Extension profile generator + benchmark for startup with profile.

tools/perf/profile_creators/extension_profile_extender.py

Index: tools/perf/profile_creators/extension_profile_extender.py
diff --git a/tools/perf/profile_creators/extension_profile_extender.py b/tools/perf/profile_creators/extension_profile_extender.py
new file mode 100644
index 0000000000000000000000000000000000000000..49527d8f03a7d31903c6975bdb4787c7aaa7b54f
--- /dev/null
+++ b/tools/perf/profile_creators/extension_profile_extender.py
@@ -0,0 +1,133 @@
+# Copyright 2016 The Chromium Authors. All rights reserved.

[robliao, 2015/07/16 21:10:04]

2015 :-)

[sydli, 2015/07/17 01:00:29]

Done.

+# Use of this source code is governed by a BSD-style license that can be
+# found in the LICENSE file.
+
+import json
+import os
+import platform
+import zipfile
+
+from catapult_base import cloud_storage
+import page_sets
+from profile_creators import fast_navigation_profile_extender
+from telemetry.core import exceptions
+
+
+# Remote target upload directory in cloud storage for extensions.
+REMOTE_DIR = 'extension_set'
+
+# Target zip file.
+ZIP_NAME = 'extensions.zip'
+
+
+class InvalidExtensionArchiveError(exceptions.Error):
+  """Exception thrown when remote archive is invalid or malformed.
+
+  Remote archive should be located at REMOTE_DIR/ZIP_NAME. Upon failure,
+  prompts user to update remote archive using update_remote_extensions
+  script.
+  """
+
+  def __init__(self, msg=''):
+    msg += ('\nTry running\n'
+            '\tpython update_remote_extensions.py -e extension_set.csv\n'
+            'in src/tools/perf/profile_creator subdirectory.')
+    super(InvalidExtensionArchiveError, self).__init__(msg)
+
+
+class ExtensionProfileExtender(
+      fast_navigation_profile_extender.FastNavigationProfileExtender):
+  """Creates a profile with many extensions."""
+
+  def __init__(self, finder_options):
+    maximum_batch_size = 1

[robliao, 2015/07/16 21:10:04]

Is this local necessary if it's not read in this constructor anymore? The next
line could just be maximum_batch_size = 1

[sydli, 2015/07/17 01:00:28]

Good catch-- done.

+    super(ExtensionProfileExtender, self).__init__(finder_options,
+                                                   maximum_batch_size)
+    self._page_set = page_sets.BlankPageSet()
+    urls = [story.url for story in self._page_set.stories]
+    self._navigation_urls = urls
+    finder_options.browser_options.disable_default_apps = False
+
+  def _DownloadRemoteExtensions(self, remote_bucket, local_extensions_dir):
+    """Downloads and unzips archive of common extensions to disk.
+
+    Args:
+        remote_bucket: bucket to download remote archive from.
+        local_extensions_dir: destination extensions directory.
+
+    Raises:
+        InvalidExtensionArchiveError if remote archive is not found.
+    """
+    remote_zip_path = os.path.join(REMOTE_DIR, ZIP_NAME)
+    local_zip_path = os.path.join(local_extensions_dir, ZIP_NAME)
+    try:
+      cloud_storage.Get(remote_bucket, remote_zip_path, local_zip_path)
+    except:
+      raise InvalidExtensionArchiveError('Can\'t find archive at gs://%s/%s..'
+                                         % (remote_bucket, remote_zip_path))
+    with zipfile.ZipFile(local_zip_path, 'r') as extensions_zip:
+      extensions_zip.extractall(local_extensions_dir)

[robliao, 2015/07/16 21:10:04]

If extractall throws an exception, the local_zip_path will be hanging around.
This may need a try-finally.

[sydli, 2015/07/17 01:00:29]

Done.

+    os.remove(local_zip_path)
+
+  def _GetExtensionInfoFromCRX(self, crxfile):

[robliao, 2015/07/16 21:10:04]

Lowercase CRX

[sydli, 2015/07/17 01:00:28]

Done.

+    """Retrieves version + name of extension from CRX archive."""
+    crx_zip = zipfile.ZipFile(crxfile)
+    manifest_contents = crx_zip.read('manifest.json')
+    decoded_manifest = json.loads(manifest_contents)
+    crx_version = decoded_manifest['version']
+    extension_name = decoded_manifest['name']
+    return (crx_version, extension_name)
+
+  def _LoadExtensions(self, local_extensions_dir, profile_dir):
+    """Loads extensions in _local_extensions_dir into user profile.
+
+    Extensions are loaded according to platform specifications at
+    https://developer.chrome.com/extensions/external_extensions.html
+
+    Args:
+        local_extensions_dir: directory containing CRX files.
+        profile_dir: target profile directory for the extensions.
+
+    Raises:
+        InvalidExtensionArchiveError if archive contains a non-CRX file.
+    """
+    ext_files = os.listdir(local_extensions_dir)
+    external_ext_dir = os.path.join(profile_dir, 'External Extensions')
+    os.makedirs(external_ext_dir)
+    for ext_file in ext_files:
+      ext_path = os.path.join(local_extensions_dir, ext_file)
+      if not ext_file.endswith('.crx'):
+        raise InvalidExtensionArchiveError('Archive contains non-crx file %s.'
+                                           % ext_file)
+      (version, name) = self._GetExtensionInfoFromCRX(ext_path)

[robliao, 2015/07/16 21:10:04]

Do we need to be robust to failures if the CRX file contains bad contents?

[sydli, 2015/07/17 01:00:29]

I don't think so. Re-running the update script should be sufficient, as long as
we can trust CRX given by the Webstore. Even so, is there a good way to check
the integrity of a CRX archive? 

[robliao, 2015/07/17 03:13:07]

I don't know of any offhand. Rerunning the script sounds good. Might be useful
to suggest that in an error message.

+      extension_info = {
+            'external_crx': ext_path,
+            'external_version': version,
+            '_comment': name
+          }
+      ext_id = os.path.splitext(os.path.basename(ext_path))[0]
+      extension_json_path = os.path.join(external_ext_dir, '%s.json' % ext_id)
+      with open(extension_json_path, 'w') as f:
+        f.write(json.dumps(extension_info))
+
+  def GetUrlIterator(self):
+    """Superclass override."""
+    return iter(self._navigation_urls)
+
+  def ShouldExitAfterBatchNavigation(self):
+    """Superclass override."""
+    return False
+
+  def Run(self):
+    # DL extensions from cloud & force-install extensions into profile.

[robliao, 2015/07/16 21:10:04]

Nit DL -> Download

[sydli, 2015/07/17 01:00:28]

Done.

+    if platform.system() != 'Darwin':
+      raise NotImplementedError(
+        'Extension profile generator on %s is not yet supported'
+        % platform.system())
+    local_extensions_dir = os.path.join(self.profile_path,
+                                        'external_extensions_crx')
+    self._DownloadRemoteExtensions(cloud_storage.PARTNER_BUCKET,
+                                   local_extensions_dir)
+    self._LoadExtensions(local_extensions_dir, self.profile_path)
+    super(ExtensionProfileExtender, self).Run()
+