def get_sentence(element, element_text, text_context, xtext=XTEXT):
indexes = find_list(text_context, element_text)
size = len(indexes)
if size == 0:
return ''
elif size == 1:
return find_sentence(text_context, indexes[0],
indexes[0] + len(element_text))
else:
parent = element.getparent()
child_index_in_parent = 0
for child in parent:
if child == element:
break
else:
temp_text = normalize(xtext(child))
# We have encountered a child that has the same text,
# so the first index is not the good one.
if temp_text.find(element_text) != -1:
child_index_in_parent += 1
if child_index_in_parent < size:
return find_sentence(
text_context, indexes[child_index_in_parent],
indexes[child_index_in_parent] + len(element_text))
else:
# Something went wrong.
return find_sentence(element_text, indexes[0],
indexes[0] + len(element_text))
def get_sentence(element, element_text, text_context, xtext=XTEXT):
indexes = find_list(text_context, element_text)
size = len(indexes)
if size == 0:
return ""
elif size == 1:
return find_sentence(text_context, indexes[0], indexes[0] + len(element_text))
else:
parent = element.getparent()
child_index_in_parent = 0
for child in parent:
if child == element:
break
else:
temp_text = normalize(xtext(child))
# We have encountered a child that has the same text,
# so the first index is not the good one.
if temp_text.find(element_text) != -1:
child_index_in_parent += 1
if child_index_in_parent < size:
return find_sentence(
text_context, indexes[child_index_in_parent], indexes[child_index_in_parent] + len(element_text)
)
else:
# Something went wrong.
return find_sentence(element_text, indexes[0], indexes[0] + len(element_text))
def get_text(self, element, complex_text=False):
'''Computes the text of this element by creating a deepcopy of the
element, removing the bad children, getting the text
representation.
This is quite inefficient, but it's the best way to get a good
representation of the text in a hierarchical section.'''
new_element = deepcopy(element)
bad_elements = self.first_filter(new_element)
for bad_element in bad_elements:
if bad_element in new_element:
new_element.remove(bad_element)
if complex_text:
text_parts = []
get_recursive_text(new_element, text_parts)
else:
text_parts = new_element.xpath('.//text()')
text = '\n'.join(text_parts)
return normalize(text)
def get_text(self, element, complex_text=False):
"""Computes the text of this element by creating a deepcopy of the
element, removing the bad children, getting the text
representation.
This is quite inefficient, but it's the best way to get a good
representation of the text in a hierarchical section."""
new_element = deepcopy(element)
bad_elements = self.first_filter(new_element)
for bad_element in bad_elements:
if bad_element in new_element:
new_element.remove(bad_element)
if complex_text:
text_parts = []
get_recursive_text(new_element, text_parts)
else:
text_parts = new_element.xpath(".//text()")
text = "\n".join(text_parts)
return normalize(text)
def get_text_from_parent(self, parent, index=0, complex_text=False):
text = ''
elem = self.get_element(parent, index)
if elem is not None:
text = self.get_text(elem, complex_text)
return normalize(text)
def get_text(self, element):
return normalize(self.xtext(element))
def get_text_from_parent(self, parent, index=0):
text = ''
elem = self.get_element(parent, index)
if elem is not None:
text = self.xtext(elem)
return normalize(text)
def get_text(element, xtext=XTEXT):
return normalize(xtext(element))
def get_text(element, xtext=XTEXT):
return normalize(xtext(element))
def get_text_from_parent(self, parent, index=0, complex_text=False):
text = ""
elem = self.get_element(parent, index)
if elem is not None:
text = self.get_text(elem, complex_text)
return normalize(text)
def get_text(self, element):
return normalize(self.xtext(element))
def get_text_from_parent(self, parent, index=0):
text = ""
elem = self.get_element(parent, index)
if elem is not None:
text = self.xtext(elem)
return normalize(text)
