55 lines
2.0 KiB
Python
55 lines
2.0 KiB
Python
import argparse
|
|
import codecs
|
|
import re
|
|
|
|
def write(output_file, line):
|
|
output_file.write(line + '\n')
|
|
|
|
def write_paragraph(output_file, output_map):
|
|
if (output_map is not None):
|
|
write(output_file, output_map['paragraph'])
|
|
write(output_file, output_map['sentence'])
|
|
write(output_file, '# text = ' + ' '.join(output_map['texts']))
|
|
for (index, token_line) in enumerate(output_map['tokens'], start=1):
|
|
write(output_file, '\t'.join([str(index)] + token_line.split('\t')[1:]))
|
|
write(output_file, '')
|
|
|
|
def tweak(input_file_name, output_file_name):
|
|
|
|
output_file = codecs.open(output_file_name, 'w')
|
|
input_file = codecs.open(input_file_name, 'r')
|
|
|
|
output_map = None
|
|
for line in input_file:
|
|
if (line[0].isdigit()):
|
|
output_map['tokens'].append(line.strip())
|
|
else:
|
|
match = re.search('^# (.+?) = (.+)$', line)
|
|
if (match):
|
|
(name, value) = match.groups()
|
|
if (name == 'newpar id'):
|
|
write_paragraph(output_file, output_map)
|
|
paragraph_line = re.sub('^(# newpar id = )(\d+)$', r'\1p\2', line.strip())
|
|
output_map = {'paragraph': paragraph_line, 'sentence':None, 'texts':[], 'tokens':[]}
|
|
elif (name == 'sent_id'):
|
|
if (value.endswith('.1')):
|
|
output_map['sentence'] = re.sub('^(# sent_id = )(\d+\.1)$', r'\1s\2', line.strip())
|
|
elif (name == 'text'):
|
|
output_map['texts'].append(value)
|
|
write_paragraph(output_file, output_map)
|
|
|
|
input_file.close()
|
|
output_file.close()
|
|
|
|
|
|
if (__name__ == '__main__'):
|
|
|
|
arg_parser = argparse.ArgumentParser(description='Fix invalid XML ids.')
|
|
arg_parser.add_argument('-infile', type=str, help='Input file')
|
|
arg_parser.add_argument('-outfile', type=str, help='Output file')
|
|
arguments = arg_parser.parse_args()
|
|
input_file_name = arguments.infile
|
|
output_file_name = arguments.outfile
|
|
|
|
tweak(input_file_name, output_file_name)
|