diff options
| author | Leonard Richardson <leonardr@segfault.org> | 2015-06-24 17:03:40 -0400 |
|---|---|---|
| committer | Leonard Richardson <leonardr@segfault.org> | 2015-06-24 17:03:40 -0400 |
| commit | 905f3b0bb450d9304810c712d6d1280b449af041 (patch) | |
| tree | 84e3e6be11f1a980458e83d41747fe61dd98f0a3 /bs4/builder/_html5lib.py | |
| parent | 8fb4389df679d38add64014f5a1b71d6d4820e1f (diff) | |
If the initial <html> tag contains a CDATA list attribute such as
'class', the html5lib tree builder will now turn its value into a
list, as it would with any other tag. [bug=1296481]
Diffstat (limited to 'bs4/builder/_html5lib.py')
| -rw-r--r-- | bs4/builder/_html5lib.py | 14 |
1 files changed, 12 insertions, 2 deletions
diff --git a/bs4/builder/_html5lib.py b/bs4/builder/_html5lib.py index ea8ff43..ad3c6ef 100644 --- a/bs4/builder/_html5lib.py +++ b/bs4/builder/_html5lib.py @@ -9,7 +9,10 @@ from bs4.builder import ( HTML_5, HTMLTreeBuilder, ) -from bs4.element import NamespacedAttribute +from bs4.element import ( + NamespacedAttribute, + whitespace_re, +) import html5lib from html5lib.constants import namespaces from bs4.element import ( @@ -103,7 +106,13 @@ class AttrList(object): def __iter__(self): return list(self.attrs.items()).__iter__() def __setitem__(self, name, value): - "set attr", name, value + # If this attribute is a multi-valued attribute for this element, + # turn its value into a list. + list_attr = HTML5TreeBuilder.cdata_list_attributes + if (name in list_attr['*'] + or (self.element.name in list_attr + and name in list_attr[self.element.name])): + value = whitespace_re.split(value) self.element[name] = value def items(self): return list(self.attrs.items()) @@ -180,6 +189,7 @@ class Element(html5lib.treebuilders._base.Node): return AttrList(self.element) def setAttributes(self, attributes): + if attributes is not None and len(attributes) > 0: converted_attributes = [] |