handle-sgm.py

# made by Alexander Rosenberg Johansen
# BSD-3 license

import glob
import subprocess

test_paths_2014 = [
    'data/test-full/newstest2014-deen-src.en.sgm',
    'data/test-full/newstest2014-deen-src.de.sgm']

test_paths = test_paths_2014 

to_paths_2014 = [
    'data/test-full/newstest2014.deen.en',
    'data/test-full/newstest2014.deen.de']

to_paths = to_paths_2014

for test_path, to_path in zip(test_paths, to_paths):
    call = 'grep "^<seg" %s | sed "s/<\/\?[^>]\+>//g" > %s' %(test_path, to_path)
    subprocess.call(call, shell=True)