count of sentences in a set of markdown files

sentences.py

# A simple script to calculate the number of sentences per markdown
# file in a given directory. 
#
# Reference:
# - http://nltk.googlecode.com/svn/trunk/doc/howto/portuguese_en.html
# Author: Alexandre Rademaker

import os
import glob
import re
import nltk

sent_tokenizer = nltk.data.load('tokenizers/punkt/portuguese.pickle')

def parseFile(filename):
    raw = open(filename).read().decode('utf8')
    para = []
    current = []
    for x in raw.split("\n"):
        if len(x) > 0:
            current.append(x)
        else:
            para.append( " ".join(current) )
            current = []
    raw_text =  "\n".join(para)
    sentences = sent_tokenizer.tokenize(raw_text)
    return sentences 

for myf in glob.glob("*.md"):
    print myf, len(parseFile(myf))

You can use glob('*.md') (after importing with from glob import glob) instead of getFiles('.').

Thanks Alvaro! Python is no more my most used programming language! Unfortunately, I needed the NLTK module... I have forgotten the glob module!