reduce.py

import json
import re
import os

def is_empty(contents):
  for line in contents.splitlines():
    line = line.strip()
    if line and not line.startswith('#'):
      return False
  return True


def read_mapping(file_name):
  with open(file_name) as f:
    mapping = json.load(f)['dirs']
  return mapping

original = read_mapping('/home/nodir/tmp/dirmd/original.json')
reduced = read_mapping('/home/nodir/tmp/dirmd/reduced.json')

for dir, orig_md in original.iteritems():
  if dir.startswith('tools/binary_size/libsupersize/testdata/'):
    continue

  reduced_md = reduced.get(dir) or {}
  strip_monorail = 'monorail' in orig_md and 'monorail' not in reduced_md
  strip_os = 'os' in orig_md and 'os' not in reduced_md
  strip_email = 'teamEmail' in orig_md and 'teamEmail' not in reduced_md
  if strip_monorail or strip_os or strip_email:
    file_name = os.path.join('/home/nodir/chromium/src', dir, 'DIR_METADATA')
    keep_file = True
    if not os.path.exists(file_name):
      continue
    with open(file_name, 'r+') as f:
      data = f.read()
      if strip_monorail:
        data = re.sub('monorail:?\s*{[^}]+}\n?', '', data, flags=re.M)
      if strip_email:
        data = re.sub('team_email:.+\n?', '', data)
      if strip_os:
        data = re.sub('os:.+\n?', '', data)

      # Replace new double blank lines
      data = data.replace('\n\n\n', '\n\n')

      keep_file = not is_empty(data)
      if keep_file:
        f.seek(0)
        f.write(data)
        f.truncate()
    if not keep_file:
      os.remove(file_name)
    print 'wrote or deleted %s' % file_name