut> yeah
19:04:59 <stepkut> I removed the FromData stuff from the guestbook example too