look up phonetic spelling Apple's Dictionary.app through DictionaryServices

mac-dict.py

#!/usr/bin/python2.7
# -*- coding: utf-8 -*-

"""
@file mac-dict.py -- lookup entries in the Mac dictionary
@author Tim van Werkhoven
@date 20111220
@copyright Copyright (c) 2011 Tim van Werkhoven

Use DictionaryServices to look up words in Apple's Dictionary.app.

Loosely based on work by Alex Kontsevoy, see <http://tech-kovoy.blogspot.com/2008/07/how-to-make-use-of-dictionary-services.html>

This file is licensed under the Creative Commons Attribution-Share Alike
license versions 3.0 or higher, see
http://creativecommons.org/licenses/by-sa/3.0/
"""

import DictionaryServices
import re
import unittest
import os, sys

# Define some contants
AUTHOR = "Tim van Werkhoven <[email protected]>"
DATE = "20111220"
VERBOSITY = 0

def main():
	## First check if we need tests (do this before argparse because otherwise we might run into conflicts)
	if ('--test' in sys.argv):
		print "Bypassing regular program, running tests now!"
		# Remove '--test' string from sys.argv, keep all others
		sys.argv = [el for el in sys.argv if el not in ['--test']]
		return unittest.main()

	## Parse arguments check options
	(parser, args) = parsopts()

	remlist = "',.:;\"\n’"
	if (args.infile):
		if (VERBOSITY > 1): print "main(): args.infile"
		fd = open(args.infile, "r")
		for line in fd.xreadlines():
			if (VERBOSITY > 2): print "main(): line = %s" % (line)
			for i in remlist:
				line = line.replace(i, "")
			if (VERBOSITY > 2): print "main(): procline = %s" % (line)
			for word in line.split(" "):
				print dict_IPA(word),
			print
		fd.close()
	else:
		if (VERBOSITY > 1): print "main(): else branch"
		for word in args.words:
			print dict_IPA(word),

def dict_IPA(word):
	"""Given a word, return the pronunciation in IPA"""
	# Pre-process
	word = word.strip()
	if (len(word) < 1):
		return

	# Lookup word
	if (VERBOSITY > 1): print "dict_IPA(word = %s)" % (word)
	entry = DictionaryServices.DCSCopyTextDefinition(None, word, (0,len(word) ))
	if (not entry):
		return "n/a"

	if (VERBOSITY > 2): print "dict_IPA(): %s" % (entry)

# Find word within entry, IPA should be the next line
# 	#IPA_beg = entry.index(word+"\n") + len(word) + 1
# 	IPA_beg = re.search(word+"\s*\d*\s*\n", entry, re.IGNORECASE)
# 	if (not IPA_beg and word[-1] == "s"):
# 		if (VERBOSITY > 2): print "dict_IPA(): word[-1] == s"
# 		IPA_beg = re.search(word[:-1]+".\s*\d*\s*\n", entry, re.IGNORECASE)
#
# 	if (IPA_beg):
# 		IPA_beg = IPA_beg.end()
# 	else:
# 		return
# 	#IPA_beg2 = re.search("\n|.*|\n", entry[IPA_beg1:]).end() + IPA_beg1

	IPA_beg = entry.index("\n") + 1
	IPA_end = entry.index("\n", IPA_beg)
	if (VERBOSITY > 2): print "dict_IPA(): %d -- %d" % (IPA_beg, IPA_end)
	# Extract IPA, remove vertical bars
	IPA = entry[IPA_beg+1:IPA_end-1]
	if (VERBOSITY > 2): print "dict_IPA(): %s" % (IPA)
	return IPA

def parsopts():
	"""Parse program options, check sanity and return results"""
	import argparse
	parser = argparse.ArgumentParser(description="Lookup words from Apple's Dictonary.app. You can change the language by opening Dictionary.app's preferences and only enabling the dictionary you want to use.", epilog='Comments & bugreports to %s' % (AUTHOR))

	parser.add_argument('words', metavar='W', type=str, nargs='*', default=[],
						help='words to lookup.')
	parser.add_argument('--infile', metavar='F', type=str,
						help='read words from F.')

	parser.add_argument('-v', dest='debug', action='append_const', const=1,
						help='increase verbosity')
	parser.add_argument('-q', dest='debug', action='append_const', const=-1,
						help='decrease verbosity')

	args = parser.parse_args()

	# Check & fix some options
	checkopts(parser, args)

	# Return results
	return (parser, args)

def checkopts(parser, args):
	"""Check program options sanity"""
	if (args.infile):
		if (not os.path.exists(args.infile)):
			print "Error: infile does not exist!"
			parser.print_usage()
			sys.exit(0)
		elif (not os.path.isfile(args.infile)):
			print "Warning: infile is not a regular file!"
	elif (not len(args.words)):
		print "Error: need at least one word or an input file"
		parser.print_usage()
		sys.exit(0)
	if (args.debug):
		global VERBOSITY
		VERBOSITY = sum(args.debug)
		print "checkopts(): verbosity: %d" % (VERBOSITY)

### Tests start here
class TestWords(unittest.TestCase):
	def setUp(self):
		"""Make list of words"""
		self.words = [u"worse", u"pronunciation", u"Terpsichore"]
		self.ipa_uk = [u"wəːs", u"prənʌnsɪˈeɪʃ(ə)n", u"təːpˈsɪkəri"]
		self.ipa_us = [u"wərs", u"prəˌnənsiˈeɪʃən", u"tərpˈsɪkəri"]

	# Explore environment / settings
	def test0a_detect_lang(self):
		"""Detect which dictionary we are using (US or UK)"""
		if (dict_IPA("worse").strip() == u"wərs"):
			self.lang = "US"
			print "Got US dictionary"
		elif (dict_IPA("worse").strip() == u"wəːs"):
			self.lang = "UK"
			print "Got UK dictionary"
		else:
			raise RuntimeError("Unknown dictionary language!")

	# Static tests
	def test1a_ipa(self):
		"""Test pre-defined words for pronunciation"""
		this_ipa = self.ipa_uk
		print self.lang
		if (self.lang == "US"):
			this_ipa = self.ipa_us

		for word, ipa in zip(self.words, this_ipa):
			testipa = dict_IPA(word).strip()
			print u"%s=%s" % (testipa, ipa)
			self.assertEqual(dict_IPA(word).strip(), ipa)

# This must be the final part of the file, code after this won't be executed
if __name__ == "__main__":
	sys.exit(main())

### Scratch space here

print dict_IPA("worse")
print dict_IPA("pronunciation")
print dict_IPA("Terpsichore")
print dict_IPA("Terpsichore")

### EOF