def localize_misc(dom, path, prefix='misc'):
'''
A few standalone resources to localize, including an xml, an ico and an iframe
'''
misc_path = os.path.join(path, prefix)
if not os.path.exists(misc_path):
os.mkdir(misc_path)
re_search = re.compile(
'<link rel="search" type="application/opensearchdescription\+xml" \
href="(?P<url>http://.*?\.xml)" title="Facebook">')
m_search = re_search.search(dom)
url = m_search.group('url')
file = FBParser.url_to_file(url)
FBParser.save_resource(url, misc_path, file)
dom = dom.replace(url, os.path.join(prefix, file))
re_ico = re.compile(
'<link rel="shortcut icon" href="(?P<url>http://.*?\.ico)">')
m_ico = re_ico.search(dom)
url = m_ico.group('url')
file = FBParser.url_to_file(url)
FBParser.save_resource(url, misc_path, file)
dom = dom.replace(url, os.path.join(prefix, file))
re_uicif = re.compile('<iframe src="(?P<url>http://.*?\.html)"')
m_uicif = re_uicif.search(dom)
if m_uicif:
url = m_uicif.group('url')
file = FBParser.url_to_file(url)
FBParser.save_resource(url, misc_path, file)
dom = dom.replace(url, os.path.join(prefix, file))
# redirect the rest of the hrefs to about:blank (most of them are hyperlinks)
# (call this after localize_css!)
dom = re.sub('href="http://.+?"', 'href="about:blank"', dom)
return dom
def anonym_images(dom, path, filename):
'''
Anonymize images and regenerate file names.
'''
img_files = FBParser.get_content(
os.path.join(path, filename.rstrip('html') + 'img_list'),
encoding='ascii')
img_files = img_files.split('\n')
images = {}
if os.path.isfile(
os.path.join(path, filename.rstrip('html') + 'img_mapping')):
# already have garbled images, only replace filenames in dom
img_mapping = FBParser.get_content(
os.path.join(path, filename.rstrip('html') + 'img_mapping'),
encoding='ascii')
img_mapping = img_mapping.split('\n')
for mapping in img_mapping:
img_file, new_file = mapping.split(': ')
dom = dom.replace(img_file, new_file)
else: # garbled image not generated yet
for img_file in img_files:
ext = os.path.splitext(img_file)[1]
prefix = os.path.split(img_file)[0]
new_file = prefix + '/anonym_' + str(random.getrandbits(40)) + ext
images[img_file] = new_file
dom = dom.replace(img_file, new_file)
FBParser.save_resource(os.path.join(path, img_file), path, new_file)
st_mapping = []
for key, val in images.items():
st_mapping.append(key + ': ' + val)
FBParser.save_content(
'\n'.join(st_mapping),
os.path.join(path, filename.rstrip('html') + 'img_mapping'),
encoding='ascii')
for image in images:
images[image] = os.path.join(path, images[image])
FBParser.garble_image.garble(images.values())
return dom
def anonym_images(dom, path, filename):
'''
Anonymize images and regenerate file names.
'''
img_files = FBParser.get_content(
os.path.join(path, filename.rstrip('html') + 'img_list'),
encoding='ascii')
img_files = img_files.split('\n')
images = {}
for img_file in img_files:
ext = os.path.splitext(img_file)[1]
prefix = os.path.split(img_file)[0]
new_file = prefix + '/anonym_' + str(random.getrandbits(40)) + ext
images[img_file] = new_file
dom = dom.replace(img_file, new_file)
FBParser.save_resource(os.path.join(path, img_file), path, new_file)
os.remove(os.path.join(path, img_file))
st_mapping = []
for image in images:
images[image] = os.path.join(path, images[image])
FBParser.garble_image.garble(images.values())
return dom