<?xml version="1.0" encoding="utf-8"?>
<rss xmlns:atom="http://www.w3.org/2005/Atom" version="2.0"><channel><title>Disqus - Latest Comments for leethal</title><link>http://disqus.com/by/leethal/</link><description></description><atom:link href="http://disqus.com/leethal/comments.rss" rel="self"></atom:link><language>en</language><lastBuildDate>Mon, 12 Oct 2009 09:01:14 -0000</lastBuildDate><item><title>Re: August Lilleaas</title><link>http://august.lilleaas.net/s3_upload_snow_leopard_service#comment-19892689</link><description>&lt;p&gt;I have no idea how you were able to do this. When I run the script, it asks&lt;br&gt;me for a bucket name. The only explanation I can think of is that you have a&lt;br&gt;~/.s3_service_bucket file on your system already (the file where the script&lt;br&gt;stores the bucket name), containing the string "public".&lt;/p&gt;</description><dc:creator xmlns:dc="http://purl.org/dc/elements/1.1/">August Lilleaas</dc:creator><pubDate>Mon, 12 Oct 2009 09:01:14 -0000</pubDate></item><item><title>Re: August Lilleaas</title><link>http://august.lilleaas.net/s3_upload_snow_leopard_service#comment-16169588</link><description>&lt;p&gt;Perhaps the input got garbled somehow? Open the keychain, look for a key named "s3_service". Open it and double check if the credentials are OK.&lt;/p&gt;</description><dc:creator xmlns:dc="http://purl.org/dc/elements/1.1/">August Lilleaas</dc:creator><pubDate>Tue, 08 Sep 2009 01:25:28 -0000</pubDate></item><item><title>Re: August Lilleaas</title><link>http://august.lilleaas.net/s3_upload_snow_leopard_service#comment-15757314</link><description>&lt;p&gt;Your ~/.s3_service_bucket is probably blank. That's where the script stores the bucket name. I just tested it locally with a blank ~/.s3_service_bucket and got the same error message you did.&lt;/p&gt;&lt;p&gt;Added a fix for this, download the latest version above (v3), and try again!&lt;/p&gt;</description><dc:creator xmlns:dc="http://purl.org/dc/elements/1.1/">August Lilleaas</dc:creator><pubDate>Wed, 02 Sep 2009 08:55:31 -0000</pubDate></item><item><title>Re: August Lilleaas</title><link>http://august.lilleaas.net/s3_upload_snow_leopard_service#comment-15756443</link><description>&lt;p&gt;Perhaps the input is off somehow. Open the service in automator, replace the shell script with `puts STDIN.read.inspect`, and paste what you get in the clipboard here. It might contain your S3 access credentials, so be sure to censor those before you paste it.&lt;/p&gt;</description><dc:creator xmlns:dc="http://purl.org/dc/elements/1.1/">August Lilleaas</dc:creator><pubDate>Wed, 02 Sep 2009 08:24:17 -0000</pubDate></item><item><title>Re: August Lilleaas</title><link>http://august.lilleaas.net/s3_upload_snow_leopard_service#comment-15756013</link><description>&lt;p&gt;Open a terminal and run these two commands:&lt;/p&gt;&lt;p&gt;# curl -O &lt;a href="http://pastie.org/602867.txt" rel="nofollow noopener" target="_blank" title="http://pastie.org/602867.txt"&gt;http://pastie.org/602867.txt&lt;/a&gt;&lt;br&gt;# ruby 602867.txt&lt;/p&gt;&lt;p&gt;What output do you get? Open the .txt file and try a few different values on the `input` part on the top. E.g your actual access key and so on.&lt;/p&gt;</description><dc:creator xmlns:dc="http://purl.org/dc/elements/1.1/">August Lilleaas</dc:creator><pubDate>Wed, 02 Sep 2009 08:06:18 -0000</pubDate></item><item><title>Re: August Lilleaas</title><link>http://august.lilleaas.net/s3_upload_snow_leopard_service#comment-15629096</link><description>&lt;p&gt;Whoops, thanks for pointing this out. Updated the article accordingly.&lt;/p&gt;</description><dc:creator xmlns:dc="http://purl.org/dc/elements/1.1/">August Lilleaas</dc:creator><pubDate>Mon, 31 Aug 2009 01:17:37 -0000</pubDate></item><item><title>Re: August Lilleaas</title><link>http://august.lilleaas.net/s3_upload_snow_leopard_service#comment-15618161</link><description>&lt;p&gt;I noticed this when testing. Didn't find a way to set the strata/level  &lt;br&gt;of the dialogue. Will investigate though.&lt;/p&gt;</description><dc:creator xmlns:dc="http://purl.org/dc/elements/1.1/">August Lilleaas</dc:creator><pubDate>Sun, 30 Aug 2009 19:26:02 -0000</pubDate></item><item><title>Re: Authlogic and rescue_from ActionController::RoutingError</title><link>http://www.mrkris.com/2009/08/21/authlogic-and-rescue_from-actioncontroller-routingerror/#comment-15261896</link><description>&lt;p&gt;Thanks! Found ya on google, had the same problem when rescue_from-ing to handle 404's.&lt;/p&gt;</description><dc:creator xmlns:dc="http://purl.org/dc/elements/1.1/">August Lilleaas</dc:creator><pubDate>Sun, 23 Aug 2009 11:15:34 -0000</pubDate></item><item><title>Re: August Lilleaas</title><link>http://august.lilleaas.net/procs_and_blocks_and_anonymous_functions#comment-15255496</link><description>&lt;p&gt;Well, it's not really a reference to that method, it creates a proc.&lt;/p&gt;&lt;p&gt;&lt;/p&gt;&lt;pre&gt;module Foo&lt;br&gt;  def self.bar&lt;br&gt;  end&lt;br&gt;end&lt;br&gt;&lt;br&gt;p Foo.method(:bar).object_id&lt;br&gt;# =&amp;gt; 85280&lt;br&gt;p Foo.method(:bar).object_id&lt;br&gt;# =&amp;gt; 85220&lt;/pre&gt;&lt;p&gt;&lt;/p&gt;&lt;p&gt;So even though the functionality is very similar, it isn't a reference to a pointer or anything like that; it converts the method into a proc every time you call Object#method.&lt;/p&gt;</description><dc:creator xmlns:dc="http://purl.org/dc/elements/1.1/">August Lilleaas</dc:creator><pubDate>Sun, 23 Aug 2009 05:16:28 -0000</pubDate></item><item><title>Re: August Lilleaas</title><link>http://august.lilleaas.net/where_to_put_those_codes#comment-14436689</link><description>&lt;p&gt;D'oh, you were talking about include, not require. I apologize for the  &lt;br&gt;brainfart.&lt;/p&gt;</description><dc:creator xmlns:dc="http://purl.org/dc/elements/1.1/">August Lilleaas</dc:creator><pubDate>Fri, 07 Aug 2009 12:25:47 -0000</pubDate></item><item><title>Re: August Lilleaas</title><link>http://august.lilleaas.net/where_to_put_those_codes#comment-14435374</link><description>&lt;p&gt;That's wrong. When you refer to `MyModule` of `lib/my_module.rb` in `app/models/my_model.rb`, Rails will require `lib/my_module.rb` for you, thanks to ActiveSupport::Dependencies, and it's generally better to rely on auto requiring in case of differences between Ruby's and Rails' load path.&lt;/p&gt;</description><dc:creator xmlns:dc="http://purl.org/dc/elements/1.1/">August Lilleaas</dc:creator><pubDate>Fri, 07 Aug 2009 11:56:25 -0000</pubDate></item><item><title>Re: August Lilleaas</title><link>http://august.lilleaas.net/deadlier_and_simpler_rails_deployment#comment-11549310</link><description>&lt;p&gt;I must admit that using rake tasks to do the SSH-ing for me didn't cross my mind. The only benefit pushmaster provides is that you boil it all down to one command -- &lt;code&gt;git push&lt;/code&gt;, and also the fact that you don't have to write any SSH-ing rake tasks ; ) I kinda like not having to do anything other than pushing, though. I like systems that doesn't get in my way, and that Just Works (tm) so that I don't have to think.&lt;/p&gt;&lt;p&gt;I would expect that the most frequent use of pushmaster is not to deploy webpages, but to provide 3rd party services that isn't in Github's official list of services with Git repository metadata, such as posting to your own properitary project manager whenever something happens on Github.&lt;/p&gt;</description><dc:creator xmlns:dc="http://purl.org/dc/elements/1.1/">August Lilleaas</dc:creator><pubDate>Mon, 22 Jun 2009 04:39:43 -0000</pubDate></item><item><title>Re: August Lilleaas</title><link>http://august.lilleaas.net/combining_named_scopes#comment-10648996</link><description>&lt;p&gt;Some credit should be given to the person/people that implemented named scopes. All you have to do is to define a class method and then call other class methods. Super simple.&lt;/p&gt;&lt;p&gt;Some credit should also be given to Rails itself. Allthough it's kind of trivial, &lt;code&gt;metaclass&lt;/code&gt; is a shortcut for &lt;code&gt;(class &amp;lt;&amp;lt; self; self end)&lt;/code&gt;, which makes the code a whole lot better looking. &lt;code&gt;define_method&lt;/code&gt; is a instance method, so it has to be called on instances, not classes. In this case, the instance of &lt;code&gt;Class&lt;/code&gt; for that particular class, or the metaclass as it's commonly called.&lt;/p&gt;&lt;p&gt;But indeed, Ruby should be given the remaining credit, which is a lot. Such a flexible and lovable language.&lt;/p&gt;</description><dc:creator xmlns:dc="http://purl.org/dc/elements/1.1/">August Lilleaas</dc:creator><pubDate>Tue, 09 Jun 2009 09:06:03 -0000</pubDate></item><item><title>Re: August Lilleaas</title><link>http://august.lilleaas.net/procs_and_blocks_and_anonymous_functions#comment-10414904</link><description>&lt;p&gt;Thanks for the heads up! I'm not that into parsers, so I've only noticed this behaviour but haven't given much thought to _why_ it behaves like this. I'll see if I can find a way to integrate this information in the article.&lt;/p&gt;</description><dc:creator xmlns:dc="http://purl.org/dc/elements/1.1/">August Lilleaas</dc:creator><pubDate>Wed, 03 Jun 2009 04:07:18 -0000</pubDate></item><item><title>Re: August Lilleaas</title><link>http://august.lilleaas.net/extending_ruby_and_rails#comment-10414851</link><description>&lt;p&gt;Indeed. This article is about extending core classes within your own applications, so it doesn't really apply there. But if you're writing a gem or a plugin or anything similar, it definetely makes sense to namespace your core extensions. If you're writing a gem, you don't want to override String#camelize. It will render your gem impossible to use in the context of Rails, because your own String#camelize you use in the gem probably isn't compatible with Rails' String#camelize — effectively breaking Rails.&lt;/p&gt;</description><dc:creator xmlns:dc="http://purl.org/dc/elements/1.1/">August Lilleaas</dc:creator><pubDate>Wed, 03 Jun 2009 04:01:59 -0000</pubDate></item><item><title>Re: Review: Figure Out When</title><link>http://quiteuseful.co.uk/post/116751843#comment-10390830</link><description>&lt;p&gt;Thanks for the review!&lt;/p&gt;&lt;p&gt;I just rolled out an update: The calendar now tries to pick a sensible first month. E.g. if you click 2 dates in january and 5 in february, february is the month you see when you open the calendar, regardless of the time of year you access the calendar. Thanks for the suggestion :)&lt;/p&gt;</description><dc:creator xmlns:dc="http://purl.org/dc/elements/1.1/">August Lilleaas</dc:creator><pubDate>Tue, 02 Jun 2009 12:54:50 -0000</pubDate></item><item><title>Re: August Lilleaas</title><link>http://august.lilleaas.net/extending_ruby_and_rails#comment-10353163</link><description>&lt;p&gt;Neat! I've seen gems that adds their own String#camelize, String#underscore and so on, that are incompatible with the ones in ActiveSupport, making it impossible to use the gem in a Rails context. Made ActionController look for postsController instead of Admin::PostsController and other nasty things.&lt;/p&gt;&lt;p&gt;With that said: In a Rails app, you're in control, so you won't really need to do it there. But when you're creating a gem, namespacing core exts a module like this is definitely sensible — almost obligatory, I'd say. Nevertheless, thanks for mentioning this =)&lt;/p&gt;&lt;p&gt;Seems like there are ome caveats, though: &lt;a href="http://pastie.org/496618" rel="nofollow noopener" target="_blank" title="http://pastie.org/496618"&gt;http://pastie.org/496618&lt;/a&gt;. You have to inherit from the top-level class, and you have to use the class constructors and not literals. Having to use &lt;a href="http://String.new" rel="nofollow noopener" target="_blank" title="String.new"&gt;String.new&lt;/a&gt;("foo") instead of just "foo" all over the library/gem seems a bit tedious.&lt;/p&gt;</description><dc:creator xmlns:dc="http://purl.org/dc/elements/1.1/">August Lilleaas</dc:creator><pubDate>Mon, 01 Jun 2009 11:44:42 -0000</pubDate></item><item><title>Re: Magnus Holm - Nokogirl</title><link>http://oldblog.judofyr.net/posts/nokogirl.html#comment-8604879</link><description>&lt;p&gt;I usually type "concubine", fraudian slip I guess&lt;/p&gt;</description><dc:creator xmlns:dc="http://purl.org/dc/elements/1.1/">August Lilleaas</dc:creator><pubDate>Thu, 23 Apr 2009 09:46:07 -0000</pubDate></item><item><title>Re: stuff - rails - Rest For Rails</title><link>http://stuff.lilleaas.net/rest_for_rails#comment-6471733</link><description>&lt;p&gt;Thanks! Added a few examples.&lt;/p&gt;</description><dc:creator xmlns:dc="http://purl.org/dc/elements/1.1/">August Lilleaas</dc:creator><pubDate>Sat, 21 Feb 2009 17:31:36 -0000</pubDate></item><item><title>Re: stuff - ruby - Things You Can Do In Ruby</title><link>http://stuff.lilleaas.net/things_you_can_do_in_ruby#comment-6203928</link><description>&lt;p&gt;Hmm, on 1.8.6, this happens:&lt;/p&gt;&lt;p&gt;&lt;code&gt;h = Hash.new {|hash, key| hash[key] = [] }&lt;br&gt;h['wtf'] &amp;lt;&amp;lt; 'hi'&lt;br&gt;puts h.inspect&lt;br&gt;# =&amp;gt; {"wtf"=&amp;gt;["hi"]}&lt;br&gt;&lt;br&gt;h = Hash.new {|hash, key| [] }&lt;br&gt;h['wtf'] &amp;lt;&amp;lt; 'hi'&lt;br&gt;puts h.inspect&lt;br&gt;# =&amp;gt; {}&lt;/code&gt;&lt;/p&gt;&lt;p&gt;It doesn't seem to check the arity either, &lt;code&gt;Hash.new { [] }&lt;/code&gt; (without any block parameters) produces the same result.&lt;/p&gt;</description><dc:creator xmlns:dc="http://purl.org/dc/elements/1.1/">August Lilleaas</dc:creator><pubDate>Thu, 12 Feb 2009 06:49:08 -0000</pubDate></item><item><title>Re: HTML formatting in a comment?</title><link>https://disqus.com/home/discussion/disqus/html_formatting_in_a_comment/#comment-6203890</link><description>&lt;p&gt;Seems l ike a code tag with multiple lines are pre-formatted.&lt;/p&gt;&lt;p&gt;&lt;code&gt;hi&lt;br&gt;allright&lt;/code&gt;&lt;/p&gt;</description><dc:creator xmlns:dc="http://purl.org/dc/elements/1.1/">August Lilleaas</dc:creator><pubDate>Thu, 12 Feb 2009 06:44:49 -0000</pubDate></item><item><title>Re: stuff - ruby - Case Or Hash</title><link>http://stuff.lilleaas.net/case_or_hash#comment-4829981</link><description>&lt;p&gt;I just learned something new about hashes, thanks!&lt;/p&gt;&lt;p&gt;I'll leave the article as it is, though, it's still a valid use case. Say, running default_path is horribly expensive so you only want to run it when you have to.&lt;/p&gt;</description><dc:creator xmlns:dc="http://purl.org/dc/elements/1.1/">August Lilleaas</dc:creator><pubDate>Fri, 02 Jan 2009 10:24:59 -0000</pubDate></item><item><title>Re: stuff - meta - Please Welcome Disqus</title><link>http://stuff.lilleaas.net/please_welcome_disqus#comment-4703645</link><description>&lt;p&gt;Oh, isn't that great, I just re-bumped all the items in my RSS feed. Sowwy =/&lt;/p&gt;</description><dc:creator xmlns:dc="http://purl.org/dc/elements/1.1/">August Lilleaas</dc:creator><pubDate>Sun, 28 Dec 2008 20:51:05 -0000</pubDate></item><item><title>Re: stuff - meta - Please Welcome Disqus</title><link>http://stuff.lilleaas.net/please_welcome_disqus#comment-4685214</link><description>&lt;p&gt;First, lol!&lt;/p&gt;</description><dc:creator xmlns:dc="http://purl.org/dc/elements/1.1/">August Lilleaas</dc:creator><pubDate>Sun, 28 Dec 2008 17:16:56 -0000</pubDate></item><item><title>Re: ReinH &amp;mdash; Bad Designers Steal&amp;nbsp;Too</title><link>http://reinh.com/blog/2008/12/03/bad-designers-steal-too.html#comment-4164602</link><description>&lt;p&gt;ReinH, what's up with leaving out the fact that it's openly available on github under the label "blogging engine" and then after swombat tweets you three times you post a tweet that mentions this blog post? It seems like you're the troublemaker here. Lame.&lt;/p&gt;</description><dc:creator xmlns:dc="http://purl.org/dc/elements/1.1/">August Lilleaas</dc:creator><pubDate>Thu, 04 Dec 2008 08:21:42 -0000</pubDate></item></channel></rss>