def scour_xml(doc):
in_string = doc.toxml()
doc.unlink()
options = scour.sanitizeOptions(
Values({
"remove_descriptive_elements": True,
"enable_viewboxing": True,
"strip_ids": True,
"protect_ids_list": "ayah_markers,content",
}))
# scour the string
out_string = scour.scourString(in_string, options)
# prepare the output xml.dom.minidom object
doc = minidom.parseString(out_string.encode("utf-8"))
# since minidom does not seem to parse DTDs properly
# manually declare all attributes with name "id" to be of type ID
# (otherwise things like doc.getElementById() won't work)
all_nodes = doc.getElementsByTagName("*")
for node in all_nodes:
try:
node.setIdAttribute("id")
except NotFoundErr:
pass
return doc
def get_scour_options():
options = scour.sanitizeOptions()
options.remove_descriptions = True
options.remove_descriptive_elements = True
options.remove_metadata = True
options.remove_titles = True
options.shorten_ids = False
options.strip_comments = True
options.strip_xml_prolog = True
options.strip_xml_space_attribute = True
return options
def clean_path_data(path_data, accuracy):
# noinspection PyPep8Naming, PyMethodMayBeStatic
class ElementWrapper:
def __init__(self, data):
self.data = data
def getAttribute(self, attr):
if attr == 'd':
return self.data
else:
return ''
def hasAttribute(self, attr):
if attr == 'd':
return True
else:
return False
def setAttribute(self, attr, value):
if attr == 'd':
self.data = value
else:
return
def nodeType(self):
return Node.ELEMENT_NODE
import scour
scour.scour._num_path_segments_removed = 0
scour.scour._num_bytes_saved_in_path_data = 0
context = decimal.Context(prec=accuracy)
scour.scour.scouringContext = context
scour.scour.scouringContextC = context
wrapper = ElementWrapper(path_data)
cleanPath(wrapper, sanitizeOptions(None))
return wrapper.data
def saveJpg(dest, fileout, quality):
# The image quality, on a scale from 0 (worst) to 95 (best), the default is 75
if quality < 1:
quality = 75
elif quality > 95:
quality = 95
dest.convert('RGB').save(fileout, 'JPEG', optimize=True, subsampling=0, quality=quality)
# python-scour
if ".svg" in fileout:
from scour.scour import sanitizeOptions, start
options = sanitizeOptions()
options.strip_xml_prolog = True # --strip-xml-prolog
options.remove_metadata = True # --remove-metadata
options.strip_comments = True # --enable-comment-stripping
options.strip_ids = True # --enable-id-stripping
options.indent_type = None # --indent=none
start(options, open(filein, 'rb'), open(fileout, 'wb'))
# python-pil
else:
from PIL import Image, ImageSequence
source = Image.open(filein)
size = calcSize(source, size)
if source.format == 'GIF' and ".gif" in fileout:
dest = resizeAnimatedGif(source, size, fixed)
saveGif(dest, fileout, quality)