Experimental IRC log happs-2008-08-06

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.

10:44:45<MarcWeber>chr1s ping
10:45:30<MarcWeber>I've sucessfully installed happs-hsp now. hsp-xml still fails due to a compilation error.
10:50:45<chr1s>MarcWeber: me too, stepcut sent a patch where he commented out all the hinotify stuff.
10:53:46<MarcWeber>Which is the best way learning about how to use hsp ?
10:56:09<chr1s>I really have no idea, I don't know anything about HSP
10:57:00<MarcWeber>But you'd like to learn about it as well?
10:57:06<MarcWeber>Or why have your tried compiling it?
15:46:18<Saul>I want to use some datatypes in a library with happs state, so I need to make them an instance of Typeable, however I can't do that using deriving since it's in a library
15:46:33<Saul>Is there a Template Haskell function for doing that?
15:47:05<Lemmih>You can use standalone deriving or wrap it in a newtype.
15:49:04<Saul>ok
15:49:57<Saul>thanks
15:51:23<Saul>May I ask why there is hardly any documentation out there?
15:57:29<Lemmih>Because HAppS is fairly experimental.
16:01:59<Saul>ok, so I can expect that to change once version 1.0 comes out?
16:02:57<Saul>still I think it might be useful to add some haddock comments to the functions
16:06:06<dcoutts_>Lemmih: it'd be worth trying the latest haddock release
16:38:28<Lemmih>Saul: Patches are welcome.
16:41:04<Lemmih>dcoutts_: Same error with haddock-2.2.2
16:41:26<dcoutts_>Lemmih: report it to waern
16:42:08<MarcWeber>Saul: Really, can't you ? doesn't ghc-6.8.2 even support that?
16:48:32<Saul>Lemmih: I have a hard enough time figuring out how to use happs, let alone patch it
16:49:21<Saul>MarcWeber: ghc supports deriving typeable, but that only works on datatypes that you define yourself
16:50:16<Saul>MarcWeber: Just like it does for Eq and Ord and such
16:51:56<MarcWeber>http://www.haskell.org/ghc/docs/latest/html/users_guide/deriving.html @ Saul 8.5.2
16:59:40<Saul>MarcWeber: Thanks, that works like a charm
17:00:55<MarcWeber>You've been right about ghc 6.6
17:04:55<Saul>I was using 6.8.2 already, I just didn't know about the option
18:26:40<chr1s>Lemmih: experimental? seriously? we're trying to build a website here!
18:26:42<chr1s>:p
18:27:52<chr1s>also, I don't think the documentation problem with HAppS is in the API-documentation, that's fairly well documented. The types help a lot.
18:28:16<chr1s>I think we mainly need a good tutorial / documented examples.
18:28:48<chr1s>I think Eelco is working on that currently
18:29:24<chr1s>Lemmih: do you only work on HAppS as a developer, or do you also develop using HAppS?
18:48:09<MarcWeber>Is utf8-string the library of choice when encoding utf-8?
19:21:27<Saul>chr1s: I just read your blog post on formlets, seemed interesting
19:21:50<Saul>chr1s: I've worked on something similar myself, but your method seems a bit more elegant
19:22:44<MarcWeber>How do you deal with utf-8?
19:23:09<MarcWeber>I mean the bytestring library doesn't at all.. So should it be used at all here?
19:23:33<Saul>chr1s: But do you allow checks using monadic data, and how do/will you handle Maybe types?
19:24:06<Saul>MarcWeber: I don't know, maybe you should ask on #haskell, there might be someone there who knows
19:24:24<MarcWeber>I've already done that..
19:24:52<MarcWeber>I've tried converting parts of WASH using lazy bytestrings.. and there is a function which escapes some special chars..
19:25:12<MarcWeber>but if the string already contains UTF8 then it won't work in all cases as expected
19:25:15<Saul>I see
19:25:33<Saul>Well I don't have any experience with UTF-8, so I can't help you
19:30:25<Lemmih>chr1s: The former.
19:31:22<Lemmih>chr1s: (re building a website) How very brave of you, sir! (:
19:32:09<Saul>Lemmih: Does that mean I'm brave too? :P
19:35:06<Lemmih>You're a bunch of lambda eating lions.
19:36:04<jgrimes>haha..
20:00:44<Saul>I'm having some problems with getting my stateful functions to work
20:00:53<Saul>I get weird type errors
20:02:38<Lemmih>We're here to help.
20:04:13<Saul>I pasted at http://hpaste.org/9434
20:05:04<Saul>I get errors involving UpdateEvent and QueryEvent (and also MonadReader and MonadState)
20:05:31<Lemmih>Saul: update $ AddGame gd
20:05:39<Lemmih>Same for the query.
20:08:00<Saul>ok thanks
20:13:30<chr1s>Saul: cool!
20:14:10<chr1s>Saul: I think formlets should work just fine with Maybe types.
20:14:58<chr1s>Lemmih: I do have to say, now that we are finally really understanding how to work with HAppS, it's pretty awesome.
20:15:36<chr1s>I really like it so far. We're also building sort of a framework on top of HAppS, hopefully we can release that sometime soon.
20:16:00<Lemmih>That's good to hear.
20:16:03<Saul>chr1s: And can you check data using monadic actions?
20:16:35<chr1s>Saul: well, currently not. However, it would be fairly trivial to extend the library to do that.
20:16:37<Saul>chr1s: And how do you catch invalid input (like when someone doesn't use a browser and some fields do not have a value)
20:16:58<chr1s>Saul: currently, that gives an error.
20:17:17<chr1s>That's a bug (but not a hard one to fix).
20:18:18<Saul>chr1s: And can you use multiple values as well (which is needed for menus and radiobuttons)
20:18:26<Saul>?
20:19:24<chr1s>Saul: sure!
20:19:56<chr1s>so one example would be this:
20:20:26<chr1s>inputRadio :: [(a, String)] -> Formlet a
20:21:02<chr1s>where a can be *any* type. You could then generate a radiogroup that has the second component as label (the String), and the index of the list as value.
20:22:01<Saul>Although the strings don't need to be unique
20:22:21<Saul>Oh wait, you would number them internally
20:22:39<chr1s>and then in your validate method you convert the POST'ed value (a String) into an Integer idx, and then do an "fst $ ls !! idx", where ls is the list you originally passed in.
20:22:47<MarcWeber>Lemmih: I'd like to write an web application as well.. but I'm struggling which html library to use and which character representation..
20:23:29<chr1s>MarcWeber: fwiw, we choose Xhtml.Strict. It's been very nice, although larger html fragments are a bit of a mess.
20:23:45<Saul>MarcWeber: I'm using that as well
20:24:00<chr1s>I'm going to see if we could replace it by stepcut's hsp-template
20:24:57<Saul>I used hsp when I tried wash, it seemed pretty nice
20:25:52<MarcWeber>Have you had a look at WASH at well?
20:26:04<Saul>Although I didn't really like how it did forms
20:26:06<MarcWeber>I like the idea of having checked correct output
20:26:45<chr1s>I think the formlets are an awesome idea.
20:26:51<Saul>chr1s: Me too
20:26:57<chr1s>for a full explanation, see http://groups.inf.ed.ac.uk/links/formlets/
20:30:11<MarcWeber>I need some training on Data.Applicative
20:32:24<Saul>Those are pretty cool
20:34:48<chr1s>MarcWeber: http://www.soi.city.ac.uk/~ross/papers/Applicative.html
20:34:52<chr1s>ACTION is away now.
21:48:05<Saul>I'm getting an infinite loop when I submit my form :(
21:50:34<Saul>ah nevermind, my read couldn't parse the input

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