html2text.py

from readability import Document
import os
import sys
import io
for fn in os.listdir(sys.argv[1]):
    try:
        with io.open(os.path.join(sys.argv[1], fn), encoding='utf-8') as f:
            text = f.read()
            if 'batch_fails' not in fn : 
                doc = Document(text)
                readable_article = doc.summary()
                readable_title = doc.short_title()
                text = readable_title + '\n' + readable_article
                
            with io.open(os.path.join(sys.argv[2], fn), 'w', encoding='utf-8') as writer:
                 writer.write(text)
    except:
        pass