You can not select more than 25 topics
Topics must start with a letter or number, can include dashes ('-') and can be up to 35 characters long.
39 lines
1.5 KiB
39 lines
1.5 KiB
import argparse
|
|
import lxml.etree as lxml
|
|
|
|
|
|
def xpath_find(element,expression):
|
|
return element.xpath(expression, namespaces={'tei':'http://www.tei-c.org/ns/1.0'})
|
|
|
|
|
|
def count_tokens(paragraph):
|
|
return len(xpath_find(paragraph, './/tei:w|.//tei:pc'))
|
|
|
|
|
|
def split(input_file_name, single_file_name, multiple_file_name):
|
|
|
|
tree = lxml.parse(input_file_name)
|
|
root = tree.getroot()
|
|
paragraphs = xpath_find(root, './/tei:p')
|
|
for paragraph in paragraphs:
|
|
if (count_tokens(paragraph) > 1):
|
|
paragraph.getparent().remove(paragraph)
|
|
tree.write(single_file_name, encoding='UTF-8', pretty_print=True)
|
|
|
|
tree = lxml.parse(input_file_name)
|
|
root = tree.getroot()
|
|
paragraphs = xpath_find(root, './/tei:p')
|
|
for paragraph in paragraphs:
|
|
if (count_tokens(paragraph) == 1):
|
|
paragraph.getparent().remove(paragraph)
|
|
tree.write(multiple_file_name, encoding='UTF-8', pretty_print=True)
|
|
|
|
|
|
if (__name__ == '__main__'):
|
|
arg_parser = argparse.ArgumentParser(description='Split input TEI into single and multiple token units.')
|
|
arg_parser.add_argument('-infile', type=str, required=True, help='Input TEI file')
|
|
arg_parser.add_argument('-single', type=str, required=True, help='Output single token TEI file')
|
|
arg_parser.add_argument('-multiple', type=str, required=True, help='Output multiple token TEI file')
|
|
arguments = arg_parser.parse_args()
|
|
split(arguments.infile, arguments.single, arguments.multiple)
|