added pysed and itunes_diff

bashrc

@@ -4,6 +4,7 @@ export EDITOR=vim
 export PAGER=w3m
 export COWPATH=/opt/local/share/cowsay
 export LDFLAGS=-L/opt/local/lib/
+export LESS=-R
 
 stty erase ^H
 

bin/README

@@ -2,4 +2,6 @@ irssinp: AppleScript that returns a string with the current iTunes playing song.
 flac2mp3: extract flac metadata, encode to -V 2 mp3, set id3 tags with id3v2
 latest_additions.php: RSS feed of changes to a bugzilla query.  as-is it shows changes in the firefox 4 blocker list.  you can pretty easily change this to any bugzilla or query.
 treediff.py: show files not common to two paths
+pysed.py: primitive sed with python regex.
+itunes_diff.py: given an itunes plist and a dirtree shows you files in dirtree not in itunes.  use it to get rid of duplicate songs in itunes after a "Organize Library".
 

bin/itunes_diff.py

@@ -0,0 +1,50 @@
+import plistlib
+import sys
+import os
+import urllib
+import re
+
+stripper = re.compile("^file://localhost(.*)")
+
+#file://localhost/Volumes/ntfs/Music/Music/Apple%20Computer/Developer%20Helper%20Volume%201-%20Phi/07%20Breaking%20Through.mp3
+
+if len(sys.argv) < 3:
+   print "Usage: itunes_diff.py iTunes\ Music\ Library.xml /path/to/itunesdir [--ignore-dotfiles]"
+   sys.exit()
+
+ignore_dotfiles = False
+try:
+   if sys.argv[3] == "--ignore-dotfiles":
+      ignore_dotfiles = True
+except IndexError:
+   pass
+
+countdict = plistlib.readPlist(sys.argv[1])
+
+walker = os.walk(sys.argv[2])
+
+hard_drive_paths = set()
+itunes_library_paths = set()
+
+for directory in walker:
+   for filename in directory[2]:
+      hard_drive_paths.add(os.path.join(directory[0], filename))
+
+for track in countdict["Tracks"].iteritems():
+   if "Location" in track[1]:
+      unquoted = urllib.unquote(track[1]["Location"])
+      match = stripper.search(unquoted)
+      if match:
+         itunes_library_paths.add(match.group(1))
+
+hd_only = hard_drive_paths - itunes_library_paths
+
+for x in hd_only:
+   if not ignore_dotfiles:
+      print x
+   else:
+      if not os.path.basename(x).startswith('.'):
+         print x
+
+#hopefully there aren't any newlines in your mp3 file names!
+

bin/pysed.py

@@ -0,0 +1,47 @@
+#!/usr/bin/env python
+import re
+import fileinput
+import sys, os
+
+cheatsheet = """You need two args!
+
+pysed.py: regex replacement [file1, file2...]
+
+*: match 0 or more
++: match 1 or more
+?: match 0 or 1
+*?, +?, ??: non-greedy matching
+
+\d , \D: any digit, any not digit
+\s , \S: any whitespace, any not whitespace
+\w , \W: any number, letter or _, any not number, letter or _
+
+{n}: match n repetitions
+{m,n}: match m to n repetitions
+
+(?iLmsux): one or more chars from set adjust group matching.
+   i: ignore case
+   L: locale dependent
+   m: does not do what you might expect
+   s: . also matches newlines
+   u: unicode dependent
+   x: ignore spaces except when escaped, comments after #
+(?:): group can't be backreferenced
+(?P<foo>): you can backreference group with \gfoo in substitution or (?P=foo) in regex
+(?(backref)regex|regex): ternary if backref exists"""
+
+if len(sys.argv) < 3:
+   print >> sys.stderr, cheatsheet
+   sys.exit()
+
+regex = re.compile(sys.argv[1])
+replacement = sys.argv[2]
+
+sedfiles = []
+if len(sys.argv) > 3:
+   for sedfile in sys.argv[3:]:
+      if os.path.isfile(sedfile):
+         sedfiles.append(sedfile)
+
+for line in fileinput.input(sedfiles):
+   sys.stdout.write(re.sub(regex, replacement, line))