HomoFast-eSpeak-Persian/mkdictlist

#!/usr/bin/python
#
# Copyright (C) 2011 Reece H. Dunn
# Licence: GPLv3
#
# A script for generating the dictionary Makefile rules from the files in dictsource.

import sys
import os

voices = set()
dictionaries = {}

# Map voice names to dictionaries when these do not match.
special_voices = {
	'zh-yue': 'zhy',
}

def find_voices(path):
	for filename in os.listdir(path):
		voice_path = os.path.join(path, filename)
		if os.path.isdir(voice_path):
			if not filename in ['!v', 'mb']:
				find_voices(voice_path)
		else:
			if filename in special_voices.keys():
				voices.add(special_voices[filename])
			else:
				voices.add(filename)

find_voices('espeak-data/voices')

for filename in os.listdir('dictsource'):
	if filename.endswith('_rules') or filename.endswith('_list') or filename.endswith('_listx'):
		dic, cat = filename.split('_')
		if dic in voices:
			if not dic in dictionaries.keys():
				dictionaries[dic] = []
			dictionaries[dic].append('dictsource/%s' % filename)

print '##### dictionaries:'
print
print 'dictionaries: \\'
for n, name in enumerate(sorted(dictionaries.keys())):
	if not name in ['bo']: # espeak fails to read these voices
		if n == len(dictionaries.keys()) - 1:
			print '\tespeak-data/%s_dict' % name
		else:
			print '\tespeak-data/%s_dict \\' % name
for name, files in sorted(dictionaries.items()):
	print
	print '%s: espeak-data/%s_dict' % (name, name)
	print 'dictsource/%s_extra:' % name
	print '\ttouch dictsource/%s_extra' % name
	print 'espeak-data/%s_dict: src/espeak espeak-data/phontab %s dictsource/%s_extra' % (name, ' '.join(sorted(files)), name)
	print '\tcd dictsource && ../src/espeak --compile=%s && cd ..' % name