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

grit/gather/chrome_html.py

+#!/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

[Jói, 2012/05/23 10:22:46]

For triple quotes, we prefer """ (in grit at least, and I think this is Google
style as well).  For regular quotes, prefer ' but can use " if it helps you
avoid escaping.

[flackr, 2012/05/23 19:34:14]

Done, thanks.

+high DPI resources.
+
+This is a small script that takes a HTML file, looks for src attributes

[Jói, 2012/05/23 10:22:46]

small script -> small gatherer (?)

[flackr, 2012/05/23 19:34:14]

Done.

+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_]*')
+
+
+def InsertImageSet(
+    src_match, base_path, scale_factors, distribution):
+  '''Regex replace function which inserts -webkit-image-set.
+
+  Takes a regex match for url('path'). If the file is local, checks for
+  files of the same name in folders corresponding to the supported scale
+  factors. If the file is from a chrome://theme/ source, inserts the
+  supported @Nx scale factor request. In either case inserts a
+  -webkit-image-set rule to fetch the appropriate image for the current
+  scale factor.
+
+  Args:
+    src_match: regex match object with 'filename' named capturing group

[Jói, 2012/05/23 10:22:46]

I would suggest making the regex a lazy_re.compile'd regex at the top level, so
that you can refer to it here, i.e. that the matches come from that regex or one
with the same groups.

[flackr, 2012/05/23 19:34:14]

Done.

+    base_path: path to look for relative file paths in
+    scale_factors: a list of the supported scale factors (i.e. ['2x'])
+    distribution: string that should replace %DISTRIBUTION%.
+
+  Returns:
+    string
+  '''
+  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:

[Jói, 2012/05/23 10:22:46]

why not do 
if filename.find('http://') != -1 and filename.find('https://') != -1:
  ...

(I presume we're not worried about Gopher URLs ;-) although if we do need to
match e.g. ftp URLs we could use a regex to check)

[flackr, 2012/05/23 19:34:14]

This is the same condition as used in html_inline, the code that follows expects
that the filename is a relative file path. We weed to at least worry about
chrome:// as well.

[Jói, 2012/05/23 22:59:54]

OK.  Fine by me to leave as is.

+    # 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])

[Jói, 2012/05/23 10:22:46]

Have you tested this on Windows?  A more platform-agnostic way to do it would be

scale_image_path = os.path.join(scale_path[0], sc, scale_path[1])

I guess in CSS we probably always use / but just a thought.

[flackr, 2012/05/23 19:34:14]

Done. Good point, here we want an OS path but below we want a CSS path. I have
commented the following case as such. I will test on Windows to make sure
everything works as expected.

+    if os.path.isfile(scale_image_path):
+      scale_image_name = re.sub('(?P<path>(.*/)?)(?P<file>[^/]*)',

[Jói, 2012/05/23 10:22:46]

Again here you may need to be careful with / vs. \ and might want to use
os.path.join.  If this is not a practical concern, a line of documentation
somewhere above like 'CSS always uses forward slashed paths' might make sense.

[flackr, 2012/05/23 19:34:14]

Done.

+                                '\\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.'''

[Jói, 2012/05/23 10:22:46]

I think this docstring is copy-pasted and needs to be updated.

[flackr, 2012/05/23 19:34:14]

Done.

+  # 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).decode('utf-8').encode('ascii', 'ignore')
+
+
+class ChromeHtml(interface.GathererBase):
+  '''Represents an HTML document processed for Chrome WebUI.
+
+  HTML documents used in Chrome WebUI have local resources inlined and
+  automatically insert references to high DPI assets used in CSS properties
+  with the use of the -webkit-image-set value. This does not generate any
+  translateable messages and instead generates a single DataPack resource.'''

[Jói, 2012/05/23 10:22:46]

For multi-line docstrings, prefer to put the closing """ on a separate line.

[flackr, 2012/05/23 19:34:14]

Done.

+
+  def __init__(self, html):
+    '''Creates a new object that represents the file 'html'.
+    Args:
+      html: 'filename.html'
+    '''
+    super(type(self), self).__init__()
+    self.filename_ = html
+    self.inlined_text_ = None
+    self.scale_factors_ = []

[Jói, 2012/05/23 10:22:46]

Should this include 100 right off the bat? I guess for that case we don't add an
image set so maybe not?

[flackr, 2012/05/23 19:34:14]

Yes, for 1x/100 we don't add an image set.

+
+  def SetDefines(self, defines):
+    if 'scale_factors' in defines:
+      self.scale_factors_ = defines['scale_factors'].split(',')
+
+  def GetText(self):
+    '''Returns inlined text of the HTML document'''
+    return self.inlined_text_
+
+  def GetData(self, lang, encoding):
+    '''Returns 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):
+    '''Parses and inlines the represented file.'''
+    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)