- Haskell-Cafe - Haskell.org

Re: [web-devel] Re: [Haskell-cafe] Google Summer of Code: BlazeHTML RFC
by Jasper Van der Jeugt 03 Jun '10

03 Jun '10

On Thu, May 27, 2010 at 8:31 PM, Matt Parker <moonmaster9000(a)gmail.com> wrote: > will it be possible to easily interleave IO values into the HTML? like > instead of the [1,2,3] > > ul $ forM_ [1, 2, 3] (li . string . show) > > what if it was a function that returned IO [1,2,3] (maybe 1,2,3 came out of > a database). will the forM_ handle that OK? Depends on how you write the code. More precisely, it depends on in which monad the forM_ takes place. Also, I would not recommend interleaving IO values into the HTML, because the HTML is pure -- you usually don't want to mix that with IO. I'd rather write something like: numbers <- getFromDataBase return $ ul $ forM_ numbers (li . string . show) Or it could be possible to store your templates in a separate modules, and they would all have a pure type signature: templateX :: Arg1 -> Arg2 -> Html a I think this would be the preferred way, because it allows a separation between the pure View code and the IO-interleaved Controller code in a web application. Kind regards, Jasper Van der Jeugt On Thu, May 27, 2010 at 8:31 PM, Matt Parker <moonmaster9000(a)gmail.com> wrote: > will it be possible to easily interleave IO values into the HTML? like > instead of the [1,2,3] > > ul $ forM_ [1, 2, 3] (li . string . show) > > what if it was a function that returned IO [1,2,3] (maybe 1,2,3 came out of > a database). will the forM_ handle that OK? > > > On Thu, May 27, 2010 at 12:19 PM, Matt Parker <moonmaster9000(a)gmail.com> > wrote: >> >> Blaze HTML looks wonderful to me (especially with the do notation), and >> better then Hamlet for one reason: writing idiomatic haskell for stuff like >> this: >> >> ul $ forM_ [1, 2, 3] (li . string . show) >> >> instead of the Hamlet style: >> >> %ul >> $forall children.person child >> %li $child$ >> >> which, i don't know, for some reason, made me feel like i was back in PHP >> land. >> >> -matt >> >> >> On Thu, May 27, 2010 at 7:08 AM, Alberto G. Corona <agocorona(a)gmail.com> >> wrote: >>> >>> As a user, I have too many HTML generators, a few of them with Ajax and >>> none with server-side event handling (like ASPX or JSPX). Ajax is >>> complicated but server side event handling is what I really miss because it >>> is simple from the user point of view, my ervents could be handled in >>> haskell code rather than in javaScript and I implicitly could use the >>> advantages of dinamic HTML and Ajax without the need to know them at all. >>> Imagine a dynamic Web application with 100% haskell code made with >>> dynamic widgets created by third party developers. >>> So, anyone want to create a HTML templating system with server side event >>> handling? It is not terribly hard to do. (I refer to ASP.NET documentation >>> or the JavaServer Faces framework). >>> By the way, I vote for XML templating or else, combinator templating that >>> produce XHML templating because it can be handled by a future graphical IDE. >>> >>> 2010/5/27 Jasper Van der Jeugt <jaspervdj(a)gmail.com> >>>> >>>> Hey Bas, >>>> >>>> > How about also providing an enumerator back-end? >>>> > >>>> > http://hackage.haskell.org/packages/archive/iteratee/0.3.5/doc/html/Data-It… >>>> > >>>> > Then your library can integrate more easily with the snap framework: >>>> > http://snapframework.com >>>> >>>> Sure, I can do that. But I already tested integration with the snap >>>> framework, the best path here seems to call the `writeLBS` function >>>> from the snap framework on the `L.ByteString` that BlazeHtml produces >>>> (`writeLBS` internally uses an enumerator). >>>> >>>> Kind regards, >>>> Jasper Van der Jeugt >>>> >>>> On Thu, May 27, 2010 at 10:38 AM, Bas van Dijk <v.dijk.bas(a)gmail.com> >>>> wrote: >>>> > Q14: Do you see any problems with respect to integrating BlazeHtml in >>>> > your favourite web-framework/server? >>>> > >>>> > How about also providing an enumerator back-end? >>>> > >>>> > http://hackage.haskell.org/packages/archive/iteratee/0.3.5/doc/html/Data-It… >>>> > >>>> > Then your library can integrate more easily with the snap framework: >>>> > http://snapframework.com >>>> > >>>> > Regards, >>>> > >>>> > Bas >>>> > >>>> _______________________________________________ >>>> Haskell-Cafe mailing list >>>> Haskell-Cafe(a)haskell.org >>>> http://www.haskell.org/mailman/listinfo/haskell-cafe >>> >>> >>> _______________________________________________ >>> web-devel mailing list >>> web-devel(a)haskell.org >>> http://www.haskell.org/mailman/listinfo/web-devel >>> >> > >

1 0

Happy GLR parser creation problem
by Paul Jensen 02 Jun '10

02 Jun '10

[Apologies if this is the wrong place to ask -- there appears to be no Happy-specific list.] I'm interested in experimenting with using a GLR parser to parse some data that could be interpreted in different ways. I'm trying to use Happy's GLR mode but I am encountering an error building one of the examples (specifically http://darcs.haskell.org/happy/examples/glr/expr-eval/Expr.y) [Mac OS X 10.6.3] rhoda:expr-eval psj$ ghc --version The Glorious Glasgow Haskell Compilation System, version 6.12.1 rhoda:expr-eval psj$ happy --version Happy Version 1.18.4 Copyright (c) 1993-1996 Andy Gill, Simon Marlow (c) 1997-2005 Simon Marlow rhoda:expr-eval psj$ happy --glr --ghc Expr.y shift/reduce conflicts: 9 happy: src/ProduceGLRCode.lhs:190:11-76: Irrefutable pattern failed for pattern (pre, drop_me : post) It's the same error someone else encountered last year, described at http://www.haskell.org/pipermail/haskell-cafe/2009-July/063770.html . Does anyone have any insights into this? If Happy's GLR support isn't working just now, are there any other parser generators that produce GLR parsers that anyone would recommend having a look at? Many thanks, psj

2 1

Re: [web-devel] Google Summer of Code: BlazeHTML RFC
by Matt Parker 02 Jun '10

02 Jun '10

> Q3: Which HTML version would you preferably use? > HTML 5. google summer of code should be about pushing the new and exciting.

7 8

Resource module
by Arie Peterson 02 Jun '10

02 Jun '10

Dear list, For some time, I have maintained a small private module centred around the following type of "resource": > newtype Resource cap m = Resource { with :: forall a. (cap -> m a) -> m a } Interpretation: @Resource cap m@ is a resource, providing a "capability" of type @cap@, which does administration (like opening and closing handles) in the monad @m@. I would like to use a standard module instead of this howe-grown one, so I can publish things that depend on it. Does anyone know a package that provides this? Things I tried: - The type is suspiciously similar to the continuation monad transformer; in fact @Resource cap m@ is isomorphic to @forall t. ContT t m cap@. However, I can't use this latter type directly, because I would like functions on Resources like the instance > instance (Monoid cap) => Monoid (Resource cap m) , which is not possible without a newtype wrapper. - I tried using the "regions" package, since it has a very similar purpose, but this seems impossible: some resources I could not express in the form required for its class 'Resource' (methods 'open' and 'close'). If something like this is not yet around, I'll upload my version, but I'd like to reuse, if possible. Regards, Arie

2 6

ANN: threads-0.2
by Bas van Dijk 02 Jun '10

02 Jun '10

Hello, I just released threads-0.2. The library which lets you fork threads and wait for their result. INSTALL: $ cabal update $ cabal install threads API DOCS: http://hackage.haskell.org/package/threads-0.2 DEVELOPMENT: darcs get http://code.haskell.org/~basvandijk/code/threads CHANGES: * I removed the regular 'Control.Concurrent.ThreadId' out of our own 'ThreadId α' type and renamed the type to 'Result α': data ThreadId α = ThreadId { result ∷ TMVar (Either SomeException α) , threadId ∷ Control.Concurrent.ThreadId } to: newtype Result α = Result { unResult ∷ TMVar (Either SomeException α) } * The various fork functions now return the regular 'ThreadId' paired with a 'Result α' value: forkIO ∷ IO α → IO (ThreadId α) to: forkIO ∷ IO α → IO (ThreadId, Result α) * The various wait functions now just receive a 'Result α' value: wait ∷ ThreadId α → IO (Either SomeException α) to: wait ∷ Result α → IO (Either SomeException α) * I removed: throwTo ∷ Exception e ⇒ ThreadId α → e → IO () and: killThread ∷ ThreadId α → IO () because they aren't needed anymore since we have direct access to the regular 'ThreadId'. * When you build the package with GHC the GHC specific 'forkOnIO' function is additionally build in: forkOnIO ∷ Int → IO α → IO (ThreadId, Result α) * For the rest I added documentation and made some internal changes like a simplified implementation of ThreadGroups and made small stylistic changes all over the place. FUTURE PLANS: Inspired by the 'parIO' combinator from the 'async' package, I'm thinking about adding a module which offers similar functionality: data ParGroup α new ∷ IO (ParGroup α) forkIO ∷ ParGroup α → IO α → IO ThreadId wait ∷ ParGroup α → IO (Maybe α) The idea is that when you have multiple IO computations, that each compute the same answer but differ in how fast they compute that answer, you can compute them in parallel using 'forkIO'. Finally 'wait' blocks until one of the threads has computed an answer and will return Just that answer. When all of the threads fail to compute an answer (throw an exception) wait returns Nothing. Regards, Bas

1 0

TagSoup 0.9
by Ralph Hodgson 02 Jun '10

02 Jun '10

Hello Neil , I was using TagSoup 0.8 with great success. On upgrading to 0.9 I have this error: TQ\TagSoup\TagSoupExtensions.lhs:29:17: `Tag' is not applied to enough type arguments Expected kind `*', but `Tag' has kind `* -> *' In the type synonym declaration for `Bundle' Failed, modules loaded: TQ.Common.TextAndListHandling. where line 29 is the type declaration for 'bundle' in the following code: > module TQ.TagSoup.TagSoupExtensions where > import TQ.Common.TextAndListHandling > import Text.HTML.TagSoup > import Text.HTML.Download > import Control.Monad > import Data.List > import Data.Char > type Bundle = [Tag] [snip] > tagsOnPage :: String -> IO(String) > tagsOnPage url = do > tags <- liftM parseTags $ openURL url > let results = unlines $ map(show) $ tags > return (results) > extractTags :: Tag -> Tag -> [Tag] -> [Tag] > extractTags fromTag toTag tags = takeWhile (~/= toTag ) $ dropWhile (~/= fromTag ) tags > extractTagsBetween :: Tag -> [Tag] -> [Tag] > extractTagsBetween _ [] = [] > extractTagsBetween markerTag tags = if startTags == [] > then [] > else [head startTags] ++ (takeWhile (~/= markerTag ) $ tail startTags) > where > startTags = dropWhile (~/= markerTag ) tags I need to repair this code quickly. I am hoping you can quickly help me resolve this. Thanks. Ralph Hodgson, @ralphtq <http://twitter.com/ralphtq>

7 22

Re: The state of Hackage: what are we doing about it?
by Bas van Dijk 01 Jun '10

01 Jun '10

On Tue, Jun 1, 2010 at 7:09 PM, Matthias Kilian <kili(a)outback.escape.de> wrote: > - "Liveness" of a library, that is: does it still get updates? Does it > build with recent versions of GHC? Note that Hackage already shows the upload date and for which versions of GHC the package does and doesn't build. > - Reverse dependencies -- how many other libraries and/or programs are > using this library (Maybe something for Hackage-2.0)? My brother Roel created a patch for this. See ticket #576: http://hackage.haskell.org/trac/hackage/ticket/576 Also see our mirror Hackage that implements that patch: http://bifunctor.homelinux.net/~roel/hackage/packages/hackage.html The server automatically synchronizes with Hackage daily. It would be great to have this in the regular Hackage however! Regards, Bas

1 0

Re: Language Shootout reverse-complement benchmark
by Isaac Gouy 01 Jun '10

01 Jun '10

On Tue, Jun 1, 2010 at 10:25 AM, David Leimbach <leimy2k <at> gmail.com> wrote: > I'm still trying to figure out what the point of the shootout really is. >From one point of view - http://shootout.alioth.debian.org/help.php#why > If there's no dedicated folks working with a language there, trying to > make things run faster, a language will come out looking inefficient > potentially. If there isn't even one effective programmer willing to contribute tiny programs ... > There's a lot of compile flags and optimizations that can make a > difference in probably all of the languages listed on that page. The compile flags and optimizations used are shown below the program source code.

1 0

Pugs with GHC 6.12
by Simon Thompson 01 Jun '10

01 Jun '10

Has anyone successfully compiled Pugs with GHC 6.12? The Pugs page suggests that it compiles, but it falls to link (and therefore build) when I try it. Thanks! Simon T.

3 2

Haskell + CouchDB, anyone?
by Clint Moore 01 Jun '10

01 Jun '10

Hi there, I'm trying to use any version of couchdb with Haskell and Database.CouchDB and I'm not having a ton of luck. I've talked to others in the past that have used it without a problem so I'm wondering what it is that I'm doing wrong. I can connect, and create new databases without a problem - do any of you couchdb-using haskell people see a problem with this line? runCouchDB' $ newDoc (db "testify") $ JSObject $ toJSObject [("name",JSString $ toJSString $ "value")] I figured I would start small and just try to put something basic into a document. The rest of this email is also at a gist at http://gist.github.com/417442 Here is a list of the revisions of couchdb and the different errors I'm getting back as well as some of the code I'm using to try to test this.. Hopefully it will help. {- CouchDB 0.11.0: [error] [<0.95.0>] attempted upload of invalid JSON undefined [info] [<0.95.0>] 127.0.0.1 - - 'PUT' /honks/cmoore? 400 CouchDB Trunk as of 5/28 - 10:26 PST: [info] [<0.95.0>] 127.0.0.1 - - 'PUT' /honks? 201 [debug] [<0.114.0>] 'PUT' /honks/cmoore? {1,1} Headers: [{'Connection',"keep-alive"}, {'Content-Type',"application/json"}, {'Host',"localhost:5984"}, {'User-Agent',"hs-HTTP-4000.0.5"}] [debug] [<0.114.0>] OAuth Params: [] [error] [<0.114.0>] attempted upload of invalid JSON undefined [info] [<0.114.0>] 127.0.0.1 - - 'PUT' /honks/cmoore? 400 [debug] [<0.114.0>] httpd 400 error response: {"error":"bad_request","reason":"invalid UTF-8 JSON"} CouchDB rev 818083: [debug] [<0.107.0>] 'PUT' /honks/cmoore? {1,1} Headers: [{'Connection',"keep-alive"}, {'Content-Type',"application/json"}, {'Host',"localhost:5984"}, {'User-Agent',"hs-HTTP-4000.0.5"}] [debug] [<0.107.0>] OAuth Params: [] [debug] [<0.107.0>] Minor error in HTTP request: {invalid_json,undefined} [debug] [<0.107.0>] Stacktrace: [{mochijson2,decode,1}, {couch_httpd_db,db_doc_req,3}, {couch_httpd_db,do_db_req,2}, {couch_httpd,handle_request,5}, {mochiweb_http,headers,5}, {proc_lib,init_p_do_apply,3}] [info] [<0.107.0>] 127.0.0.1 - - 'PUT' /honks/cmoore? 500 [debug] [<0.107.0>] httpd 500 error response: {"error":"invalid_json","reason":"undefined"} CouchDB 0.10.1: [error] [<0.278.0>] attempted upload of invalid JSON undefined [info] [<0.278.0>] 127.0.0.1 - - 'POST' /testify? 400 -} module Main where import Text.JSON import Database.CouchDB import Database.CouchDB.JSON data Person = Person { name :: String, email :: String } deriving (Eq,Show) instance JSON Person where showJSON (Person nm em) = JSObject $ toJSObject [("name", showJSON nm),("email", showJSON em)] readJSON val = do obj <- jsonObject val nm <- jsonField "name" obj em <- jsonField "email" obj return (Person nm em) create = runCouchDB "localhost" 5984 $ createDB "honks" main :: IO () main = do _ <- runCouchDB "localhost" 5984 $ newNamedDoc (db "honks") (doc "cmoore") $ Person "cmoore" "cmoore(a)w.com" putStrLn "Honk." --- Later, against 0.10.1: cmoore@cathedral:~/tmp$ ghci GHCi, version 6.12.2: http://www.haskell.org/ghc/ :? for help Loading package ghc-prim ... linking ... done. Loading package integer-gmp ... linking ... done. Loading package base ... linking ... done. Loading package ffi-1.0 ... linking ... done. Prelude> :m +Database.CouchDB Prelude Database.CouchDB> :m +Text.JSON Prelude Database.CouchDB Text.JSON> :set prompt "> " > runCouchDB' $ createDB "testify" Loading package array-0.3.0.0 ... linking ... done. Loading package syb-0.1.0.2 ... linking ... done. Loading package base-3.0.3.2 ... linking ... done. Loading package bytestring-0.9.1.6 ... linking ... done. Loading package mtl-1.1.0.2 ... linking ... done. Loading package parsec-2.1.0.1 ... linking ... done. Loading package network-2.2.1.5 ... linking ... done. Loading package old-locale-1.0.0.2 ... linking ... done. Loading package old-time-1.0.0.4 ... linking ... done. Loading package HTTP-4000.0.8 ... linking ... done. Loading package containers-0.3.0.0 ... linking ... done. Loading package pretty-1.0.1.1 ... linking ... done. Loading package json-0.4.4 ... linking ... done. Loading package CouchDB-0.10.2 ... linking ... done. > runCouchDB' $ newDoc (db "testify") $ JSObject $ toJSObject [("name",JSString $ toJSString $ "value")] *** Exception: HTTP/1.1 400 Bad Request Server: CouchDB/0.10.1 (Erlang OTP/R13B) Date: Fri, 28 May 2010 18:46:36 GMT Content-Type: text/plain;charset=utf-8 Content-Length: 54 Cache-Control: must-revalidate

2 1