Experimental IRC log happs-2007-10-22

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.

02:11:10<stepcut>is there an example on how to upload and save a file to HAppS ?
02:15:29<alexj>stepcut: upload is basically defined by input.
02:15:36<alexj>no example
02:15:49<alexj>there is an s3 lib for saving.
02:15:53<alexj>but it is not up-to-date.
02:19:02<stepcut>ok, I'll poke around and figure out how to use the FileSaver stuff
02:19:15<stepcut>alexj: I had some other quick questions, since you are here
02:20:33<stepcut>if I defined a type like, data Foo = Bar | Baz and then do:
02:20:35<stepcut>*Ticket> toPairs Bar
02:20:35<stepcut>[]
02:20:52<stepcut>it seems odd that I get an empty list
02:21:50<stepcut>I also found it odd that this never seems to work:
02:22:00<stepcut>*Ticket> fromString Rigid (toString Bar) :: Maybe Foo
02:22:00<stepcut>Nothing
02:22:34<stepcut>though, this does,
02:22:35<stepcut>> runIdentity (fromString Flexible (toString Bar)) :: Foo
02:22:35<stepcut>Bar
02:23:13<alexj>thats interesting. igloo has been working on Flexible vs Rigid parsing.
02:23:32<alexj>can you post it as a bug?
02:24:07<stepcut>ok
02:24:21<alexj>the toPairs thing seems more reasonable. not sure what toPairs is supposed to do with your Foo type.
02:24:37<stepcut>yeah, I am not really sure either
02:26:50<stepcut>poking around more, I think toPairs is a lot more restrictive that I original thought
02:53:59<stepcut>S3 is more than I need for saving at this point in time -- local filesystem is fine. My real question is, am I supposed to be using some sort of side-runner thing to do the saving? It looks like a lot of that code was commented out recently
02:59:10<stepcut>I don't want to update the state until the file is actually writing to disk, but if the file is going to take a long time to upload, I don't want to block up the whole system until it is done...
03:10:11<stepcut>oh, it appears igloo fixed the (fromString Rigid ...) thing yesterday :)
03:14:39<stepcut>hrm, maybe I just do something like, forkIO $ do L.writeFile fp (inputValue contents) >> updateFileList, where updateFileList is one of those autogenerated functions that adds an event to the event queue
04:16:44<alexj>stepcut, yes all handlers are just IO. You can just go ahead and try to write the file and then send an update command. no need to fork.
04:34:39<stepcut>is each handler already run in a seperate thread ?
04:35:12<stepcut>ACTION is very unclear on how the ACID parts of HAppS work these days
04:35:36<alexj>stepcut: yes handlers are just doing IO.
04:35:43<alexj>(depends what you mean by handler)
04:36:11<alexj>ACID happens when you do an update
04:36:27<alexj>update $ MyCommand somearg anotherarg
04:36:47<stepcut>ok
04:38:29<stepcut>so, if someone is uploading a large file, the rest of the system is not blocked because, each incoming connection is automatically handled in a seperate thread ?
04:42:15<alexj>yes
04:42:45<alexj>if you attempt to store the file instelf inside state, things will be bad. but you shouldn't do that.
04:43:00<stepcut>ok
04:44:07<stepcut>if I do a 'query SomeCommand', and then a few lines later, do the exact same query, the results could be different then, because the data set was updated by a different thread ?
04:44:30<alexj>not sure.
04:44:48<alexj>we've been talking about making control over this issue more finegraned.
04:45:14<alexj>I think they should be different.
04:45:19<stepcut>Is there a way to be sure that several queries are in the same 'transaction'
04:45:29<alexj>yes make them a single query.
04:45:52<stepcut>ok, makes sense
04:46:41<stepcut>I remember the way that I thought the original version worked, a number years ago, so I am trying to figure out what is different now ;)
04:48:22<stepcut>thanks!
04:48:24<stepcut>ACTION goes to bed
08:03:06<profmakx>yello!
08:16:24<profmakx>I installed HAppS from darcs yesterday, but when I try to start it with runhaskell -ihaskell haskell/Main.hs i get the following error Main.hs: getAddrInfo: does not exist (servname not supported for ai_socktype). I wasnt able to find out even where this happens
08:16:42<profmakx>I suppose its related to that I have a Network package with ipv6 support installed
09:25:47<profmakx>oh well. it seems that Network is broken on FreeBSD
09:25:56<profmakx>(at least with IPv6 support enabled)
12:31:08<Laney>Can someone give me a hand embedding some IO action into a hello world program? Here's the code I'm working with atm: http://pastebin.com/d61f2d367
12:39:58<Laney>OK never mind, I got it. (curtime >>= ok)

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