def equal(element1, element2, ignoreattrs=False):
"""Return True if two elements are equal, false if they are not.
The ignoreattrs option, when True, causes this function to ignore
element attributes when checking equality. Added as a requirement for
another project (Similar Content), should find a cleaner way of
achieving the same result. """
log = logging.getLogger()
#tags
tag1 = Tag._cleantag(element1.tag)
tag2 = Tag._cleantag(element2.tag)
if tag1 != tag2:
return False
#attrs
attrib1 = element1.attrib
Attrib.cleanKeys(attrib1)
attrib2 = element2.attrib
Attrib.cleanKeys(attrib2)
if (attrib1 != attrib2) and (not ignoreattrs):
return False
#text
try:
text1 = element1.text.rstrip()
except AttributeError:
text1 = ''
try:
text2 = element2.text.rstrip()
except AttributeError:
text2 = ''
if text1 != text2:
return False
#tail
try:
tail1 = element1.tail.rstrip()
except AttributeError:
tail1 = ''
try:
tail2 = element2.tail.rstrip()
except:
tail2 = ''
if tail1 != tail2:
return False
return True
def add(element1, element2):
"""Add element2 to element1, modify element1 in place and return it.
This function modifies the node in place to make ensure there are not
too many copies created in complicated programs.
The reason this function also returns the element is to allow this function
to be used in compound statements, for example:
add(node1, invert(add(node2, node3)))"""
newtag = Tag.addTags(element1.tag, element2.tag)
#have to wrap the attributes in dict() to avoid a bus error
newattribs = Attrib.addAttribs(dict(element1.attrib), dict(element2.attrib))
element1.tag = newtag
element1.text = Text.addText(element1.text, element2.text)
element1.tail = Text.addText(element1.tail, element2.tail)
for i in element1.attrib:
del element1.attrib[i]
for key in newattribs.keys():
try:
element1.set(key, newattribs[key])
except TypeError:
log = logging.getLogger()
log.error('TypeError: %s' % str(sys.exc_info()[1]))
log.error('key = %s\tnewattribs[key] = %s' % (str(key), str(newattribs[key])))
raise
return element1
def isUnitEvent(event):
if event.getEventName() == "StartElementEvent":
if Tag.isUnitTag(event.tag) and Attrib.isUnitAttrib(event.attr) and Text.isUnitText(event.text):
return True
else:
return False
elif event.getEventName() == "EndElementEvent":
if Tag.isUnitTag(event.tag) and Text.isUnitText(event.text):
return True
else:
return False
else:
raise TypeError("Invalid event type: %s" % str(event1))
def addEvents(event1, event2):
"""Add the two events and return a new event. Raise TypeError if the event
types do not match."""
import TreeGroup.Common.Attrib as Attrib
import TreeGroup.Common.Tag as Tag
import TreeGroup.Common.Text as Text
if not type(event1) == type(event2):
print("%s != %s" % (type(event1), type(event2)))
raise TypeError
if event1.getEventName() == "StartElementEvent":
newevent = StartElementEvent(Tag.addTags(event1.tag, event2.tag), Attrib.addAttribs(event1.attr, event2.attr), Text.addText(event1.text, event2.text))
return newevent
elif event1.getEventName() == "EndElementEvent":
newevent = EndElementEvent(Tag.addTags(event1.tag, event2.tag), Text.addText(event1.text, event2.text))
return newevent
else:
raise TypeError("Invalid event type: %s" % str(event1))
def invert(element1):
"""Invert the element, modify in place and return it.
This function modifies the node in place to make ensure there are not
too many copies created in complicated programs.
The reason this function also returns the element is to allow this function
to be used in compound statements, for example:
add(node1, invert(add(node2, node3)))
"""
# inverted = lxml.etree.Element(Tag.tagInverse(element1.tag), )
element1.tag = Tag.tagInverse(element1.tag)
element1.text = Text.textInverse(element1.text)
element1.tail = Text.textInverse(element1.tail)
newattribs = Attrib.attribInverse(dict(element1.attrib))
for i in element1.attrib:
del element1.attrib[i]
for j in newattribs:
element1.set(j, newattribs[j])
return element1
def __eq__(self, o):
if type(self) == type(o) and Tag.equal(self.tag, o.tag) and Attrib.equal(self.attr, o.attr) and Text.equal(self.text, o.text):
return True
else:
return False