pymeds/cdrtoc.py

#!/usr/bin/env python

import string

types = frozenset([ 'CD_DA', 'CD_ROM', 'CD_ROMXA'])

def decodestr(i, pos):
	return decodestrend(i, pos)[0]

def decodestrend(i, pos):
	r = []
	bspos = None
	dqpos = None
	while True:
		if bspos is None or bspos == -1 or bspos < pos:
			bspos = i.find('\\', pos)
		if dqpos is None or dqpos < pos:
			dqpos = i.index('"', pos)
		if bspos >= 0 and bspos < dqpos:
			r.append(i[pos:bspos])
			c = i[bspos + 1]
			if c == '"':
				r.append('"')
				pos = bspos + 2
			elif c in string.digits:
				r.append(unichr(int(i[bspos + 1:bspos + 4], 8)))
				pos = bspos + 4
			else:
				raise ValueError('unknown escape char')
		else:
			r.append(i[pos:dqpos])
			break

	return ''.join(r), dqpos

def parsetoc(toc):
	# state machine info:
	#   0: header
	#   1: in CD_TEXT
	#   2: in LANGUAGE_MAP
	#   3: in LANGUAGE

	r = { 'tracks': {} }
	langmap = {}
	state = 0
	curlang = None
	textobj = None
	langobj = None
	track = 0
	for i in toc.split('\n'):
		i = i.strip()
		if not i:
			continue

		items = i.split()
		key = items[0]

		if state == 0:
			if i in types:
				r['type'] = i
			elif key == 'CD_TEXT':
				state = 1
				if track == 0:
					textobj = r
			elif key == 'TRACK':
				track += 1
				textobj = { 'track': track }
				r['tracks'][track] = textobj
			elif key == 'TWO_CHANNEL_AUDIO':
				textobj['channels'] = 2
			elif key == 'FOUR_CHANNEL_AUDIO':
				textobj['channels'] = 4
			elif key == 'ISRC':
				textobj['isrc'] = decodestr(i, i.index('"') + 1)
			elif key == 'COPY':
				textobj['copy'] = True
			elif items[0] == 'NO' and items[1] == 'COPY':
				textobj['copy'] = False
			elif key == 'PRE_EMPHASIS':
				textobj['preemphasis'] = True
			elif items[0] == 'NO' and items[1] == 'PRE_EMPHASIS':
				textobj['preemphasis'] = False
			elif key == 'FILE':
				pass	# XXX
			elif key == 'START':
				pass	# XXX
			elif key == '//':
				pass
			else:
				raise ValueError('unknown line: %s' % `i`)
		elif state == 1:
			if key == 'LANGUAGE_MAP':
				state = 2
			elif key == 'LANGUAGE':
				state = 3
				langobj = textobj
				# XXX - don't try to use more than one!
				#lang = items[1].strip()
				#textobj[langmap[lang]] = langobj
			elif key == '}':
				textobj = None
				state = 0
		elif state == 2:
			if key == '}':
				state = 1
			else:
				key, value = (x.strip() for x in i.split(':'))
				value = int(value)
				langmap[key] = value
		elif state == 3:
			if key == '}':
				langobj = None
				state = 1
			else:
				curl = i.find('{')
				dquo = i.find('"')
				if curl != -1 and curl < dquo:
					val = i[i.index('{') + 1:i.index('}')]
					val = ''.join(chr(int(x)) for x in
					    val.split(','))
				else:
					if dquo == -1:
						raise ValueError('no dquote')
					val = decodestr(i, dquo + 1)
				langobj[key] = val

	return r

if __name__ == '__main__':
	import sys

	for i in sys.argv[1:]:
		print 'file:', `i`
		print parsetoc(open(i).read())