00:27:26 <astroboy> stepkut, stepkut: how can I get the requested path? I'm trying with "look "PATH_INFO"" but it doesn't seem to work
00:35:26 <jaspervdj> stepkut: ping
00:36:59 <astroboy> jaspervdj: btw I'm rostayob, thanks for the quick pull :P
00:37:36 <jaspervdj> astroboy: Ah, okay, thanks
00:37:51 <jaspervdj> stepkut changed the happstack code a bit, now trying to merge
00:38:37 <astroboy> jaspervdj: well his code will probably be better, it shouldn't be difficult
00:49:46 <astroboy> i'm hooked to haskell I can't study :(
00:59:10 <MAN> goodevening, may somebody spare a few minutes to answer a question? :)
01:04:32 <astroboy> MAN_: I'll try :P
01:05:23 <MAN_> ok ,astroboy
01:05:33 <MAN_> I've installed happstack 0.6, great stuff...
01:05:50 <MAN_> but now I see happstack-auth has not been updated (yet?)
01:05:56 <MAN_> is there a plan to do so?
01:06:49 <astroboy> MAN_: I don't know, I just built a custom auth system
01:07:21 <MAN_> astroboy: from scatch?!
01:07:49 <MAN_> astroboy: that sounds like too much work for my plans... I was looking for a more standarized way.
01:07:51 <astroboy> MAN_: well yes
01:08:15 <astroboy> in my experience most of the times is simpler to do your own stuff at the end :
01:08:22 <astroboy> but yeah, I guess it can be useful
01:08:52 <MAN_> astroboy: is your code open and available?
01:09:06 <mafs> stepkut: is working on auth I believe
01:09:14 <mafs> Don't know how far along it is, though
01:09:28 <MAN_> oh, good.
01:09:49 <astroboy> MAN_: I'm writing it now it should be available soon on https://github.com/rostayob/hsnews , but it's not a Component. I guess you could look into it anyway
01:09:58 <MAN_> I've been searching around and saw there are (were) several attempts to build _the_ auth layer for happstack.
01:10:06 <astroboy> MAN_: the problem is that
01:10:19 <astroboy> building *THE* auth layer
01:10:21 <astroboy> is difficult
01:10:23 <astroboy> in every framework
01:10:52 <MAN_> astroboy: yes, I suppose so. I realize happstack is still a work in progress.
01:12:52 <MAN_> astroboy: oh, user ranks! great idea :P
01:13:16 <astroboy> MAN_: it's really simple
01:13:19 <astroboy> but that's all I need
01:13:29 <astroboy> but if you wait 10 mins
01:13:34 <astroboy> it's better now :P
01:40:34 <MAN_> astroboy: It seems building hsnews triggered a butterfly effect in my packages -_-
01:46:43 <astroboy> MAN_: ah, I don't think you can do that anyway, I'm using a modified digestive-functors
01:46:57 <astroboy> MAN_: but I can't see what damage it could've done, what happened?
01:46:58 <MAN_> astroboy: buh!
01:47:24 <astroboy> I mean, you should have most of the stuff installed anyway
01:47:38 <MAN_> astroboy: A cascade of download-build-install in my system that eventually rebuilt happstack... kinda funny.}
01:48:26 <MAN_> astroboy: yes, the packages are there. Some have been rebuilt to previous version (or something like that). I've already fixed it.
01:48:39 <astroboy> MAN_: oh ok. cabal install it's shit anyway, I don't use it anymore
01:48:41 <MAN_> where fix = reinstall
01:48:53 <MAN_> astroboy: it has it many flaws, yes.
01:48:53 <astroboy> but I don't *think* it's my fault.
01:49:09 <MAN_> astroboy: I never said that, relax XD
01:49:22 <astroboy> no it's just that I'm interested :P
01:49:30 <astroboy> if I screwed something up I want to know
01:49:34 <astroboy> since I don't use cabal install
01:49:44 <MAN_> astroboy: oh. http://cdsmith.wordpress.com/2011/01/17/the-butterfly-effect-in-cabal/
01:50:24 <MAN_> astroboy: quick read. The comments point out that doing this effectively is no stroll in the park
01:50:44 <astroboy> MAN_: what I suggest you to do
01:50:47 <astroboy> is to install archlinux
01:50:50 <astroboy> and use cabal2arch
01:51:02 <astroboy> so you're sure that only one version of an hackage package is installed
01:51:06 <astroboy> only way to avoid huge messes
01:51:17 <astroboy> in my experience
01:51:41 <MAN_> astroboy: I did, for I while. A few months back. But I kind of _want_ to be able to have several versions of the same package installed...
01:51:51 <MAN_> I just wish they'd get along better :P
01:52:33 <astroboy> MAN_: yeah I know, but apart from parsec there are few cases in which it actually happens
01:53:15 <MAN_> ok, I'm out. Thanks for the time, astroboy.
02:00:54 <astroboy> kk
03:33:04 <siracusa> Yay! Happstack 6 still works with GHC 6.10.4, that's neat.
03:52:43 <stepkut> siracusa: yep!
03:52:47 <stepkut> siracusa: not 6.8 though
03:54:55 <stepkut> I'm working on some post release stuff now, like updating hackage-server to Happstack 6
04:38:36 <siracusa> How do you get GET or POST data separately? lookInput doesn't seem to care where a value comes from.
04:43:25 <stepkut> siracusa: it is really not GET vs POST but QUERY_STRING vs request body
04:44:04 <stepkut> siracusa: http://happstack.com/docs/crashcourse/RqData.html#rqdatalimiting
04:47:10 <siracusa> Thanks, I still was in chapter 2 of the course :-)
04:58:47 <stepkut> siracusa: no worries :)
06:06:21 <siracusa> stepkut: While reading about RqData and looking at the source, it would be nice if getDataFn wouldn't return a list of error messages in case of failure, but instead a list of values of say RqDataError for better error processing.