Sync options

--HG--
extra : convert_revision : svn%3Aacbfec75-9323-0410-a652-858a13e371e0/trunk%40821

parse.py

@@ -91,11 +91,7 @@ def printOutput(parser, document, opts):
  sys.stdout.write(document.hilite("utf-8"))
  elif opts.html:
  kwargs = {}
- for opt in ['inject_meta_charset', 'strip_whitespace', 'sanitize',
- 'omit_optional_tags', 'quote_attr_values', 'quote_char',
- 'use_best_quote_char', 'minimize_boolean_attributes',
- 'use_trailing_solidus', 'escape_lt_in_attrs',
- 'escape_rcdata']:
+ for opt in serializer.HTMLSerializer.options:
  kwargs[opt] = getattr(opts,opt)
  if not kwargs['quote_char']: del kwargs['quote_char']
  tokens = treewalkers.getTreeWalker(opts.treebuilder)(document)
@@ -176,6 +172,11 @@ def getOptParser():
  default=False, dest="use_trailing_solidus",
  help="use trailing solidus")
 
+ parser.add_option("", "--space-before-trailing-solidus",
+ action="store_true", default=False,
+ dest="space_before_trailing_solidus",
+ help="add space before trailing solidus")
+
  parser.add_option("", "--escape-lt-in-attrs", action="store_true",
  default=False, dest="escape_lt_in_attrs",
  help="escape less than signs in attribute values")

src/html5lib/filters/sanitizer.py

@@ -0,0 +1,8 @@
+import _base
+from html5lib.sanitizer import HTMLSanitizerMixin
+
+class Filter(_base.Filter, HTMLSanitizerMixin):
+ def __iter__(self):
+ for token in _base.Filter.__iter__(self):
+ token = self.sanitize_token(token)
+ if token: yield token

src/html5lib/sanitizer.py

@@ -2,7 +2,7 @@
 from xml.sax.saxutils import escape, unescape
 from tokenizer import HTMLTokenizer
 
-class HTMLSanitizer(HTMLTokenizer):
+class HTMLSanitizerMixin:
  """ sanitization of XHTML+MathML+SVG and of inline style attributes."""
 
  acceptable_elements = ['a', 'abbr', 'acronym', 'address', 'area', 'b',
@@ -130,38 +130,37 @@ class HTMLSanitizer(HTMLTokenizer):
  # => <script> do_nasty_stuff() </script>
  # sanitize_html('<a href="javascript: sucker();">Click here for $100</a>')
  # => <a>Click here for $100</a>
- def __iter__(self):
- for token in HTMLTokenizer.__iter__(self):
- if token["type"] in ["StartTag", "EndTag", "EmptyTag"]:
- if token["name"] in self.allowed_elements:
- if token.has_key("data"):
- attrs = dict([(name,val) for name,val in token["data"][::-1] if name in self.allowed_attributes])
- for attr in self.attr_val_is_uri:
- if not attrs.has_key(attr): continue
- val_unescaped = re.sub("[`\000-\040\177-\240\s]+", '', unescape(attrs[attr])).lower()
- if re.match("^[a-z0-9][-+.a-z0-9]*:",val_unescaped) and (val_unescaped.split(':')[0] not in self.allowed_protocols):
- del attrs[attr]
- if attrs.has_key('style'):
- attrs['style'] = self.sanitize_css(attrs['style'])
- token["data"] = [[name,val] for name,val in attrs.items()]
- yield token
- else:
- if token["type"] == "EndTag":
- token["data"] = "</%s>" % token["name"]
- elif token["data"]:
- attrs = ''.join([' %s="%s"' % (k,escape(v)) for k,v in token["data"]])
- token["data"] = "<%s%s>" % (token["name"],attrs)
- else:
- token["data"] = "<%s>" % token["name"]
- if token["type"] == "EmptyTag":
- token["data"]=token["data"][:-1] + "/>"
- token["type"] = "Characters"
- del token["name"]
- yield token
- elif token["type"] == "Comment":
- pass
+ def sanitize_token(self, token):
+ if token["type"] in ["StartTag", "EndTag", "EmptyTag"]:
+ if token["name"] in self.allowed_elements:
+ if token.has_key("data"):
+ attrs = dict([(name,val) for name,val in token["data"][::-1] if name in self.allowed_attributes])
+ for attr in self.attr_val_is_uri:
+ if not attrs.has_key(attr): continue
+ val_unescaped = re.sub("[`\000-\040\177-\240\s]+", '', unescape(attrs[attr])).lower()
+ if re.match("^[a-z0-9][-+.a-z0-9]*:",val_unescaped) and (val_unescaped.split(':')[0] not in self.allowed_protocols):
+ del attrs[attr]
+ if attrs.has_key('style'):
+ attrs['style'] = self.sanitize_css(attrs['style'])
+ token["data"] = [[name,val] for name,val in attrs.items()]
+ return token
  else:
- yield token
+ if token["type"] == "EndTag":
+ token["data"] = "</%s>" % token["name"]
+ elif token["data"]:
+ attrs = ''.join([' %s="%s"' % (k,escape(v)) for k,v in token["data"]])
+ token["data"] = "<%s%s>" % (token["name"],attrs)
+ else:
+ token["data"] = "<%s>" % token["name"]
+ if token["type"] == "EmptyTag":
+ token["data"]=token["data"][:-1] + "/>"
+ token["type"] = "Characters"
+ del token["name"]
+ return token
+ elif token["type"] == "Comment":
+ pass
+ else:
+ return token
 
  def sanitize_css(self, style):
  # disallow urls
@@ -187,3 +186,9 @@ def sanitize_css(self, style):
  clean.append(prop + ': ' + value + ';')
 
  return ' '.join(clean)
+
+class HTMLSanitizer(HTMLTokenizer, HTMLSanitizerMixin):
+ def __iter__(self):
+ for token in HTMLTokenizer.__iter__(self):
+ token = self.sanitize_token(token)
+ if token: yield token

src/html5lib/serializer/htmlserializer.py

@@ -7,10 +7,6 @@
 import gettext
 _ = gettext.gettext
 
-from html5lib.filters.whitespace import Filter as WhitespaceFilter
-from html5lib.filters.optionaltags import Filter as OptionalTagFilter
-from html5lib.filters.inject_meta_charset import Filter as InjectMetaCharsetFilter
-
 from html5lib.constants import voidElements, booleanAttributes, spaceCharacters
 from html5lib.constants import rcdataElements
 
@@ -67,17 +63,16 @@ class HTMLSerializer(object):
  escape_lt_in_attrs = False
  escape_rcdata = False
 
- omit_optional_tags = True
-
- strip_whitespace = False
-
  inject_meta_charset = True
+ strip_whitespace = False
+ sanitize = False
+ omit_optional_tags = True
 
  options = ("quote_attr_values", "quote_char", "use_best_quote_char",
  "minimize_boolean_attributes", "use_trailing_solidus",
  "space_before_trailing_solidus", "omit_optional_tags",
  "strip_whitespace", "inject_meta_charset", "escape_lt_in_attrs",
- "escape_rcdata")
+ "escape_rcdata", 'use_trailing_solidus', "sanitize")
 
  def __init__(self, **kwargs):
  if kwargs.has_key('quote_char'):
@@ -91,13 +86,19 @@ def serialize(self, treewalker, encoding=None):
  in_cdata = False
  self.errors = []
  if encoding and self.inject_meta_charset:
- treewalker = InjectMetaCharsetFilter(treewalker, encoding)
+ from html5lib.filters.inject_meta_charset import Filter
+ treewalker = Filter(treewalker, encoding)
  # XXX: WhitespaceFilter should be used before OptionalTagFilter
  # for maximum efficiently of this latter filter
  if self.strip_whitespace:
- treewalker = WhitespaceFilter(treewalker)
+ from html5lib.filters.whitespace import Filter
+ treewalker = Filter(treewalker)
+ if self.sanitize:
+ from html5lib.filters.sanitizer import Filter
+ treewalker = Filter(treewalker)
  if self.omit_optional_tags:
- treewalker = OptionalTagFilter(treewalker)
+ from html5lib.filters.optionaltags import Filter
+ treewalker = Filter(treewalker)
  for token in treewalker:
  type = token["type"]
  if type == "Doctype":