Add chrome_html gatherer, which inlines html and automatically generates image set…

grit/gather/chrome_html.py

Index: grit/gather/chrome_html.py
diff --git a/grit/gather/chrome_html.py b/grit/gather/chrome_html.py
new file mode 100755
index 0000000000000000000000000000000000000000..5b273f9a7b1f7d104551c4031800c5cc378aee09
--- /dev/null
+++ b/grit/gather/chrome_html.py
@@ -0,0 +1,132 @@
+#!/usr/bin/env python
+# Copyright (c) 2012 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.
+
+"""Prepares a Chrome HTML file by inlining resources and adding references to high DPI resources.

[Jói, 2012/05/17 19:39:12]

80 char limit

[flackr, 2012/05/22 21:25:03]

Done.

+
+This is a small script that takes a HTML file, looks for src attributes
+and inlines the specified file, producing one HTML file with no external
+dependencies. It recursively inlines the included files. When inlining CSS
+image files this script also checks for the existence of high DPI versions
+of the inlined file including those on relevant platforms.
+"""
+
+import os
+import re
+import sys
+import types
+import base64
+import mimetypes
+
+from grit.gather import interface
+from grit.format import html_inline
+from grit import lazy_re
+from grit import util
+
+# Matches a chrome theme source URL.
+_THEME_SOURCE = lazy_re.compile('chrome://theme/IDR_[A-Z0-9_]*')
+

[Jói, 2012/05/17 19:39:12]

two blank lines between top-level elements (here and elsewhere)

[flackr, 2012/05/22 21:25:03]

Done.

+def InsertImageSet(
+    src_match, base_path, scale_factors, distribution):
+  filename = src_match.group('filename')
+  attr = src_match.group('attribute')
+  prefix = src_match.string[src_match.start():src_match.start('filename')-1]
+
+  # Any matches for which a chrome URL handler will serve all scale factors
+  # can simply request all scale factors.
+  if _THEME_SOURCE.match(filename):
+    images = ["url(\"%s\") %s" % (filename, '1x')]
+    for sc in scale_factors:
+      images.append("url(\"%s@%s\") %s" % (filename, sc, sc))
+    return "%s: -webkit-image-set(%s" % (attr, ', '.join(images))
+
+  if filename.find(':') != -1:
+    # filename is probably a URL, which we don't want to bother inlining
+    return src_match.group(0)
+
+  filename = filename.replace('%DISTRIBUTION%', distribution)
+  filepath = os.path.join(base_path, filename)
+  images = ["url(\"%s\") %s" % (filename, '1x')]
+
+  for sc in scale_factors:
+    # check for existence of file and add to image set.
+    scale_path = os.path.split(os.path.join(base_path, filename))
+    scale_image_path = "%s/%s/%s" % (scale_path[0], sc, scale_path[1])
+    if os.path.isfile(scale_image_path):
+      scale_image_name = re.sub('(?P<path>(.*/)?)(?P<file>[^/]*)',
+                                '\\g<path>' + sc + '/\\g<file>',
+                                filename)
+      images.append("url(\"%s\") %s" % (scale_image_name, sc))
+  return "%s: -webkit-image-set(%s" % (attr, ', '.join(images))
+
+def InsertImageSets(
+    filepath, text, scale_factors, distribution):
+  """Helper function that inlines external images in CSS backgrounds."""
+  # Add high DPI urls for css attributes: content, background,
+  # or *-image.
+  return re.sub('(?P<attribute>content|background|[\w-]*-image):[ ]*' +
+                'url\((?:\'|\")(?P<filename>[^"\'\)\(]*)(?:\'|\")',
+                lambda m: InsertImageSet(m, filepath, scale_factors,
+                                         distribution),
+                text).encode('ascii', 'ignore')
+
+class ChromeHtml(interface.GathererBase):
+  '''Represents an HTML document.'''

[Jói, 2012/05/17 19:39:12]

Maybe extend this to note the functionality of this type, and that you do not
extract any translateable messages (most gatherers do, and generate translated
versions of the resources they reference).

[flackr, 2012/05/22 21:25:03]

Done.

+
+  def __init__(self, html):
+    '''Creates a new object that represents 'text'.
+    Args:
+      html: 'filename.html'
+    '''
+    super(type(self), self).__init__()
+    self.filename_ = html
+    self.inlined_text_ = None
+    self.scale_factors_ = []
+
+  def SetAttributes(self, attrs):
+    '''Sets node attributes used by the gatherer.
+
+    This checks the scale_factors attribute.
+
+    Args:
+      attrs: The mapping of node attributes.
+    '''
+    if 'scale_factors' in attrs:
+      self.scale_factors_ = attrs['scale_factors'].split(' ')
+
+  def GetText(self):
+    '''Returns the original text of the HTML document'''
+    return self.inlined_text_
+
+  def GetData(self, lang, encoding):
+    '''Return inlined text of the HTML document'''
+    return self.inlined_text_
+
+  def Translate(self, lang, pseudo_if_not_available=True,
+                skeleton_gatherer=None, fallback_to_english=False):
+    '''Returns this document translated.'''
+    return self.inlined_text_
+
+  def Parse(self):
+    self.inlined_text_ = html_inline.InlineToString(self.filename_, None,
+        rewrite_function=lambda fp, t, d: InsertImageSets(
+            fp, t, self.scale_factors_, d))
+
+  @staticmethod
+  def FromFile(html, extkey=None, encoding = 'utf-8'):
+    '''Creates a ChromeHtml object for the contents of 'html'.  Returns a new
+    ChromeHtml object.
+          
+    Args:
+      html: file('') | 'filename.html'
+      extkey: ignored
+      encoding: 'utf-8' (encoding is ignored)
+
+    Return:
+      ChromeHtml(text_of_file)
+    '''
+    if not isinstance(html, types.StringTypes):
+      html = html.name
+   
+    return ChromeHtml(html)