TVHeadend Python PostProcessing script

pp.py

"""
Author: ckarrie <ckarrie@gmail.com>

Python PostProcessing script that deletes all duplicate recordings according to their
description and filesize. The script keeps the recording with the largest filesize.

Usage:

1) put this python file somewhere (i.e. /media/daten/tvheadend/pp.py)
2) Add this line to TVHeadend > Configuration > Recording > [your recording settings] > Post-processor command
   replace <path/to/this/file/pp.py> according to Point 1)

        /usr/bin/python2.7 /media/daten/tvheadend/pp.py "%f" "%d"
        
3) Record something with this recording settings

"""


import sys, os
import shutil
import re
import syslog

_p = "[TVH PP pp.py] "

#TVH_HOST = "127.0.0.1"
#TVH_PORT = 9981


args = sys.argv

current_recording_file = args[1]
current_description_file = current_recording_file + '.txt'
current_description = args[2]

descriptionfile = open(current_recording_file + ".txt", "w")
descriptionfile.write(current_description)
descriptionfile.close()

dvr_folder, dvr_filename = os.path.split(current_recording_file)

same_description = {}

print("Scanning DVR folder", dvr_folder)

syslog.syslog(_p + 'Start processing in %s' % dvr_folder)

for root, dirs, files in os.walk(dvr_folder):
    syslog.syslog(_p + 'Walking through %s' % root)
    for name in files:
        full_filename = os.path.join(root, name)
        
        if name.endswith('.mkv'):
            syslog.syslog(_p + 'Checking %s' % full_filename)
            full_txt_filename = full_filename + '.txt'
            
            if os.path.exists(full_txt_filename):            
                description_f = open(full_txt_filename, 'r')
                description = description_f.read()
                description_f.close()
                description_shortened = description[:100]
                
                if description_shortened not in same_description.keys():
                    same_description[description_shortened] = [full_filename]
                else:
                    same_description[description_shortened].append(full_filename)
            
            else:
                print("Missing description file %s, ignoring..." % full_txt_filename)

        
for descr, files in same_description.items():

    print(" >%s< has %d file(s)" %(descr[:50], len(files)))
    
    largest_size = 0
    largest_size_filename = ''
    
    for f in files:
        size = os.path.getsize(f)
        if size > largest_size:
            largest_size = size
            largest_size_filename = f
            
    if largest_size and largest_size_filename and (len(files) > 1):
        for fname in files:
            if fname != largest_size_filename:
                 print("  - Removing", fname)
                 os.remove(fname)
                 os.remove(fname + '.txt')

syslog.syslog(_p + "All done")

TODO: delete those database entries if file is missing