torrent-rename

#!/usr/bin/python2
# coding: UTF-8
# vi: set noet sw=4 ts=4 nolist :

import sys
import os
import shutil
import re
import json
import urllib2
import string
import logging
import logging.handlers
import atexit
import subprocess


try:
	_, torrentid, torrentname, torrentpath = sys.argv
	print torrentid, torrentname, torrentpath
except ValueError:
	pass


class BufferingSMTPHandler(logging.handlers.BufferingHandler):
	def __init__(self, mailhost, fromaddr, toaddrs, subject, capacity):
		logging.handlers.BufferingHandler.__init__(self, capacity)
		self.mailhost = mailhost
		self.mailport = None
		self.fromaddr = fromaddr
		self.toaddrs = toaddrs
		self.subject = subject
		self.setFormatter(logging.Formatter("%(asctime)s %(levelname)-5s %(message)s"))

	def flush(self):
		if len(self.buffer) > 0:
			try:
				import smtplib
				port = self.mailport
				if not port:
					port = smtplib.SMTP_PORT
				smtp = smtplib.SMTP(self.mailhost, port)
				msg = "From: %s\r\nTo: %s\r\nSubject: %s\r\n\r\n" % (self.fromaddr, string.join(self.toaddrs, ","), self.subject)
				for record in self.buffer:
					s = self.format(record)
					msg = msg + s + "\r\n"
				smtp.sendmail(self.fromaddr, self.toaddrs, msg)
				smtp.quit()
			except Exception, exc:
				raise
				print exc
				self.handleError(None)  # no particular record
			self.buffer = []


logger = logging.getLogger("torrent-rename")
logger.setLevel(logging.DEBUG)

# This handler emails me anything that is an error or worse.
mailhandler = BufferingSMTPHandler('localhost', 'cmiller@zippy', ['root@zippy'], 'torrent-rename abnormal events', 500)
mailhandler.setLevel(logging.INFO)
logformatter = logging.Formatter("line%(lineno)d %(message)s")
mailhandler.setFormatter(logformatter)
logger.addHandler(mailhandler)

stdouthandler = logging.StreamHandler(stream=sys.stdout)
logger.addHandler(stdouthandler)

atexit.register(logging.shutdown)


quash = ("hdtv", "x264", "avc", "dvdrip", "brrip", "mpeg-4", "[vtv]", "xvid-", "480p", "720i", "720p", "1080i", "1080p", "ws", "pdtv", "4od", "dvdscr", "albiondean", "merry xmas", "bdrip", "repack", "bluray", "proper", "yify", "dvd", "complete", "dvdrip", "linkzkraze", "tvblinkz", "tvboom", "hdtv-", "webrip", "boxxx", "remastered", "www", "rarbg", "etrg", "uncut", "hdrip", "xvid", "internal", "preair")

year = re.compile(r"\b((?:19|20)\d\d)\b")
ep = re.compile(r"\b(?:S(\d\d?)E(\d\d?)|(\d\d?)x(\d\d?))\b", re.I)

kept_extensions = set((".avi", ".mov", ".mkv", ".ogv", ".m4v", ".mp4", ".srt", ".smi", ".3gp", ".ts"))

class ConsistencyError(ValueError): pass


def smart_title(s):
	"""
	>>> smart_title("the of by")
	'The of by'
	>>> smart_title("HoUnd  spaceSpace ")
	'Hound Spacespace'
	>>> smart_title("")
	''
	>>> smart_title("01EST")
	'01est'
	>>> smart_title("TeSt (s01e99), other")
	'Test (s01e99), Other'
	>>> smart_title("Elmentary (2019) (s01e99)")
	'Elementary (2019) (s01e99)'
	>>> smart_title("(s06e11) (s06e12)")
	'(s06e11) (s06e12)'
	"""
	pieces = re.split(r"\s+", s)
	result = list()
	for i, p in enumerate(pieces):
		if not p: continue
		lc = p.lower()
		if i != 0:
			if result[-1][-1] in string.lowercase:
				if (lc in ("the", "of", "a", "from", "and", "but", "for", "nor", "or", "to", "by", "at")):
					result.append(lc)
					continue
		result.append(lc[0].upper() + lc[1:])
	return " ".join(result).replace("Elmentary ", "Elementary ")


def reformat(filename, path_prefix):
	"""
	Year always first because it's part of the name, then episode or season info.

	>>> reformat('Welcome.to.Sweden.2014.S01E09.HDTV.XviD-AFG.avi', None)
	('Welcome to Sweden (2014)/Welcome to Sweden (2014) (s01e09).avi', '.avi')
	>>> reformat('Doctor Who (2005) (S40E05).mkv', None)
	('Doctor Who (2005)/Doctor Who (2005) (s40e05).mkv', '.mkv')
	>>> reformat('Rubicon (S01E05), Connect The Dots.avi', None)
	('Rubicon/Rubicon (s01e05) Connect the Dots.avi', '.avi')
	>>> reformat('Top Gear (S05E05) (2004-11-21).avi', None)
	('Top Gear/Top Gear (2004-11-21) (s05e05).avi', '.avi')
	>>> reformat('Food Inc (2008).avi', None)
	('Food Inc (2008)/Food Inc (2008).avi', '.avi')
	>>> reformat('Top Gear (S05E10) 2005-01-02, The Best Of Season 5.avi', None)
	('Top Gear/Top Gear (2005-01-02) (s05e10) The Best of Season 5.avi', '.avi')
	>>> reformat('Rubicon.S01E01.Pilot.HDTV.XviD-FQM.avi', None)
	('Rubicon/Rubicon (s01e01) Pilot.avi', '.avi')
	>>> reformat('Rubicon.S01E11.A.Good.Days.Work.HDTV.XviD-FQM.[VTV].avi', None)
	('Rubicon/Rubicon (s01e11) A Good Days Work.avi', '.avi')
	>>> reformat("Rubicon (s01e11), A Good Day's Work.avi", None)
	("Rubicon/Rubicon (s01e11) A Good Day's Work.avi", '.avi')
	>>> reformat("Doctor Who (2005) A Christmas Carol 2010 Special.mkv", None)
	('Doctor Who (2005)/Doctor Who (2005) A Christmas Carol 2010 Special.mkv', '.mkv')
	>>> reformat("The.Direct.Article.1.avi", None)
	('The Direct Article 1/The Direct Article 1.avi', '.avi')
	>>> reformat("Dr. Horrible.Act.1.avi", None)
	('Dr Horrible Act 1/Dr Horrible Act 1.avi', '.avi')
	>>> reformat("Justified Season 1 Episode 13 Bulletville.avi", None)
	('Justified/Justified (s01e13) Bulletville.avi', '.avi')
	>>> reformat("Star Trek The Next Generation Season 3 Episode 21- Hollow Pursuits.avi", None)
	('Star Trek the Next Generation/Star Trek the Next Generation (s03e21) Hollow Pursuits.avi', '.avi')
	>>> reformat("The.Colbert.Report.2012.04.11.Michelle.Obama.HDTV.XviD-FQM.[VTV].avi", None)
	('The Colbert Report/The Colbert Report (2012-04-11) Michelle Obama.avi', '.avi')
	>>> reformat("Community Season 2, Episode 06 – Epidemiology - LinkzKraze.mp4", None)
	('Community/Community (s02e06) Epidemiology.mp4', '.mp4')
	>>> reformat("glee.404.hdtv-lol.mp4", None)
	('Glee/Glee (s04e04).mp4', '.mp4')
	>>> reformat("The Big Bang Theory 605 Hdtv-lol.mp4", None)
	('The Big Bang Theory/The Big Bang Theory (s06e05).mp4', '.mp4')
	>>> reformat("Archer - 2E11 - Jeu Monegasque.mkv", None)
	('Archer/Archer (s02e11) Jeu Monegasque.mkv', '.mkv')
	>>> reformat("Burn.Notice.S06E11E12.HDTV.x264-2HD.[VTV].mp4", None)
	('Burn Notice/Burn Notice (s06e11) (s06e12).mp4', '.mp4')
	>>> reformat("Star Trek - The Next Generation- S01 E01+2-Encounter at Fairpoint.mp4", None)
	('Star Trek the Next Generation/Star Trek the Next Generation (s01e01) (s01e02) Encounter at Fairpoint.mp4', '.mp4')
	>>> reformat('The.Powerpuff.Girls.S06.E07.Crazy.Mixed.Up.Puffs-Mizzen.In.Action-[boxXx].mkv', None)
	('The Powerpuff Girls/The Powerpuff Girls (s06e07) Crazy Mixed Up Puffs-mizzen In Action.mkv', '.mkv')
	>>> reformat('Rockford_Files-S03xE15-The_Trees_The_Bees_And_TT_Flowers_Part_2.avi', None)
	('Rockford Files/Rockford Files (s03e15) The Trees the Bees and Tt Flowers Part 2.avi', '.avi')
	>>> reformat('My Neighbor Totoro.mp4', None)
	('My Neighbor Totoro/My Neighbor Totoro.mp4', '.mp4')
	>>> reformat('My Neighbor Totoro.srt', None)
	('My Neighbor Totoro/My Neighbor Totoro.srt', '.srt')
	>>> reformat('Arrested Development - S1 E 02 - Top Banana.avi', None)
	('Arrested Development/Arrested Development (s01e02) Top Banana.avi', '.avi')
	>>> reformat('My Neighbor Totoro.srt', None)
	('My Neighbor Totoro/My Neighbor Totoro.srt', '.srt')
	>>> reformat('Magnum P I 3x01,3x02 Did You See the Sunrise (1)&(2).avi', None)
	('Magnum P I/Magnum P I (s03e01) (s03e02) Did You See the Sunrise (1)&(2).avi', '.avi')
	>>> reformat('penn.and.teller.fool.us.s02e01.phone.y.business.hdtv.x264.mp4', None)
	('Penn and Teller Fool Us/Penn and Teller Fool Us (s02e01) Phone Y Business.mp4', '.mp4')
	>>> reformat('Sicario.2015.1080p.Bluray.DTS.x264-ETRG.mkv', None)
	('Sicario (2015)/Sicario (2015).mkv', '.mkv')
	>>> reformat('Italian.srt', 'The Jungle Book 2016 1080p BluRay x264 DTS-JYK/Subs')
	('The Jungle Book (2016)/Italian.srt', '.srt')
	"""

	already_dated = False
	assert os.sep not in filename
	fragments = list()
	times = list()
	basename, extension = os.path.splitext(filename)

	is_language_file = (extension == ".srt" and path_prefix)

	if is_language_file:
		orig_filename, basename = filename, path_prefix

	want_container = True
	container_pieces = list()
	container_finished = False

	b = basename
	# S01 E01+2
	basename = re.sub(r"(?i)(?<=\w)-\s+", r" ", basename)

	basename = re.sub(r"(?i)\b(\S\d+)\s+(E\d+)\s*\+?\s*(\d+)\b", r"\1\2E\3 ", basename)
	basename = re.sub(r"(?i)\b(\S\d+)\s+(E\d+)-?\b", r"(\1\2)", basename)
	basename = re.sub(r"(?i)\bseason\s?(\d+),?\s?episode\s?(\d+)-*", r"(s\1e\2)", basename)
	basename = re.sub(r"(?i)\bseason\s?(\d+),?\s?show\s?(\d+)", r"(s\1e\2)", basename)
	basename = re.sub(r"\b(2\d\d\d)\.(\d\d?)\.(\d\d?)\b", r"\1-\2-\3", basename)
	basename = re.sub(r"(?i)(?<=\))-(?=\w)", r" ", basename)
	basename = re.sub(r"(?i)(?<=\s)-(?=\w)", r"", basename)
	basename = re.sub(r"(?i)\b(\d+)E(\d\d)\b", r"s\1e\2", basename)
	basename = re.sub(r"(?i)\bs\s*(\d+)\s*E\s*(\d\d)\b", r"s\1e\2", basename)
	basename = re.sub(r"(?i)\bs(\d+)\.E(\d\d)\b", r"s\1e\2 ", basename)
	basename = re.sub(r"(?i)-s(\d+)x?e(\d+)-", r" s\1e\2 ", basename)
	basename = re.sub(r"(?i)\b(\d+)x?(\d\d+),", r" s\1e\2 ", basename)
	if b != basename:
		want_container = True

	basename = re.sub(r"\.US\.", r".", basename)
	basename = re.sub(r"([a-zA-Z])-(\d+)\b", r"\1\2", basename)

	for piece in re.split(r"\.| +|_|\[|\]", basename):
		was_time = False
		if piece.endswith(","):
			piece = piece[:-1]

		if piece == "-": continue
		if piece == "–": continue
		if not piece.strip(): continue
		if piece.lower().startswith(quash):
			break

		if piece.lower() == "dvdscr":
			fragments.append("(screener)")
			break

		m = re.match(r"(?i)^\(S(\d+)E(\d+)\)$", piece)
		if m:
			want_container = True
			container_finished = True
			times.append("s%02de%02d" % (int(m.group(1)), int(m.group(2))))
			was_time = True
		m = re.match(r"(?i)^S(\d+)E(\d+)-?E(\d+)$", piece)
		if m:
			want_container = True
			container_finished = True
			times.append("s%02de%02d" % (int(m.group(1)), int(m.group(3))))
			times.append("s%02de%02d" % (int(m.group(1)), int(m.group(2))))
			was_time = True
		m = re.match(r"(?i)^S(\d+)E(\d+)$", piece)
		if m:
			want_container = True
			container_finished = True
			times.append("s%02de%02d" % (int(m.group(1)), int(m.group(2))))
			was_time = True
		m = re.match(r"(?i)^S(\d+)$", piece)
		if m:
			want_container = True
			container_finished = True
			times.append("s%02d" % (int(m.group(1)),))
			was_time = True
		m = re.match(r"(?i)^E(\d+)$", piece)
		if m:
			want_container = True
			container_finished = True
			times.append("e%02d" % (int(m.group(1)),))
			was_time = True
		m = re.match(r"^(\d)(\d\d)$", piece)
		if m:
			if fragments and m.group(1)+m.group(2) != "720":
				want_container = True
				container_finished = True
				times.append("s%02de%02d" % (int(m.group(1)), int(m.group(2))))
				was_time = True
			else:
				pass  # let fragments catch
				
		m = re.match(r"^(\d+)x(\d+)$", piece)
		if m:
			want_container = True
			container_finished = True
			times.append("s%02de%02d" % (int(m.group(1)), int(m.group(2))))
			was_time = True
		m = re.match(r"^\(((?:201|200|19\d)\d)\)$", piece)
		if m and not already_dated:
			want_container = True
			container_finished = True
			container_pieces.append("(" + m.group(1) + ")")
			times.append("%04d" % (int(m.group(1)),))
			was_time = True
			already_dated = True
		m = re.match(r"^((?:201|200|19\d)\d)$", piece)
		if m and not already_dated:
			want_container = True
			container_finished = True
			container_pieces.append("(" + m.group(1) + ")")
			times.append("%04d" % (int(m.group(1)),))
			was_time = True
			already_dated = True
		m = re.match("^\(((?:201|200|19\d)\d).([01]\d).([0123]\d)\)$", piece)
		if m and not already_dated:
			want_container = True
			container_finished = True
			times.append("%s-%s-%s" % (m.group(1), m.group(2), m.group(3)))
			#container_pieces.append("(" + m.group(1) + ")")
			was_time = True
			already_dated = True
		m = re.match("^((?:201|200|19\d)\d).([01]\d).([0123]\d)$", piece)
		if m and not already_dated:
			want_container = True
			container_finished = True
			times.append("%s-%s-%s" % (m.group(1), m.group(2), m.group(3)))
			#container_pieces.append("(" + m.group(1) + ")")
			was_time = True
			already_dated = True
		
		if times:
			if not was_time:
				fragments.append("(" + ") (".join(sorted(times)) + ")")
				times = list()
				fragments.append(piece)
			continue
		
		fragments.append(piece)
		if not container_finished:
			container_pieces.append(piece)

	basename = smart_title(" ".join(fragments)).strip("-")
	dirs = os.listdir("/data/%%media/sorted/torrents")
	candidates = sorted((d for d in dirs if basename in d), key=lambda s: (-len(s), s))
	
	if candidates and candidates[0] != basename:
		basename = candidates[0]
		y = basename.split()[-1]
		if y not in container_pieces:
			container_pieces.append(basename.split()[-1])

	if times:
		for time in sorted(times):
			if time not in basename:
				basename += " (" + time + ")"

	if is_language_file:
		return smart_title(" ".join(container_pieces)) + "/" + orig_filename, extension.lower()
	
	if want_container:
		return smart_title(" ".join(container_pieces)) + "/" + basename + extension.lower(), extension.lower()
	else:
		return basename + extension.lower(), extension.lower()


def rename(dummy, container, filenames):
	assert container.startswith("/data/%%media/unsorted")

	unrooted_container = container[len("/data/%%media/unsorted")+1:]

	imdbttid = None
	nfo_file_contents = None

	if unrooted_container.startswith("imdb-tt"):
		first_slash_position = unrooted_container.find("/")
		if first_slash_position == -1:
			imdbttid = int(unrooted_container[8:])
		else:
			imdbttid = int(unrooted_container[8:first_slash_position])
		nfo_file_contents = """<tvshow>\n<id moviedb="imdb">tt{}</id>\n</tvshow>""".format(imdbttid)

	renamed = list()
	for filename in filenames:
		try:
			if os.path.isfile(os.path.join(container, filename)):
				if "trailer" in filename.lower():
					logger.debug("should remove %r", filename)
					continue
				if "sample" in filename.lower():
					logger.debug("should remove %r", filename)
					continue
				new_filename, extension = reformat(filename, container)

				consistency = reformat(os.path.basename(new_filename), container)[0]
				if new_filename != consistency:
					raise ConsistencyError(new_filename + " != " + consistency +" // " + filename)

				src = os.path.join(container, filename)
				# FIXME set dst with show name
				dst = os.path.join("/data/%%media/sorted/torrents", new_filename)

				if extension not in kept_extensions:
					print("not extension", filename)
					continue

				if src == dst:
					print("srcdst", src)
					continue

				if os.sep not in new_filename:
					if os.path.exists(dst):
						logger.debug("Weird file %s   (AND REMOVING!)", filename)
						os.remove(dst)
						print("weird", filename)
						continue
					else:
						logger.debug("Just a file %s", new_filename)

				if not os.path.exists(dst):

					try:
						os.mkdir(os.path.dirname(dst))
					except OSError:
						pass
					try:
						if nfo_file_contents and re.search(r"\s\(s\d\de\d\d\)", dst):
							nfo_file = os.path.join(os.path.dirname(dst), "tvshow.nfo")
							if not os.path.exists(nfo_file):
								with open(nfo_file, "w") as f:
									f.write(nfo_file_contents)
									logger.debug("Wrote new tvshow nfo file near " + dst)
						renamed.append(dst)
						os.link(src, dst)
						print((os.path.join(container, filename)))
					except OSError, exc:
						print(src, dst)
						logger.warn("# couldn't link %s", (dst,))
						raise
						pass
			else:
				print("not file", filename)

		except ConsistencyError, ex:
			pass
			#logger.warn("ignoring inconsistent name, %s", ex)
			
	return renamed

import doctest
if doctest.testmod().failed > 0:
	raise ValueError("Test errors")

try:
	os.path.walk("/data/%%media/unsorted", rename, None)
except Exception, ex:
	logger.exception("failed to complete")
	raise

print(rename(None, "/data/%%media/unsorted/The.Karate.Kid.2010.1080p.BluRay.x264.AC3-ETRG", ["The.Karate.Kid.2010.1080p.BluRay.x264.AC3-ETRG.mp4"]))

subprocess.call(["find", "/data/%%media/sorted/torrents", "-type", "f", "-ctime", "+180", "-print", "-delete"])
subprocess.call(["find", "/data/%%media/unsorted", "-type", "f", "-links", "1", "-ctime", "+60", "-print", "-delete"])