http://www.stephenbeckeriv.comMon, 09 May 2011 07:56:19 PDTen-ushttp://stephenbeckeriv.com/main/index/Textport_0.0.4http://stephenbeckeriv.com/main/index/Textport_0.0.4Mon, 09 May 2011 14:56:04 UTCThe add-on process for mozilla I pretty slow. I understand its a volunteer group of people. I would not really gate my best feature on a group that does not have the companies best interests in mind. I had a preliminarily and needed to update. The first review was speedy but the second was taking for ever. I thought switching it to a full review might help but no luck so far. <br/> Until then here is a copy of my <a href='https://addons.mozilla.org/firefox/downloads/file/117689/textport-0.0.4-tb.xpi?src=becker'>textport 0.0.4</a>http://stephenbeckeriv.com/main/index/Textporthttp://stephenbeckeriv.com/main/index/TextportFri, 01 Apr 2011 03:42:36 UTCMy first thunderbird extension. Texport is a plugin that allows you to export your email data using a template. The template are <a href='http://embeddedjs.com/'>EJS</a>. Texport also can export the contacts from a set of emails. Action can be taken on folder or selected emails. <br/> Plugin to thunderbird that allows you to export your contact information or message bodies to text files. This is enabled in two menus. The first is the folder level menu. This will preform the action on ever message in the folder. For larger folders this can take a long time. I exported 19k message in 2 hours. The second is in the message panel level. Highlight one or more messages to preform the actions on. <br/> Source: <a href='https://github.com/sbeckeriv/textport'>https://github.com/sbeckeriv/textport</a> <br/> Thunderbird page: <a href='https://addons.mozilla.org/en-US/thunderbird/addon/textport/'>https://addons.mozilla.org/en-US/thunderbird/addon/textport/</a> http://stephenbeckeriv.com/main/index/thunderbird_3.1_and_firebug_1.7b3http://stephenbeckeriv.com/main/index/thunderbird_3.1_and_firebug_1.7b3Fri, 18 Mar 2011 01:21:31 UTCI wanted to start up a new thunderbird plugin project. I require 3 plugins. 1) extension developer. Its amazingly easy to install a dev build with this. 2) firebug. yes even for thunderbird. 3) firechrome. This gives me direct access to my running code. <br/> I prefer working with Mozilla based apps over Google's chrome. http://stephenbeckeriv.com/main/index/when_the_state_runs_things_IIIhttp://stephenbeckeriv.com/main/index/when_the_state_runs_things_IIISun, 23 Jan 2011 04:37:18 UTCAs a project I have been working on a website that scrapes data from the wa.liq state site. It was to include search, location based info, json data dump files for clients, and other fun features the state is to outdated to understand. I have had fun with selecting a brand type (Scotch) and a store (#28). I have a page which shows you the brands in stock but my page also shows you which brands they do not have that store but have at other stores. It also includes an approximate mile count to the closest store.<br/> But as of Friday the state updated their website. Hooray! Not really. The site is still poor at best. The update comes with a new look. A look I do not like. There is a slight search improvement where you can search per brand type per city (not store like mine does. I was going to offer city). The usability of this also very poor. The find store now lacks the product you were looking for. <br/> To top it all off they updated their system to some new version of ASP. It now includes value checking and javascript for form manipulations. It also holds the search part in an iframe so you have to see the ugly layout and buttons.<br/> Long and short of it is that I am rewriting my parsers in mechanize. Its been fun to learn mechanize and its Javascript shortfalls. I thought all was lost when I read the Javascript and had mechanize mimic the code. I should be reparsing data by weekend's end. I take the update to the state site as a personal challenge. I will give the drinkers of Washington more features then they can dream of. When I am done here I am turning my sites to Oregon! <br/> As long as the state is running things. http://stephenbeckeriv.com/main/index/ruby_bundle_command_line_configshttp://stephenbeckeriv.com/main/index/ruby_bundle_command_line_configsSun, 23 Jan 2011 04:22:13 UTCI use DreamHostPS. Its why this page is so fast ;). Point really is that after downloading and installing libxslt-1.1.21 locally. I needed to configure nokogiri to use it. easy enough <br/> <pre> ./configure --prefix=/home/sbecker/lib make make install gem install nokogiri -- --with-xslt-include=/home/sbecker/lib/include/ --with-xslt-lib=/home/sbecker/lib/lib/ </pre> <br/> Now comes along my 1 rails 3 app where bundle controls every gem I get to use. Ok easy. Add it to the Gemfile and move on. Nope. After searching around for stuff like 'gem bundle command line options' I found one tiny mention of bundle config. Here is the code to set up your command line options. I am not in the mood to learn what it means or why. Maybe later. <pre> bundle config build.nokogiri --with-xslt-include=/home/sbecker/lib/include/ --with-xslt-lib=/home/sbecker/lib/lib/ </pre> Simple enough. Just new and not clear.http://stephenbeckeriv.com/main/index/httpresponsecodeshttp://stephenbeckeriv.com/main/index/httpresponsecodesWen, 29 Dec 2010 16:23:20 UTCWhen developing code that uses web services I often code for the error case. What if I get back a 404 or 500? This is what I should do. I often have a hard time testing this. Some times the service provides a way to trigger errors. Those calls often are all about the payload. <br/><br/> I have started a new website httpresponsecodes.com that will take a code like 200 and give you back a 200 response. You can also set the body. For example http://www.httpresponsecodes.com/200?response_body=cow <br/><br/> Leaving of the response_body will return some details about the code. I dont have all the details on all the codes yet but 200 works fine. Its also one of the few that the browser renders fine. <br/><br/> Examples: <br/><br/> <a href='http://www.httpresponsecodes.com/200?response_body=cow'> 200 with body</a> <br/><br/> <a href='http://www.httpresponsecodes.com/200'> 200 with details of 200</a> <br/><br/> <a href='http://www.httpresponsecodes.com/500?response_body={"zip":20}'> 500 with json http://www.httpresponsecodes.com/500?response_body={"zip":20}</a>http://stephenbeckeriv.com/main/index/when_the_state_runs_things_IIhttp://stephenbeckeriv.com/main/index/when_the_state_runs_things_IISun, 12 Dec 2010 18:00:39 UTCI picked up my liq site scraping again. I reworked my db using sql. I am going to use ruby 1.9.X and rails 3. It should be up soon. http://stephenbeckeriv.com/main/index/when_the_state_runs_thingshttp://stephenbeckeriv.com/main/index/when_the_state_runs_thingsWen, 17 Mar 2010 08:13:34 UTCI live in yet another state that controls the liquor sales. PA is the other state. After talking with a coworker about some booze they showed me the state liquor store website. Turns out that they post the prices, stock and location of all the booze they carry. <br /><br /> After a night of scripting I have 2 scripts. The first one finds the brands and prices. The second one finds the quantities of each brand and what store they are at. Then it stores it in a mongodb (many reasons i am using mongo but that is not for here) <br /><br /> So after unscrewing dreamhost with rvm I finally have these scripts running on my server. But you say "big deal becker. what can you do with that data?". To which I say .. <br /><br/> The store at PORT ANGELES 1331 E FRONT STREET PT ANGELES WA 98362-4616 currently has 176 .75 liter bottles in stock of MACNAUGHTONS 36 MTH CDN TRAVELER which by the way is 80 proof! It is being sold at 12.45 including tax. Its not on sale currently... <br /><br/> I have more data but its getting late and my screens are ready to run all night.http://stephenbeckeriv.com/main/index/RVMhttp://stephenbeckeriv.com/main/index/RVMWen, 17 Mar 2010 08:03:17 UTCThere is a new-ish gem called rvm that allows you to manage ruby vms and gems with a few simple commands. I just used it to unscrew my dreamhost server out of config hell with gems. In less then an hour i rebuild ruby 1.8.7 and ruby 1.9.1 and install most of the gems dreamhost gives a standard account. <br \> I suggest looking at RVM very closely. I will never run another ruby box with out it.http://stephenbeckeriv.com/main/index/Current_projectshttp://stephenbeckeriv.com/main/index/Current_projectsMon, 07 Dec 2009 06:02:25 UTCI am currently working on a dashboard for my life. It will include things like the weather, news, movie release dates, OneBusAway information and other fun stuff. I am building it with rails and mongomapper. I am hoping to hang a screen on my wall and have it powered by a mac mini. Step one is to get it all working. <br/><br/> Other projects: Revisited IMDB and updated it to run again thanks to someone who wanted to use it. The problem with scrapers is that the code needs to keep changing. I would like to build a parser for the imdb data files but until then.. <br/><br/> Which brings me to my next project. An interface to my movie collect. I think I have finally have a handle on my movies and my system setup correctly. I am going to build a movie server (non-streaming) on top of my dvd image collection. http://stephenbeckeriv.com/main/index/Ruby_and_Divide_By_Zerohttp://stephenbeckeriv.com/main/index/Ruby_and_Divide_By_ZeroThr, 15 Oct 2009 21:37:03 UTC<pre><pre> <span class="ident">number</span> <span class="punct">=</span> <span class="punct">(</span><span class="number">1</span><span class="punct">/</span><span class="number">0</span><span class="punct">.</span><span class="ident">to_f</span><span class="punct">)</span> <span class="comment">#will not error</span> <span class="ident">number</span> <span class="punct">=</span> <span class="number">0</span> <span class="keyword">unless</span> <span class="number">1</span><span class="punct"><=></span><span class="ident">number</span> <span class="comment"># will return nil if its infinity or NaN</span> </pre></pre> No more checking number.nan? || number.infinite? http://stephenbeckeriv.com/main/index/busyhttp://stephenbeckeriv.com/main/index/busySun, 11 Oct 2009 03:34:56 UTCBeen busy. Started a new job. Gist.com Working on a few different ideas and been living. I hope to post something new soon.http://stephenbeckeriv.com/main/index/greasemonkey_day1http://stephenbeckeriv.com/main/index/greasemonkey_day1Thr, 18 Jun 2009 16:10:01 UTCI just got a G1 phone and I was playing around with google charts. Turns out they do QR bar codes. It turns out that the barcode scanner on my G1 reads the QR. So I made a greasemonkey script that adds the barcode to the bottom of every page. (After the body tag until I figure it out more) <a href="http://deathbyescalator.com/barcode.user.js">http://deathbyescalator.com/barcode.user.js</a> I can now just view a site, scan it and bookmark it on my phone.