nlp/spacy-nlp

#!/usr/bin/env python3.5
# coding=utf-8

from argparse import ArgumentParser
from xml.sax.saxutils import escape
import chardet
import spacy
import textwrap


SPACY_MODELS = {'de': 'de_core_news_sm',
                'el': 'el_core_news_sm',
                'en': 'en_core_web_sm',
                'es': 'es_core_news_sm',
                'fr': 'fr_core_news_sm',
                'it': 'it_core_news_sm',
                'nl': 'nl_core_news_sm',
                'pt': 'pt_core_news_sm'}


# Parse the given arguments
parser = ArgumentParser(description=('Tag a text file with spaCy and save it '
                                     'as a verticalized text file.'))
parser.add_argument('i', metavar='txt-sourcefile')
parser.add_argument('o', metavar='vrt-destfile')
parser.add_argument('-l', '--language', choices=SPACY_MODELS.keys(),
                    required=True)
parser.add_argument('--check-encoding', action='store_true')
args = parser.parse_args()


# If requested: Check the encoding of the text contents from the input file
# Else: Use utf-8
if args.check_encoding:
    with open(args.i, "rb") as input_file:
        bytes = input_file.read()
        encoding = chardet.detect(bytes)['encoding']
else:
    encoding = 'utf-8'


# Load the text contents from the input file
with open(args.i, encoding=encoding) as input_file:
    text = input_file.read()
    # spaCys NLP is limited to strings with maximum 1 million characters at
    # once. So we split it into suitable chunks.
    text_chunks = textwrap.wrap(text, 1000000, break_long_words=False)
    # the text variable potentially occupies a lot of system memory and is no
    # longer needed...
    del text


# Setup the spaCy toolkit by loading the chosen language model
nlp = spacy.load(SPACY_MODELS[args.language])


# Create the output file in verticalized text format
# See: http://cwb.sourceforge.net/files/CWB_Encoding_Tutorial/node3.html
output_file = open(args.o, 'w+')
output_file.write('<?xml version="1.0" encoding="UTF-8"?>\n<corpus>\n<text>\n')
for text_chunk in text_chunks:
    doc = nlp(text_chunk)
    for sent in doc.sents:
        output_file.write('<s>\n')
        for token in sent:
            # Skip whitespace tokens
            if token.text.isspace():
                continue
            output_file.write('{}'.format(escape(token.text))
                              + '\t{}'.format(escape(token.lemma_))
                              + '\t{}'.format(token.pos_)
                              + '\t{}'.format(token.tag_)
                              + '\t{}\n'.format(token.ent_type_ or 'NULL'))
        output_file.write('</s>\n')
output_file.write('</text>\n</corpus>')
output_file.close()