Rails - Apr 2005 - Using 3rd party libraries with Rails and Typo (uninitialized constants)

So I''ve got a typo blog based on the current head and everything is set
up and working great. However, I wanted to give RubyPants 
(http://chneukirchen.org/blog/static/projects/rubypants.html) a try to 
"educate" the quotes in my posts and convert hypens to em dashes and 
such.

I dropped rubypants.rb in my typo installation''s lib directory (the top
level one, along with html_engine.rb and friends) and, just for testing 
purposes, modified the _articles.rhtml partial so that instead of

<%= article.body_html %>

it''s

<%= RubyPants.new(article.body_html).to_html %>

But, when I do this, I get an Exception that RubyPants is an 
unitialized constant. Isn''t stuff in "lib" supposed to be
automagically
loaded?

Help!

- Sean

On Apr 11, 2005 9:34 PM, Sean Santry
<rails-6SqCC7WXOXpVtazZjgo0Tw@public.gmane.org>
wrote:
> So I''ve got a typo blog based on the current head and everything
is set
> up and working great. However, I wanted to give RubyPants
> (http://chneukirchen.org/blog/static/projects/rubypants.html) a try to
> "educate" the quotes in my posts and convert hypens to em dashes
and
> such.
> 
> I dropped rubypants.rb in my typo installation''s lib directory
(the top
> level one, along with html_engine.rb and friends) and, just for testing
> purposes, modified the _articles.rhtml partial so that instead of
> 
> <%= article.body_html %>
> 
> it''s
> 
> <%= RubyPants.new(article.body_html).to_html %>
> 
> But, when I do this, I get an Exception that RubyPants is an
> unitialized constant. Isn''t stuff in "lib" supposed to
be automagically
> loaded?
> 
> Help!
> 
> - Sean
> 
> _______________________________________________
> Rails mailing list
> Rails-1W37MKcQCpIf0INCOvqR/iCwEArCW2h5@public.gmane.org
> http://lists.rubyonrails.org/mailman/listinfo/rails
> 
Hi Sean

It''s automatically _available_ to be loaded/required, but is not
automatically loaded.  If you''ll be using this module everywhere in
your app, just add a require line to your application controller file;
otherwise, just add it where necessary.

Dave

-- 
Dave Goodlad
dgoodlad-Re5JQEeQqe8AvxtiuMwx3w@public.gmane.org or
dave-eHfbeeWWzZOw5LPnMra/2Q@public.gmane.org
http://david.goodlad.ca/

On Apr 11, 2005, at 11:37 PM, David Goodlad wrote:
> It''s automatically _available_ to be loaded/required, but is not
> automatically loaded.  If you''ll be using this module everywhere
in
> your app, just add a require line to your application controller file;
> otherwise, just add it where necessary.
Thanks for the quick reply David (twice in one day!). I added

require ''rubypants''

to my application.rb but no luck--same error. I also tried adding the 
require directly to the articles_controller.rb....

- Sean

On Apr 11, 2005, at 11:51 PM, Sean Santry wrote:
> On Apr 11, 2005, at 11:37 PM, David Goodlad wrote:
>
>> It''s automatically _available_ to be loaded/required, but is
not
>> automatically loaded.
>
> I also tried adding the require directly to the 
> articles_controller.rb....

Interesting. To get this working, after I killed the lighttpd instance 
under which the application was running, I had to remove the fcgi 
socket file the application was using. Once I did that, it started 
working.... bizarre...

- Sean

Sean,

On 12.4.2005, at 07:20, Sean Santry wrote:
>
>
> Interesting. To get this working, after I killed the lighttpd instance 
> under which the application was running, I had to remove the fcgi 
> socket file the application was using. Once I did that, it started 
> working.... bizarre...
>
Are you working in production environment? I got bitten by this once, 
too, until I remembered to delete the socket files (=kill the fcgi 
processes). That''s not so bizarre, actually. lighttpd just starts the 
processes and knows which sockets to use for a given app. Other than, 
they live their own life and keep your Rails app running even after 
lighttpd is killed.

Note that this should not be the case when you are running your app in 
development mode. You can select the mode e.g. in your lighttpd config 
file:

fastcgi.server = ( ".fcgi" =>
   ( "localhost" =>
       (
         "min-procs" => 5,
         "max-procs" => 5,
         "socket"   => "/tmp/diary.fcgi.socket",
         "bin-path" =>
"/Users/jarkko/Sites/diary/public/dispatch.fcgi",
         "bin-environment" => ( "RAILS_ENV" =>
"development" )
       )
   )
)

//jarkko
> - Sean
>
> _______________________________________________
> Rails mailing list
> Rails-1W37MKcQCpIf0INCOvqR/iCwEArCW2h5@public.gmane.org
> http://lists.rubyonrails.org/mailman/listinfo/rails
>
--
Jarkko Laine
http://jlaine.net
http://odesign.fi


_______________________________________________
Rails mailing list
Rails-1W37MKcQCpIf0INCOvqR/iCwEArCW2h5@public.gmane.org
http://lists.rubyonrails.org/mailman/listinfo/rails

On Apr 12, 2005, at 3:25 AM, Jarkko Laine wrote:
> On 12.4.2005, at 07:20, Sean Santry wrote:
>>
>> Interesting. To get this working, after I killed the lighttpd 
>> instance under which the application was running, I had to remove the 
>> fcgi socket file the application was using. Once I did that, it 
>> started working.... bizarre...
>
> Are you working in production environment? I got bitten by this once, 
> too, until I remembered to delete the socket files (=kill the fcgi 
> processes).
Yes, that definitely explains it. I was working under the assumption 
that killing lighttpd also killed the associated fcgi processes. I 
didn''t realize that the fcgi processes lived until the socket files 
themselves were removed.

Thanks for the explanation!

- Sean