From kevwil at gmail.com  Thu Nov  1 14:21:30 2007
From: kevwil at gmail.com (Kevin Williams)
Date: Thu, 1 Nov 2007 12:21:30 -0600
Subject: Mongrel breaking Merb?
In-Reply-To: <ef3e5d50710310858x62cbc0eg9c68f059c87092da@mail.gmail.com>
References: <683a886f0710310754m34407e6dp527d403bb692404d@mail.gmail.com>
	<ef3e5d50710310858x62cbc0eg9c68f059c87092da@mail.gmail.com>
Message-ID: <683a886f0711011121k2f601602sd96afdd7dd7eec1d@mail.gmail.com>

I just updated everything this morning, and still get the error.
Here's the message.

I'm now on Leopard using the Apple build of Ruby. I have mongrel 1.0.4
and swiftiply 0.6.1 installed. I don't think there are any other gems
that would be involved here.

/Library/Ruby/Gems/1.8/gems/mongrel-1.0.4/lib/mongrel/configurator.rb:139:in
`initialize': wrong number of arguments (5 for 4) (ArgumentError)
	from /Library/Ruby/Gems/1.8/gems/mongrel-1.0.4/lib/mongrel/configurator.rb:139:in
`new'
	from /Library/Ruby/Gems/1.8/gems/mongrel-1.0.4/lib/mongrel/configurator.rb:139:in
`listener'
	from /Library/Ruby/Gems/1.8/gems/merb-0.4.0/lib/merb/server.rb:518:in
`cloaker_'
	from /Library/Ruby/Gems/1.8/gems/mongrel-1.0.4/lib/mongrel/configurator.rb:50:in
`call'
	from /Library/Ruby/Gems/1.8/gems/mongrel-1.0.4/lib/mongrel/configurator.rb:50:in
`initialize'
	from /Library/Ruby/Gems/1.8/gems/merb-0.4.0/lib/merb/server.rb:517:in `new'
	from /Library/Ruby/Gems/1.8/gems/merb-0.4.0/lib/merb/server.rb:517:in
`mongrel_start'
	from /Library/Ruby/Gems/1.8/gems/merb-0.4.0/lib/merb/server.rb:446:in `run'
	from /Library/Ruby/Gems/1.8/gems/merb-0.4.0/bin/merb:6
	from /usr/bin/merb:16:in `load'
	from /usr/bin/merb:16

Thanks for looking into it. :)


On 10/31/07, Wayne E. Seguin <wayneeseguin at gmail.com> wrote:
> On 10/31/07, Kevin Williams <kevwil at gmail.com> wrote:
> > Has anyone had troubles from Mongrel > 1.0.1? Just curious, because
> > the last time I tried to run Merb it gave me some Mongrel error about
> > the wrong number of arguments, or something like that.
> >
> > I can post the exact message tonight when I get home.
> >
>
> Kevin,
>
> There have been a few changes so exact message would be most helpful.
>
>  Thank you,
>
>   ~Wayne


-- 
Cheers,

Kevin Williams
http://kevwil.com/
http://www.almostserio.us/
http://kevwil.jaiku.com/

From ez at engineyard.com  Thu Nov  1 21:56:17 2007
From: ez at engineyard.com (Ezra Zygmuntowicz)
Date: Thu, 1 Nov 2007 18:56:17 -0700
Subject: Mongrel breaking Merb?
In-Reply-To: <683a886f0711011121k2f601602sd96afdd7dd7eec1d@mail.gmail.com>
References: <683a886f0710310754m34407e6dp527d403bb692404d@mail.gmail.com>
	<ef3e5d50710310858x62cbc0eg9c68f059c87092da@mail.gmail.com>
	<683a886f0711011121k2f601602sd96afdd7dd7eec1d@mail.gmail.com>
Message-ID: <31564A68-60AB-4CCD-9BCA-34A7091B76EF@engineyard.com>


	This is a known probelm in swiftiply, the newest mongrel release  
takes 5 args for the constructor of a new http server but the  
evented_mongrel still uses 4. Kirk will be releasing an update  
sometime soon.

Thanks
-Ezra


On Nov 1, 2007, at 11:21 AM, Kevin Williams wrote:

> I just updated everything this morning, and still get the error.
> Here's the message.
>
> I'm now on Leopard using the Apple build of Ruby. I have mongrel 1.0.4
> and swiftiply 0.6.1 installed. I don't think there are any other gems
> that would be involved here.
>
> /Library/Ruby/Gems/1.8/gems/mongrel-1.0.4/lib/mongrel/ 
> configurator.rb:139:in
> `initialize': wrong number of arguments (5 for 4) (ArgumentError)
> 	from /Library/Ruby/Gems/1.8/gems/mongrel-1.0.4/lib/mongrel/ 
> configurator.rb:139:in
> `new'
> 	from /Library/Ruby/Gems/1.8/gems/mongrel-1.0.4/lib/mongrel/ 
> configurator.rb:139:in
> `listener'
> 	from /Library/Ruby/Gems/1.8/gems/merb-0.4.0/lib/merb/server.rb:518:in
> `cloaker_'
> 	from /Library/Ruby/Gems/1.8/gems/mongrel-1.0.4/lib/mongrel/ 
> configurator.rb:50:in
> `call'
> 	from /Library/Ruby/Gems/1.8/gems/mongrel-1.0.4/lib/mongrel/ 
> configurator.rb:50:in
> `initialize'
> 	from /Library/Ruby/Gems/1.8/gems/merb-0.4.0/lib/merb/server.rb: 
> 517:in `new'
> 	from /Library/Ruby/Gems/1.8/gems/merb-0.4.0/lib/merb/server.rb:517:in
> `mongrel_start'
> 	from /Library/Ruby/Gems/1.8/gems/merb-0.4.0/lib/merb/server.rb: 
> 446:in `run'
> 	from /Library/Ruby/Gems/1.8/gems/merb-0.4.0/bin/merb:6
> 	from /usr/bin/merb:16:in `load'
> 	from /usr/bin/merb:16
>
> Thanks for looking into it. :)
>
>
> On 10/31/07, Wayne E. Seguin <wayneeseguin at gmail.com> wrote:
>> On 10/31/07, Kevin Williams <kevwil at gmail.com> wrote:
>>> Has anyone had troubles from Mongrel > 1.0.1? Just curious, because
>>> the last time I tried to run Merb it gave me some Mongrel error  
>>> about
>>> the wrong number of arguments, or something like that.
>>>
>>> I can post the exact message tonight when I get home.
>>>
>>
>> Kevin,
>>
>> There have been a few changes so exact message would be most helpful.
>>
>> Thank you,
>>
>>  ~Wayne
>
>
> -- 
> Cheers,
>
> Kevin Williams
> http://kevwil.com/
> http://www.almostserio.us/
> http://kevwil.jaiku.com/
> _______________________________________________
> Merb-devel mailing list
> Merb-devel at rubyforge.org
> http://rubyforge.org/mailman/listinfo/merb-devel

Ezra Zygmuntowicz
Founder & Ruby Hacker
ezra at engineyard.com
EngineYard.com



From kevwil at gmail.com  Thu Nov  1 22:00:16 2007
From: kevwil at gmail.com (Kevin Williams)
Date: Thu, 1 Nov 2007 20:00:16 -0600
Subject: Mongrel breaking Merb?
In-Reply-To: <31564A68-60AB-4CCD-9BCA-34A7091B76EF@engineyard.com>
References: <683a886f0710310754m34407e6dp527d403bb692404d@mail.gmail.com>
	<ef3e5d50710310858x62cbc0eg9c68f059c87092da@mail.gmail.com>
	<683a886f0711011121k2f601602sd96afdd7dd7eec1d@mail.gmail.com>
	<31564A68-60AB-4CCD-9BCA-34A7091B76EF@engineyard.com>
Message-ID: <683a886f0711011900x4ba78c7anb85530418b1a12a5@mail.gmail.com>

Sweet. Thanks for the clarification!

On 11/1/07, Ezra Zygmuntowicz <ez at engineyard.com> wrote:
>
>         This is a known probelm in swiftiply, the newest mongrel release
> takes 5 args for the constructor of a new http server but the
> evented_mongrel still uses 4. Kirk will be releasing an update
> sometime soon.
>
> Thanks
> -Ezra
>
>
> On Nov 1, 2007, at 11:21 AM, Kevin Williams wrote:
>
> > I just updated everything this morning, and still get the error.
> > Here's the message.
> >
> > I'm now on Leopard using the Apple build of Ruby. I have mongrel 1.0.4
> > and swiftiply 0.6.1 installed. I don't think there are any other gems
> > that would be involved here.
> >
> > /Library/Ruby/Gems/1.8/gems/mongrel-1.0.4/lib/mongrel/
> > configurator.rb:139:in
> > `initialize': wrong number of arguments (5 for 4) (ArgumentError)
> >       from /Library/Ruby/Gems/1.8/gems/mongrel-1.0.4/lib/mongrel/
> > configurator.rb:139:in
> > `new'
> >       from /Library/Ruby/Gems/1.8/gems/mongrel-1.0.4/lib/mongrel/
> > configurator.rb:139:in
> > `listener'
> >       from /Library/Ruby/Gems/1.8/gems/merb-0.4.0/lib/merb/server.rb:518:in
> > `cloaker_'
> >       from /Library/Ruby/Gems/1.8/gems/mongrel-1.0.4/lib/mongrel/
> > configurator.rb:50:in
> > `call'
> >       from /Library/Ruby/Gems/1.8/gems/mongrel-1.0.4/lib/mongrel/
> > configurator.rb:50:in
> > `initialize'
> >       from /Library/Ruby/Gems/1.8/gems/merb-0.4.0/lib/merb/server.rb:
> > 517:in `new'
> >       from /Library/Ruby/Gems/1.8/gems/merb-0.4.0/lib/merb/server.rb:517:in
> > `mongrel_start'
> >       from /Library/Ruby/Gems/1.8/gems/merb-0.4.0/lib/merb/server.rb:
> > 446:in `run'
> >       from /Library/Ruby/Gems/1.8/gems/merb-0.4.0/bin/merb:6
> >       from /usr/bin/merb:16:in `load'
> >       from /usr/bin/merb:16
> >
> > Thanks for looking into it. :)
> >
> >
> > On 10/31/07, Wayne E. Seguin <wayneeseguin at gmail.com> wrote:
> >> On 10/31/07, Kevin Williams <kevwil at gmail.com> wrote:
> >>> Has anyone had troubles from Mongrel > 1.0.1? Just curious, because
> >>> the last time I tried to run Merb it gave me some Mongrel error
> >>> about
> >>> the wrong number of arguments, or something like that.
> >>>
> >>> I can post the exact message tonight when I get home.
> >>>
> >>
> >> Kevin,
> >>
> >> There have been a few changes so exact message would be most helpful.
> >>
> >> Thank you,
> >>
> >>  ~Wayne
> >
> >
> > --
> > Cheers,
> >
> > Kevin Williams
> > http://kevwil.com/
> > http://www.almostserio.us/
> > http://kevwil.jaiku.com/
> > _______________________________________________
> > Merb-devel mailing list
> > Merb-devel at rubyforge.org
> > http://rubyforge.org/mailman/listinfo/merb-devel
>
> Ezra Zygmuntowicz
> Founder & Ruby Hacker
> ezra at engineyard.com
> EngineYard.com
>
>
>


-- 
Cheers,

Kevin Williams
http://kevwil.com/
http://www.almostserio.us/
http://kevwil.jaiku.com/

From kevwil at gmail.com  Fri Nov  2 08:19:00 2007
From: kevwil at gmail.com (Kevin Williams)
Date: Fri, 2 Nov 2007 06:19:00 -0600
Subject: respond_to API
In-Reply-To: <20071019054009.GA19544@uk.tiscali.com>
References: <11409D88-BED6-48EC-8FF3-17217B5B55A4@gweezlebur.com>
	<683a886f0710161003h7aa903fch532e2e9f5de67253@mail.gmail.com>
	<2CD1E79E-D567-4770-935B-5A3FF6BC1613@gmail.com>
	<20071019054009.GA19544@uk.tiscali.com>
Message-ID: <683a886f0711020519l50e5e159u720b56fc2b979335@mail.gmail.com>

http://merb.devjavu.com/changeset/795

crossing_fingers.do
  ::Manually.refactor_controllers
  views.each {|v| mv "#{v.name}.#{v.extension}"
"#{v.name}.#{v.format}.#{v.extension}"}
end

I got one controller refactored, and the html format didn't break yet,
so we'll see how the :xml and :json formats go.


On 10/18/07, Brian Candler <B.Candler at pobox.com> wrote:
> On Thu, Oct 18, 2007 at 02:11:24PM -0400, Stephen Caudill wrote:
> > in the case of a rendered template, you do end up with a win though:
> >
> > class MyController
> >    provides :html
> >    def index
> >      provides :json, :yaml
> >    end
> > end
> >
> > vs.
> >
> > class MyController
> >    def index
> >      respond_to do |format|
> >        format.html
> >        format.json
> >        format.yaml
> >      end
> >    end
> > end
>
> But in that simple case, couldn't you just match the format to the template
> extension automatically?
>
> That is,
>
>    format.foo { ... }
>
> tells you how to render a foo request, but if not specified, it defaults to
> render action.rfoo
>
> The 'provides' would be implied by the existence of the template with the
> right extension.
>
> B.
> _______________________________________________
> Merb-devel mailing list
> Merb-devel at rubyforge.org
> http://rubyforge.org/mailman/listinfo/merb-devel
>


-- 
Cheers,

Kevin Williams
http://kevwil.com/
http://www.almostserio.us/
http://kevwil.jaiku.com/

From ivey at gweezlebur.com  Fri Nov  2 08:43:51 2007
From: ivey at gweezlebur.com (Michael D. Ivey)
Date: Fri, 2 Nov 2007 07:43:51 -0500
Subject: respond_to API
In-Reply-To: <683a886f0711020519l50e5e159u720b56fc2b979335@mail.gmail.com>
References: <11409D88-BED6-48EC-8FF3-17217B5B55A4@gweezlebur.com>
	<683a886f0710161003h7aa903fch532e2e9f5de67253@mail.gmail.com>
	<2CD1E79E-D567-4770-935B-5A3FF6BC1613@gmail.com>
	<20071019054009.GA19544@uk.tiscali.com>
	<683a886f0711020519l50e5e159u720b56fc2b979335@mail.gmail.com>
Message-ID: <B9A83B62-8A17-44FD-8851-9537B70F70B5@gweezlebur.com>

On Nov 2, 2007, at 7:19 AM, Kevin Williams wrote:
> http://merb.devjavu.com/changeset/795
>
> crossing_fingers.do
>   ::Manually.refactor_controllers
>   views.each {|v| mv "#{v.name}.#{v.extension}"
> "#{v.name}.#{v.format}.#{v.extension}"}
> end
>
> I got one controller refactored, and the html format didn't break yet,
> so we'll see how the :xml and :json formats go.

For the record, I didn't *have* to drop the legacy template names,  
but Ez wanted me to get it in the same changeset so everything broke  
at once.

HTML-only apps will probably not notice, other than the template rename.

respond_to based apps will be a little harder, but really shouldn't  
be too tough.

If anyone runs into trouble converting, hit me up on IRC, or on the  
Mailing List, or the Google Group, or in Charlotte Saturday...I'll be  
more than happy to help.

From kevwil at gmail.com  Fri Nov  2 10:33:01 2007
From: kevwil at gmail.com (Kevin Williams)
Date: Fri, 2 Nov 2007 08:33:01 -0600
Subject: respond_to API
In-Reply-To: <B9A83B62-8A17-44FD-8851-9537B70F70B5@gweezlebur.com>
References: <11409D88-BED6-48EC-8FF3-17217B5B55A4@gweezlebur.com>
	<683a886f0710161003h7aa903fch532e2e9f5de67253@mail.gmail.com>
	<2CD1E79E-D567-4770-935B-5A3FF6BC1613@gmail.com>
	<20071019054009.GA19544@uk.tiscali.com>
	<683a886f0711020519l50e5e159u720b56fc2b979335@mail.gmail.com>
	<B9A83B62-8A17-44FD-8851-9537B70F70B5@gweezlebur.com>
Message-ID: <683a886f0711020733x6c43d898j45fbb00342e6d73@mail.gmail.com>

Everything seems smooth so far, but I've only got one (restful)
controller so far. I do have about a dozen view files which needed
renaming, but that wasn't too bad.

I'm not complaining - this is life on the edge, things will break. I'm
looking forward to the next release.

On 11/2/07, Michael D. Ivey <ivey at gweezlebur.com> wrote:
> On Nov 2, 2007, at 7:19 AM, Kevin Williams wrote:
> > http://merb.devjavu.com/changeset/795
> >
> > crossing_fingers.do
> >   ::Manually.refactor_controllers
> >   views.each {|v| mv "#{v.name}.#{v.extension}"
> > "#{v.name}.#{v.format}.#{v.extension}"}
> > end
> >
> > I got one controller refactored, and the html format didn't break yet,
> > so we'll see how the :xml and :json formats go.
>
> For the record, I didn't *have* to drop the legacy template names,
> but Ez wanted me to get it in the same changeset so everything broke
> at once.
>
> HTML-only apps will probably not notice, other than the template rename.
>
> respond_to based apps will be a little harder, but really shouldn't
> be too tough.
>
> If anyone runs into trouble converting, hit me up on IRC, or on the
> Mailing List, or the Google Group, or in Charlotte Saturday...I'll be
> more than happy to help.
> _______________________________________________
> Merb-devel mailing list
> Merb-devel at rubyforge.org
> http://rubyforge.org/mailman/listinfo/merb-devel
>


-- 
Cheers,

Kevin Williams
http://kevwil.com/
http://www.almostserio.us/
http://kevwil.jaiku.com/

From frazer.list at gmail.com  Sat Nov  3 16:11:13 2007
From: frazer.list at gmail.com (frazer horn)
Date: Sat, 3 Nov 2007 16:11:13 -0400
Subject: Render js
Message-ID: <f472410b0711031311x7c42af3bxe66b3347be84a6ec@mail.gmail.com>

Hi all

Not sure if I'm missing something, render :js => "alert('hello');" does
return the alert but also displays alert('hello') in the template as text.
This is evident for all javascript returned via render js (inline or in
.js.erb).

Any help appreciated

Thanks

Frazer
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://rubyforge.org/pipermail/merb-devel/attachments/20071103/8f5bbc0a/attachment.html 

From kevwil at gmail.com  Mon Nov  5 22:34:07 2007
From: kevwil at gmail.com (Kevin Williams)
Date: Mon, 5 Nov 2007 20:34:07 -0700
Subject: what did I do?
Message-ID: <683a886f0711051934r21f1ee79td55776accb4e41d5@mail.gmail.com>

I'm following merb trunk. This morning I updated my gems, sync'd merb
source and installed the merb gem from source. I don't know if I did
something wrong, or if a new gem messed things up, or if there was a
merb change which broke something. Now I get this error:

Request: REQUEST_URI: /  (2007-11-05 20:27:04)
Params: {"action"=>"index", "controller"=>"articles"}
Cookies: {"_session_id"=>"fe8eea1f497d002ee39ad61cbd868a2b"}
Setting up session
undefined method `keys' for #<Article:0x16b2de8> - (NoMethodError)
/Users/kevwil/dev/git/rubyrubyroo/app/views/articles/index.html.haml:4
/Library/Ruby/Gems/1.8/gems/activerecord-1.15.5/lib/active_record/base.rb:1863:in
`method_missing'
/Library/Ruby/Gems/1.8/gems/merb-0.4.0/lib/merb/mixins/controller.rb:65:in `url'
/Library/Ruby/Gems/1.8/gems/merb-0.4.0/lib/merb/view_context.rb:73:in `send'
/Library/Ruby/Gems/1.8/gems/merb-0.4.0/lib/merb/view_context.rb:73:in
`method_missing'
(eval):11:in `_render_haml_1'
(eval):8:in `each'
(eval):8:in `_render_haml_1'
/Library/Ruby/Gems/1.8/gems/haml-1.7.1/lib/haml/engine.rb:435:in `send'
/Library/Ruby/Gems/1.8/gems/haml-1.7.1/lib/haml/engine.rb:435:in `compile'
/Library/Ruby/Gems/1.8/gems/haml-1.7.1/lib/haml/engine.rb:186:in `to_html'
/Library/Ruby/Gems/1.8/gems/merb-0.4.0/lib/merb/template/haml.rb:44:in
`transform'
/Library/Ruby/Gems/1.8/gems/merb-0.4.0/lib/merb/mixins/render.rb:209:in `render'
/Users/kevwil/dev/git/rubyrubyroo/app/controllers/articles.rb:10:in `index'
/Library/Ruby/Gems/1.8/gems/merb-0.4.0/lib/merb/abstract_controller.rb:57:in
`send'
/Library/Ruby/Gems/1.8/gems/merb-0.4.0/lib/merb/abstract_controller.rb:57:in
`call_action'
/Library/Ruby/Gems/1.8/gems/merb-0.4.0/lib/merb/abstract_controller.rb:32:in
`dispatch'
/Library/Ruby/Gems/1.8/gems/merb-0.4.0/lib/merb/controller.rb:80:in `dispatch'
/Library/Ruby/Gems/1.8/gems/merb-0.4.0/lib/merb/dispatcher.rb:53:in
`dispatch_action'
/Library/Ruby/Gems/1.8/gems/merb-0.4.0/lib/merb/dispatcher.rb:35:in `handle'
/Library/Ruby/Gems/1.8/gems/merb-0.4.0/lib/merb/mongrel_handler.rb:86:in
`process'
/Library/Ruby/Gems/1.8/gems/mongrel-1.1/lib/mongrel.rb:155:in `process_client'
/Library/Ruby/Gems/1.8/gems/mongrel-1.1/lib/mongrel.rb:154:in `each'
/Library/Ruby/Gems/1.8/gems/mongrel-1.1/lib/mongrel.rb:154:in `process_client'
/Library/Ruby/Gems/1.8/gems/mongrel-1.1/lib/mongrel.rb:281:in `run'
/Library/Ruby/Gems/1.8/gems/mongrel-1.1/lib/mongrel.rb:281:in `initialize'
/Library/Ruby/Gems/1.8/gems/mongrel-1.1/lib/mongrel.rb:281:in `new'
/Library/Ruby/Gems/1.8/gems/mongrel-1.1/lib/mongrel.rb:281:in `run'
/Library/Ruby/Gems/1.8/gems/mongrel-1.1/lib/mongrel.rb:264:in `initialize'
/Library/Ruby/Gems/1.8/gems/mongrel-1.1/lib/mongrel.rb:264:in `new'
/Library/Ruby/Gems/1.8/gems/mongrel-1.1/lib/mongrel.rb:264:in `run'
/Library/Ruby/Gems/1.8/gems/mongrel-1.1/lib/mongrel/configurator.rb:282:in `run'
/Library/Ruby/Gems/1.8/gems/mongrel-1.1/lib/mongrel/configurator.rb:281:in
`each'
/Library/Ruby/Gems/1.8/gems/mongrel-1.1/lib/mongrel/configurator.rb:281:in `run'
/Library/Ruby/Gems/1.8/gems/merb-0.4.0/lib/merb/server.rb:557:in `cloaker_'
/Library/Ruby/Gems/1.8/gems/mongrel-1.1/lib/mongrel/configurator.rb:50:in `call'
/Library/Ruby/Gems/1.8/gems/mongrel-1.1/lib/mongrel/configurator.rb:50:in
`initialize'
/Library/Ruby/Gems/1.8/gems/merb-0.4.0/lib/merb/server.rb:548:in `new'
/Library/Ruby/Gems/1.8/gems/merb-0.4.0/lib/merb/server.rb:548:in `mongrel_start'
/Library/Ruby/Gems/1.8/gems/merb-0.4.0/lib/merb/server.rb:477:in `run'
/Library/Ruby/Gems/1.8/gems/merb-0.4.0/bin/merb:6
/usr/bin/merb:16:in `load'
/usr/bin/merb:16
undefined method `each' for #<Merb::MemorySession:0x20f5404> - (NoMethodError)
(erubis):144:in `dispatch_default_exception'
/Library/Ruby/Gems/1.8/gems/merb-0.4.0/lib/merb/dispatcher.rb:95:in
`dispatch_default_exception'
/Library/Ruby/Gems/1.8/gems/merb-0.4.0/lib/merb/dispatcher.rb:77:in
`dispatch_exception'
/Library/Ruby/Gems/1.8/gems/merb-0.4.0/lib/merb/dispatcher.rb:39:in `handle'
/Library/Ruby/Gems/1.8/gems/merb-0.4.0/lib/merb/mongrel_handler.rb:86:in
`process'
/Library/Ruby/Gems/1.8/gems/mongrel-1.1/lib/mongrel.rb:155:in `process_client'
/Library/Ruby/Gems/1.8/gems/mongrel-1.1/lib/mongrel.rb:154:in `each'
/Library/Ruby/Gems/1.8/gems/mongrel-1.1/lib/mongrel.rb:154:in `process_client'
/Library/Ruby/Gems/1.8/gems/mongrel-1.1/lib/mongrel.rb:281:in `run'
/Library/Ruby/Gems/1.8/gems/mongrel-1.1/lib/mongrel.rb:281:in `initialize'
/Library/Ruby/Gems/1.8/gems/mongrel-1.1/lib/mongrel.rb:281:in `new'
/Library/Ruby/Gems/1.8/gems/mongrel-1.1/lib/mongrel.rb:281:in `run'
/Library/Ruby/Gems/1.8/gems/mongrel-1.1/lib/mongrel.rb:264:in `initialize'
/Library/Ruby/Gems/1.8/gems/mongrel-1.1/lib/mongrel.rb:264:in `new'
/Library/Ruby/Gems/1.8/gems/mongrel-1.1/lib/mongrel.rb:264:in `run'
/Library/Ruby/Gems/1.8/gems/mongrel-1.1/lib/mongrel/configurator.rb:282:in `run'
/Library/Ruby/Gems/1.8/gems/mongrel-1.1/lib/mongrel/configurator.rb:281:in
`each'
/Library/Ruby/Gems/1.8/gems/mongrel-1.1/lib/mongrel/configurator.rb:281:in `run'
/Library/Ruby/Gems/1.8/gems/merb-0.4.0/lib/merb/server.rb:557:in `cloaker_'
/Library/Ruby/Gems/1.8/gems/mongrel-1.1/lib/mongrel/configurator.rb:50:in `call'
/Library/Ruby/Gems/1.8/gems/mongrel-1.1/lib/mongrel/configurator.rb:50:in
`initialize'
/Library/Ruby/Gems/1.8/gems/merb-0.4.0/lib/merb/server.rb:548:in `new'
/Library/Ruby/Gems/1.8/gems/merb-0.4.0/lib/merb/server.rb:548:in `mongrel_start'
/Library/Ruby/Gems/1.8/gems/merb-0.4.0/lib/merb/server.rb:477:in `run'
/Library/Ruby/Gems/1.8/gems/merb-0.4.0/bin/merb:6
/usr/bin/merb:16:in `load'
/usr/bin/merb:16


Any suggestions?

--
Cheers,

Kevin Williams
http://kevwil.com

From frazer.list at gmail.com  Wed Nov  7 07:22:38 2007
From: frazer.list at gmail.com (frazer horn)
Date: Wed, 7 Nov 2007 07:22:38 -0500
Subject: merb 0.4.0 session issue
Message-ID: <f472410b0711070422n7c75712dt207892b05b36465f@mail.gmail.com>

Not sure if anyone else has had this problem.

I was running on merb edge fine a few days ago.  Last night I went to
merb 0.4.0 via gem install and now any call to session[:foo] causes an
error >> undefined method '[]' for nil:NilClass

I have uninstalled and reinstalled merb with no luck.

dependencies.rb
use_orm :activerecord

merb.yml
:session_store: active_record

Gems
merb 0.4.0
merb_activerecord 0.4.2

Mac OSX

Any help appreciated.

From jonathan at new-bamboo.co.uk  Wed Nov  7 09:56:08 2007
From: jonathan at new-bamboo.co.uk (Jonathan Conway)
Date: Wed, 7 Nov 2007 14:56:08 +0000
Subject: merb 0.4.0 session issue
In-Reply-To: <f472410b0711070422n7c75712dt207892b05b36465f@mail.gmail.com>
References: <f472410b0711070422n7c75712dt207892b05b36465f@mail.gmail.com>
Message-ID: <62E17284-42CB-4CBA-92F2-30C8E524A5DA@new-bamboo.co.uk>

A few things are different with how you configure sessions in Merb  
0.4.0.

Try changing :session_store: active_record to :sql_session: true

Hope this helps

Jonathan

----
http://www.jaikoo.com


On 7 Nov 2007, at 12:22, frazer horn wrote:

> Not sure if anyone else has had this problem.
>
> I was running on merb edge fine a few days ago.  Last night I went to
> merb 0.4.0 via gem install and now any call to session[:foo] causes an
> error >> undefined method '[]' for nil:NilClass
>
> I have uninstalled and reinstalled merb with no luck.
>
> dependencies.rb
> use_orm :activerecord
>
> merb.yml
> :session_store: active_record
>
> Gems
> merb 0.4.0
> merb_activerecord 0.4.2
>
> Mac OSX
>
> Any help appreciated.
> _______________________________________________
> Merb-devel mailing list
> Merb-devel at rubyforge.org
> http://rubyforge.org/mailman/listinfo/merb-devel


From merb at dusty.name  Wed Nov  7 10:05:41 2007
From: merb at dusty.name (Dusty Doris)
Date: Wed, 7 Nov 2007 10:05:41 -0500
Subject: merb/sequel mutex on or off? proxy_balancer or no?
Message-ID: <e9e3ad7b0711070705v768068i576c2be0bcd16c0f@mail.gmail.com>

Just a quick question to those of you using merb and sequel or
datamapper.  How do you setup your merb app?

For example:

merb -X off -c 5, then run mod_proxy_balancer or nginx to load balance
between the 5 (or whatever #) of merb nodes?

merb -X off -d, then simply use something like proxypass

merb -c 5, keep the mutex lock and cluster it

haproxy, swiftiply, evented mongrel......


What I'm curious about is if you are using sequel, do you typically
remove those mutex locks?  If so, is there a benefit still in running
a clustered mongrel setup?  Just curious to how others are setting up
their applications.  Right now I'm removing the mutex lock and setting
up a cluster of 5 behind proxy_balancer and its working great.
However, I was just curious to what others are doing.

Thanks.  Its looking great.  I especially love the new provides :xml,
:text, ... in the responder.  Really nice idea.

-Dusty Doris

From ivey at gweezlebur.com  Wed Nov  7 10:18:34 2007
From: ivey at gweezlebur.com (Michael D. Ivey)
Date: Wed, 7 Nov 2007 09:18:34 -0600
Subject: merb 0.4.0 session issue
In-Reply-To: <f472410b0711070422n7c75712dt207892b05b36465f@mail.gmail.com>
References: <f472410b0711070422n7c75712dt207892b05b36465f@mail.gmail.com>
Message-ID: <FF27824A-E1D9-4981-911A-431FDC172240@gweezlebur.com>

On Nov 7, 2007, at 6:22 AM, frazer horn wrote:
> I was running on merb edge fine a few days ago.  Last night I went to
> merb 0.4.0 via gem install and now any call to session[:foo] causes an
> error >> undefined method '[]' for nil:NilClass
>
> I have uninstalled and reinstalled merb with no luck.

I just installed merb and merb_activerecord from gems, created a new  
app, turned on AR sessions,
did the migration, and sessions work.

So to figure out why they don't work in your app, gonna need to see  
more info.

Can you reply with relevant chunks of your merb.yml, dependencies.rb,  
schema.rb, controller, etc?

Also, do you have merb_activerecord installed?  The name changed from  
merb_active_record to merb_activerecord right before 0.4.0 released.

From kevwil at gmail.com  Wed Nov  7 10:36:00 2007
From: kevwil at gmail.com (Kevin Williams)
Date: Wed, 7 Nov 2007 08:36:00 -0700
Subject: what did I do?
In-Reply-To: <683a886f0711051934r21f1ee79td55776accb4e41d5@mail.gmail.com>
References: <683a886f0711051934r21f1ee79td55776accb4e41d5@mail.gmail.com>
Message-ID: <683a886f0711070736r4fa9548by601fa299cea1c801@mail.gmail.com>

Seems to be fixed. Odd that I got no reply whatsoever, but hey, I'm
glad it's fixed!

Hooray for 0.4 being released! (and swiftiply being fixed) I can stop
following trunk and stop bothering everyone with dumb questions. :)


On Nov 5, 2007 8:34 PM, Kevin Williams <kevwil at gmail.com> wrote:
> I'm following merb trunk. This morning I updated my gems, sync'd merb
> source and installed the merb gem from source. I don't know if I did
> something wrong, or if a new gem messed things up, or if there was a
> merb change which broke something. Now I get this error:
>
> Request: REQUEST_URI: /  (2007-11-05 20:27:04)
> Params: {"action"=>"index", "controller"=>"articles"}
> Cookies: {"_session_id"=>"fe8eea1f497d002ee39ad61cbd868a2b"}
> Setting up session
> undefined method `keys' for #<Article:0x16b2de8> - (NoMethodError)
> /Users/kevwil/dev/git/rubyrubyroo/app/views/articles/index.html.haml:4
> /Library/Ruby/Gems/1.8/gems/activerecord-1.15.5/lib/active_record/base.rb:1863:in
> `method_missing'
> /Library/Ruby/Gems/1.8/gems/merb-0.4.0/lib/merb/mixins/controller.rb:65:in `url'
> /Library/Ruby/Gems/1.8/gems/merb-0.4.0/lib/merb/view_context.rb:73:in `send'
> /Library/Ruby/Gems/1.8/gems/merb-0.4.0/lib/merb/view_context.rb:73:in
> `method_missing'
> (eval):11:in `_render_haml_1'
> (eval):8:in `each'
> (eval):8:in `_render_haml_1'
> /Library/Ruby/Gems/1.8/gems/haml-1.7.1/lib/haml/engine.rb:435:in `send'
> /Library/Ruby/Gems/1.8/gems/haml-1.7.1/lib/haml/engine.rb:435:in `compile'
> /Library/Ruby/Gems/1.8/gems/haml-1.7.1/lib/haml/engine.rb:186:in `to_html'
> /Library/Ruby/Gems/1.8/gems/merb-0.4.0/lib/merb/template/haml.rb:44:in
> `transform'
> /Library/Ruby/Gems/1.8/gems/merb-0.4.0/lib/merb/mixins/render.rb:209:in `render'
> /Users/kevwil/dev/git/rubyrubyroo/app/controllers/articles.rb:10:in `index'
> /Library/Ruby/Gems/1.8/gems/merb-0.4.0/lib/merb/abstract_controller.rb:57:in
> `send'
> /Library/Ruby/Gems/1.8/gems/merb-0.4.0/lib/merb/abstract_controller.rb:57:in
> `call_action'
> /Library/Ruby/Gems/1.8/gems/merb-0.4.0/lib/merb/abstract_controller.rb:32:in
> `dispatch'
> /Library/Ruby/Gems/1.8/gems/merb-0.4.0/lib/merb/controller.rb:80:in `dispatch'
> /Library/Ruby/Gems/1.8/gems/merb-0.4.0/lib/merb/dispatcher.rb:53:in
> `dispatch_action'
> /Library/Ruby/Gems/1.8/gems/merb-0.4.0/lib/merb/dispatcher.rb:35:in `handle'
> /Library/Ruby/Gems/1.8/gems/merb-0.4.0/lib/merb/mongrel_handler.rb:86:in
> `process'
> /Library/Ruby/Gems/1.8/gems/mongrel-1.1/lib/mongrel.rb:155:in `process_client'
> /Library/Ruby/Gems/1.8/gems/mongrel-1.1/lib/mongrel.rb:154:in `each'
> /Library/Ruby/Gems/1.8/gems/mongrel-1.1/lib/mongrel.rb:154:in `process_client'
> /Library/Ruby/Gems/1.8/gems/mongrel-1.1/lib/mongrel.rb:281:in `run'
> /Library/Ruby/Gems/1.8/gems/mongrel-1.1/lib/mongrel.rb:281:in `initialize'
> /Library/Ruby/Gems/1.8/gems/mongrel-1.1/lib/mongrel.rb:281:in `new'
> /Library/Ruby/Gems/1.8/gems/mongrel-1.1/lib/mongrel.rb:281:in `run'
> /Library/Ruby/Gems/1.8/gems/mongrel-1.1/lib/mongrel.rb:264:in `initialize'
> /Library/Ruby/Gems/1.8/gems/mongrel-1.1/lib/mongrel.rb:264:in `new'
> /Library/Ruby/Gems/1.8/gems/mongrel-1.1/lib/mongrel.rb:264:in `run'
> /Library/Ruby/Gems/1.8/gems/mongrel-1.1/lib/mongrel/configurator.rb:282:in `run'
> /Library/Ruby/Gems/1.8/gems/mongrel-1.1/lib/mongrel/configurator.rb:281:in
> `each'
> /Library/Ruby/Gems/1.8/gems/mongrel-1.1/lib/mongrel/configurator.rb:281:in `run'
> /Library/Ruby/Gems/1.8/gems/merb-0.4.0/lib/merb/server.rb:557:in `cloaker_'
> /Library/Ruby/Gems/1.8/gems/mongrel-1.1/lib/mongrel/configurator.rb:50:in `call'
> /Library/Ruby/Gems/1.8/gems/mongrel-1.1/lib/mongrel/configurator.rb:50:in
> `initialize'
> /Library/Ruby/Gems/1.8/gems/merb-0.4.0/lib/merb/server.rb:548:in `new'
> /Library/Ruby/Gems/1.8/gems/merb-0.4.0/lib/merb/server.rb:548:in `mongrel_start'
> /Library/Ruby/Gems/1.8/gems/merb-0.4.0/lib/merb/server.rb:477:in `run'
> /Library/Ruby/Gems/1.8/gems/merb-0.4.0/bin/merb:6
> /usr/bin/merb:16:in `load'
> /usr/bin/merb:16
> undefined method `each' for #<Merb::MemorySession:0x20f5404> - (NoMethodError)
> (erubis):144:in `dispatch_default_exception'
> /Library/Ruby/Gems/1.8/gems/merb-0.4.0/lib/merb/dispatcher.rb:95:in
> `dispatch_default_exception'
> /Library/Ruby/Gems/1.8/gems/merb-0.4.0/lib/merb/dispatcher.rb:77:in
> `dispatch_exception'
> /Library/Ruby/Gems/1.8/gems/merb-0.4.0/lib/merb/dispatcher.rb:39:in `handle'
> /Library/Ruby/Gems/1.8/gems/merb-0.4.0/lib/merb/mongrel_handler.rb:86:in
> `process'
> /Library/Ruby/Gems/1.8/gems/mongrel-1.1/lib/mongrel.rb:155:in `process_client'
> /Library/Ruby/Gems/1.8/gems/mongrel-1.1/lib/mongrel.rb:154:in `each'
> /Library/Ruby/Gems/1.8/gems/mongrel-1.1/lib/mongrel.rb:154:in `process_client'
> /Library/Ruby/Gems/1.8/gems/mongrel-1.1/lib/mongrel.rb:281:in `run'
> /Library/Ruby/Gems/1.8/gems/mongrel-1.1/lib/mongrel.rb:281:in `initialize'
> /Library/Ruby/Gems/1.8/gems/mongrel-1.1/lib/mongrel.rb:281:in `new'
> /Library/Ruby/Gems/1.8/gems/mongrel-1.1/lib/mongrel.rb:281:in `run'
> /Library/Ruby/Gems/1.8/gems/mongrel-1.1/lib/mongrel.rb:264:in `initialize'
> /Library/Ruby/Gems/1.8/gems/mongrel-1.1/lib/mongrel.rb:264:in `new'
> /Library/Ruby/Gems/1.8/gems/mongrel-1.1/lib/mongrel.rb:264:in `run'
> /Library/Ruby/Gems/1.8/gems/mongrel-1.1/lib/mongrel/configurator.rb:282:in `run'
> /Library/Ruby/Gems/1.8/gems/mongrel-1.1/lib/mongrel/configurator.rb:281:in
> `each'
> /Library/Ruby/Gems/1.8/gems/mongrel-1.1/lib/mongrel/configurator.rb:281:in `run'
> /Library/Ruby/Gems/1.8/gems/merb-0.4.0/lib/merb/server.rb:557:in `cloaker_'
> /Library/Ruby/Gems/1.8/gems/mongrel-1.1/lib/mongrel/configurator.rb:50:in `call'
> /Library/Ruby/Gems/1.8/gems/mongrel-1.1/lib/mongrel/configurator.rb:50:in
> `initialize'
> /Library/Ruby/Gems/1.8/gems/merb-0.4.0/lib/merb/server.rb:548:in `new'
> /Library/Ruby/Gems/1.8/gems/merb-0.4.0/lib/merb/server.rb:548:in `mongrel_start'
> /Library/Ruby/Gems/1.8/gems/merb-0.4.0/lib/merb/server.rb:477:in `run'
> /Library/Ruby/Gems/1.8/gems/merb-0.4.0/bin/merb:6
> /usr/bin/merb:16:in `load'
> /usr/bin/merb:16
>
>
> Any suggestions?
>
> --
> Cheers,
>
> Kevin Williams
> http://kevwil.com
>

From ivey at gweezlebur.com  Wed Nov  7 10:39:55 2007
From: ivey at gweezlebur.com (Michael D. Ivey)
Date: Wed, 7 Nov 2007 09:39:55 -0600
Subject: what did I do?
In-Reply-To: <683a886f0711070736r4fa9548by601fa299cea1c801@mail.gmail.com>
References: <683a886f0711051934r21f1ee79td55776accb4e41d5@mail.gmail.com>
	<683a886f0711070736r4fa9548by601fa299cea1c801@mail.gmail.com>
Message-ID: <09F74F63-580F-44C0-ACFA-BFE7D22D77C8@gweezlebur.com>

On Nov 7, 2007, at 9:36 AM, Kevin Williams wrote:
> Seems to be fixed. Odd that I got no reply whatsoever, but hey, I'm
> glad it's fixed!

Yeah Ez pointed out your email right before we released.  I didn't  
recognize the error you were getting as a url() issue.  After someone  
else hit the same thing, I figured out what it was.

From luislavena at gmail.com  Thu Nov  8 08:04:53 2007
From: luislavena at gmail.com (Luis Lavena)
Date: Thu, 8 Nov 2007 10:04:53 -0300
Subject: Back to ruby2ruby?
Message-ID: <71166b3b0711080504q51618cb2xd3961bbf28539488@mail.gmail.com>

Hello guys.

I missed a lot of the development before 0.4 got out (btw, excellent work!).

I checkout last night the merb-trunk code and run the specs, found
that everything passes with 1 pending, but no errors or failures about
missing ruby2ruby.

There is a ticket (#274)[1] that discuss this, but I remember we
discussed this back in September about depending on RubyInline...

Also, the merb gem dependencies didn't state it, nor the documentation
as ticket mention...

ry dahl wrote a proof of concept extracting the count parameters part
of ParseTree, what happened to that?

Anyway, excellent progress and great looking site (still a few issues
with small fonts on Firefox ;-)

Regards,

[1] http://merb.devjavu.com/ticket/274
[2] http://rubyforge.org/pipermail/merb-devel/2007-September/000209.html

-- 
Luis Lavena
Multimedia systems
-
Leaders are made, they are not born. They are made by hard effort,
which is the price which all of us must pay to achieve any goal that
is worthwhile.
Vince Lombardi

From ivey at gweezlebur.com  Thu Nov  8 10:22:10 2007
From: ivey at gweezlebur.com (Michael D. Ivey)
Date: Thu, 8 Nov 2007 09:22:10 -0600
Subject: Back to ruby2ruby?
In-Reply-To: <71166b3b0711080504q51618cb2xd3961bbf28539488@mail.gmail.com>
References: <71166b3b0711080504q51618cb2xd3961bbf28539488@mail.gmail.com>
Message-ID: <8D490CB7-899A-4B89-BA50-D4B507360CC3@gweezlebur.com>

On Nov 8, 2007, at 7:04 AM, Luis Lavena wrote:
> I checkout last night the merb-trunk code and run the specs, found
> that everything passes with 1 pending, but no errors or failures about
> missing ruby2ruby.

It's only required if you want to use parameterized actions.

From ivey at gweezlebur.com  Mon Nov 12 14:33:54 2007
From: ivey at gweezlebur.com (Michael D. Ivey)
Date: Mon, 12 Nov 2007 13:33:54 -0600
Subject: Sessions broken in 0.4.0 (includes hotfix)
Message-ID: <D1BD2681-F11D-405A-9674-E444A2AAC27E@gweezlebur.com>

We've gotten several reports on #merb that each request gets a new  
session ID in 0.4.0.  I quickly tested and verified.  scoundrel  
tracked it down, and I think it's fixed.

If you need to apply this by hand to your 0.4.0 install, it's a 1  
line fix:


Index: /trunk/lib/merb/controller.rb
===================================================================
--- /trunk/lib/merb/controller.rb (revision 904)
+++ /trunk/lib/merb/controller.rb (revision 961)
@@ -12,5 +12,5 @@
      self._subclasses = []

-    self._session_id_key = :_session_id
+    self._session_id_key = '_session_id'
      self._session_expiry = Time.now + Merb::Const::WEEK * 2


This will be in 0.4.1 when it comes out RSN.

From ez at engineyard.com  Mon Nov 12 16:27:00 2007
From: ez at engineyard.com (Ezra Zygmuntowicz)
Date: Mon, 12 Nov 2007 13:27:00 -0800
Subject: merb/sequel mutex on or off? proxy_balancer or no?
In-Reply-To: <e9e3ad7b0711070705v768068i576c2be0bcd16c0f@mail.gmail.com>
References: <e9e3ad7b0711070705v768068i576c2be0bcd16c0f@mail.gmail.com>
Message-ID: <1E8CEEBF-2849-48BF-AD6F-F114E4AF7373@engineyard.com>

Dusty-

	I usually turn off the mutex and use evented mongrel. gem install  
swiftiply then when you start merb:

EVENT=1 merb -X off

Cheers-
-Ezra

On Nov 7, 2007, at 7:05 AM, Dusty Doris wrote:

> Just a quick question to those of you using merb and sequel or
> datamapper.  How do you setup your merb app?
>
> For example:
>
> merb -X off -c 5, then run mod_proxy_balancer or nginx to load balance
> between the 5 (or whatever #) of merb nodes?
>
> merb -X off -d, then simply use something like proxypass
>
> merb -c 5, keep the mutex lock and cluster it
>
> haproxy, swiftiply, evented mongrel......
>
>
> What I'm curious about is if you are using sequel, do you typically
> remove those mutex locks?  If so, is there a benefit still in running
> a clustered mongrel setup?  Just curious to how others are setting up
> their applications.  Right now I'm removing the mutex lock and setting
> up a cluster of 5 behind proxy_balancer and its working great.
> However, I was just curious to what others are doing.
>
> Thanks.  Its looking great.  I especially love the new provides :xml,
> :text, ... in the responder.  Really nice idea.
>
> -Dusty Doris
> _______________________________________________
> Merb-devel mailing list
> Merb-devel at rubyforge.org
> http://rubyforge.org/mailman/listinfo/merb-devel

Ezra Zygmuntowicz
Founder & Ruby Hacker
ezra at engineyard.com
EngineYard.com



From kevwil at gmail.com  Tue Nov 13 12:53:23 2007
From: kevwil at gmail.com (Kevin Williams)
Date: Tue, 13 Nov 2007 10:53:23 -0700
Subject: nitpicking :)
Message-ID: <683a886f0711130953o35d847dfhe2cdc930b6f6d083@mail.gmail.com>

The http://merbivore.com/get_merb.html#trunk page mentions the 0.4.0
version, but 0.4.1 is out. Is this a part of the site which will be
constantly updated as part of the release cycle?

The roadmap on merb.devjavu.com has outstanding tickets for 0.4.0 and
0.4.1, yet both versions are out. Were those issues fixed, or should
they be bumped forward?

Yeah, this is really anal crap, I know. I would be surprised if
sending this doesn't piss someone off. That said, some days the little
things are distracting enough to speak up about, you know?

--
Cheers,

Kevin Williams
http://kevwil.com

From ivey at gweezlebur.com  Tue Nov 13 13:25:37 2007
From: ivey at gweezlebur.com (Michael D. Ivey)
Date: Tue, 13 Nov 2007 12:25:37 -0600
Subject: nitpicking :)
In-Reply-To: <683a886f0711130953o35d847dfhe2cdc930b6f6d083@mail.gmail.com>
References: <683a886f0711130953o35d847dfhe2cdc930b6f6d083@mail.gmail.com>
Message-ID: <FF49CACC-3BAC-4E2F-9574-2DAB60F9A751@gweezlebur.com>

> The http://merbivore.com/get_merb.html#trunk page mentions the 0.4.0
> version, but 0.4.1 is out. Is this a part of the site which will be
> constantly updated as part of the release cycle?

Ez, who all has write to merbivore?

> The roadmap on merb.devjavu.com has outstanding tickets for 0.4.0 and
> 0.4.1, yet both versions are out. Were those issues fixed, or should
> they be bumped forward?

I've rolled them all to either 0.4.2 or Future, but others should  
feel free to disagree with my classification.  Some *may* be fixed  
already.


From ivey at gweezlebur.com  Thu Nov 15 21:26:47 2007
From: ivey at gweezlebur.com (Michael D. Ivey)
Date: Thu, 15 Nov 2007 20:26:47 -0600
Subject: Challenge
Message-ID: <FBBA3C6F-B38C-42DA-A84C-7B0A77B5965C@gweezlebur.com>

The original:
http://pastie.textmate.org/private/lqvrlyyvkv2kbugoxtiz6w

Send your submissions to me by email or on #merb.  I'll blog it, and  
pick a winner.  The winner does not mean it makes it into Merb, just  
means I personally like it best.

But it might make it into Merb.



# Scenario:
# You have a simple application that lists all the monkeys at the zoo.
# Every monkey has a name, and that name is unique.
# It is currently working.
#
# Your boss comes and asks you to change the "ugly" /monkeys/44 URLs
# to be "pretty" and "SEO optimized" URLs like /monkeys/koko, with the
# monkey's name in the URL.
#
# Challenge:
# Design your perfect API for doing this, using any combination of
# controller, model, and router changes.  Submit it as a diff of this  
pastie.
#
# Degree of difficulty:
# Don't complain that doing /monkeys/koko is dumb, or what happens
# when there are two monkeys named koko.
#
# Submit your diffs to #merb


# router:
r.resources :monkeys

class Monkey < MythicalORM

end


class Monkeys < Application
   def show(id)
     @monkey = Monkey.find(id)
   end

   def create(monkey)
     @monkey = Monkey.create(monkey)
     redirect url(:monkey, at monkey)
   end
end

From chris at oxdi.eu  Fri Nov 16 04:14:02 2007
From: chris at oxdi.eu (Chris Farmiloe)
Date: Fri, 16 Nov 2007 09:14:02 +0000
Subject: Challenge
In-Reply-To: <FBBA3C6F-B38C-42DA-A84C-7B0A77B5965C@gweezlebur.com>
References: <FBBA3C6F-B38C-42DA-A84C-7B0A77B5965C@gweezlebur.com>
Message-ID: <80175437-9079-460D-A1A8-EF38BF401806@oxdi.eu>

# psudo-untested-ruby-while-having-coffee code version by chrisfarms
#
# produces urls like /monkeys/3-your-keywords
#
# works for SEO, while maintaining all the benifits of a solid numeric  
ID
# plus points for SEO... if the keywords are changed, 302 redirects  
will automatically
# move the client to the right page.... solving the duplicate content  
issue or
# 404 errors after fixing a typo in a 'slug'
#
# controller code should be untouched
http://pastie.textmate.org/private/gcet24o9er4milpqemjjq


On 16 Nov 2007, at 02:26, Michael D. Ivey wrote:

> The original:
> http://pastie.textmate.org/private/lqvrlyyvkv2kbugoxtiz6w
>
> Send your submissions to me by email or on #merb.  I'll blog it, and
> pick a winner.  The winner does not mean it makes it into Merb, just
> means I personally like it best.
>
> But it might make it into Merb.
>
>
>
> # Scenario:
> # You have a simple application that lists all the monkeys at the zoo.
> # Every monkey has a name, and that name is unique.
> # It is currently working.
> #
> # Your boss comes and asks you to change the "ugly" /monkeys/44 URLs
> # to be "pretty" and "SEO optimized" URLs like /monkeys/koko, with the
> # monkey's name in the URL.
> #
> # Challenge:
> # Design your perfect API for doing this, using any combination of
> # controller, model, and router changes.  Submit it as a diff of this
> pastie.
> #
> # Degree of difficulty:
> # Don't complain that doing /monkeys/koko is dumb, or what happens
> # when there are two monkeys named koko.
> #
> # Submit your diffs to #merb
>
>
> # router:
> r.resources :monkeys
>
> class Monkey < MythicalORM
>
> end
>
>
> class Monkeys < Application
>   def show(id)
>     @monkey = Monkey.find(id)
>   end
>
>   def create(monkey)
>     @monkey = Monkey.create(monkey)
>     redirect url(:monkey, at monkey)
>   end
> end
> _______________________________________________
> Merb-devel mailing list
> Merb-devel at rubyforge.org
> http://rubyforge.org/mailman/listinfo/merb-devel

-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://rubyforge.org/pipermail/merb-devel/attachments/20071116/8724d694/attachment.html 

From B.Candler at pobox.com  Fri Nov 16 04:48:51 2007
From: B.Candler at pobox.com (Brian Candler)
Date: Fri, 16 Nov 2007 09:48:51 +0000
Subject: Challenge
In-Reply-To: <FBBA3C6F-B38C-42DA-A84C-7B0A77B5965C@gweezlebur.com>
References: <FBBA3C6F-B38C-42DA-A84C-7B0A77B5965C@gweezlebur.com>
Message-ID: <20071116094851.GA847@uk.tiscali.com>

On Thu, Nov 15, 2007 at 08:26:47PM -0600, Michael D. Ivey wrote:
> # Scenario:
> # You have a simple application that lists all the monkeys at the zoo.
> # Every monkey has a name, and that name is unique.
> # It is currently working.

Presumably it also has something like

  def index
    @monkeys = Monkey.find(:all)
  end

> # Your boss comes and asks you to change the "ugly" /monkeys/44 URLs
> # to be "pretty" and "SEO optimized" URLs like /monkeys/koko, with the
> # monkey's name in the URL.
> #
> # Challenge:
> # Design your perfect API for doing this

What do you mean by 'API' here? Or is the idea to explore all the
possibilities which might be implied?

As far as I can see:

(1) you could use an ORM which permits use of an arbitrary field
    as the primary key; in this case only the model would need to change.

class Monkey < MythicalORM
  self.primary_key = 'name'
end

(In the case of ActiveRecord, I believe that non-integer ID fields are not
supported. Also, if you intend retaining the old integer 'id' column in
parallel with the monkey name primary key, beware that AR may no longer
generate the next integer id for you automatically, so you may need to do
this explicitly in a before_create hook)

(2) modify Merb so the url() helper can use an arbitrary model field as the
resource id, e.g.

r.resources :monkeys, :resource_id => :name

class Monkeys < Application
  def show(id)
    @monkey = Monkey.find_by_name(id) || (raise "No such monkey")
  end
end

The :resource_id is used by the url() helper to pick the correct attribute
to put into the URL.

But if the idea is to add some feature to Merb for this, I would suggest
first a generalisation of the problem to allow for composite primary keys.
e.g. if each monkey has a first name and last name, and the tuple (first
name, last name) is unique, then allow resource URLs of the form

    /monkeys/koko/pops
or  /monkeys/koko;pops

This introduces further difficulties (*), but is realistic when wanting to
map a web application onto an existing schema which was designed in this
way.

(*) in particular: the database may have primary key tuples of the form
("","foo") and/or (null,"foo")

> # Don't complain that doing /monkeys/koko is dumb, or what happens
> # when there are two monkeys named koko

Or what happens when a monkey's name contains a forward slash. (If you
replace monkeys with car parts, IDs of the form xxx/yyy are not uncommon;
with Rails at least, even encoding as %2F is treated as a path separator)

Which reminds me of: http://xkcd.com/327/

Incidentally, when I wanted to do something similar to this in a Rails app,
I ended up using URLs of the form /monkeys?name=koko

Then if a single monkey matches, I redirect to /monkeys/44; if multiple
monkeys match, I list them. Something like this:

  MAX_RESULT = 10

  def index
    if params[:name]
      @monkeys = Monkey.find(:all, :conditions=>['name=?',params[:name]],
	:limit => MAX_RESULT, :order => 'name')
      render_monkeys('Monkeys by name')
    else
      @monkeys = Monkey.find(:all)
    end
  end

  def render_monkeys(title = 'Monkeys')
    @title = title
    if @monkeys.size == 1
      redirect_to monkey_url(@monkeys.first)
    elsif @monkey.size == 0
      flash.now[:notice] = 'Monkey not found'
    else if @monkeys.size == MAX_RESULT
      flash.now[:notice] = "Only the first #{MAX_RESULT} monkeys shown"
    end
  end

You can extend the 'if' statement in index to allow for other ways of
referring to the monkey, e.g. /monkeys?asset_tag=12345.

Now, this suggests a question back to the boss: after the proposed change of
URLs, what happens if you want to fetch a monkey given its original ID of
44?

Would we need to provide a second URL of the form /monkeys?id=44 ? Or will
the old ID just become a normal data attribute? Or is this change of
"primary key" going to be so pervasive that we will delete the old id column
from the database entirely?

If multiple 'primary keys' are to be exposed, you could argue that this
should be done orthogonally, e.g.

   /monkeys/id/44
   /monkeys/name/koko
   /monkeys/asset_tag/12345

But this still begs the question of which form will be generated by the
url() helper. So ultimately one of these is going to take precedence over
all the others.

Regards,

Brian.

From ivey at gweezlebur.com  Fri Nov 16 09:52:13 2007
From: ivey at gweezlebur.com (Michael D. Ivey)
Date: Fri, 16 Nov 2007 08:52:13 -0600
Subject: Challenge
In-Reply-To: <80175437-9079-460D-A1A8-EF38BF401806@oxdi.eu>
References: <FBBA3C6F-B38C-42DA-A84C-7B0A77B5965C@gweezlebur.com>
	<80175437-9079-460D-A1A8-EF38BF401806@oxdi.eu>
Message-ID: <D37DA473-BCA0-425F-8E43-D922B04C7A4A@gweezlebur.com>

On Nov 16, 2007, at 3:14 AM, Chris Farmiloe wrote:
> http://pastie.textmate.org/private/gcet24o9er4milpqemjjq


Thanks, Chris, got it on the list.

I wasn't going to comment on entries, but since this is only a sorta- 
contest (there will be a prize!) I figured it was OK.

I had never thought of pushing it all into the model.  It may be a  
"violation of MVC" to raise controller exceptions from inside models,  
but is that a bad thing?  Raising "Moved" with the new name is  
excellent.

So I guess the MVC purist in me was horrified, and the Resource guy  
in me thought this was super sexy.

Have you used a pattern like this in production yet?

BTW:  Merb already defines MovedPermanently and MovedTemporarily, but  
they don't take the arg for the new URL.  We should add that:
    raise MovedPermanently.new(new_url)


From chris at oxdi.eu  Fri Nov 16 10:15:41 2007
From: chris at oxdi.eu (Chris Farmiloe)
Date: Fri, 16 Nov 2007 15:15:41 +0000
Subject: Challenge
In-Reply-To: <D37DA473-BCA0-425F-8E43-D922B04C7A4A@gweezlebur.com>
References: <FBBA3C6F-B38C-42DA-A84C-7B0A77B5965C@gweezlebur.com>
	<80175437-9079-460D-A1A8-EF38BF401806@oxdi.eu>
	<D37DA473-BCA0-425F-8E43-D922B04C7A4A@gweezlebur.com>
Message-ID: <EEB44A1A-6287-4B1A-8456-4A589391359C@oxdi.eu>


On 16 Nov 2007, at 14:52, Michael D. Ivey wrote:
> I had never thought of pushing it all into the model.  It may be a
> "violation of MVC" to raise controller exceptions from inside models,
> but is that a bad thing?  Raising "Moved" with the new name is
> excellent.

I've been playing with a fork of merb for one production app, and for  
me ControllerExceptions are more like ApplicationExceptions not  
specific to controller-land

> Have you used a pattern like this in production yet?

On a few rails apps, I've used something similar, although without  
merb's exception controllers its a bit messy.

try:

http://www.africacollection.co.uk/pages/1-holidays-to-southern-africa-the-indian-ocean

then

http://www.africacollection.co.uk/pages/1-holidays-to-southern-africa-blah-blah-blah


> BTW:  Merb already defines MovedPermanently and MovedTemporarily, but
> they don't take the arg for the new URL.  We should add that:
>    raise MovedPermanently.new(new_url)

A built in handler for the Moved__ error classes would be nice too.

raise MovedTemporaily.new(url)

would be roughly equal to

redirect(url) and throw(:halt)


-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://rubyforge.org/pipermail/merb-devel/attachments/20071116/7f7c3761/attachment-0001.html 

From ivey at gweezlebur.com  Fri Nov 16 10:19:11 2007
From: ivey at gweezlebur.com (Michael D. Ivey)
Date: Fri, 16 Nov 2007 09:19:11 -0600
Subject: Challenge
In-Reply-To: <20071116094851.GA847@uk.tiscali.com>
References: <FBBA3C6F-B38C-42DA-A84C-7B0A77B5965C@gweezlebur.com>
	<20071116094851.GA847@uk.tiscali.com>
Message-ID: <061D0910-BF8D-48B0-B6D5-488C36D07847@gweezlebur.com>

On Nov 16, 2007, at 3:48 AM, Brian Candler wrote:
> Presumably it also has something like
>   def index
>     @monkeys = Monkey.find(:all)
>   end

Yep just left them out for brevity.  They're the standard CRUD actions.

> What do you mean by 'API' here? Or is the idea to explore all the
> possibilities which might be implied?

By "Perfect API" I meant, "Assume Merb is your Perfect Framework, and  
works exactly as it should.
What would you have to change in your app, using Merb's perfect API,  
to accomplish your task?"

> (2) modify Merb so the url() helper can use an arbitrary model  
> field as the
> resource id, e.g.
>
> r.resources :monkeys, :resource_id => :name

You're at least the third person to suggest attaching this to the  
routes, so that seems like an approach worth exploring for Merb.


> But if the idea is to add some feature to Merb for this

Well, it's both.  Merb needs some way to handle not using the id  
attribute of a model in URLs.  Especially in the case where the  
models aren't even database backed.  But this is also a contest, for  
making up cool and pretty APIs.


> I would suggest first a generalisation of the problem to allow for  
> composite primary keys.
> e.g. if each monkey has a first name and last name, and the tuple  
> (first name, last name) is unique, then allow resource URLs of the  
> form
>
>     /monkeys/koko/pops
> or  /monkeys/koko;pops

Tuples as URL segments...interesting.  My brain immediately went to
/monkeys/[koko,pops] but that's not legal.  After spending too much  
time in the spec when I should be getting ready for work, I came up  
with this:

/monkeys;koko,pops

; signifies params for a path segment....so it says "the collection  
of monkeys, limited by this tuple" ... legal, semantic....kinda  
pretty if you're a URI geek, too.

I can't decide without more linking and parsing if / 
monkeys/;koko,pops would also be legal.


Anyway, interesting stuff, thanks for the email.  I didn't see an  
official entry into the contest, though.  :)

From B.Candler at pobox.com  Fri Nov 16 11:16:32 2007
From: B.Candler at pobox.com (Brian Candler)
Date: Fri, 16 Nov 2007 16:16:32 +0000
Subject: Challenge
In-Reply-To: <061D0910-BF8D-48B0-B6D5-488C36D07847@gweezlebur.com>
References: <FBBA3C6F-B38C-42DA-A84C-7B0A77B5965C@gweezlebur.com>
	<20071116094851.GA847@uk.tiscali.com>
	<061D0910-BF8D-48B0-B6D5-488C36D07847@gweezlebur.com>
Message-ID: <20071116161632.GA22771@uk.tiscali.com>

On Fri, Nov 16, 2007 at 09:19:11AM -0600, Michael D. Ivey wrote:
> Tuples as URL segments...interesting.  My brain immediately went to
> /monkeys/[koko,pops] but that's not legal.  After spending too much  
> time in the spec when I should be getting ready for work, I came up  
> with this:
> 
> /monkeys;koko,pops
> 
> ; signifies params for a path segment....so it says "the collection  
> of monkeys, limited by this tuple" ... legal, semantic....kinda  
> pretty if you're a URI geek, too.

FWIW, Rails have dropped their use of the semicolon, apparently because of
some compatibility problems. i.e. they now use /monkeys/123/squawk instead
of /monkeys/123;squawk

See:
http://weblog.rubyonrails.org/2007/10/5/rails-1-2-4-maintenance-release

Given that a tuple is always of fixed size, I would be inclined towards
/monkeys/koko/pops, if only because (a) in the degenerate case of the
single-key tuple it's the same as now, and (b) it doesn't introduce any
more characters which need escaping if they occur within the key columns.

Given a composite key, perhaps the router should set the 'id' parameter to
an array? This would be convenient if the ORM supports composite keys in
this way, as you could still do

    Monkey.find(id)       # id = ['koko','pops']

I don't know which, if any, ORMs would support this. But in any case you can
always do

    Monkey.find_by_firstname_and_lastname(*id) || (raise "No banana")

But in any case maybe it's better to name the key components explicitly,
e.g.

    r.resources :monkeys, :resource_id=>[:firstname,:lastname]

which could just build a pattern of monkeys/:firstname/:lastname and the
routing code wouldn't have to change.

> Anyway, interesting stuff, thanks for the email.  I didn't see an  
> official entry into the contest, though.  :)

I don't know what the prize is, but I can probably live without it :-)

Regards,

Brian.

From ivey at gweezlebur.com  Fri Nov 16 11:20:23 2007
From: ivey at gweezlebur.com (Michael D. Ivey)
Date: Fri, 16 Nov 2007 10:20:23 -0600
Subject: Challenge
In-Reply-To: <EEB44A1A-6287-4B1A-8456-4A589391359C@oxdi.eu>
References: <FBBA3C6F-B38C-42DA-A84C-7B0A77B5965C@gweezlebur.com>
	<80175437-9079-460D-A1A8-EF38BF401806@oxdi.eu>
	<D37DA473-BCA0-425F-8E43-D922B04C7A4A@gweezlebur.com>
	<EEB44A1A-6287-4B1A-8456-4A589391359C@oxdi.eu>
Message-ID: <1B33FCC6-3337-442A-8C63-B71AB0D2BD97@gweezlebur.com>

On Nov 16, 2007, at 9:15 AM, Chris Farmiloe wrote:
>> BTW:  Merb already defines MovedPermanently and MovedTemporarily, but
>> they don't take the arg for the new URL.  We should add that:
>>    raise MovedPermanently.new(new_url)
>
> A built in handler for the Moved__ error classes would be nice too.
>
> raise MovedTemporaily.new(url)
>
> would be roughly equal to
>
> redirect(url) and throw(:halt)

Yep, that should be in the default generated Exceptions controller.

What sensible default should MovedPermanently use for the url?  I  
guess just nil, and if you don't specify one moved_permanently would  
display the error.

I'll work on this next week if no one else does.


From chris at oxdi.eu  Fri Nov 16 11:29:37 2007
From: chris at oxdi.eu (Chris Farmiloe)
Date: Fri, 16 Nov 2007 16:29:37 +0000
Subject: uploads and swiftiply
Message-ID: <1BD6F334-513E-4D7A-984D-29B6107B10C0@oxdi.eu>


This is actually REALLY obvious, but I just thought i'd post it up  
here since it took a few brain clicks to dawn on me.



Evented mongrels don't play nice with upload-progress systems.



(although with the new nginx patch and a couple of mongrels it  
wouldn't be so bad.)

</public-service-announcement>



From B.Candler at pobox.com  Fri Nov 16 16:00:06 2007
From: B.Candler at pobox.com (Brian Candler)
Date: Fri, 16 Nov 2007 21:00:06 +0000
Subject: Challenge
In-Reply-To: <20071116161632.GA22771@uk.tiscali.com>
References: <FBBA3C6F-B38C-42DA-A84C-7B0A77B5965C@gweezlebur.com>
	<20071116094851.GA847@uk.tiscali.com>
	<061D0910-BF8D-48B0-B6D5-488C36D07847@gweezlebur.com>
	<20071116161632.GA22771@uk.tiscali.com>
Message-ID: <20071116210006.GA4435@uk.tiscali.com>

I just remembered a major problem with the boss' suggestion: what if a
monkey's name collides with a controller action?

For example, is /monkeys/new a 'new' action on the whole collection, or a
'show' action on an individual monkey whose name is 'new'?

This pretty much *requires* IDs to be numeric, unless we accept a further
rejigging of URLs, such as

    /monkeys/new
    /monkeys/show/koko
    /monkeys/edit/koko

or

    /monkeys/*/new
    /monkeys/koko
    /monkeys/koko/edit

or even

    /monkeys/new
    /monkey/koko
    /monkey/koko/edit

The first of these most closely retains the controller/action semantic, but
does not match the boss' URL spec, and it moves away fm the idea of an
object as a 'resource' like a file in a filesystem.

For another example of ambiguity, have a look at this bit of routing code,
and consider what happens if a monkey's name is 'edit':

        next_level.match("/:id/edit").to_route.name(:"edit_#{route_singular_name}")
        next_level.match("/:action/:id").to_route.name(:"custom_#{route_singular_name}")

(Unlikely? Well 'Edith' is a name, so I can imagine that 'Edit' might be a
name in some other languages :-)

I think these problems go away if semicolons are required for named actions,
i.e.

    /monkeys;new
    /monkeys/koko
    /monkeys/koko;edit

but at the moment Merb and Rails accept both ; and /

Regards,

Brian.

From ivey at gweezlebur.com  Fri Nov 16 18:41:29 2007
From: ivey at gweezlebur.com (Michael D. Ivey)
Date: Fri, 16 Nov 2007 17:41:29 -0600
Subject: Challenge
In-Reply-To: <20071116210006.GA4435@uk.tiscali.com>
References: <FBBA3C6F-B38C-42DA-A84C-7B0A77B5965C@gweezlebur.com>
	<20071116094851.GA847@uk.tiscali.com>
	<061D0910-BF8D-48B0-B6D5-488C36D07847@gweezlebur.com>
	<20071116161632.GA22771@uk.tiscali.com>
	<20071116210006.GA4435@uk.tiscali.com>
Message-ID: <648E28BC-2812-4130-9909-8804B5C55D13@gweezlebur.com>

>     /monkeys/new
>     /monkey/koko
>     /monkey/koko/edit

I've always preferred this style.  The collection resource and the  
individual resource are different things.

What about:

Collection:     /monkeys
New form:       /new-monkey
Member:         /monkey/koko
Edit form:      /monkey-editor/koko


We're way off track from the Challenge, but I love talking about URLs.

From ethan at echoditto.com  Mon Nov 26 00:21:08 2007
From: ethan at echoditto.com (Ethan Winn)
Date: Mon, 26 Nov 2007 00:21:08 -0500
Subject: Best Way to Write Helpers Using Markaby?
Message-ID: <14AF6FEA-1DE3-4F4D-9BD9-3C6E99ED6142@echoditto.com>

Hey there Merbers,

I'm a Merb neophyte checking it out with a simple app exploration  
after being inspired by Mr. Katz's presentation @ JQueryCamp and with  
the desire to use Markaby if at all possible (I heart camping). I've  
done a little exploration of how to use Markaby in helper code, but  
any guidance would definitely be appreciated (I definitely do not  
understand the "Merb way").

At first I tried just putting the `tag_name do` blocks right in my  
helpers, but this didn't work out.  What I've found does work is  
wrapping this code in `Markaby::Builder.new.html do` blocks, such as  
this:

     def tasks_table tasks
       Markaby::Builder.new.html do
         p "update tasks below"
         table do
           tasks.each do |task|
             task_row(task)
           end
         end
       end
     end

While this works, I'm pretty sure that making a new builder instance  
for each helper call.

What's the "right" way?

Thanks.

--
Ethan Winn
ethan at echoditto.com



-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://rubyforge.org/pipermail/merb-devel/attachments/20071126/07cd40a9/attachment.html 

From mattaimonetti at gmail.com  Mon Nov 26 01:56:32 2007
From: mattaimonetti at gmail.com (Matt Aimonetti)
Date: Sun, 25 Nov 2007 22:56:32 -0800
Subject: transition form respond_to to provides
Message-ID: <f58d8dcb0711252256s7cfd793bk21f709c12a7a9854@mail.gmail.com>

I really like the new mime type handler however I was trying to convert some
code and couldn't figure out how.

My rails action was handle 2 formats, html and xml

      respond_to do |format|
        format.html # show.html.erb
        format.xml do
          if @account.active? && @account.valid_referer?(session[:referer])
            render :xml => @account.to_xml(:except => [:id, :created_at])
         else
            render :text => 'your account has been disable'
         end
       end

Now that I'm trying to switch this controller to Merb, I can't figure how to
have a custom xml builder and how to have a condition to the mime type
render.

Any help would be greatly appreciated.

Thanks

-Matt
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://rubyforge.org/pipermail/merb-devel/attachments/20071125/30d8c973/attachment.html 

From ivey at gweezlebur.com  Mon Nov 26 10:05:39 2007
From: ivey at gweezlebur.com (Michael D. Ivey)
Date: Mon, 26 Nov 2007 09:05:39 -0600
Subject: transition form respond_to to provides
In-Reply-To: <f58d8dcb0711252256s7cfd793bk21f709c12a7a9854@mail.gmail.com>
References: <f58d8dcb0711252256s7cfd793bk21f709c12a7a9854@mail.gmail.com>
Message-ID: <27DFEEB7-5F17-4B6D-86AE-BDDFC41B3FA1@gweezlebur.com>

>       respond_to do |format|
>         format.html # show.html.erb
>         format.xml do
>           if @account.active? && @account.valid_referer?(session 
> [:referer])
>             render :xml => @account.to_xml(:except =>  
> [:id, :created_at])
>          else
>             render :text => 'your account has been disable'
>          end
>        end


Here's an untested quick rewrite.

provides :xml
if content_type == :xml
   if @account.active? && @account.valid_referer?(session[:referer])
     render :xml => @account.to_xml(:except => [:id, :created_at])
   else
     "Your account has been disabled."
   end
else
   render
end

From wayneeseguin at gmail.com  Mon Nov 26 11:48:38 2007
From: wayneeseguin at gmail.com (Wayne E. Seguin)
Date: Mon, 26 Nov 2007 11:48:38 -0500
Subject: [ANN] WNY Ruby Users Group, first meeting
Message-ID: <ef3e5d50711260848v6f11cfb1k1c43f75a8b64a007@mail.gmail.com>

Hello friends,

When: Nov 26th Meetup (Wed 7pm)

Where: Spot Coffee in downtown Buffalo, side room
( http://maps.google.com/maps?f=q&hl=en&geocode=&time=&date=&ttype=&q=227+Delaware+Ave,+Buffalo,+Erie,+New+York+14202,+United+States&sll=42.890586,-78.876839&sspn=0.05886,0.090294&ie=UTF8&z=17&iwloc=addr&om=1
)

Duration: 1-2 hours

http://www.wnyruby.com/

I hope to see you all there!

  ~Wayne

From john at smokinggun.com  Mon Nov 26 13:47:09 2007
From: john at smokinggun.com (John Weir)
Date: Mon, 26 Nov 2007 13:47:09 -0500
Subject: Loading a specific version of MERB gem
Message-ID: <D7FD1D81-A7BF-44AC-8CDE-F7A7D20C360F@smokinggun.com>

greetings,

i have a project based around merb 0.3.7 and I do not want to fiddle  
with it one bit.  and now I want to start a new project running  
0.4.2(and beyond).  what is the recommend method of starting merb from  
a specific gem version?

thanks - jw



From B.Candler at pobox.com  Mon Nov 26 15:03:31 2007
From: B.Candler at pobox.com (Brian Candler)
Date: Mon, 26 Nov 2007 20:03:31 +0000
Subject: Loading a specific version of MERB gem
In-Reply-To: <D7FD1D81-A7BF-44AC-8CDE-F7A7D20C360F@smokinggun.com>
References: <D7FD1D81-A7BF-44AC-8CDE-F7A7D20C360F@smokinggun.com>
Message-ID: <20071126200331.GA27554@uk.tiscali.com>

On Mon, Nov 26, 2007 at 01:47:09PM -0500, John Weir wrote:
> i have a project based around merb 0.3.7 and I do not want to fiddle  
> with it one bit.  and now I want to start a new project running  
> 0.4.2(and beyond).  what is the recommend method of starting merb from  
> a specific gem version?

rubygems provides such a mechanism: try

   merb _0.3.7_ [any other args]

Type the underscores literally - they are not just there for emphasis :-)

Here's how it works:

$ cat /usr/bin/merb
#!/usr/bin/env ruby
#
# This file was generated by RubyGems.
#
# The application 'merb' is installed as part of a gem, and
# this file is here to facilitate running it.
#

require 'rubygems'
version = "> 0"
if ARGV.first =~ /^_(.*)_$/ and Gem::Version.correct? $1 then
  version = $1
  ARGV.shift
end
gem 'merb', version
load 'merb'

From gthoppae at gmail.com  Fri Nov 30 05:12:16 2007
From: gthoppae at gmail.com (Gnanasekaran Thoppae)
Date: Fri, 30 Nov 2007 11:12:16 +0100
Subject: Running merb app via fcgi
Message-ID: <1AA6EBC0-AC34-40B3-962C-E5714D7DA46A@gmail.com>

Hi all,

I would like to know if there is a .htaccess template
config to run Merb apps via fcgi under apache.

If yes, where I can I find it?

Thanks.

-Gnana