nest_word_count.py

import sys, re
from pathlib import Path


def preprocessor(text):
    # Remove comments
    text = re.sub(r'\/\/.+', '', text)
    text = re.sub(r'\/\*(\*(?!\/)|[^*])*\*\/', '', text)
    # Remove the package name
    text = re.sub(r'package\s\w+(\.\w+)+;', '', text)
    # Remove imports
    text = re.sub(r'import\s\w+(\.\w+)+;', '', text)
    # Replace numbers
    text = re.sub(r'-?\d+(\.\d+)?(e-?\d+)?', '#', text)
    return text


def split(text):
    text = preprocessor(text)
    words = re.split(r'[^\w\{\}]', text)
    return list(filter(None, words))


def count(dir):
    pathlist = Path(dir).glob('**/*.java')
    result = []
    for path in pathlist:
        path_in_str = str(path)
        file = open(path_in_str, 'r')
        text = file.read()
        words = split(text)
        weight = 0
        for word in words:
            if word == '{':
                weight += 1
            elif word == '}':
                weight -= 1
            else:
                if len(result) <= weight:
                    result.append(0)
                result[weight] += 1
    return result


if __name__ == '__main__':
    if len(sys.argv) == 1:
        print("Usage: %s [directory]" % (sys.argv[0]))
        exit(0)
    print(count(sys.argv[1]))