[Telemetry] Add script to snapshot page's HTML

telemetry/bin/snap_page

Index: telemetry/bin/snap_page
diff --git a/telemetry/bin/snap_page b/telemetry/bin/snap_page
new file mode 100755
index 0000000000000000000000000000000000000000..75ed8444e2cef66239b25ca1bca96acea9192b51
--- /dev/null
+++ b/telemetry/bin/snap_page
@@ -0,0 +1,73 @@
+#!/usr/bin/env python
+# Copyright (c) 2017 The Chromium Authors. All rights reserved.
+# Use of this source code is governed by a BSD-style license that can be
+# found in the LICENSE file.
+
+import codecs
+import os
+import sys
+
+TELEMETRY_DIR = os.path.join(os.path.abspath(os.path.dirname(__file__)), '..')
+sys.path.append(TELEMETRY_DIR)
+
+from telemetry.util import wpr_modes
+from telemetry.core import util
+from telemetry.internal.browser import browser_finder
+from telemetry.internal.browser import browser_options
+from telemetry.internal.util import binary_manager
+
+
+def SnapPage(finder_options, url, interactive, snapshot_path):

[pdr., 2017/09/05 21:30:41]

WDYT of adding a simple end-to-end integration test of this? I.e., just snap a
simple static hello world page?

+  """ Save the HTML snapshot of the page whose address is |url| to
+  |snapshot_path|.
+  """
+  snapshot_path = os.path.abspath(snapshot_path)
+  binary_manager.InitDependencyManager([])
+  possible_browser = browser_finder.FindBrowser(finder_options)
+  browser = possible_browser.Create(finder_options)
+  try:
+    tab = browser.tabs[0]
+    tab.Navigate(url)
+    tab.WaitForDocumentReadyStateToBeComplete()
+    if interactive:
+      raw_input(
+          'Activating interactive mode. Press enter after you finish '
+          "interacting with the page to snapshot the page's DOM content.")
+    with open(os.path.join(util.GetTelemetryThirdPartyDir(), 'snap-it',
+        'HTMLSerializer.js')) as f:
+      snapit_script = f.read()
+    tab.ExecuteJavaScript(snapit_script)
+    tab.ExecuteJavaScript(
+        '''
+        var serializedDomArray;
+        var htmlSerializer = new HTMLSerializer();
+        htmlSerializer.processDocument(document);
+        htmlSerializer.fillHolesAsync(document, function(s) {
+          serializedDomArray = s.html;
+        });
+        ''')
+    print 'Snapshotting content of %s. This could take a while...' % url
+    tab.WaitForJavaScriptCondition('serializedDomArray !== undefined')
+    serialized_dom = ''.join(tab.EvaluateJavaScript('serializedDomArray'))

[pdr., 2017/09/05 21:30:41]

What is ''.join for? Is this converting the result to a string? Is that
necessary?

+    with codecs.open(snapshot_path, 'w', 'utf-8') as f:
+      f.write(serialized_dom)
+    print 'Successfully saved snapshot to file://%s' % snapshot_path
+  finally:
+    browser.Close()
+
+
+def main(args):
+  options = browser_options.BrowserFinderOptions()
+  options.browser_options.wpr_mode = wpr_modes.WPR_OFF
+  parser = options.CreateParser(
+      usage='Create a static HTML snapshot of the page')
+  parser.add_option('--url', help='URL of the web page to record')
+  parser.add_option('--interactive', action="store_true", default=False,
+      help='Activate interactive mode after loading the page')
+  parser.add_option('--snapshot-path', help='Where to save the snapshot',
+      default='snapshot.html')
+  parser.parse_args(args)
+  SnapPage(options, options.url, options.interactive, options.snapshot_path)
+
+if __name__ == '__main__':
+  sys.exit(main(sys.argv[1:]))