So I was trying to get my pc laptop setup for fudcon which meant a triple boot setup between windows, ubuntu and fedora. I had some troubles importing various important information about my music within iTunes.

Here is a script to remove duplicate songs in an iTunes folder:

import hashlib
import os
import re
import sys

def hash(fnam):
	fobj = open(fnam, "r")
	hobj = hashlib.sha256()
	
	while (1):
		data = fobj.read(2**20)
		
		if (not data):
			break
		
		hobj.update(data)
	
	return hobj.hexdigest()

fdic = {}

while (1):
	file = sys.stdin.readline()
	
	if (not file):
		break
	
	file = file.strip()
	uniq = hash(file)
	
	if (uniq in fdic.keys()):
		print("removing:",file)
		os.unlink(file)
	
	else:
		print("found:",file)
		fdic[uniq] = file

This script attempts to merge an old iTunes library xml file with a newly imported one. The result is a merged iTunes library xml file that can then be re-imported into iTunes thus restoring some various meta-data about your music. Note: Before you re-import the merged xml playlist file, make sure you delete the current one and disable the check-box stating to copy songs into the media folder.

import os
import re
import sys

if (len(sys.argv) < 3):
	print("Usage: %s <source> <merge>" % (sys.argv[0]))
	sys.exit(0)

def outp(ordr, dict, line):
	for item in ordr:
		sys.stdout.write(dict[item][1])
	
	sys.stdout.write(line)
	sys.stdout.flush()

def xmlr(fnam, sdic={}):
	alen = len(sys.argv)
	slen = len(sdic.keys())
	xord = []; xdic = {}
	xobj = open(fnam, "r")
	
	xdat = {}
	
	while (1):
		xlin = xobj.readline()
		
		if (not xlin):
			break
		
		slin = xlin.strip("\0\t\r\n ")
		xreg = re.match("^<key>([^<]+)</key>(<[^>]+>[^<]+<[^>]+>)$", slin)
		
		try:
			xkey = (xdic["Artist"][0] + xdic["Album"][0] + xdic["Name"][0])
		
		except:
			xkey = ""
		
		if (slin == "<dict>"):
			if (slen > 0):
				outp(xord, xdic, xlin)
			
			xord = []; xdic = {}
		
		elif (xreg):
			if (xreg.group(1) not in xord):
				xord.append(xreg.group(1))
			
			xdic[xreg.group(1)] = [xreg.group(2), xlin]
		
		elif (slin == "</dict>"):
			if (slen < 1):
				xdat[xkey] = xdic
			
			else:
				for x in range(3, alen):
					try:
						xdic[sys.argv[x]] = sdic[xkey][sys.argv[x]]
						
						if (sys.argv[x] not in xord):
							xord.append(sys.argv[x])
					
					except:
						#print("error:",sys.exc_info())
						pass
				
				outp(xord, xdic, xlin)
			
			xord = []; xdic = {}
		
		else:
			if (slen > 0):
				outp(xord, xdic, xlin)
			
			xord = []; xdic = {}
	
	return xdat

sdat = xmlr(sys.argv[1])
ddat = xmlr(sys.argv[2], sdic=sdat)

Here’s an example of how to use the command:

python.exe /cygdrive/c/Users/jon/Desktop/itml.py /cygdrive/g/tmp/itunes/iTunes\ Music\ Library0.xml /cygdrive/g/tmp/itunes/iTunes\ Music\ Library.xml 'Date Added' 'Play Count' | unix2dos.exe | tee /cygdrive/g/tmp/itunes/iTunes\ Music\ Library2.xml && cp /cygdrive/g/tmp/itunes/iTunes\ Music\ Library2.xml /cygdrive/g/tmp/itunes/iTunes\ Music\ Library3.xml