Add script to find unused Polymer elements

third_party/polymer/v1_0/find_unused_elements.py

Index: third_party/polymer/v1_0/find_unused_elements.py
diff --git a/third_party/polymer/v1_0/find_unused_elements.py b/third_party/polymer/v1_0/find_unused_elements.py
new file mode 100755
index 0000000000000000000000000000000000000000..58999937f520387d0e8f1d5d0c83f3d036168e48
--- /dev/null
+++ b/third_party/polymer/v1_0/find_unused_elements.py
@@ -0,0 +1,165 @@
+#!/usr/bin/python2
+
+# Copyright 2016 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.
+
+"""Identifies Polymer elements that downloaded but not used by Chrome.
+
+Only finds "first-order" unused elements; re-run after removing unused elements
+to check if other elements have become unused.
+"""
+
+import os
+import re
+import subprocess
+
+
+class UnusedElementsDetector:
+  """Finds unused Polymer elements."""
+
+  # Unused elements to ignore because we plan to use them soon.
+  __whitelist = (
+    # TODO(dschuyler): Use element or remove from whitelist.
+    'carbon-route',
+    # TODO(tsergeant): Use element or remove from whitelist.
+    'iron-scroll-threshold',
+    # Necessary for closure.
+    'polymer-externs',
+  )
+
+  @staticmethod
+  def __StripHtmlComments(filename):
+    """Returns the contents of an HTML file with <!-- --> comments stripped.
+
+    Not a real parser.
+
+    Args:
+      filename: The name of the file to read.
+
+    Returns:
+      A string consisting of the file contents with comments removed.
+    """
+    with open(filename) as f:
+      return re.sub('<!--.*?-->', '', f.read(), flags=re.MULTILINE|re.DOTALL)
+
+  @staticmethod
+  def __StripJsComments(filename):
+    """Returns the minified contents of a JavaScript file with comments and
+    grit directives removed.
+
+    Args:
+      filename: The name of the file to read.
+
+    Returns:
+      A string consisting of the minified file contents with comments and grit
+      directives removed.
+    """
+    with open(filename) as f:
+      text = f.read()
+    text = re.sub('<if .*?>', '', text)
+    text = re.sub('</if>', '', text)
+
+    proc = subprocess.Popen(['uglifyjs', filename], stdout=subprocess.PIPE)
+    return proc.stdout.read()
+
+  @staticmethod
+  def __StripComments(filename):
+    """Returns the contents of a JavaScript or HTML file with comments removed.
+
+    Args:
+      filename: The name of the file to read.
+
+    Returns:
+      A string consisting of the file contents processed via
+      __StripHtmlComments or __StripJsComments.
+    """
+    if filename.endswith('.html'):
+      text = UnusedElementsDetector.__StripHtmlComments(filename)
+    elif filename.endswith('.js'):
+      text = UnusedElementsDetector.__StripJsComments(filename)
+    else:
+      assert False, 'Invalid filename: %s' % filename
+    return text
+
+  @staticmethod
+  def Run():
+    """Finds unused Polymer elements and prints a summary."""
+    proc = subprocess.Popen(
+      ['git', 'rev-parse', '--show-toplevel'],
+      stdout=subprocess.PIPE)
+    src_dir = proc.stdout.read().strip()
+    polymer_dir = os.path.dirname(os.path.realpath(__file__))
+    components_dir = os.path.join(polymer_dir, 'components-chromium')
+
+    elements = []
+    for name in os.listdir(components_dir):
+      path = os.path.join(components_dir, name)
+      if os.path.isdir(path):
+        elements.append(name)
+
+    relevant_src_dirs = (
+      os.path.join(src_dir, 'chrome'),
+      os.path.join(src_dir, 'ui'),
+      os.path.join(src_dir, 'components'),
+      components_dir
+    )
+
+    for element in elements:
+      if element in UnusedElementsDetector.__whitelist:
+        continue

[Dan Beam, 2016/05/03 19:22:08]

what does this do?

[michaelpg, 2016/05/03 21:29:19]

removed

+
+    unused_elements = []
+    for element in elements:
+      if (element not in UnusedElementsDetector.__whitelist and
+          not UnusedElementsDetector.__IsImported(element, relevant_src_dirs)):
+        unused_elements.append(element)
+
+    if len(unused_elements):

[Dan Beam, 2016/05/03 19:22:08]

i think this can be just

  if unused_elements:

[michaelpg, 2016/05/03 21:29:19]

Done.

+      print 'Found unused elements: %s\nRemove from bower.json and re-run ' \
+        'reproduce.sh, or add to whitelist in %s' % (
+          ', '.join(unused_elements), os.path.basename(__file__))
+
+  @staticmethod
+  def __IsImported(element_dir, dirs):
+    """Returns whether the element directory is used in HTML or JavaScript.
+
+    Args:
+      element_dir: The name of the element's directory.
+      dirs: The directories in which to check for usage.
+
+    Returns:
+      True if the element's directory is used in |dirs|.
+    """
+    polymer_dir = os.path.dirname(os.path.realpath(__file__))
+    components_dir = os.path.join(polymer_dir, 'components-chromium')
+    for path in dirs:
+      # Find an import or script referencing the tag's directory.
+      for (dirpath, _, filenames) in os.walk(path):
+        # Ignore the element's own files.
+        if dirpath.startswith(os.path.join(components_dir, element_dir)):
+          continue
+
+        for filename in filenames:
+          if not filename.endswith('.html') and not filename.endswith('.js'):
+            continue
+
+          # Skip generated files that may include the element source.
+          if filename in ('crisper.js', 'vulcanized.html'):
+            continue
+
+          with open(os.path.join(dirpath, filename)) as f:
+            text = f.read()
+          if not re.search('/%s/' % element_dir, text):
+            continue
+
+          # Check the file again, ignoring comments (e.g. example imports and
+          # scripts).
+          if re.search('/%s' % element_dir,
+                       UnusedElementsDetector.__StripComments(
+                         os.path.join(dirpath, filename))):
+            return True
+    return False
+
+
+UnusedElementsDetector.Run()