First stab at trying to pre-render content for search crawlers.

seo-snapshots/make_snapshot.py

@@ -0,0 +1,29 @@
+import subprocess
+import urllib
+
+from BeautifulSoup import BeautifulSoup
+
+
+BASE_URL = 'http://localhost:5000'
+OUTPUT_PATH = 'snapshots/'
+
+URLS = [
+  ('/', 'index.html')
+]
+
+for url, output in URLS:
+  final_url = BASE_URL + url
+
+  out_html = subprocess.check_output(['phantomjs', 'phantomjs-runner.js',
+                                      final_url])
+
+  # Remove script tags
+  soup = BeautifulSoup(out_html)
+  to_extract = soup.findAll('script')
+  for item in to_extract:
+      item.extract()
+
+  to_write = OUTPUT_PATH + output
+
+  with open(to_write, 'w') as output_file:
+    output_file.write(soup.prettify())

seo-snapshots/phantomjs-runner.js

@@ -0,0 +1,23 @@
+var system = require('system');
+var url = system.args[1] || '';
+if(url.length > 0) {
+  var page = require('webpage').create();  
+  page.open(url, function (status) {
+    if (status == 'success') {
+      var delay, checker = (function() {
+        var html = page.evaluate(function () {
+          var ready = document.getElementsByClassName('ready-indicator')[0];
+          if(ready.getAttribute('data-status') == 'ready') {
+            return document.getElementsByTagName('html')[0].outerHTML;
+          }
+        });
+        if(html) {
+          clearTimeout(delay);
+          console.log(html);
+          phantom.exit();
+        }
+      });
+      delay = setInterval(checker, 100);
+    }
+  });
+}