adds rake task to scrape sites for spec examples

Rakefile

@@ -1 +1,44 @@
 require "bundler/gem_tasks"
+require "nokogiri"
+require "open-uri"
+require "pp"
+
+task :update_spec_cases do
+  sources = [
+    { urls: ["http://microformats.org/wiki/microformats-2"],
+      html_selector: ".source-html4strict",
+      json_selector: ".source-javascript",
+      html_method: "inner_text"
+    },
+    { urls: ["http://microformat2-node.jit.su/h-adr.html"],
+      html_selector: ".e-x-microformat",
+      json_selector: ".language-json",
+      html_method: "inner_html"
+    }
+  ]
+
+  sources.each do |source|
+    source[:urls].each do |url|
+      document = Nokogiri::HTML(open(url).read)
+      html = document.css(source[:html_selector]).map { |e| e.send(source[:html_method]) }
+      json = document.css(source[:json_selector]).map { |e| e.inner_text }
+
+      filename = url.split("/").last.gsub(/[.]\w+/, "")
+      filepath = "spec/support/cases/"
+
+      ([html.length, json.length].min).times do |index|
+
+        File.open("#{filepath}#{filename}-#{index}.html", "w") do |f|
+          f.write "<!-- #{url} -->\n"
+          f.write html[index]
+        end
+
+        File.open("#{filepath}#{filename}-#{index}.js", "w") do |f|
+          f.write "// #{url}\n"
+          f.write json[index]
+        end
+      end
+
+    end
+  end
+end

spec/support/cases/h-adr-0.html

@@ -0,0 +1,3 @@
+<!-- http://microformat2-node.jit.su/h-adr.html -->
+
+<p class="h-adr">665 3rd St. Suite 207 San Francisco, CA 94107 U.S.A.</p>

spec/support/cases/h-adr-0.js

@@ -0,0 +1,10 @@
+// http://microformat2-node.jit.su/h-adr.html
+
+{
+    "items": [{
+        "type": ["h-adr"],
+        "properties": {
+            "name": ["665 3rd St. Suite 207 San Francisco, CA 94107 U.S.A."]
+        }
+    }]
+}

spec/support/cases/h-adr-1.html

@@ -0,0 +1,10 @@
+<!-- http://microformat2-node.jit.su/h-adr.html -->
+
+<p class="h-adr">
+    <span class="p-street-address">665 3rd St.</span>  
+    <span class="p-extended-address">Suite 207</span>  
+    <span class="p-locality">San Francisco</span>,  
+    <span class="p-region">CA</span>  
+    <span class="p-postal-code">94107</span>  
+    <span class="p-country-name">U.S.A.</span>  
+</p>

spec/support/cases/h-adr-1.js

@@ -0,0 +1,16 @@
+// http://microformat2-node.jit.su/h-adr.html
+
+{
+    "items": [{
+        "type": ["h-adr"],
+        "properties": {
+            "street-address": ["665 3rd St."], 
+            "extended-address": ["Suite 207"], 
+            "locality": ["San Francisco"], 
+            "region": ["CA"], 
+            "postal-code": ["94107"], 
+            "country-name": ["U.S.A."], 
+            "name": ["665 3rd St. Suite 207 San Francisco, CA 94107 U.S.A."]
+        }
+    }]
+}

spec/support/cases/h-adr-2.html

@@ -0,0 +1,12 @@
+<!-- http://microformat2-node.jit.su/h-adr.html -->
+
+<p class="h-adr">
+    <span class="p-name">Bricklayer's Arms</span>
+    <span class="p-label"> 
+        <span class="p-street-address">3 Charlotte Road</span>,  
+        <span class="p-locality">City of London</span>,  
+        <span class="p-postal-code">EC2A 3PE</span>, 
+        <span class="p-country-name">UK</span> 
+    </span> – 
+    Geo:(<span class="p-geo">51.526421;-0.081067</span>) 
+</p>

spec/support/cases/h-adr-2.js

@@ -0,0 +1,16 @@
+// http://microformat2-node.jit.su/h-adr.html
+
+{
+    "items": [{
+        "type": ["h-adr"],
+        "properties": {
+            "name": ["Bricklayer's Arms"],
+            "label": ["3 Charlotte Road, City of London, EC2A 3PE, UK"],
+            "street-address": ["3 Charlotte Road"],
+            "locality": ["City of London"],
+            "postal-code": ["EC2A 3PE"],
+            "country-name": ["UK"],
+            "geo": ["51.526421;-0.081067"]
+        }
+    }]
+}

spec/support/cases/h-adr-3.html

@@ -0,0 +1,6 @@
+<!-- http://microformat2-node.jit.su/h-adr.html -->
+
+<p class="h-adr">
+    <a class="p-name u-geo" href="geo:51.526421;-0.081067;crs=wgs84;u=40">Bricklayer's Arms</a>, 
+    <span class="p-locality">London</span> 
+</p>

spec/support/cases/h-adr-3.js

@@ -0,0 +1,12 @@
+// http://microformat2-node.jit.su/h-adr.html
+
+{
+    "items": [{
+        "type": ["h-adr"],
+        "properties": {
+            "name": ["Bricklayer's Arms"],
+            "geo": ["geo:51.526421;-0.081067;crs=wgs84;u=40"],
+            "locality": ["London"]
+        }
+    }]
+}

spec/support/cases/microformats-2-0.html

@@ -0,0 +1,2 @@
+<!-- http://microformats.org/wiki/microformats-2 -->
+<span class="h-card">Frances Berriman</span>

spec/support/cases/microformats-2-0.js

@@ -0,0 +1,9 @@
+// http://microformats.org/wiki/microformats-2
+{
+  "items": [{ 
+    "type": ["h-card"],
+    "properties": {
+      "name": ["Frances Berriman"] 
+    }
+  }]
+}

spec/support/cases/microformats-2-1.html

@@ -0,0 +1,2 @@
+<!-- http://microformats.org/wiki/microformats-2 -->
+<a class="h-card" href="http://benward.me">Ben Ward</a>

spec/support/cases/microformats-2-1.js

@@ -0,0 +1,10 @@
+// http://microformats.org/wiki/microformats-2
+{
+  "items": [{ 
+    "type": ["h-card"],
+    "properties": {
+      "name": ["Ben Ward"],
+      "url": ["http://benward.me"]
+    }
+  }]
+}

spec/support/cases/microformats-2-2.html

@@ -0,0 +1,5 @@
+<!-- http://microformats.org/wiki/microformats-2 -->
+<a class="h-card" href="http://rohit.khare.org/">
+ <img alt="Rohit Khare"
+      src="https://s3.amazonaws.com/twitter_production/profile_images/53307499/180px-Rohit-sq_bigger.jpg" />
+</a>

spec/support/cases/microformats-2-2.js

@@ -0,0 +1,11 @@
+// http://microformats.org/wiki/microformats-2
+{
+  "items": [{ 
+    "type": ["h-card"],
+    "properties": {
+      "name": ["Rohit Khare"],
+      "url": ["http://rohit.khare.org"],
+      "photo": ["https://s3.amazonaws.com/twitter_production/profile_images/53307499/180px-Rohit-sq_bigger.jpg"]
+    }
+  }]
+}

spec/support/cases/microformats-2-3.html

@@ -0,0 +1,17 @@
+<!-- http://microformats.org/wiki/microformats-2 -->
+<div class="h-card">
+  <img class="u-photo" alt="photo of Mitchell"
+       src="https://webfwd.org/content/about-experts/300.mitchellbaker/mentor_mbaker.jpg"/>
+  <a class="p-name u-url"
+     href="http://blog.lizardwrangler.com/" 
+    >Mitchell Baker</a>
+ (<a class="u-url" 
+     href="https://twitter.com/MitchellBaker"
+    >@MitchellBaker</a>)
+  <span class="p-org">Mozilla Foundation</span>
+  <p class="p-note">
+    Mitchell is responsible for setting the direction and scope of the Mozilla Foundation and its activities.
+  </p>
+  <span class="p-category">Strategy</span>
+  <span class="p-category">Leadership</span>
+</div>

spec/support/cases/microformats-2-3.js

@@ -0,0 +1,20 @@
+// http://microformats.org/wiki/microformats-2
+{
+  "items": [{ 
+    "type": ["h-card"],
+    "properties": {
+      "photo": ["https://webfwd.org/content/about-experts/300.mitchellbaker/mentor_mbaker.jpg"],
+      "name": ["Mitchell Baker"],
+      "url": [
+        "http://blog.lizardwrangler.com/",
+        "https://twitter.com/MitchellBaker"
+      ],
+      "org": ["Mozilla Foundation"],
+      "note": ["Mitchell is responsible for setting the direction and scope of the Mozilla Foundation and its activities."],
+      "category": [
+        "Strategy",
+        "Leadership"
+      ]
+    }
+  }]
+}

spec/support/cases/microformats-2-4.html

@@ -0,0 +1,16 @@
+<!-- http://microformats.org/wiki/microformats-2 -->
+<div class="h-event">
+  <a class="p-name u-url" href="http://indiewebcamp.com/2012">
+    IndieWebCamp 2012
+  </a>
+  from <time class="dt-start">2012-06-30</time> 
+  to <time class="dt-end">2012-07-01</time> at 
+  <span class="p-location h-card">
+    <a class="p-name p-org u-url" href="http://geoloqi.com/">
+      Geoloqi
+    </a>, 
+    <span class="p-street-address">920 SW 3rd Ave. Suite 400</span>, 
+    <span class="p-locality">Portland</span>, 
+    <abbr class="p-region" title="Oregon">OR</abbr>
+  </span>
+</div>

spec/support/cases/microformats-2-4.js

@@ -0,0 +1,24 @@
+// http://microformats.org/wiki/microformats-2
+{
+  "items": [{ 
+    "type": ["h-event"],
+    "properties": {
+      "name": ["IndieWebCamp 2012"],
+      "url": ["http://indiewebcamp.com/2012"],
+      "start": ["2012-06-30"],
+      "end": ["2012-07-01"],
+      "location": [{
+        "value": "Geoloqi, 920 SW 3rd Ave. Suite 400, Portland, OR",
+        "type": ["h-card"],
+        "properties": {
+          "name": ["Geoloqi"],
+          "org": ["Geoloqi"],
+          "url": ["http://geoloqi.com/"],
+          "street-address": ["920 SW 3rd Ave. Suite 400"],
+          "locality": ["Portland"],
+          "region": ["Oregon"]
+        }
+      }]
+    }
+  }]
+}

spec/support/cases/microformats-2-5.html

@@ -0,0 +1,7 @@
+<!-- http://microformats.org/wiki/microformats-2 -->
+<div class="h-card">
+  <a class="p-name u-url"
+     href="http://blog.lizardwrangler.com/" 
+    >Mitchell Baker</a> 
+  (<span class="p-org">Mozilla Foundation</span>)
+</div>

spec/support/cases/microformats-2-5.js

@@ -0,0 +1,11 @@
+// http://microformats.org/wiki/microformats-2
+{
+  "items": [{ 
+    "type": ["h-card"],
+    "properties": {
+      "name": ["Mitchell Baker"],
+      "url": ["http://blog.lizardwrangler.com/"],
+      "org": ["Mozilla Foundation"]
+    }
+  }]
+}

spec/support/cases/microformats-2-6.html

@@ -0,0 +1,9 @@
+<!-- http://microformats.org/wiki/microformats-2 -->
+<div class="h-card">
+  <a class="p-name u-url"
+     href="http://blog.lizardwrangler.com/" 
+    >Mitchell Baker</a> 
+  (<a class="p-org h-card" 
+      href="http://mozilla.org/"
+     >Mozilla Foundation</a>)
+</div>

spec/support/cases/microformats-2-6.js

@@ -0,0 +1,18 @@
+// http://microformats.org/wiki/microformats-2
+{
+  "items": [{ 
+    "type": ["h-card"],
+    "properties": {
+      "name": ["Mitchell Baker"],
+      "url": ["http://blog.lizardwrangler.com/"],
+      "org": [{
+        "value": "Mozilla Foundation",
+        "type": ["h-card"],
+        "properties": {
+          "name": ["Mozilla Foundation"],
+          "url": ["http://mozilla.org/"]
+        }
+      }]
+    }
+  }]
+}

spec/support/cases/microformats-2-7.html

@@ -0,0 +1,9 @@
+<!-- http://microformats.org/wiki/microformats-2 -->
+<div class="h-card">
+  <a class="p-name u-url"
+     href="http://blog.lizardwrangler.com/" 
+    >Mitchell Baker</a> 
+  (<a class="p-org h-card h-org" 
+      href="http://mozilla.org/"
+     >Mozilla Foundation</a>)
+</div>

spec/support/cases/microformats-2-7.js

@@ -0,0 +1,18 @@
+// http://microformats.org/wiki/microformats-2
+{
+  "items": [{ 
+    "type": ["h-card"],
+    "properties": {
+      "name": ["Mitchell Baker"],
+      "url": ["http://blog.lizardwrangler.com/"],
+      "org": [{
+        "value": "Mozilla Foundation",
+        "type": ["h-card", "h-org"],
+        "properties": {
+          "name": ["Mozilla Foundation"],
+          "url": ["http://mozilla.org/"]
+        }
+      }]
+    }
+  }]
+}

spec/support/cases/microformats-2-8.html

@@ -0,0 +1,9 @@
+<!-- http://microformats.org/wiki/microformats-2 -->
+<div class="h-card">
+  <a class="p-name u-url"
+     href="http://blog.lizardwrangler.com/" 
+    >Mitchell Baker</a> 
+  (<a class="h-org h-card" 
+      href="http://mozilla.org/"
+     >Mozilla Foundation</a>)
+</div>

spec/support/cases/microformats-2-8.js

@@ -0,0 +1,17 @@
+// http://microformats.org/wiki/microformats-2
+{
+  "items": [{ 
+    "type": ["h-card"],
+    "properties": {
+      "name": ["Mitchell Baker"],
+      "url": ["http://blog.lizardwrangler.com/"]
+    },
+    "children": [{
+      "type": ["h-card","h-org"],
+      "properties": {
+        "name": ["Mozilla Foundation"],
+        "url": ["http://mozilla.org/"]
+      }  
+    }]
+  }]
+}

spec/support/cases/microformats-2-9.html

@@ -0,0 +1,9 @@
+<!-- http://microformats.org/wiki/microformats-2 -->
+<div class="h-card">
+  <a class="p-name u-url"
+     href="http://blog.lizardwrangler.com/" 
+    >Mitchell Baker</a> 
+  (<a class="h-card" 
+      href="http://mozilla.org/"
+     >Mozilla Foundation</a>)
+</div>

spec/support/cases/microformats-2-9.js

@@ -0,0 +1,17 @@
+// http://microformats.org/wiki/microformats-2
+{
+  "items": [{ 
+    "type": ["h-card"],
+    "properties": {
+      "name": ["Mitchell Baker"],
+      "url": ["http://blog.lizardwrangler.com/"]
+    },
+    "children": [{
+      "type": ["h-card"],
+      "properties": {
+        "name": ["Mozilla Foundation"],
+        "url": ["http://mozilla.org/"]
+      }  
+    }]
+  }]
+}