def load_mrd(dir):
print('parsing source dictionary file...')
mrd_source = MrdDataSource(os.path.join(dir,'morphs.utf8.mrd'),
os.path.join(dir,'gramtab.utf8.mrd'),
strip_EE=True)
mrd_source.load()
print('calculating rule frequencies...')
mrd_source.calculate_rule_freq()
return mrd_source
coding, 'utf8')
convert_file(os.path.join(src_dir, 'Morph', gramtab),
os.path.join(dest_dir, lang, 'gramtab.utf8.mrd'),
coding, 'utf8')
def cleanup_after_convert(dir):
print('cleaning up...')
os.unlink(os.path.join(dir, 'morphs.utf8.mrd'))
os.unlink(os.path.join(dir, 'gramtab.utf8.mrd'))
print("========")
if __name__ == '__main__':
MrdDataSource.setup_psyco()
src_dir = 'dicts/src/Dicts'
dest_dir = 'dicts/converted'
LANGUAGES = 'en', 'ru'
FORMATS = 'cdb', 'sqlite', # 'tinycdb', 'cdblib', # 'shelve',
for lang in LANGUAGES:
convert_dicts(src_dir, dest_dir, lang)
dest = os.path.join(dest_dir, lang)
mrd = load_mrd(dest)
make_pickled(dest, mrd)
for fmt in FORMATS: