- web-devel - Haskell.org

GGHandler and MonadPeelIO
by Aur Saraf 08 Apr '11

08 Apr '11

Hello, ----------------->>>>>>>>>><<<<<<<<<-------------------- runDB: need liftIOHandler ⁋For the impatient: stick liftIOHandler at the beginning of runDB like so: runDB db = liftIOHandler $ fmap connPool getYesod >>= Settings.runConnectionPool dbNo comments Add a commentName: ⁋For everyone else: WAI 0.2 contained the request body as an Enumerator in the Request datatype, and an Application was Request -> IO Response. In WAI 0.3, an Application is Request -> Iteratee ByteString IO Response. Long story short: all code in your Handler function now lives on top of an Iteratee monad so that it can access the request body.No comments Add a commentName: ⁋This works great most of the time. The only downside is when you need to deal with exceptions. It's impossible to define a MonadPeelIO instance for Iteratee. Therefore, in Yesod 0.7, we have a new datatype called GGHandler, which is just a generalization of GHandler to allow an arbitrary inner monad.No comments Add a commentName: ⁋GHandler defaults to having an Iteratee on the inside. But when you need a MonadPeelIO (like we do in Persistent), you need to have an IO on the inside. Eventually though, you'll need to convert your GGHandler IO to a GHandler. That's what liftIOHandler does.No comments ----------------->>>>>>>>>><<<<<<<<<-------------------- Could anyone please explain this better, and what it means to someone upgrading an application? Specifically, what happens to: -- | A useful synonym; most of the DB handler functions are of this type type DB = SqlPersist Handler ? Thanks :-), Aur

2 1

Passing a MongoDB connection through to Yesod
by Xavier Shay 08 Apr '11

08 Apr '11

Hello, I am having trouble wiring up Persistent.MongoDB into a Yesod app - I can't quite wrap my head around how it fits together. I have this test file: https://gist.github.com/909264 I know withMongoDBConn wiring is incorrect, but I can't find a good example to base it off. Also not sure what the correct `select` call is to use - Database.MongoDB or Database.Persist.MongoDB. Cheers, Xavier PS I'm using: https://github.com/gregwebs/persistent/tree/MongoDB5

2 1

Problem installing hamlet 0.7.2 series
by Aur Saraf 06 Apr '11

06 Apr '11

This happens with hamlet 0.7.2*: [sonoflilit@localhost fr]$ cabal install hamlet-0.7.2 Resolving dependencies... Downloading hamlet-0.7.2... Configuring hamlet-0.7.2... Preprocessing library hamlet-0.7.2... Preprocessing executables for hamlet-0.7.2... Building hamlet-0.7.2... [1 of 9] Compiling Text.MkSizeType ( Text/MkSizeType.hs, dist/build/Text/MkSizeType.o ) Text/MkSizeType.hs:44:8: Warning: This binding for `show' shadows the existing binding imported from Prelude [2 of 9] Compiling Text.Shakespeare ( Text/Shakespeare.hs, dist/build/Text/Shakespeare.o ) [3 of 9] Compiling Text.Hamlet.Parse ( Text/Hamlet/Parse.hs, dist/build/Text/Hamlet/Parse.o ) [4 of 9] Compiling Text.Hamlet.Quasi ( Text/Hamlet/Quasi.hs, dist/build/Text/Hamlet/Quasi.o ) Text/Hamlet/Quasi.hs:34:43: Warning: In the use of `string' (imported from Text.Blaze, but defined in Text.Blaze.Internal): Deprecated: "Use Blaze.Html.toHtml" Text/Hamlet/Quasi.hs:129:4: Warning: Fields of `QuasiQuoter' not initialised: quotePat In the expression: QuasiQuoter {quoteExp = hamletFromString set} In the definition of `hamletWithSettings': hamletWithSettings set = QuasiQuoter {quoteExp = hamletFromString set} Text/Hamlet/Quasi.hs:138:4: Warning: Fields of `QuasiQuoter' not initialised: quotePat In the expression: QuasiQuoter {quoteExp = \ s -> do { x <- hamletFromString set s; .... }} In the definition of `hamletWithSettings'': hamletWithSettings' set = QuasiQuoter {quoteExp = \ s -> do { ... }} [5 of 9] Compiling Text.Cassius ( Text/Cassius.hs, dist/build/Text/Cassius.o ) Loading package ghc-prim ... linking ... done. Loading package integer-gmp ... linking ... done. Loading package base ... linking ... done. Loading package syb-0.1.0.2 ... linking ... done. Loading package base-3.0.3.2 ... linking ... done. Loading package parsec-2.1.0.1 ... linking ... done. Loading package bytestring-0.9.1.7 ... linking ... done. Loading package array-0.3.0.1 ... linking ... done. Loading package containers-0.3.0.0 ... linking ... done. Loading package deepseq-1.1.0.2 ... linking ... done. Loading package text-0.11.0.6 ... linking ... done. Loading package blaze-builder-0.2.1.4 ... linking ... done. Loading package transformers-0.2.2.0 ... linking ... done. Loading package enumerator-0.4.9.1 ... linking ... done. Loading package blaze-builder-enumerator-0.2.0.2 ... linking ... done. Loading package json-types-0.1 ... linking ... done. Loading package json-enumerator-0.0.1.1 ... linking ... done. Loading package failure-0.1.0.1 ... linking ... done. Loading package blaze-html-0.4.1.0 ... linking ... done. Loading package extensible-exceptions-0.1.1.1 ... linking ... done. Loading package filepath-1.1.0.4 ... linking ... done. Loading package old-locale-1.0.0.2 ... linking ... done. Loading package old-time-1.0.0.5 ... linking ... done. Loading package unix-2.4.0.2 ... linking ... done. Loading package directory-1.0.1.1 ... linking ... done. Loading package pretty-1.0.1.1 ... linking ... done. Loading package process-1.0.1.3 ... linking ... done. Loading package Cabal-1.8.0.6 ... linking ... done. Loading package ghc-binary-0.5.0.2 ... linking ... done. Loading package bin-package-db-0.0.0.0 ... linking ... done. Loading package hpc-0.5.0.5 ... linking ... done. Loading package template-haskell ... linking ... done. Loading package ghc-6.12.3 ... linking ... done. Loading package mtl-2.0.1.0 ... linking ... done. Loading package time-1.1.4 ... linking ... done. Loading package random-1.0.0.2 ... linking ... done. Loading package QuickCheck-2.4.0.1 ... linking ... done. Loading package ffi-1.0 ... linking ... done. Text/Cassius.hs:202:4: Warning: This binding for `parseComment' shadows the existing binding defined at Text/Cassius.hs:148:0 Text/Cassius.hs:213:10: Warning: Fields of `QuasiQuoter' not initialised: quotePat In the expression: QuasiQuoter {quoteExp = cassiusFromString} In the definition of `cassius': cassius = QuasiQuoter {quoteExp = cassiusFromString} Text/Cassius.hs:319:16: Warning: Pattern match(es) are non-exhaustive In a case alternative: Patterns not matched: [] (ghc-prim:GHC.Types.C# #x) : _ with #x `notElem` ['c', 'e', 'i', 'm', 'p', '%'] [ghc-prim:GHC.Types.C# 'c'] (ghc-prim:GHC.Types.C# 'c') : ((ghc-prim:GHC.Types.C# #x) : _) with #x `notElem` ['m'] ... [6 of 9] Compiling Text.Julius ( Text/Julius.hs, dist/build/Text/Julius.o ) Text/Julius.hs:100:9: Warning: Fields of `QuasiQuoter' not initialised: quotePat In the expression: QuasiQuoter {quoteExp = juliusFromString} In the definition of `julius': julius = QuasiQuoter {quoteExp = juliusFromString} [7 of 9] Compiling Text.Hamlet.RT ( Text/Hamlet/RT.hs, dist/build/Text/Hamlet/RT.o ) [8 of 9] Compiling Text.Hamlet.Debug ( Text/Hamlet/Debug.hs, dist/build/Text/Hamlet/Debug.o ) [9 of 9] Compiling Text.Hamlet ( Text/Hamlet.hs, dist/build/Text/Hamlet.o ) [1 of 9] Compiling Text.MkSizeType ( Text/MkSizeType.hs, dist/build/Text/MkSizeType.p_o ) Text/MkSizeType.hs:44:8: Warning: This binding for `show' shadows the existing binding imported from Prelude [2 of 9] Compiling Text.Shakespeare ( Text/Shakespeare.hs, dist/build/Text/Shakespeare.p_o ) [3 of 9] Compiling Text.Hamlet.Parse ( Text/Hamlet/Parse.hs, dist/build/Text/Hamlet/Parse.p_o ) [4 of 9] Compiling Text.Hamlet.Quasi ( Text/Hamlet/Quasi.hs, dist/build/Text/Hamlet/Quasi.p_o ) Text/Hamlet/Quasi.hs:34:43: Warning: In the use of `string' (imported from Text.Blaze, but defined in Text.Blaze.Internal): Deprecated: "Use Blaze.Html.toHtml" Text/Hamlet/Quasi.hs:129:4: Warning: Fields of `QuasiQuoter' not initialised: quotePat In the expression: QuasiQuoter {quoteExp = hamletFromString set} In the definition of `hamletWithSettings': hamletWithSettings set = QuasiQuoter {quoteExp = hamletFromString set} Text/Hamlet/Quasi.hs:138:4: Warning: Fields of `QuasiQuoter' not initialised: quotePat In the expression: QuasiQuoter {quoteExp = \ s -> do { x <- hamletFromString set s; .... }} In the definition of `hamletWithSettings'': hamletWithSettings' set = QuasiQuoter {quoteExp = \ s -> do { ... }} [5 of 9] Compiling Text.Cassius ( Text/Cassius.hs, dist/build/Text/Cassius.p_o ) Loading package ghc-prim ... linking ... done. Loading package integer-gmp ... linking ... done. Loading package base ... linking ... done. Loading package syb-0.1.0.2 ... linking ... done. Loading package base-3.0.3.2 ... linking ... done. Loading package parsec-2.1.0.1 ... linking ... done. Loading package bytestring-0.9.1.7 ... linking ... done. Loading package array-0.3.0.1 ... linking ... done. Loading package containers-0.3.0.0 ... linking ... done. Loading package deepseq-1.1.0.2 ... linking ... done. Loading package text-0.11.0.6 ... linking ... done. Loading package blaze-builder-0.2.1.4 ... linking ... done. Loading package transformers-0.2.2.0 ... linking ... done. Loading package enumerator-0.4.9.1 ... linking ... done. Loading package blaze-builder-enumerator-0.2.0.2 ... linking ... done. Loading package json-types-0.1 ... linking ... done. Loading package json-enumerator-0.0.1.1 ... linking ... done. Loading package failure-0.1.0.1 ... linking ... done. Loading package blaze-html-0.4.1.0 ... linking ... done. Loading package extensible-exceptions-0.1.1.1 ... linking ... done. Loading package filepath-1.1.0.4 ... linking ... done. Loading package old-locale-1.0.0.2 ... linking ... done. Loading package old-time-1.0.0.5 ... linking ... done. Loading package unix-2.4.0.2 ... linking ... done. Loading package directory-1.0.1.1 ... linking ... done. Loading package pretty-1.0.1.1 ... linking ... done. Loading package process-1.0.1.3 ... linking ... done. Loading package Cabal-1.8.0.6 ... linking ... done. Loading package ghc-binary-0.5.0.2 ... linking ... done. Loading package bin-package-db-0.0.0.0 ... linking ... done. Loading package hpc-0.5.0.5 ... linking ... done. Loading package template-haskell ... linking ... done. Loading package ghc-6.12.3 ... linking ... done. Loading package mtl-2.0.1.0 ... linking ... done. Loading package time-1.1.4 ... linking ... done. Loading package random-1.0.0.2 ... linking ... done. Loading package QuickCheck-2.4.0.1 ... linking ... done. Loading package ffi-1.0 ... linking ... done. Text/Cassius.hs:202:4: Warning: This binding for `parseComment' shadows the existing binding defined at Text/Cassius.hs:148:0 Text/Cassius.hs:213:10: Warning: Fields of `QuasiQuoter' not initialised: quotePat In the expression: QuasiQuoter {quoteExp = cassiusFromString} In the definition of `cassius': cassius = QuasiQuoter {quoteExp = cassiusFromString} Text/Cassius.hs:319:16: Warning: Pattern match(es) are non-exhaustive In a case alternative: Patterns not matched: [] (ghc-prim:GHC.Types.C# #x) : _ with #x `notElem` ['c', 'e', 'i', 'm', 'p', '%'] [ghc-prim:GHC.Types.C# 'c'] (ghc-prim:GHC.Types.C# 'c') : ((ghc-prim:GHC.Types.C# #x) : _) with #x `notElem` ['m'] ... [6 of 9] Compiling Text.Julius ( Text/Julius.hs, dist/build/Text/Julius.p_o ) Text/Julius.hs:100:9: Warning: Fields of `QuasiQuoter' not initialised: quotePat In the expression: QuasiQuoter {quoteExp = juliusFromString} In the definition of `julius': julius = QuasiQuoter {quoteExp = juliusFromString} [7 of 9] Compiling Text.Hamlet.RT ( Text/Hamlet/RT.hs, dist/build/Text/Hamlet/RT.p_o ) [8 of 9] Compiling Text.Hamlet.Debug ( Text/Hamlet/Debug.hs, dist/build/Text/Hamlet/Debug.p_o ) [9 of 9] Compiling Text.Hamlet ( Text/Hamlet.hs, dist/build/Text/Hamlet.p_o ) Registering hamlet-0.7.2... [ 1 of 11] Compiling Text.Shakespeare ( Text/Shakespeare.hs, dist/build/hamlet6to7/hamlet6to7-tmp/Text/Shakespeare.p_o ) [ 2 of 11] Compiling Old.Utf8 ( Old/Utf8.hs, dist/build/hamlet6to7/hamlet6to7-tmp/Old/Utf8.p_o ) [ 3 of 11] Compiling Text.MkSizeType ( Text/MkSizeType.hs, dist/build/hamlet6to7/hamlet6to7-tmp/Text/MkSizeType.p_o ) [ 4 of 11] Compiling Text.Hamlet.Parse ( Text/Hamlet/Parse.hs, dist/build/hamlet6to7/hamlet6to7-tmp/Text/Hamlet/Parse.p_o ) [ 5 of 11] Compiling Text.Hamlet.Quasi ( Text/Hamlet/Quasi.hs, dist/build/hamlet6to7/hamlet6to7-tmp/Text/Hamlet/Quasi.p_o ) Text/Hamlet/Quasi.hs:34:43: Warning: In the use of `string' (imported from Text.Blaze, but defined in Text.Blaze.Internal): Deprecated: "Use Blaze.Html.toHtml" Text/Hamlet/Quasi.hs:129:4: Warning: Fields of `QuasiQuoter' not initialised: quotePat In the expression: QuasiQuoter {quoteExp = hamletFromString set} In the definition of `hamletWithSettings': hamletWithSettings set = QuasiQuoter {quoteExp = hamletFromString set} Text/Hamlet/Quasi.hs:138:4: Warning: Fields of `QuasiQuoter' not initialised: quotePat In the expression: QuasiQuoter {quoteExp = \ s -> do { x <- hamletFromString set s; .... }} In the definition of `hamletWithSettings'': hamletWithSettings' set = QuasiQuoter {quoteExp = \ s -> do { ... }} [ 6 of 11] Compiling Text.Cassius ( Text/Cassius.hs, dist/build/hamlet6to7/hamlet6to7-tmp/Text/Cassius.p_o ) Loading package ghc-prim ... linking ... done. Loading package integer-gmp ... linking ... done. Loading package base ... linking ... done. Loading package syb-0.1.0.2 ... linking ... done. Loading package base-3.0.3.2 ... linking ... done. Loading package parsec-2.1.0.1 ... linking ... done. Loading package bytestring-0.9.1.7 ... linking ... done. Loading package array-0.3.0.1 ... linking ... done. Loading package containers-0.3.0.0 ... linking ... done. Loading package deepseq-1.1.0.2 ... linking ... done. Loading package text-0.11.0.6 ... linking ... done. Loading package blaze-builder-0.2.1.4 ... linking ... done. Loading package transformers-0.2.2.0 ... linking ... done. Loading package enumerator-0.4.9.1 ... linking ... done. Loading package blaze-builder-enumerator-0.2.0.2 ... linking ... done. Loading package json-types-0.1 ... linking ... done. Loading package json-enumerator-0.0.1.1 ... linking ... done. Loading package failure-0.1.0.1 ... linking ... done. Loading package blaze-html-0.4.1.0 ... linking ... done. Loading package extensible-exceptions-0.1.1.1 ... linking ... done. Loading package filepath-1.1.0.4 ... linking ... done. Loading package old-locale-1.0.0.2 ... linking ... done. Loading package old-time-1.0.0.5 ... linking ... done. Loading package unix-2.4.0.2 ... linking ... done. Loading package directory-1.0.1.1 ... linking ... done. Loading package pretty-1.0.1.1 ... linking ... done. Loading package process-1.0.1.3 ... linking ... done. Loading package Cabal-1.8.0.6 ... linking ... done. Loading package ghc-binary-0.5.0.2 ... linking ... done. Loading package bin-package-db-0.0.0.0 ... linking ... done. Loading package hpc-0.5.0.5 ... linking ... done. Loading package template-haskell ... linking ... done. Loading package ghc-6.12.3 ... linking ... done. Loading package mtl-2.0.1.0 ... linking ... done. Loading package time-1.1.4 ... linking ... done. Loading package random-1.0.0.2 ... linking ... done. Loading package QuickCheck-2.4.0.1 ... linking ... done. Loading package ffi-1.0 ... linking ... done. Text/Cassius.hs:1:0: cannot find normal object file `dist/build/hamlet6to7/hamlet6to7-tmp/Text/MkSizeType.o' while linking an interpreted expression cabal: Error: some packages failed to install: hamlet-0.7.2 failed during the building phase. The exception was: ExitFailure 1 I need this to be able to convert my app to 0.7... Can anyone help? Thanks, Aur

3 3

ANN: Hakyll 3.1
by Jasper Van der Jeugt 06 Apr '11

06 Apr '11

Hello all, I've just uploaded Hakyll 3.1.0.0 [1] to Hackage. It contains some backwards-incompatible changes (but they are really for the greater good). The most important change is the matching done in the rules DSL. What used to be written as: route "posts/*" someRoute compile "posts/*" someCompiler Is now written as: match "posts/*" $ do route someRoute compile someCompiler This new style is way more expressive, as you can define custom predicates such as: match (predicate (\i -> matches "foo/*" i && not (matches "foo/bar" i))) $ do ... For more information, see the Pattern documentation [2]. Another slight change is that the `Hakyll.Web.Tags` [3] module now uses an association list instead of a `Data.Map` -- in case you want to change the order in which tags are displayed. If you have any questions or trouble migrating, feel free to drop a mail to me or here [4] or visit us in #hakyll on Freenode. [1]: http://hackage.haskell.org/package/hakyll [2]: http://jaspervdj.be/hakyll/reference/Hakyll-Core-Identifier-Pattern.html [3]: http://jaspervdj.be/hakyll/reference/Hakyll-Web-Tags.html [4]: https://groups.google.com/forum/#!forum/hakyll Cheers, Jasper

1 0

Quitting Yesod
by Aur Saraf 06 Apr '11

06 Apr '11

Hello, I'm trying to profile a yesod app, and it seems like Haskell's profiler only writes to file when the program exists. Gracefully. So Ctrl+C twice is out of the question. What I'm doing: $ ~/Downloads/spawn-fcgi-1.6.3/src/spawn-fcgi -n -s /tmp/f.socket -- ~/projects/f/dist/build/fastcgi/fastcgi +RTS -p Then I enter a page that I want to profile. Then my system starts swapping because the page takes 100% RAM (I'm suspecting a space leak). Then I hit Ctrl+C Ctrl+C to quit the yesod program. Then I look at fastcgi.prof, and find it empty. How can I get profiling info to be written to file? Am I wrong in something I'm doing? Thanks, Aur

3 3

Re: [web-devel] [Haskell-beginners] happstack versus yesod
by Michael Snoyman 06 Apr '11

06 Apr '11

This probably isn't the right forum for expressing this thought, but here goes anyway: What exactly is the freedom that Yesod takes away from you? I know this idea gets passed around a lot: Yesod isn't "really" Haskell, Yesod is its own language, etc. But all of that is really just talking about some sugar that Yesod puts in place for you: under the surface, it's just functions, data types and type classes. You can go ahead and write a complete Yesod application without using Template Haskell of QuasiQuotation. So what's the restriction that Yesod has that Happstack/Snap don't? Yesod requires you to provide some datatypes. That's it. What's the purpose of the datatypes? Mostly, it's about type-safe URLs. I'm not exactly sure why the term "restriction" or "freedom" even applies here: Yesod has a *feature* that Happstack/Snap do not include out of the box. Let me phrase this in a different way: "But I don't like Haskell, because the idea of Python in more interesting for me. Python gives you some freedom." I really do see this as the same argument. You can go ahead and right some code without static type checking in Python, and you might even write it faster than you would have in Haskell. But go ahead and start making changes, and unless you have some very good unit tests, you're in for a world of hurt. In Haskell, often times the compiler will save you from yourself. The same is true in Yesod. I've refactored incredibly large codebases, with hundreds of different resource patterns, and *never* had to worry about 404s. Maybe some people consider that giving up freedom. I consider it giving up the freedom to shoot yourself in the foot. Oh, and if you are absolutely dependent on the idea of getting rid of this data-type requirements, and want all the freedom of dynamic routes, you can try out a 44 line module[1] that I mentioned recently[2]. OK, enough of this rant. I'm not trying to attack you, I'm just tired of this notion that Yesod is a ball-and-chain around your ankles. People can have legitimate reasons to prefer Happstack or Snap to Yesod, I've never claimed that it was a perfect framework for everyone. I just don't think this is one of those legitimate reasons. Michael [1] https://gist.github.com/884802 <https://gist.github.com/884802>[2] http://osdir.com/ml/general/2011-04/msg02160.html On Tue, Apr 5, 2011 at 8:35 PM, Alexey G <kreed131(a)gmail.com> wrote: > Hello. It's very interesting theme for me and I want to express my > opinion. > Yesod - very fast web-framework( > http://www.yesodweb.com/blog/preliminary-warp-cross-language-benchmarks) > But I don't like Yesod, because idea of Happstack more interesting for me. > Happstack gives to you some freedom. > > I trying to use Snap Framework now. And it's very good for me. Try it > http://snapframework.com/! > > Sorry for my english. > > 2011/4/5 Ramy Abdel-Azim <ramy.abdelazim(a)gmail.com> > >> I already posted about this a little bit but i think i should let the old >> thread get back to it's regularly scheduled conversaion. on to my question: >> >> So I know there are a million easier ways to createa blog (wordpress, >> django, drupal, etc. etc.) but which would you recommend I use to build a >> beginner website that is simliar to a blog: happstack or yesod? >> _Ramy >> >> _______________________________________________ >> Beginners mailing list >> Beginners(a)haskell.org >> http://www.haskell.org/mailman/listinfo/beginners >> >> > > _______________________________________________ > Beginners mailing list > Beginners(a)haskell.org > http://www.haskell.org/mailman/listinfo/beginners > >

3 3

Yesod CRUD Help
by Francisco Jose CHAVES ALONSO 05 Apr '11

05 Apr '11

Hi In order to know how the CRUD works, I was trying the blog example of the Screencast, updated to yesod 7 but the mkToForm gives me the following error: No instance for (PersistEntity [Database.Persist.Base.EntityDef]) arising from a use of `mkToForm' Possible fix: add an instance declaration for (PersistEntity [Database.Persist.Base.EntityDef]) In the expression: mkToForm In the first argument of `share', namely `[mkToForm, mkPersist, mkMigrate "migrateAll"]' In the expression: share [mkToForm, mkPersist, mkMigrate "migrateAll"] [Database.Persist.Base.EntityDef "Entry" [] [("title", "String", []), ("day", "Day", ["Desc"]), ....] [] []] Any suggestion? Thanks in advance Francisco The code is: {-# LANGUAGE TypeFamilies, QuasiQuotes, GeneralizedNewtypeDeriving, TemplateHaskell, MultiParamTypeClasses #-} import Yesod import Yesod.Helpers.Crud import Database.Persist import Database.Persist.Sqlite import Database.Persist.TH import Data.Time (Day) import Network.Wai.Middleware.Debug (debug) import Network.Wai.Handler.Warp (run) share [mkToForm, mkPersist, mkMigrate "migrateAll"] [persist| Entry title String day Day Desc content Html deriving |] -- Necesary for CRUD instance Item Entry where itemTitle = entryTitle data Blog = Blog { pool :: ConnectionPool } type EntryCrud = Crud Blog Entry mkYesod "Blog" [parseRoutes| / RootR GET /entry/#EntryId EntryR GET -- Subsite /admin AdminR EntryCrud defaultCrud |] instance Yesod Blog where approot _ = "http://localhost:3000" instance YesodPersist Blog where type YesodDB Blog = SqlPersist runDB db = liftIOHandler $ fmap pool getYesod >>= runSqlPool db getRootR = do entries <- runDB $ selectList [] [EntryDayDesc] 0 0 defaultLayout $ do setTitle $ string "Yesod Blog Tutorial Homepage" [hamlet| <h1> Archive <ul> $forall entry <- entries <li> <a href="@{EntryR (fst entry)}">#{entryTitle (snd entry)} <p> <a href="@{AdminR CrudListR}">Admin |] getEntryR entryid = do entry <- runDB $ get404 entryid defaultLayout $ do setTitle $ string $ entryTitle entry [hamlet| <h1> #{entryTitle entry} <h2> #{show (entryDay entry)} #{entryContent entry} |] withBlog :: (Application -> IO a) -> IO a withBlog f = withSqlitePool "blog.db3" 8 $ \pool -> do flip runSqlPool pool $ do (runMigration migrateAll) insert $ Entry "First Entry" (read "2011-04-04") $ preEscapedString "<h3>First Entry</h3>" let h = Blog pool toWaiApp h >>= f main = withBlog $ run 3000 . debug

2 1

Auth plugin usage questions
by Ian Duncan 05 Apr '11

05 Apr '11

I currently have a function of type Widget (), and I need to find get the user's authorization (or not) in order to render the appropriate confidential information. How exactly do I get the authorization information? There's currently not really any documentation for this stuff, so I thought I'd ask and see what the best way to do this is.

2 1

Re: [web-devel] [Haskell-beginners] happstack versus yesod
by Michael Snoyman 05 Apr '11

05 Apr '11

I'm not sure where you got that implication from. Happstack is a good framework, and using it will definitely help you learn Haskell. But the same is true of Yesod or Snap. I guess you might say that if your goal is to learn Haskell, and not to create websites, then you shouldn't be using a framework at all, and should be writing code that reads/sends ByteStrings over sockets. You'll definitely learn more Haskell that way... Michael On Tue, Apr 5, 2011 at 9:50 PM, Ramy Abdel-Azim <ramy.abdelazim(a)gmail.com>wrote: > well, i'll add my $.02 here (hopefully without throwing fuel on the fire) > and say that - without knowing much about either, I asked the question for > pete's sake! - is that from what you're saying, if I'dlike to use one of > these frameworks to not only build my website but to also solidify what > knowledge of haskell i've acquired, I should go with Happstack. > > > On Tue, Apr 5, 2011 at 2:32 PM, Michael Snoyman <michael(a)snoyman.com>wrote: > >> This probably isn't the right forum for expressing this thought, but here >> goes anyway: >> >> What exactly is the freedom that Yesod takes away from you? I know this >> idea gets passed around a lot: Yesod isn't "really" Haskell, Yesod is its >> own language, etc. But all of that is really just talking about some sugar >> that Yesod puts in place for you: under the surface, it's just functions, >> data types and type classes. You can go ahead and write a complete Yesod >> application without using Template Haskell of QuasiQuotation. >> >> So what's the restriction that Yesod has that Happstack/Snap don't? Yesod >> requires you to provide some datatypes. That's it. What's the purpose of the >> datatypes? Mostly, it's about type-safe URLs. I'm not exactly sure why the >> term "restriction" or "freedom" even applies here: Yesod has a *feature* >> that Happstack/Snap do not include out of the box. >> >> Let me phrase this in a different way: "But I don't like Haskell, because >> the idea of Python in more interesting for me. Python gives you some >> freedom." I really do see this as the same argument. You can go ahead and >> right some code without static type checking in Python, and you might even >> write it faster than you would have in Haskell. But go ahead and start >> making changes, and unless you have some very good unit tests, you're in for >> a world of hurt. In Haskell, often times the compiler will save you from >> yourself. >> >> The same is true in Yesod. I've refactored incredibly large codebases, >> with hundreds of different resource patterns, and *never* had to worry about >> 404s. Maybe some people consider that giving up freedom. I consider it >> giving up the freedom to shoot yourself in the foot. >> >> Oh, and if you are absolutely dependent on the idea of getting rid of this >> data-type requirements, and want all the freedom of dynamic routes, you can >> try out a 44 line module[1] that I mentioned recently[2]. >> >> OK, enough of this rant. I'm not trying to attack you, I'm just tired of >> this notion that Yesod is a ball-and-chain around your ankles. People can >> have legitimate reasons to prefer Happstack or Snap to Yesod, I've never >> claimed that it was a perfect framework for everyone. I just don't think >> this is one of those legitimate reasons. >> >> Michael >> >> [1] https://gist.github.com/884802 >> <https://gist.github.com/884802>[2] >> http://osdir.com/ml/general/2011-04/msg02160.html >> >> >> On Tue, Apr 5, 2011 at 8:35 PM, Alexey G <kreed131(a)gmail.com> wrote: >> >>> Hello. It's very interesting theme for me and I want to express my >>> opinion. >>> Yesod - very fast web-framework( >>> http://www.yesodweb.com/blog/preliminary-warp-cross-language-benchmarks >>> ). >>> But I don't like Yesod, because idea of Happstack more interesting for >>> me. Happstack gives to you some freedom. >>> >>> I trying to use Snap Framework now. And it's very good for me. Try it >>> http://snapframework.com/! >>> >>> Sorry for my english. >>> >>> 2011/4/5 Ramy Abdel-Azim <ramy.abdelazim(a)gmail.com> >>> >>>> I already posted about this a little bit but i think i should let the >>>> old thread get back to it's regularly scheduled conversaion. on to my >>>> question: >>>> >>>> So I know there are a million easier ways to createa blog (wordpress, >>>> django, drupal, etc. etc.) but which would you recommend I use to build a >>>> beginner website that is simliar to a blog: happstack or yesod? >>>> _Ramy >>>> >>>> _______________________________________________ >>>> Beginners mailing list >>>> Beginners(a)haskell.org >>>> http://www.haskell.org/mailman/listinfo/beginners >>>> >>>> >>> >>> _______________________________________________ >>> Beginners mailing list >>> Beginners(a)haskell.org >>> http://www.haskell.org/mailman/listinfo/beginners >>> >>> >> >> _______________________________________________ >> Beginners mailing list >> Beginners(a)haskell.org >> http://www.haskell.org/mailman/listinfo/beginners >> >> >

1 0

yesod-dynamic
by Michael Snoyman 04 Apr '11

04 Apr '11

Hi all, There's an inherent difference in how Yesod handles routing, by default, and how Happstack/Snap handle routing, by default. (Apologies if I am mistaken in my understanding of the latter.) Yesod is built in with the concept of type-safe URLs. This forms the basis of routing in Yesod: the list of path segments is first converted to a type-safe URL value, and then the appropriate handler is chosen from that. On the other hand, Happstack/Snap both allow you to perform routing directly on the path segments. What many people don't realize is that it's entirely possible to do segment-based routing in Yesod. I don't recommend the practice in general, since it defeats a lot of the type safety Yesod gives you. However, in some situations it's a necessity. For example, yesod-auth allows arbitrary authentication plugins, which would not mesh up well with type-safe URLs. The trick to bypassing the type safety is to use multi-pieces. Instead of describing how this is possible, I will merely point to an example: https://gist.github.com/884802 The Yesod.Dynamic module defined there is really a general way to allow Yesod routing based on path segments. Since the question has come up a number of times in the past, I'm curious if people would want this available on hackage. I'm sure a nice set of routing operators would be useful here as well. Michael

4 4