January 2011 - web-devel

web-routes-quasi gotcha
by Nubis 21 Jan '11

21 Jan '11

Hi guys, I was just investigating an issue in my yesod app that prevented me from running the devel-server. It was a warning about route patterns overlapping S4M/Server/Controller.hs:33:0: Warning: Pattern match(es) are overlapped In the definition of `parse[awnQ]': parse[awnQ] : "con" : "f" [] = ... parse[awnQ] : "con" : "h" [] = ... parse[awnQ] : "con" : "m" : "list" [] = ... parse[awnQ] : "con" : "m" : "c" [] = ... The issue seems to come up when part of the route is a single character (in this case, f, h and m) like /con/f/ /con/h/ /con/m/list/ I ended up just changing my routes to be more 'expressive' which is fine by me. Haven't looked at the source of the issue yet, I can reproduce it on 0.6.2 but haven't tried 0.6.3 so I apologize in advance if it is already fixed :) Just wanted to get it out here so I don't forget to look into it later (I'm heading home now) Cheers! ----nubis :)

2 1

Directory listings
by Michael Snoyman 19 Jan '11

19 Jan '11

Hi everyone, I'm working on a static file serving WAI app[1], which can either be used from a library, or combined with Warp as a standalone executable. One of its features will be directory listings. Is any interested in designing a directory listing page? Extra points if you can style it using some Haskell colors. Thanks, Michael [1] https://github.com/snoyberg/wai-app-static

2 2

Changes implemented in Hamlet
by Michael Snoyman 15 Jan '11

15 Jan '11

Hi everyone, I wanted to give you all an update on the changes in Hamlet 0.7[1]. So far, the following have been implemented: * Adding angle-bracket syntax * Remove period-delimited variables, and allow qualified functions. For example, $Data.List.sort myString$ * Change forall and maybe to use <-, eg $forall person <- people I haven't tried it on real life code yet, but I have converted the entire Hamlet test suite to use this, and I'm quite happy with the result. Overall, I think things are much more readable. Just for the record: the second item (period-delimited variables) affects Cassius and Julius as well. So now for the last few decisions: * Do we rename maybe to just? I am leaning towards no on this. * For Cassius, do we switch from a whitespace sensitive syntax to brackets? * Do we convert from $my variable$ to #{my variable}. * And if we do this, can we consider #{my (nested call)} and #{my $ nested call} the same? Please see the wiki page[1] for more information on all of these. As far as the I18N note at the bottom: I think we'll be able to release such a feature without introducing any breaking changes. Most likely, a "message" will be a specialization of standard variable interpolation, eg: #?{myI18NMessage} In other words, I'm in no rush to get that feature in the 0.7.0 release, it can wait until 0.7.1. I *do*, however, want to get this Hamlet released soon so that it doesn't hold up Yesod 0.7 any longer: since Yesod 0.6 will not work with WAI 0.3, it also won't work with Warp, and I think a lot of people will want to be able to use Warp soon. Michael [1] http://wiki.yesodweb.com/Hamlet%200.7%20changes

1 0

[Yesod] an API which return Root Relative URL.
by いとうかつとし 07 Jan '11

07 Jan '11

Hi I want an api which return Root Relative URL. I've been developing wiki clone. When a writer select and click an image from photo galary in order to insert a image, click event insert a wiki image link like as "![image_name]( http://localhost:3000/uploaded/1)". But I think this - wiki contents has self Absolute URL - is not good. Is there the API? I want the api, but i remove approot string from Absolute URL, now.

2 1

lazy RESTful Content
by Greg Weber 07 Jan '11

07 Jan '11

lazy and REST seem to go together :) Perhaps this will expose my ignorance of laziness/strictness and isn't at all Yesod specific. I have a handler that can serve html or json. When an html request comes in it still executes the json code. I assume this is because the json code uses liftIO. What is the strategy for avoiding executing the json code?

2 3

Yesod routes and Persistent entities in separate files
by Michael Snoyman 05 Jan '11

05 Jan '11

Hi all, The Yesod scaffolding tool currently does precisely one thing: produce a scaffolded site. It does *not* provide any facilities for adding new routes or entities after the fact, or creating basic CRUD interfaces the way Rails or Django do. Nubis has proposed adding this functionality. The current roadblock is that the Yesod routes and Persistent entities are defined with quasi-quotation within the code. I believe that technically speaking it should be possible to make this kind of feature work, but it would make more sense to do it as a separate file. I think the proposal would essentially change: mkYesod "MyApp" [$parseRoutes|...|] to mkYesod "MyApp" "MyApp.routes" with the ... being in MyApp.routes, and a similar change for the $persist quasi-quotation. What does everyone think of this? I see the following advantages: * Makes it more straight-forward to write these kinds of scaffolding tools. * Gets away from quasi-quotation for those who have an aversion to it. On the other hand, the downsides are: * One more file to deal with. * The standard issue of Haskell files not automatically recompiling when an external files are changed. This can be worked around with wai-handler-devel. Another possibility is offering *both* options, but as much as possible I'm trying to avoid creating multiple approaches. Michael

2 2

Re: [web-devel] Wrapping the widget generated by the authEmail backend.
by Nubis 04 Jan '11

04 Jan '11

Exactly! I was looking for addWidget. I'll give it a shot at providing my own html also, I'll let you know my final solution. Thank you all guys. cheers ----nubis :) On Tue, Jan 4, 2011 at 9:29 AM, Matt Brown <matt(a)softmechanics.net> wrote: > Regarding your error, I think you want addWidget instead of addHamlet. > This compiles for me: > > loginHandler = do > let tm = liftHandler getRouteToMaster > let email = tm >>= apLogin authEmail > defaultLayout $ do > setTitle $ string "Login" > addWidget $(hamletFile "login") > > -matt > >

1 0

Re: [web-devel] Wrapping the widget generated by the authEmail backend.
by Christian Klauser 04 Jan '11

04 Jan '11

---------- Forwarded message ---------- From: Christian Klauser <ch27k89(a)gmail.com> Date: 4 January 2011 11:35 Subject: Re: [web-devel] Wrapping the widget generated by the authEmail backend. To: Nubis <nubis(a)woobiz.com.ar> Hi, I hit the very same problem a couple of days ago. You can certainly just build a "compatible" form (with the same field names) and that will just work fine. If you want to use the hamlet provided by the plugins, here is what I did: loginHandler = defaultLayout $ do setTitle $ string "Login" tm <- liftHandler getRouteToMaster let renderPlugin :: AuthPlugin YourSiteHere -> SubWidget Auth () -- necessary renderPlugin = flip apLogin tm openidForm = renderPlugin authOpenId emailForm = renderPlugin authEmail addWidget $(widgetFile "login") In 'login.hamlet' I can then use ^openidForm^ and ^emailForm^ like any other widget. Neat as it is, you'll still have a hard time styling these login forms, as don't have any CSS classes assigned to them. It will be difficult to distinguish between %input!type=text and %input!type=submit, so you might end up re-creating those forms anyway :-( Hope this helps Regards Chris On 4 January 2011 03:38, Nubis <nubis(a)woobiz.com.ar> wrote: > Hi guys, > I'm trying to build a good looking login page, but I noticed the > loginHandler that comes with yesod-auth just concatenates all the widgets > from all the used auth backends. To make matters worse, the markup some of > the backends generate is not even my preferred brand! :) > > So I try to at least wrap the authEmail widget inside my own html, with a > login.hamlet file that looks a bit like this: > .box > %h1 Login > .wrapper > ^email^ > > I manage to get the widget out of the authEmail backend, but then I can't > use it inside my hamlet file that I add using addHamlet b/c it doesn't > typecheck. > > loginHandler = do > let tm = liftHandler getRouteToMaster > let email = tm >>= apLogin authEmail > defaultLayout $ do > setTitle $ string "Login" > addHamlet $(Settings.hamletFile "login") > > Is there any way around this? Am I approaching this the wrong way? > This could also be solved if instead of a widget I could get the url and > fieldname information from the backend, so I can use those things inside my > ad-hoc html. > Moreso, as a way to customize the html I'm tempted to override or redefine > apLogin inside my app's domain so I can customize the html it generates. But > just to keep taking advantage of haskell's strenghts it would be nice to > formalize this idea a bit more (and I don't know how to approach that). > > thanks in advance for your help > cheers! > ----nubis :) > > _______________________________________________ > web-devel mailing list > web-devel(a)haskell.org > http://www.haskell.org/mailman/listinfo/web-devel > >

2 1

Wrapping the widget generated by the authEmail backend.
by Nubis 03 Jan '11

03 Jan '11

Hi guys, I'm trying to build a good looking login page, but I noticed the loginHandler that comes with yesod-auth just concatenates all the widgets from all the used auth backends. To make matters worse, the markup some of the backends generate is not even my preferred brand! :) So I try to at least wrap the authEmail widget inside my own html, with a login.hamlet file that looks a bit like this: .box %h1 Login .wrapper ^email^ I manage to get the widget out of the authEmail backend, but then I can't use it inside my hamlet file that I add using addHamlet b/c it doesn't typecheck. loginHandler = do let tm = liftHandler getRouteToMaster let email = tm >>= apLogin authEmail defaultLayout $ do setTitle $ string "Login" addHamlet $(Settings.hamletFile "login") Is there any way around this? Am I approaching this the wrong way? This could also be solved if instead of a widget I could get the url and fieldname information from the backend, so I can use those things inside my ad-hoc html. Moreso, as a way to customize the html I'm tempted to override or redefine apLogin inside my app's domain so I can customize the html it generates. But just to keep taking advantage of haskell's strenghts it would be nice to formalize this idea a bit more (and I don't know how to approach that). thanks in advance for your help cheers! ----nubis :)

2 1

Hamlet variables: (foo bar) baz or foo (bar baz)
by Michael Snoyman 03 Jan '11

03 Jan '11

I thought I would start this in a separate thread. Right now in Hamlet, $foo.bar.baz$ (or equivalently $foo bar baz$ gets converted to the Haskell code "foo (bar baz)". This makes a fair bit of sense for the period-delimited syntax, but feels very wrong for the space-delimited syntax. So the question is: should we change things around? That would mean we would need to modify our current templates to include extra parentheses, eg: $foo.bar.baz$ -> $foo (bar baz)$ $a.b.c.d$ -> $a (b (c d))$ I suppose that, in theory, if we actually change the variable interpolation character to a percent sign, we could take back the dollar sign to work like Haskell, eg: $a.b.c.d$ -> %a $ b $ c d% And just since I brought *that* up: we could consider using a different symbol for variable interpolation than the percent sign, such as a hash or ampersand. However, both of those already have special meaning (hash == id, ampersand == HTML escape character), so I don't know how well that would work. There was talk once upon a time of merging the syntax for dollar-sign and caret interpolation, which now that we've added some more polymorphism it might be possible. Caveat: I haven't actually written any code for this, so I don't know how feasible it is. I'm basically throwing out ideas as they come to me. Michael

5 6