Enhanced white_spacey.py

gistfile1.py

'''
Created on Apr 19, 2014

@author: Chris

Edited by Dan on Jul 21, 2014
'''

import os 
import re
import codecs
from functools import partial

def get_src_files(root, extensions=['.c', '.cpp', '.h', '.hpp']):
  for root, _, files in os.walk(root):
    yield from (os.path.join(root, f) for f in files
                if any(f.endswith(ext) for ext in extensions))
    
def read_files(filenames):
  for filename in filenames:
    try:
      with codecs.open(filename, 'r') as f:
        yield from f.readlines()
    except ValueError:
      print("Couldn't read %s" % filename)
      pass

def extract_forloops(lines):
  yield from (
      line.strip() for line in lines
      if 'for (' in line
      # One of those god forsaken 'enhanced' for loops. Ignore
      and ':' not in line
  )

class Codebase:
    def __init__(self, name, path):
        self.name = name
        self.path = path

    def get_loops(self):
        return extract_forloops(read_files(get_src_files(self.path)))

if __name__ == '__main__':
  spaced_apart = partial(re.search, '[a-zA-Z]\s\=')
  grouped_together = partial(re.search, '[a-zA-Z]\=')

  codebases = [
    Codebase("Linux kernel", "linux-3.15.6"),
    Codebase("Git", "git"),
    Codebase("OpenSSL", "openssl-1.0.1h"),
    Codebase("Python 3", "Python-3.4.1"),
    Codebase("Ruby", "ruby-2.1.2"),
    ]
  
  for code in codebases:
      forloops = code.get_loops()

      counts = {'spaced': 0, 'nonspaced':0}
      for loop in forloops:
        if grouped_together(loop):
          counts['nonspaced'] += 1
        elif spaced_apart(loop):
          counts['spaced'] += 1

      print(code.name)
      print(counts)