def getAdjacentTextElement(current,
topMostNode,
forward=None,
traversingUp=False):
if forward is None:
forward = topMostNode
topMostNode = None
res = None
#print "getAdjacentTextElement", current, topMostNode, forward, traversingUp
# If traversingUp, then the children have already been processed
if not traversingUp:
if DOM.getChildCount(current) > 0:
if forward:
node = DOM.getFirstChild(current)
else:
node = DOM.getLastChild(current)
if DOM.getNodeType(node) == DOM.TEXT_NODE:
res = node
else:
# Depth first traversal, the recursive call deals with
# siblings
res = getAdjacentTextElement(node, topMostNode, forward, False)
if res is None:
if forward:
node = current.nextSibling
else:
node = current.previousSibling
# Traverse siblings
if node is not None:
if DOM.getNodeType(node) == DOM.TEXT_NODE:
res = node
else:
#print node, DOM.getNodeType(node), node.innerHTML
# Depth first traversal, the recursive call deals with
# siblings
res = getAdjacentTextElement(node, topMostNode, forward, False)
# Go up and over if still not found
if (res is None) and (not DOM.compare(current, topMostNode)):
node = current.parentNode
# Stop at document (technically could stop at "html" tag)
if (node is not None) and \
(DOM.getNodeType(node) != DOM.DOCUMENT_NODE):
res = getAdjacentTextElement(node, topMostNode, forward, True)
return res
def getAdjacentTextElement(current, topMostNode, forward=None, traversingUp=False):
if forward is None:
forward = topMostNode
topMostNode = None
res = None
#print "getAdjacentTextElement", current, topMostNode, forward, traversingUp
# If traversingUp, then the children have already been processed
if not traversingUp:
if DOM.getChildCount(current) > 0:
if forward:
node = DOM.getFirstChild(current)
else:
node = DOM.getLastChild(current)
if DOM.getNodeType(node) == DOM.TEXT_NODE:
res = node
else:
# Depth first traversal, the recursive call deals with
# siblings
res = getAdjacentTextElement(node, topMostNode,
forward, False)
if res is None:
if forward:
node = current.nextSibling
else:
node = current.previousSibling
# Traverse siblings
if node is not None:
if DOM.getNodeType(node) == DOM.TEXT_NODE:
res = node
else:
#print node, DOM.getNodeType(node), node.innerHTML
# Depth first traversal, the recursive call deals with
# siblings
res = getAdjacentTextElement(node, topMostNode,
forward, False)
# Go up and over if still not found
if (res is None) and (not DOM.compare(current, topMostNode)):
node = current.parentNode
# Stop at document (technically could stop at "html" tag)
if (node is not None) and \
(DOM.getNodeType(node) != DOM.DOCUMENT_NODE):
res = getAdjacentTextElement(node, topMostNode,
forward, True)
return res
