Experimental IRC log happs-2007-08-27

Available formats: content-negotiated html turtle (see SIOC for the vocabulary)

Back to channel and daily index: content-negotiated html turtle

These logs are provided as an experiment in indexing discussions using IRCHub.py, Irc2RDF.hs, and SIOC.

00:59:04<shapr>ACTION tries to write a wiki with the new HAppS...
01:03:16<stepcut>nice
01:03:26<shapr>Except that I don't understand everything yet...
01:03:32<stepcut>perhaps one with fermats last margin built in
01:03:41<shapr>Yeah, that's the plan...
01:03:53<shapr>Though there's some impedance mismatch between FLM and HAppS
01:04:45<shapr>If I keep all the annotations in WikiState, I can't easily use darcs to do version control.
01:05:36<shapr>I'd like to use darcs for disconnected wiki changes, and then push/pull new changes later.
01:05:41<shapr>But I don't see how to mix that with HAppS
01:06:32<shapr>stepcut: Any ideas?
01:07:09<stepcut>not off-hand
01:08:10<shapr>Only way I see to do that is to have HAppS state only hold which repos are local, and their filesystem location, and then use fileserve for the actual wikipages.
01:29:21<shapr>I thought of a hackish solution...
02:05:54<shapr>Anyone hacking on anything HAppS at the moment?
02:10:09<stepcut>soon, I hope
02:10:17<stepcut>but not right now :(
02:10:52<shapr>ok, just wondering if there were any questions.
02:10:59<Saizan>i was going to port hpaste (or my app) to 0.9, but i fear that there's no way to migrate the state files?
02:11:10<shapr>Saizan: Have you seen the State migration stuff?
02:11:31<shapr>I'm not sure if it works yet, but it hopefully illustrates the idea.
02:11:51<Saizan>yes, but i'm not sure if it works with 0.8.8 serialization
02:12:01<shapr>I don't know, worth trying.
02:12:32<Saizan>yup, Serialize is still there, so maybe yes
02:57:16<Saizan>shapr: why keep an HAppS.hs inside HAppS-Begin instead of substituting the one in HAppS-Server? the one in Server lacks some useful exports like HAppS.StdMain..
02:57:53<Saizan>importing HAppS outside HAppS-Begin and get "not in scope" errors was very confusing :)
03:07:25<shapr>I agree with you....
03:17:49<shapr>Submit a patch!
03:25:54<shapr>Saizan: Along the same lines, there's an Atom.hs in -Begin and in -Data
08:18:39<tuxplorer>Is there a function to add JS and CSS files? or should I be doing that too by XSLT only? Instead of tieing up the styles with the XML its better to have it in the transformation code right? So, should I be linking the CSS files in the XSLT side, or is there any function in HAppS that would do it?
16:25:18<Saizan>deriveAll creates a lot of functions like userDataType1Constr :: Data.Generics.SYB.WithClass.Basics.Constr
16:25:21<Saizan>userDataType :: Data.Generics.SYB.WithClass.Basics.DataType
16:26:30<Saizan>are those used anywhere besides the derived instance declarations?
16:27:15<Igloo>If you mean "could they be local" then probably
16:27:29<Saizan>yeah
16:29:10<Igloo>Except they are used in multiple declarations in the class, so they'dd have to be duplicated
16:30:19<Saizan>thought so
21:19:56<shapr>@yow !
21:49:12<sorear>shapr: now you can :)
22:06:26<shapr>@yow !
22:06:27<lambdabot>Why are these athletic shoe salesmen following me??
22:06:28<shapr>w00!
22:30:11<jmelesky>so, i'm having trouble following along with http://happs.org/HAppS/HAppS-Begin/public/
22:30:12<lambdabot>Title: How To Make a blog using HappS
22:30:40<jmelesky>i have sp installed, but it's complaining about Control.hs missing
22:31:14<jmelesky>if i turn around and try to install the various HAppS projects individually, HAppS-Data complains about syb-with-class-any missing
22:31:29<shapr>Ah, you'll need syb-with-class
22:32:14<shapr>darcs get http://happs.org/syb-with-class/
22:32:18<jmelesky>ah. duh. "syb-with-class-any" doesn't show up on google. should have managed that extra step
22:32:39<shapr>The version on Hackage is old.
22:32:59<shapr>You should install them in this order: Util, Data, IxSet, State, Server, Begin
22:33:38<jmelesky>wow. lots of warnings on Data
22:33:48<jmelesky>yeah, i figured the order from the cabal files
22:34:07<shapr>I wish cabal-install could install from darcs repos.
22:34:26<jmelesky>this is a fresh darcs pull from about fifteen minutes ago
22:34:27<shapr>No point in uploading the HAppS packages yet, since bugs have been found.
22:34:28<shapr>spiffy
22:34:39<shapr>that is, uploading to hackage
22:37:15<jmelesky>about how many LOC do you think?
22:37:31<shapr>About 7kloc
22:38:08<jmelesky>that's not too bad
22:38:24<shapr>It'll be smaller when we clean out some dead code.
22:38:31<shapr>And it'll be larger when some new features work :-)
22:38:46<jmelesky>i was noticing some of the promised features
22:39:05<shapr>Nice list, eh?
22:39:06<jmelesky>"HAppS takes care of multimaster replication so everything stays in sync. !!! this will be working soon."
22:39:28<shapr>I think it does already work, but it has a corner case...
22:39:48<jmelesky>grr. cabal didn't cleanly uninstall HAppS-0.8.8
22:39:59<shapr>If disconnections happen before a transaction is committed, it causes problems, iirc
22:40:47<jmelesky>that's a pretty common corner case
22:41:45<Saizan>jmelesky: no new package is called HAppS, so it doesn't have anything to "uninstall"
22:42:09<jmelesky>ah. point. so, manually unregister it, then?
22:42:18<Saizan>yes, or hide it
22:44:38<jmelesky>hmmm
22:44:43<jmelesky>unretrieved: http://happs.org/HAppS/HAppS-Util/src/HAppS/Crypto/SHA1.hs
22:44:48<jmelesky>that sounds non-lethal
22:45:40<jmelesky>cool. up and running the sample app.
22:46:52<jmelesky>it looks like the documentation line: sp runhaskell -i.hs .hs/Main.hs
22:47:02<jmelesky>should be: sp runhaskell -ihaskell haskell/Main.hs
22:55:18<shapr>yeah
22:55:32<shapr>I don't know why SHA1 isn't available... hmm
22:57:31<jmelesky>are patches accepted?
22:57:49<shapr>Yeah, send 'em in.
22:57:56<jmelesky>what address?
22:58:27<shapr>alex@vo.com, igloo@vo.com, shapr@vo.com, lemmih@vo.com
22:58:34<shapr>Or you could use the HAppS group?
22:59:00<jmelesky>HAppS@googlegroups.com ?
22:59:12<jmelesky>it's just a doc patch based on my little install adventure
22:59:29<Saizan>ah, i've sent couple to alexjacobson at happs.org
23:41:11<Saizan>is there any static guarantee that a query will be used in an app with the right state type?
23:43:48<Saizan>mmh i don't even see how is it fixed, by having the stateFuns in the hlist passed to stdMain?

Back to channel and daily index: content-negotiated html turtle