')
if use_html_entities:
text1 = entifyString(text)
text = attrquot_step_2(text1)
debug('attrquot', 'step 2 strings were different?', text==text1)
stream.write(text)
stream.close()
def appendText(node, text):
"""Parse the text as XML and insert it into the document right
after the given node.
Returns the inserted node.
"""
if type(text) == unicode:
text = text.encode('utf-8')
doc = xml.dom.minidom.parseString(text)
return node.parentNode.insertBefore(doc.documentElement, node.nextSibling)
def hasHTMLClass(node, cls):
"""
Does this node or one of its ancestors have a "class" attribute with
cls as one of the classes?
"""
if node.nodeType == node.DOCUMENT_NODE:
return False
cls_str = node.getAttribute("class")
classes = cls_str.split(" ")
if cls in classes:
return True
return hasHTMLClass(node.parentNode, cls)
def HTMLClasses(node):
"""
Yield sequence of each HTML class word, in order, going outward.
"""
if node.nodeType == node.DOCUMENT_NODE:
return
cls_str = node.getAttribute("class")
classes = cls_str.split(" ")
for cls in classes:
yield cls
for cls in HTMLClasses(node.parentNode):
yield cls
def followedByText(startNode, matchText):
"""
Is this node followed by this character (after any whitespace)?
(Look up and down the tree [using nextLeaf] to figure this out.)
There's some questionable whitespace logic... Some whitespace is
dropped. It's good enough for me for now. Maybe next stage would
be to do a regexp match, but I don't know how to tell if having
more text would satisfy the regexp (and thus we should gather more
text.)
It might also be nice to have a flag/version for seeing/matching
the markup, but that's a different thing.
>>> tree = parseString('
booba (ping)baz')
>>> node = elementById(tree, "start")
>>> followedByText(node, "(")
True
>>> followedByText(node, "(ping)")
True
>>> followedByText(node, "(pong)")
False
>>> followedByText(node, " ")
False
"""
text = ""
node = startNode
while True:
node = nextLeaf(node)
if (node.nodeType == node.TEXT_NODE or
node.nodeType == node.CDATA_SECTION):
text += node.data.strip()
else:
raise RuntimeError("Unexpected node"+repr(node))
if len(text) < len(matchText):
if text == matchText[0:len(text)]:
continue # matching so far...
else:
return False # it'll never match, give up
else:
return text[0:len(matchText)] == matchText
def nextLeaf(node):
"""
Return the next node without children, in an in-order traversal of
the tree. This can be used to navigate among TEXT_EMEMENTS, etc.
>>> tree = parseString('
boobarpingbaz')
>>> node = elementById(tree, "start")
>>> next = nextLeaf(node)
>>> next
>>> next = nextLeaf(next)
>>> next
>>> next = nextLeaf(next)
>>> next
>>> next = nextLeaf(next)
>>> next is None
True
"""
next = node.nextSibling
while next is None:
node = node.parentNode
if node is None:
return None # there is no nextLeaf
next = node.nextSibling
while next.hasChildNodes():
next = next.firstChild
return next
if __name__ == "__main__":
import doctest, sys
doctest.testmod(sys.modules[__name__])