takes a directory of files (in this case PDF targets, and creates a distribute list of N processing files (to push out to a poor man's cluster using something like <script src="https://gist.github.com/tarxvf/ba49560220e69a90d1df379a3f309150.js"></script>

process_list_prep.py

import os
rootdir = '/tmp/'
import re

count=0

for subdir, dirs, files in os.walk(rootdir):
    for file in files:
      #print file
      m = re.search("[0-9]+\.pdf$",file)
      
      if hasattr(m, 'group'):
        if m.group(0):
          #print os.path.join(subdir, file)
          count=count+1 
          file_name="/tmp/process_group_{}.txt".format(int(count/501))
          print "filename={0} count={1}".format(file_name,count)
          with open(file_name, "a") as text_file:
              text_file.write("{0}\n".format(file))