def french_count():
f = FST('french')
f.add_state('start')
f.initial_state = 'start'
for ii in xrange(10):
f.add_arc('start', 'start', str(ii), [kFRENCH_TRANS[ii]])
f.set_final('start')
return f
def french_count():
f = FST("french")
f.add_state("start")
f.initial_state = "start"
for ii in xrange(10):
f.add_arc("start", "start", str(ii), [kFRENCH_TRANS[ii]])
f.set_final("start")
return f
def truncate_to_three_digits():
"""
Create an FST that will truncate a soundex string to three digits
"""
# Ok so now let's do the second FST, the one that will truncate
# the number of digits to 3
f2 = FST('soundex-truncate')
# Indicate initial and final states
f2.add_state('1')
f2.add_state('2')
f2.add_state('3')
f2.add_state('4')
f2.initial_state = '1'
f2.set_final('1')
f2.set_final('2')
f2.set_final('3')
f2.set_final('4')
# Adds letters from input string of 'A###0000'
for letter in string.letters:
f2.add_arc('1', '1', (letter), (letter))
# Adds numbers from first FST of range 0-9
for n in range(10):
f2.add_arc('1', '2', str(n), (str(n)))
f2.add_arc('2', '3', str(n), (str(n)))
f2.add_arc('3', '4', str(n), (str(n)))
f2.add_arc('4', '4', str(n), ())
return f2
def truncate_to_three_digits():
"""
Create an FST that will truncate a soundex string to three digits
"""
# Ok so now let's do the second FST, the one that will truncate
# the number of digits to 3
f2 = FST('soundex-truncate')
# Indicate initial and final states
states = ['1', 'd1', 'd2', 'd3']
for state in states:
f2.add_state(state)
f2.initial_state = '1'
for state in ['d1', 'd2', 'd3']:
f2.set_final(state)
# Add the arcs
for letter in string.letters:
f2.add_arc('1', '1', (letter), (letter))
for index, state in enumerate(states):
if index > 0:
for n in range(10):
f2.add_arc(states[index-1], states[index], str(n), str(n))
for n in range(10):
f2.add_arc('d3', 'd3', str(n), ())
return f2
def letters_to_numbers():
"""
Returns an FST that converts letters to numbers as specified by
the soundex algorithm
"""
# Let's define our first FST
f1 = FST('soundex-generate')
# Indicate that '1' is the initial state
f1.add_state('start')
f1.add_state('next')
f1.initial_state = 'start'
# Set all the final states
f1.set_final('next')
# Add the rest of the arcs
for letter in string.ascii_lowercase:
f1.add_arc('start', 'next', (letter), (letter))
f1.add_arc('next', 'next', (letter), '0')
return f1
def truncate_to_three_digits():
"""
Create an FST that will truncate a soundex string to three digits
"""
# Ok so now let's do the second FST, the one that will truncate
# the number of digits to 3
f2 = FST('soundex-truncate')
# Indicate initial and final states
f2.add_state('1')
f2.initial_state = '1'
f2.set_final('1')
# Add the arcs
for letter in string.letters:
f2.add_arc('1', '1', (letter), (letter))
for n in range(10):
f2.add_arc('1', '1', str(n), str(n))
return f2
def add_zero_padding():
# Now, the third fst - the zero-padding fst
f3 = FST('soundex-padzero')
states = ['1', '2', '3', '4']
for state in states:
f3.add_state(state)
f3.initial_state = '1'
f3.set_final('4')
for letter in string.letters:
f3.add_arc('1', '1', letter, letter)
for number in range(1, 10):
f3.add_arc('1', '2', str(number), str(number))
f3.add_arc('2', '3', str(number), str(number))
f3.add_arc('3', '4', str(number), str(number))
f3.add_arc('2', '4', (), '00')
f3.add_arc('3', '4', (), '0')
return f3
def letters_to_numbers():
"""
Returns an FST that converts letters to numbers as specified by
the soundex algorithm
"""
# Let's define our first FST
f1 = FST('soundex-generate')
# Indicate that '1' is the initial state
f1.add_state('start')
f1.add_state('next')
f1.add_state('one')
f1.add_state('two')
f1.add_state('three')
f1.add_state('four')
f1.add_state('five')
f1.add_state('six')
f1.initial_state = 'start'
# Set all the final states
f1.set_final('next')
f1.set_final('one')
f1.set_final('two')
f1.set_final('three')
f1.set_final('four')
f1.set_final('five')
f1.set_final('six')
list_one = ['b', 'f', 'p', 'v']
list_two = ['c', 'g', 'j', 'k', 'q', 's', 'x', 'z']
list_three = ['d', 't']
list_four = ['l']
list_five = ['m', 'n']
list_six = ['r']
vowels = ['a', 'e', 'h', 'i', 'o', 'u', 'w', 'y']
# Add the rest of the arcs
# changed string.ascii_lowercase to string.letters
for letter in string.letters:
f1.add_arc('start', 'next', (letter), (letter))
for letter in string.letters:
if letter in list_one:
f1.add_arc('next', 'one', (letter), '1')
elif letter in list_two:
f1.add_arc('next', 'two', (letter), '2')
elif letter in list_three:
f1.add_arc('next', 'three', (letter), '3')
elif letter in list_four:
f1.add_arc('next', 'four', (letter), '4')
elif letter in list_five:
f1.add_arc('next', 'five', (letter), '5')
elif letter in list_six:
f1.add_arc('next', 'six', (letter), '6')
else:
f1.add_arc('next', 'next', (letter), ())
for letter in string.letters:
if letter in list_two:
f1.add_arc('one', 'two', (letter), '2')
elif letter in list_three:
f1.add_arc('one', 'three', (letter), '3')
elif letter in list_four:
f1.add_arc('one', 'four', (letter), '4')
elif letter in list_five:
f1.add_arc('one', 'five', (letter), '5')
elif letter in list_six:
f1.add_arc('one', 'six', (letter), '6')
else:
f1.add_arc('one', 'one', (letter), ())
for letter in string.letters:
if letter in list_one:
f1.add_arc('two', 'one', (letter), '1')
elif letter in list_three:
f1.add_arc('two', 'three', (letter), '3')
elif letter in list_four:
f1.add_arc('two', 'four', (letter), '4')
elif letter in list_five:
f1.add_arc('two', 'five', (letter), '5')
elif letter in list_six:
f1.add_arc('two', 'six', (letter), '6')
else:
f1.add_arc('two', 'two', (letter), ())
for letter in string.letters:
if letter in list_one:
f1.add_arc('three', 'one', (letter), '1')
elif letter in list_two:
f1.add_arc('three', 'two', (letter), '2')
elif letter in list_four:
f1.add_arc('three', 'four', (letter), '4')
elif letter in list_five:
f1.add_arc('three', 'five', (letter), '5')
elif letter in list_six:
f1.add_arc('three', 'six', (letter), '6')
else:
f1.add_arc('three', 'three', (letter), ())
for letter in string.letters:
if letter in list_one:
f1.add_arc('four', 'one', (letter), '1')
elif letter in list_two:
f1.add_arc('four', 'two', (letter), '2')
elif letter in list_three:
f1.add_arc('four', 'three', (letter), '3')
elif letter in list_five:
f1.add_arc('four', 'five', (letter), '5')
elif letter in list_six:
f1.add_arc('four', 'six', (letter), '6')
else:
f1.add_arc('four', 'four', (letter), ())
for letter in string.letters:
if letter in list_one:
f1.add_arc('five', 'one', (letter), '1')
elif letter in list_two:
f1.add_arc('five', 'two', (letter), '2')
elif letter in list_three:
f1.add_arc('five', 'three', (letter), '3')
elif letter in list_four:
f1.add_arc('five', 'four', (letter), '4')
elif letter in list_six:
f1.add_arc('five', 'six', (letter), '6')
else:
f1.add_arc('five', 'five', (letter), ())
for letter in string.letters:
if letter in list_one:
f1.add_arc('six', 'one', (letter), '1')
elif letter in list_two:
f1.add_arc('six', 'two', (letter), '2')
elif letter in list_three:
f1.add_arc('six', 'three', (letter), '3')
elif letter in list_four:
f1.add_arc('six', 'four', (letter), '4')
elif letter in list_five:
f1.add_arc('six', 'five', (letter), '5')
else:
f1.add_arc('six', 'six', (letter), ())
return f1
def letters_to_numbers():
"""
Returns an FST that converts letters to numbers as specified by
the soundex algorithm
"""
# Let's define our first FST
f1 = FST('soundex-generate')
# Indicate that '1' is the initial state
f1.add_state('start')
#f1.add_state('next')
f1.add_state('grp0')
f1.add_state('grp1')
f1.add_state('grp2')
f1.add_state('grp3')
f1.add_state('grp4')
f1.add_state('grp5')
f1.add_state('grp6')
f1.initial_state = 'start'
# Set all the final states
#f1.set_final('next')
f1.set_final('grp0')
f1.set_final('grp1')
f1.set_final('grp2')
f1.set_final('grp3')
f1.set_final('grp4')
f1.set_final('grp5')
f1.set_final('grp6')
# Add the rest of the arcs
for letter in string.ascii_lowercase:
#f1.add_arc('start', 'next', (letter), (letter)) # do we need to move this to end ????
'''
if letter in "aeiouhwy":
f1.add_arc('next', 'next', (letter), ())
f1.add_arc('grp1', 'grp1', (letter), ())
f1.add_arc('grp2', 'grp2', (letter), ())
f1.add_arc('grp3', 'grp3', (letter), ())
f1.add_arc('grp4', 'grp4', (letter), ())
f1.add_arc('grp5', 'grp5', (letter), ())
f1.add_arc('grp6', 'grp6', (letter), ())
if letter in "bfpv":
f1.add_arc('next', 'grp1', (letter), ('1'))
f1.add_arc('grp1', 'grp1', (letter), ())
f1.add_arc('grp2', 'grp1', (letter), ('1'))
f1.add_arc('grp3', 'grp1', (letter), ('1'))
f1.add_arc('grp4', 'grp1', (letter), ('1'))
f1.add_arc('grp5', 'grp1', (letter), ('1'))
f1.add_arc('grp6', 'grp1', (letter), ('1'))
if letter in "cgjkqsxz":
f1.add_arc('next', 'grp2', (letter), ('2'))
f1.add_arc('grp1', 'grp2', (letter), ('2'))
f1.add_arc('grp2', 'grp2', (letter), ())
f1.add_arc('grp3', 'grp2', (letter), ('2'))
f1.add_arc('grp4', 'grp2', (letter), ('2'))
f1.add_arc('grp5', 'grp2', (letter), ('2'))
f1.add_arc('grp6', 'grp2', (letter), ('2'))
if letter in "dt":
f1.add_arc('next', 'grp3', (letter), ('3'))
f1.add_arc('grp1', 'grp3', (letter), ('3'))
f1.add_arc('grp2', 'grp3', (letter), ('3'))
f1.add_arc('grp3', 'grp3', (letter), ())
f1.add_arc('grp4', 'grp3', (letter), ('3'))
f1.add_arc('grp5', 'grp3', (letter), ('3'))
f1.add_arc('grp6', 'grp3', (letter), ('3'))
if letter in "l":
f1.add_arc('next', 'grp4', (letter), ('4'))
f1.add_arc('grp1', 'grp4', (letter), ('4'))
f1.add_arc('grp2', 'grp4', (letter), ('4'))
f1.add_arc('grp3', 'grp4', (letter), ('4'))
f1.add_arc('grp4', 'grp4', (letter), ())
f1.add_arc('grp5', 'grp4', (letter), ('4'))
f1.add_arc('grp6', 'grp4', (letter), ('4'))
if letter in "mn":
f1.add_arc('next', 'grp5', (letter), ('5'))
f1.add_arc('grp1', 'grp5', (letter), ('5'))
f1.add_arc('grp2', 'grp5', (letter), ('5'))
f1.add_arc('grp3', 'grp5', (letter), ('5'))
f1.add_arc('grp4', 'grp5', (letter), ('5'))
f1.add_arc('grp5', 'grp5', (letter), ())
f1.add_arc('grp6', 'grp5', (letter), ('5'))
if letter in "r":
f1.add_arc('next', 'grp6', (letter), ('6'))
f1.add_arc('grp1', 'grp6', (letter), ('6'))
f1.add_arc('grp2', 'grp6', (letter), ('6'))
f1.add_arc('grp3', 'grp6', (letter), ('6'))
f1.add_arc('grp4', 'grp6', (letter), ('6'))
f1.add_arc('grp5', 'grp6', (letter), ('6'))
f1.add_arc('grp6', 'grp6', (letter), ())
f1.add_arc('start', 'next', (letter), (letter))
'''
if letter in "aeiouhwy":
f1.add_arc('start', 'grp0', (letter), (letter))
f1.add_arc('grp0', 'grp0', (letter), ())
f1.add_arc('grp1', 'grp0', (letter), ())
f1.add_arc('grp2', 'grp0', (letter), ())
f1.add_arc('grp3', 'grp0', (letter), ())
f1.add_arc('grp4', 'grp0', (letter), ())
f1.add_arc('grp5', 'grp0', (letter), ())
f1.add_arc('grp6', 'grp0', (letter), ())
if letter in "bfpv":
f1.add_arc('start', 'grp1', (letter), (letter))
f1.add_arc('grp0', 'grp1', (letter), ('1'))
f1.add_arc('grp1', 'grp1', (letter), ())
f1.add_arc('grp2', 'grp1', (letter), ('1'))
f1.add_arc('grp3', 'grp1', (letter), ('1'))
f1.add_arc('grp4', 'grp1', (letter), ('1'))
f1.add_arc('grp5', 'grp1', (letter), ('1'))
f1.add_arc('grp6', 'grp1', (letter), ('1'))
if letter in "cgjkqsxz":
f1.add_arc('start', 'grp2', (letter), (letter))
f1.add_arc('grp0', 'grp2', (letter), ('2'))
f1.add_arc('grp1', 'grp2', (letter), ('2'))
f1.add_arc('grp2', 'grp2', (letter), ())
f1.add_arc('grp3', 'grp2', (letter), ('2'))
f1.add_arc('grp4', 'grp2', (letter), ('2'))
f1.add_arc('grp5', 'grp2', (letter), ('2'))
f1.add_arc('grp6', 'grp2', (letter), ('2'))
if letter in "dt":
f1.add_arc('start', 'grp3', (letter), (letter))
f1.add_arc('grp0', 'grp3', (letter), ('3'))
f1.add_arc('grp1', 'grp3', (letter), ('3'))
f1.add_arc('grp2', 'grp3', (letter), ('3'))
f1.add_arc('grp3', 'grp3', (letter), ())
f1.add_arc('grp4', 'grp3', (letter), ('3'))
f1.add_arc('grp5', 'grp3', (letter), ('3'))
f1.add_arc('grp6', 'grp3', (letter), ('3'))
if letter in "l":
f1.add_arc('start', 'grp4', (letter), (letter))
f1.add_arc('grp0', 'grp4', (letter), ('4'))
f1.add_arc('grp1', 'grp4', (letter), ('4'))
f1.add_arc('grp2', 'grp4', (letter), ('4'))
f1.add_arc('grp3', 'grp4', (letter), ('4'))
f1.add_arc('grp4', 'grp4', (letter), ())
f1.add_arc('grp5', 'grp4', (letter), ('4'))
f1.add_arc('grp6', 'grp4', (letter), ('4'))
if letter in "mn":
f1.add_arc('start', 'grp5', (letter), (letter))
f1.add_arc('grp0', 'grp5', (letter), ('5'))
f1.add_arc('grp1', 'grp5', (letter), ('5'))
f1.add_arc('grp2', 'grp5', (letter), ('5'))
f1.add_arc('grp3', 'grp5', (letter), ('5'))
f1.add_arc('grp4', 'grp5', (letter), ('5'))
f1.add_arc('grp5', 'grp5', (letter), ())
f1.add_arc('grp6', 'grp5', (letter), ('5'))
if letter in "r":
f1.add_arc('start', 'grp6', (letter), (letter))
f1.add_arc('grp0', 'grp6', (letter), ('6'))
f1.add_arc('grp1', 'grp6', (letter), ('6'))
f1.add_arc('grp2', 'grp6', (letter), ('6'))
f1.add_arc('grp3', 'grp6', (letter), ('6'))
f1.add_arc('grp4', 'grp6', (letter), ('6'))
f1.add_arc('grp5', 'grp6', (letter), ('6'))
f1.add_arc('grp6', 'grp6', (letter), ())
return f1
def parse(self, word):
"""Parse a word morphologically
e.g.
p = Parser()
word = ['p','a','n','i','c','k','i','n','g']
p.parse(word)
---> 'panic+present participle form'
"""
# Ok so now let's do the second FST
f2 = FST('morphology-parse') # Detecting affixes ('ing' and 'ed')
# Indicate initial and final states
states = ['start', 'i', 'n', 'g', 'e', 'd']
for state in states:
f2.add_state(state)
f2.initial_state = 'start'
f2.set_final('d')
f2.set_final('g')
for letter in string.ascii_lowercase:
if not letter in 'ie':
f2.add_arc('start', 'start', letter, letter)
if not letter == 'n':
f2.add_arc('i', 'start', letter, ('i', letter))
if not letter == 'd':
f2.add_arc('e', 'start', letter, ['e', letter])
if not letter == 'g':
f2.add_arc('n', 'start', letter, ['i', 'n', letter])
f2.add_arc('start', 'i', 'i', '')
f2.add_arc('i', 'n', 'n', '')
f2.add_arc('n', 'g', 'g', '+present participle form')
f2.add_arc('start', 'e', 'e', '')
f2.add_arc('e', 'd', 'd', '+past form')
f3 = FST('morphology-parse') # K-deletion
# Indicate initial and final states
f3.add_state('start')
f3.add_state('vowel')
f3.add_state('consonant')
f3.add_state('c')
f3.add_state('k')
f3.add_state('lick_l')
f3.add_state('lick_i')
f3.add_state('lick_c')
f3.add_state('lick_k')
f3.add_state('parse')
f3.initial_state = 'start'
f3.set_final('parse')
vowels = 'aeiou'
for vowel in vowels:
f3.add_arc('start', 'vowel', vowel, vowel)
f3.add_arc('vowel', 'vowel', vowel, vowel)
f3.add_arc('consonant', 'vowel', vowel, vowel)
f3.add_arc('c', 'vowel', vowel, vowel)
for letter in string.ascii_lowercase:
f3.add_arc('parse', 'parse', letter, letter)
if letter in vowels:
continue
if not letter == 'c':
f3.add_arc('vowel', 'consonant', letter, letter)
if not letter == 'l':
f3.add_arc('start', 'consonant', letter, letter)
if not letter == 'k':
f3.add_arc('c', 'consonant', letter, letter)
f3.add_arc('consonant', 'consonant', letter, letter)
f3.add_arc('vowel', 'c', 'c', 'c')
f3.add_arc('c', 'k', 'k', '')
f3.add_arc('start', 'lick_l', 'l', 'l')
f3.add_arc('lick_l', 'lick_i', 'i', 'i')
f3.add_arc('lick_i', 'lick_c', 'c', 'c')
f3.add_arc('lick_c', 'lick_k', 'k', 'k')
f3.add_arc('lick_k', 'parse', '+', '+')
f3.add_arc('k', 'parse', '+', '+')
f3.add_arc('consonant', 'parse', '+', '+')
f3.add_arc('parse', 'parse', ' ', ' ')
output = compose(word, f2, f3)[0]
return "".join(output)
def letters_to_numbers():
"""
Returns an FST that converts letters to numbers as specified by
the soundex algorithm
"""
# Let's define our first FST
f1 = FST('soundex-generate')
# Indicate that '1' is the initial state
states = ['q1', 'q2', 'n1', 'n2', 'n3', 'n4', 'n5', 'n6']
for state in states:
f1.add_state(state)
f1.initial_state = 'q1'
# Set all the final states
for state in ['q2', 'n1', 'n2', 'n3', 'n4', 'n5', 'n6']:
f1.set_final(state)
# Add the rest of the arcs
for letter in string.ascii_lowercase:
f1.add_arc('q1', 'q2', (letter), (letter))
if letter in set('aehiouwy'):
for state in ['q2', 'n1', 'n2', 'n3', 'n4', 'n5', 'n6']:
f1.add_arc(state, state, (letter), ())
else:
if letter in set('bfpv'):
for state in ['q2', 'n2', 'n3', 'n4', 'n5', 'n6']:
f1.add_arc(state, 'n1', (letter), ('1'))
f1.add_arc('n1', 'n1', (letter), ())
elif letter in set('cgjkqsxz'):
for state in ['q2', 'n1', 'n3', 'n4', 'n5', 'n6']:
f1.add_arc(state, 'n2', (letter), ('2'))
f1.add_arc('n2', 'n2', (letter), ())
elif letter in set('dt'):
for state in ['q2', 'n1', 'n2', 'n4', 'n5', 'n6']:
f1.add_arc(state, 'n3', (letter), ('3'))
f1.add_arc('n3', 'n3', (letter), ())
elif letter in set('l'):
for state in ['q2', 'n1', 'n2', 'n3', 'n5', 'n6']:
f1.add_arc(state, 'n4', (letter), ('4'))
f1.add_arc('n4', 'n4', (letter), ())
elif letter in set('mn'):
for state in ['q2', 'n1', 'n2', 'n3', 'n4', 'n6']:
f1.add_arc(state, 'n5', (letter), ('5'))
f1.add_arc('n5', 'n5', (letter), ())
elif letter in set('r'):
for state in ['q2', 'n1', 'n2', 'n3', 'n4', 'n5']:
f1.add_arc(state, 'n6', (letter), ('6'))
f1.add_arc('n6', 'n6', (letter), ())
return f1
def truncate_to_three_digits():
"""
Create an FST that will truncate a soundex string to three digits
"""
# Ok so now let's do the second FST, the one that will truncate
# the number of digits to 3
f2 = FST('soundex-truncate')
# Indicate initial and final states
f2.add_state('1')
f2.add_state('2')
f2.add_state('3')
f2.add_state('4')
f2.initial_state = '1'
f2.set_final('4')
# Add the arcs
for letter in string.letters:
f2.add_arc('1', '1', (letter), (letter))
f2.add_arc('1', '4', (), ())
for n in range(10):
f2.add_arc('1', '2', (str(n)), (str(n)))
f2.add_arc('2', '4', (), ())
for n in range(10):
f2.add_arc('2', '3', (str(n)), (str(n)))
f2.add_arc('3', '4', (), ())
for n in range(10):
f2.add_arc('3', '4', (str(n)), (str(n)))
for n in range(10):
f2.add_arc('4', '4', (str(n)), ())
return f2
def letters_to_numbers():
"""
Returns an FST that converts letters to numbers as specified by
the soundex algorithm
"""
# Let's define our first FST
f1 = FST('soundex-generate')
# Indicate that '1' is the initial state
f1.add_state('start')
f1.add_state('1')
f1.add_state('2')
f1.add_state('3')
f1.add_state('4')
f1.add_state('5')
f1.add_state('6')
f1.add_state('7')
f1.add_state('next')
f1.initial_state = 'start'
# Set all the final states
f1.set_final('7')
#setting the rules
non_in = [
'a', 'e', 'i', 'o', 'u', 'h', 'w', 'y', 'A', 'E', 'I', 'O', 'U', 'H',
'W', 'Y'
]
rep1 = ['b', 'f', 'p', 'v', 'B', 'F', 'P', 'V']
rep2 = [
'c', 'g', 'j', 'k', 'q', 's', 'x', 'z', 'C', 'G', 'J', 'K', 'Q', 'S',
'X', 'Z'
]
rep3 = ['d', 't', 'D', 'T']
rep4 = ['l', 'L']
rep5 = ['m', 'n', 'M', 'N']
rep6 = ['r', 'R']
# Add the rest of the arcs
for letter in string.ascii_letters:
if letter in non_in:
f1.add_arc('start', 'next', (letter), (letter))
if letter in rep1:
f1.add_arc('start', '1', (letter), (letter))
if letter in rep2:
f1.add_arc('start', '2', (letter), (letter))
if letter in rep3:
f1.add_arc('start', '3', (letter), (letter))
if letter in rep4:
f1.add_arc('start', '4', (letter), (letter))
if letter in rep5:
f1.add_arc('start', '5', (letter), (letter))
if letter in rep6:
f1.add_arc('start', '6', (letter), (letter))
for letter in string.ascii_letters:
if letter in non_in:
f1.add_arc('next', 'next', (letter), ())
if letter in rep1:
f1.add_arc('next', '1', (letter), ('1'))
if letter in rep2:
f1.add_arc('next', '2', (letter), ('2'))
if letter in rep3:
f1.add_arc('next', '3', (letter), ('3'))
if letter in rep4:
f1.add_arc('next', '4', (letter), ('4'))
if letter in rep5:
f1.add_arc('next', '5', (letter), ('5'))
if letter in rep6:
f1.add_arc('next', '6', (letter), ('6'))
f1.add_arc('next', '7', (), ())
for letter in string.ascii_letters:
if letter in non_in:
f1.add_arc('1', 'next', (letter), ())
if letter in rep1:
f1.add_arc('1', '1', (letter), ())
if letter in rep2:
f1.add_arc('1', '2', (letter), ('2'))
if letter in rep3:
f1.add_arc('1', '3', (letter), ('3'))
if letter in rep4:
f1.add_arc('1', '4', (letter), ('4'))
if letter in rep5:
f1.add_arc('1', '5', (letter), ('5'))
if letter in rep6:
f1.add_arc('1', '6', (letter), ('6'))
f1.add_arc('1', '7', (), ())
for letter in string.ascii_letters:
if letter in non_in:
f1.add_arc('2', 'next', (letter), ())
if letter in rep1:
f1.add_arc('2', '1', (letter), ('1'))
if letter in rep2:
f1.add_arc('2', '2', (letter), ())
if letter in rep3:
f1.add_arc('2', '3', (letter), ('3'))
if letter in rep4:
f1.add_arc('2', '4', (letter), ('4'))
if letter in rep5:
f1.add_arc('2', '5', (letter), ('5'))
if letter in rep6:
f1.add_arc('2', '6', (letter), ('6'))
f1.add_arc('2', '7', (), ())
for letter in string.ascii_letters:
if letter in non_in:
f1.add_arc('3', 'next', (letter), ())
if letter in rep1:
f1.add_arc('3', '1', (letter), ('1'))
if letter in rep2:
f1.add_arc('3', '2', (letter), ('2'))
if letter in rep3:
f1.add_arc('3', '3', (letter), ())
if letter in rep4:
f1.add_arc('3', '4', (letter), ('4'))
if letter in rep5:
f1.add_arc('3', '5', (letter), ('5'))
if letter in rep6:
f1.add_arc('3', '6', (letter), ('6'))
f1.add_arc('3', '7', (), ())
for letter in string.ascii_letters:
if letter in non_in:
f1.add_arc('4', 'next', (letter), ())
if letter in rep1:
f1.add_arc('4', '1', (letter), ('1'))
if letter in rep2:
f1.add_arc('4', '2', (letter), ('2'))
if letter in rep3:
f1.add_arc('4', '3', (letter), (''))
if letter in rep4:
f1.add_arc('4', '4', (letter), ())
if letter in rep5:
f1.add_arc('4', '5', (letter), ('5'))
if letter in rep6:
f1.add_arc('4', '6', (letter), ('6'))
f1.add_arc('4', '7', (), ())
for letter in string.ascii_letters:
if letter in non_in:
f1.add_arc('5', 'next', (letter), ())
if letter in rep1:
f1.add_arc('5', '1', (letter), ('1'))
if letter in rep2:
f1.add_arc('5', '2', (letter), ('2'))
if letter in rep3:
f1.add_arc('5', '3', (letter), (''))
if letter in rep4:
f1.add_arc('5', '4', (letter), ('4'))
if letter in rep5:
f1.add_arc('5', '5', (letter), ())
if letter in rep6:
f1.add_arc('5', '6', (letter), ('6'))
f1.add_arc('5', '7', (), ())
for letter in string.ascii_letters:
if letter in non_in:
f1.add_arc('6', 'next', (letter), ())
if letter in rep1:
f1.add_arc('6', '1', (letter), ('1'))
if letter in rep2:
f1.add_arc('6', '2', (letter), ('2'))
if letter in rep3:
f1.add_arc('6', '3', (letter), (''))
if letter in rep4:
f1.add_arc('6', '4', (letter), ('4'))
if letter in rep5:
f1.add_arc('6', '5', (letter), ('5'))
if letter in rep6:
f1.add_arc('6', '6', (letter), ())
f1.add_arc('6', '7', (), ())
return f1
def french_count():
f = FST('french')
# Indicate initial and final states
f.add_state('start')
f.initial_state = 'start'
f.add_state('znod1')
f.add_state('znod2')
f.add_state('0')
f.add_state('units')
f.add_state('tens')
f.add_state('hundreds')
f.add_state('unod1')
f.add_state('unod2')
f.add_state('1s')
f.add_state('1e')
f.add_state('2-9s')
f.add_state('tnod1')
f.add_state('11s')
f.add_state('11e')
f.add_state('10-60s')
f.add_state('10-60e')
f.add_state('11-16s')
f.add_state('12-16e')
f.add_state('7090s')
f.add_state('80s')
f.add_state('100-900s')
f.add_state('100-900t')
f.add_state('100-900e')
# add arcs to first level
f.add_arc('start', 'znod1', ('0'), ())
f.add_arc('znod1', 'znod2', ('0'), ())
f.add_arc('znod2', '0', ('0'), [kFRENCH_TRANS[0]])
f.set_final('0')
f.add_arc('start', 'units', (), ())
f.add_arc('start', 'tens', (), ())
f.add_arc('start', 'hundreds', (), ())
# covers 1
f.add_arc('units', 'unod1', ('0'), ())
f.add_arc('unod1', 'unod2', ('0'), ())
f.add_arc('unod2', '1s', (), ())
f.add_arc('1s', '1e', ('1'), [kFRENCH_TRANS[1]])
f.set_final('1e')
# Covers 2-9
for i in range(2, 10):
f.add_arc('unod2', '2-9s', (str(i)), [kFRENCH_TRANS[i]])
f.set_final('2-9s')
# covers 10-60
f.add_arc('tens', 'tnod1', ('0'), ())
for i in range(1, 7):
f.add_arc('tnod1', '10-60s', (str(i)), [kFRENCH_TRANS[i * 10]])
f.add_arc('10-60s', '10-60e', ('0'), ())
f.add_arc('10-60s', 'unod2', (), ())
f.add_arc('10-60s', '1s', (), [kFRENCH_AND])
f.set_final('10-60e')
# covers 11-16
f.add_arc('tnod1', '11-16s', ('1'), ())
#f.add_arc('11-16s', 'units', (), ())
# covers 11
f.add_arc('11-16s', '11s', (), ())
f.add_arc('11s', '11e', ('1'), [kFRENCH_TRANS[11]])
f.set_final('11e')
# covers 12 - 19
for i in range(2, 7):
f.add_arc('11-16s', '12-16e', (str(i)), [kFRENCH_TRANS[i + 10]])
f.set_final('12-16e')
# covers 70-90
f.add_arc('tnod1', '7090s', ('7'), [kFRENCH_TRANS[60]])
f.add_arc('tnod1', '7090s', ('9'),
[kFRENCH_TRANS[4] + " " + kFRENCH_TRANS[20]])
f.add_arc('7090s', 'unod2', (), [kFRENCH_TRANS[10]])
f.add_arc('7090s', '11-16s', (), ())
f.add_arc('7090s', '10-60e', ('0'), [kFRENCH_TRANS[10]])
f.add_arc('tnod1', '80s', ('8'),
[kFRENCH_TRANS[4] + " " + kFRENCH_TRANS[20]])
f.add_arc('80s', 'unod2', (), ())
f.add_arc('80s', '10-60e', ('0'), ())
# Covers 100 - 900
f.add_arc('hundreds', '100-900s', ('1'), [kFRENCH_TRANS[100]])
for i in range(2, 10):
f.add_arc('hundreds', '100-900s', (str(i)),
[kFRENCH_TRANS[i] + " " + kFRENCH_TRANS[100]])
f.add_arc('100-900s', 'tnod1', (), ())
f.add_arc('100-900s', '100-900t', ('0'), ())
f.add_arc('100-900t', 'unod2', (), ())
f.add_arc('100-900t', '100-900e', ('0'), ())
f.set_final('100-900e')
return f
def letters_to_numbers():
"""
Returns an FST that converts letters to numbers as specified by
the soundex algorithm
"""
remove = ['a', 'e', 'h', 'i', 'o', 'u', 'w', 'y']
group1 = ['b', 'f', 'p', 'v']
group2 = ['c', 'g', 'j', 'k', 'q', 's', 'x', 'z']
group3 = ['d', 't']
group4 = ['l']
group5 = ['m', 'n']
group6 = ['r']
# Let's define our first FST
f1 = FST('soundex-generate')
# Indicate that 'start' is the initial state
f1.add_state('start')
f1.add_state('1')
f1.add_state('2')
f1.add_state('3')
f1.add_state('4')
f1.add_state('5')
f1.add_state('6')
f1.add_state('7')
f1.add_state('8')
f1.add_state('9')
f1.add_state('10')
f1.add_state('11')
f1.add_state('12')
f1.add_state('13')
f1.add_state('14')
f1.add_state('15')
f1.add_state('16')
f1.add_state('17')
f1.add_state('18')
f1.add_state('19')
f1.add_state('20')
f1.initial_state = 'start'
# Set all the final states
f1.set_final('16')
f1.set_final('17')
f1.set_final('18')
f1.set_final('19')
f1.set_final('20')
for letter in string.ascii_letters:
f1.add_arc('start', '1', (letter), (letter))
if letter in remove:
f1.add_arc('1', '1', (letter), ())
f1.add_arc('2', '8', (letter), ())
f1.add_arc('3', '8', (letter), ())
f1.add_arc('4', '8', (letter), ())
f1.add_arc('5', '8', (letter), ())
f1.add_arc('6', '8', (letter), ())
f1.add_arc('7', '8', (letter), ())
f1.add_arc('8', '8', (letter), ())
f1.add_arc('9', '15', (letter), ())
f1.add_arc('10', '15', (letter), ())
f1.add_arc('11', '15', (letter), ())
f1.add_arc('12', '15', (letter), ())
f1.add_arc('13', '15', (letter), ())
f1.add_arc('14', '15', (letter), ())
f1.add_arc('15', '15', (letter), ())
f1.add_arc('16', '16', (letter), ())
elif letter in group1:
f1.add_arc('1', '2', (letter), ('1'))
f1.add_arc('2', '2', (letter), ())
f1.add_arc('3', '9', (letter), ('1'))
f1.add_arc('4', '9', (letter), ('1'))
f1.add_arc('5', '9', (letter), ('1'))
f1.add_arc('6', '9', (letter), ('1'))
f1.add_arc('7', '9', (letter), ('1'))
f1.add_arc('8', '9', (letter), ('1'))
f1.add_arc('9', '9', (letter), ())
f1.add_arc('10', '16', (letter), ('1'))
f1.add_arc('11', '16', (letter), ('1'))
f1.add_arc('12', '16', (letter), ('1'))
f1.add_arc('13', '16', (letter), ('1'))
f1.add_arc('14', '16', (letter), ('1'))
f1.add_arc('15', '16', (letter), ('1'))
f1.add_arc('16', '16', (letter), ('1'))
elif letter in group2:
f1.add_arc('1', '3', (letter), ('2'))
f1.add_arc('3', '3', (letter), ())
f1.add_arc('2', '10', (letter), ('2'))
f1.add_arc('4', '10', (letter), ('2'))
f1.add_arc('5', '10', (letter), ('2'))
f1.add_arc('6', '10', (letter), ('2'))
f1.add_arc('7', '10', (letter), ('2'))
f1.add_arc('8', '10', (letter), ('2'))
f1.add_arc('10', '10', (letter), ())
f1.add_arc('9', '16', (letter), ('2'))
f1.add_arc('11', '16', (letter), ('2'))
f1.add_arc('12', '16', (letter), ('2'))
f1.add_arc('13', '16', (letter), ('2'))
f1.add_arc('14', '16', (letter), ('2'))
f1.add_arc('15', '16', (letter), ('2'))
f1.add_arc('16', '16', (letter), ('2'))
elif letter in group3:
f1.add_arc('1', '4', (letter), ('3'))
f1.add_arc('4', '4', (letter), ())
f1.add_arc('2', '11', (letter), ('3'))
f1.add_arc('3', '11', (letter), ('3'))
f1.add_arc('5', '11', (letter), ('3'))
f1.add_arc('6', '11', (letter), ('3'))
f1.add_arc('7', '11', (letter), ('3'))
f1.add_arc('8', '11', (letter), ('3'))
f1.add_arc('11', '11', (letter), ())
f1.add_arc('9', '16', (letter), ('3'))
f1.add_arc('10', '16', (letter), ('3'))
f1.add_arc('12', '16', (letter), ('3'))
f1.add_arc('13', '16', (letter), ('3'))
f1.add_arc('14', '16', (letter), ('3'))
f1.add_arc('15', '16', (letter), ('3'))
f1.add_arc('16', '16', (letter), ('3'))
elif letter in group4:
f1.add_arc('1', '5', (letter), ('4'))
f1.add_arc('5', '5', (letter), ())
f1.add_arc('2', '12', (letter), ('4'))
f1.add_arc('3', '12', (letter), ('4'))
f1.add_arc('4', '12', (letter), ('4'))
f1.add_arc('6', '12', (letter), ('4'))
f1.add_arc('7', '12', (letter), ('4'))
f1.add_arc('8', '12', (letter), ('4'))
f1.add_arc('12', '12', (letter), ())
f1.add_arc('9', '16', (letter), ('4'))
f1.add_arc('10', '16', (letter), ('4'))
f1.add_arc('11', '16', (letter), ('4'))
f1.add_arc('13', '16', (letter), ('4'))
f1.add_arc('14', '16', (letter), ('4'))
f1.add_arc('15', '16', (letter), ('4'))
f1.add_arc('16', '16', (letter), ('4'))
elif letter in group5:
f1.add_arc('1', '6', (letter), ('5'))
f1.add_arc('6', '6', (letter), ())
f1.add_arc('2', '13', (letter), ('5'))
f1.add_arc('3', '13', (letter), ('5'))
f1.add_arc('4', '13', (letter), ('5'))
f1.add_arc('5', '13', (letter), ('5'))
f1.add_arc('7', '13', (letter), ('5'))
f1.add_arc('8', '13', (letter), ('5'))
f1.add_arc('13', '13', (letter), ())
f1.add_arc('9', '16', (letter), ('5'))
f1.add_arc('10', '16', (letter), ('5'))
f1.add_arc('11', '16', (letter), ('5'))
f1.add_arc('12', '16', (letter), ('5'))
f1.add_arc('14', '16', (letter), ('5'))
f1.add_arc('15', '16', (letter), ('5'))
f1.add_arc('16', '16', (letter), ('5'))
elif letter in group6:
f1.add_arc('1', '7', (letter), ('6'))
f1.add_arc('7', '7', (letter), ())
f1.add_arc('2', '14', (letter), ('6'))
f1.add_arc('3', '14', (letter), ('6'))
f1.add_arc('4', '14', (letter), ('6'))
f1.add_arc('5', '14', (letter), ('6'))
f1.add_arc('6', '14', (letter), ('6'))
f1.add_arc('8', '14', (letter), ('6'))
f1.add_arc('14', '14', (letter), ())
f1.add_arc('9', '16', (letter), ('6'))
f1.add_arc('10', '16', (letter), ('6'))
f1.add_arc('11', '16', (letter), ('6'))
f1.add_arc('12', '16', (letter), ('6'))
f1.add_arc('13', '16', (letter), ('6'))
f1.add_arc('15', '16', (letter), ('6'))
f1.add_arc('16', '16', (letter), ('6'))
else:
f1.add_arc('1', '17', (), ())
f1.add_arc('2', '18', (), ())
f1.add_arc('3', '18', (), ())
f1.add_arc('4', '18', (), ())
f1.add_arc('5', '18', (), ())
f1.add_arc('6', '18', (), ())
f1.add_arc('7', '18', (), ())
f1.add_arc('8', '18', (), ())
f1.add_arc('9', '19', (), ())
f1.add_arc('10', '19', (), ())
f1.add_arc('11', '19', (), ())
f1.add_arc('12', '19', (), ())
f1.add_arc('13', '19', (), ())
f1.add_arc('14', '19', (), ())
f1.add_arc('15', '19', (), ())
f1.add_arc('16', '20', (), ())
return f1
def letters_to_numbers():
"""
Returns an FST that converts letters to numbers as specified by
the soundex algorithm
"""
# Let's define our first FST
f1 = FST('soundex-generate')
# Indicate that '1' is the initial state
f1.add_state('start')
f1.initial_state = 'start'
f1.add_state('0')
f1.set_final('0')
for letter in string.ascii_letters:
f1.add_arc('start', '0', (letter), (letter)) # for 1st letter in IP
removal_letters_set = {'a', 'e', 'i', 'o', 'u', 'h', 'w', 'y'}
for removeChar in list(removal_letters_set):
f1.add_arc('0', '0', (removeChar), ())
soundex_letter_lkp = [(['b', 'f', 'p', 'v'], '1'),
(['c', 'g', 'j', 'k', 'q', 's', 'x', 'z'], '2'),
(['d', 't'], '3'), (['l'], '4'), (['m', 'n'], '5'),
(['r'], '6')]
soundex_chars_set = set()
for charList, state in soundex_letter_lkp:
soundex_chars_set = soundex_chars_set.union(set(charList))
f1.add_state(state)
f1.set_final(state)
#build automata
for charList, state in soundex_letter_lkp:
for char in charList:
f1.add_arc('0', state, (char), (state)) #1st jump
f1.add_arc(state, state, (char), ()) # self loop
for char in list(removal_letters_set): # for vowelsset
f1.add_arc(state, '0', (char), ())
for char in list(
soundex_chars_set.difference(set(charList))
): # any other char from different group will cause return to 0 state, with OP
for returnCharList, returnState in soundex_letter_lkp:
if char in returnCharList:
f1.add_arc(state, returnState, (char), (returnState))
return f1
def french_count():
f = FST('french')
list26 = [2, 3, 4, 5, 6]
list16 = [1, 2, 3, 4, 5, 6]
list79 = [7, 8, 9]
list19 = [1, 2, 3, 4, 5, 6, 7, 8, 9]
list29 = [2, 3, 4, 5, 6, 7, 8, 9]
f.add_state('1')
f.add_state('2')
f.add_state('3')
f.add_state('4')
f.add_state('5')
f.add_state('6')
f.add_state('7')
f.add_state('8')
f.add_state('9')
f.add_state('10')
f.add_state('11')
f.add_state('12')
f.add_state('13')
f.add_state('14')
f.add_state('15')
f.add_state('16')
f.add_state('17')
f.add_state('18')
f.add_state('19')
f.add_state('20')
f.add_state('21')
f.initial_state = '1'
f.set_final('2')
f.set_final('4')
f.set_final('7')
f.set_final('8')
f.set_final('9')
f.set_final('10')
f.set_final('12')
f.set_final('14')
f.set_final('15')
f.set_final('19')
f.set_final('21')
element_z = [0]
element_o = [1]
for i in element_o:
i = str(i)
f.add_arc('1', '6', i, [kFRENCH_TRANS[int(i) * 10]])
f.add_arc('18', '6', i, [kFRENCH_TRANS[int(i) * 10]])
f.add_arc('1', '3', i, ())
f.add_arc('18', '3', i, ())
f.add_arc('11', '12', i, [kFRENCH_AND, kFRENCH_TRANS[int(i) + 10]])
f.add_arc('13', '15', i, [kFRENCH_TRANS[20], kFRENCH_TRANS[int(i)]])
f.add_arc('17', '21', i,
[kFRENCH_TRANS[20], kFRENCH_TRANS[int(i) + 10]])
f.add_arc('1', '18', i, [kFRENCH_TRANS[100]])
f.add_arc('5', '8', i, [kFRENCH_AND, kFRENCH_TRANS[int(i)]])
for i in element_z:
i = str(i)
f.add_arc('1', '10', i, [kFRENCH_TRANS[int(i)]])
f.add_arc('11', '12', i, [kFRENCH_TRANS[int(i) + 10]])
f.add_arc('18', '20', i, ())
f.add_arc('6', '9', i, ())
f.add_arc('20', '10', i, ())
f.add_arc('13', '14', i, [kFRENCH_TRANS[20]])
f.add_arc('17', '19', i, [kFRENCH_TRANS[20], kFRENCH_TRANS[10]])
f.add_arc('1', '1', i, ())
f.add_arc('5', '7', i, ())
for i in list16:
i = str(i)
f.add_arc('3', '4', i, [kFRENCH_TRANS[int(i) + 10]])
for i in list29:
i = str(i)
f.add_arc('13', '15', i, [kFRENCH_TRANS[20], kFRENCH_TRANS[int(i)]])
f.add_arc('5', '8', i, [kFRENCH_TRANS[int(i)]])
f.add_arc('1', '18', i, [kFRENCH_TRANS[int(i)], kFRENCH_TRANS[100]])
for i in list26:
i = str(i)
f.add_arc('11', '12', i, [kFRENCH_TRANS[int(i) + 10]])
f.add_arc('17', '21', i,
[kFRENCH_TRANS[20], kFRENCH_TRANS[int(i) + 10]])
f.add_arc('18', '5', i, [kFRENCH_TRANS[int(i) * 10]])
f.add_arc('1', '5', i, [kFRENCH_TRANS[int(i) * 10]])
for i in list19:
i = str(i)
f.add_arc('20', '2', i, [kFRENCH_TRANS[int(i)]])
f.add_arc('1', '2', i, [kFRENCH_TRANS[int(i)]])
for i in list79:
i = str(i)
f.add_arc('11', '12', i, [kFRENCH_TRANS[10], kFRENCH_TRANS[int(i)]])
f.add_arc('6', '9', i, [kFRENCH_TRANS[int(i)]])
f.add_arc(
'17', '21', i,
[kFRENCH_TRANS[20], kFRENCH_TRANS[10], kFRENCH_TRANS[int(i)]])
list7 = [7]
list8 = [8]
list9 = [9]
y = int(60)
x = int(4)
for i in list7:
i = str(i)
f.add_arc('18', '11', i, [kFRENCH_TRANS[y]])
f.add_arc('1', '11', i, [kFRENCH_TRANS[y]])
for i in list8:
i = str(i)
f.add_arc('1', '13', i, [kFRENCH_TRANS[x]])
f.add_arc('18', '13', i, [kFRENCH_TRANS[x]])
for i in list9:
i = str(i)
f.add_arc('1', '17', i, [kFRENCH_TRANS[x]])
f.add_arc('18', '17', i, [kFRENCH_TRANS[x]])
return f
def add_zero_padding():
#===========================================================================
# #Now, the third fst - the zero-padding fst
# f3 = FST('soundex-padzero')
#
# # Indicate initial and final states
# f3.add_state('1')
# f3.initial_state = '1'
#
# f3.add_state('2')
# f3.add_state('3')
# f3.add_state('4')
# f3.add_state('5')
#
# #f3.set_final('2')
# #f3.set_final('3')
# #f3.set_final('4')
# f3.set_final('5')
#
# # Add the arcs
# possible_chars = string.digits+ string.letters
# for letter in possible_chars:
# f3.add_arc('1', '2', (letter), (letter))
#
# f3.add_arc('2', '3', (letter), (letter))
#
#
# f3.add_arc('3', '4', (letter), (letter))
#
#
# f3.add_arc('4', '5', (letter), (letter))
#
#
# f3.add_arc('5', '5', (letter), (letter))
#
# #===========================================================================
# #f3.add_arc('1', '2', (), ('0'))
# f3.add_arc('2', '3', (), ('0'))
# f3.add_arc('3', '4', (), ('0'))
# f3.add_arc('4', '5', (), ('0'))
# #===========================================================================
# return f3
#===========================================================================
f3 = FST('soundex-padzero')
f3.add_state('1')
f3.add_state('1a')
f3.add_state('1b')
f3.add_state('2')
f3.initial_state = '1'
f3.set_final('2')
for letter in string.letters:
f3.add_arc('1', '1', (letter), (letter))
for number in xrange(10):
f3.add_arc('1', '1', (str(number)), (str(number)))
f3.add_arc('1', '1a', (), ('0'))
f3.add_arc('1a', '1b', (), ('0'))
f3.add_arc('1b', '2', (), ('0'))
return f3
def letters_to_numbers():
"""
Returns an FST that converts letters to numbers as specified by
the soundex algorithm
"""
# Let's define our first FST
f1 = FST('soundex-generate')
# Indicate that '1' is the initial state
f1.add_state('start')
f1.add_state('next')
f1.initial_state = 'start'
# Set all the final states
f1.set_final('next')
# Add the rest of the arcs
for letter in string.ascii_lowercase:
# f1.add_arc('start', 'next', (letter), (letter))
# f1.add_arc('next', 'next', (letter), ('0'))
f1.add_arc('start', 'next', (letter), (letter))
if letter in vowels:
f1.add_arc('next', 'next', (letter), ())
elif letter in grp1:
f1.add_arc('next', 'next', (letter), ('1'))
elif letter in grp2:
f1.add_arc('next', 'next', (letter), ('2'))
elif letter in grp3:
f1.add_arc('next', 'next', (letter), ('3'))
elif letter in grp4:
f1.add_arc('next', 'next', (letter), ('4'))
elif letter in grp5:
f1.add_arc('next', 'next', (letter), ('5'))
elif letter in grp6:
f1.add_arc('next', 'next', (letter), ('6'))
else:
continue
#wtf
return f1
def french_count():
f = FST('french')
f.add_state('start')
f.initial_state = 'start'
for index in range(10):
f.add_state('H'+str(index))
f.add_state('T'+str(index))
f.add_state('U'+str(index))
f.add_state('U1'+str(index))
f.set_final('U'+str(index))
f.set_final('U1'+str(index))
f.add_state('H:(T0-T9)_COMMON')
f.add_state('T:(U2-U9)_COMMON')
f.add_state('T:(U11-U19)_COMMON')
# Prepare 100s
for i in range(10):
if i==0: opList= []
elif i ==1: opList = [ kFRENCH_TRANS[100] ]
else: opList = [ kFRENCH_TRANS[i] + ' '+kFRENCH_TRANS[100] ]
f.add_arc('start','H'+str(i), [str(i)], opList )
if i !=0: f.add_arc('H'+str(i), 'H:(T0-T9)_COMMON', [], []) # 0 has special case, connections made in line 58
#- special case 0
f.add_state('0_case')
f.add_arc('H0','0_case',['0'],[])
f.add_arc('0_case', 'U0',['0'],[kFRENCH_TRANS[0]])
f.add_arc('0_case', 'U0',['1'],[kFRENCH_TRANS[1]])
f.add_arc('0_case', 'T:(U2-U9)_COMMON',[],[])
# Prepare 10s - arcs to reach state
for i in range(10):
if i==0 or i==1: opList = []
elif i<=6: opList = [ kFRENCH_TRANS[i*10] ]
elif i==7: opList = [ kFRENCH_TRANS[60] ]
else: opList = [ kFRENCH_TRANS[4]+' '+kFRENCH_TRANS[20] ]
f.add_arc('H:(T0-T9)_COMMON','T'+str(i), [str(i)], opList )
if i !=0: f.add_arc('H0', 'T'+str(i), [str(i)], opList )
# Prepare 10s - arcs to exit to 1s - Major Mapping BEGIN
f.add_arc('T0', 'U0', ['0'], [])
f.add_arc('T0', 'U1', ['1'], [kFRENCH_TRANS[1]])
f.add_arc('T0', 'T:(U2-U9)_COMMON', [], [])
f.add_arc('T1', 'T:(U11-U19)_COMMON', [], [])
f.add_arc('T1', 'U10', ['0'], [kFRENCH_TRANS[10]])
f.add_arc('T2', 'U0', ['0'], [])
f.add_arc('T2', 'U1', ['1'], [kFRENCH_AND+' '+kFRENCH_TRANS[1]])
f.add_arc('T2', 'T:(U2-U9)_COMMON', [], [])
f.add_arc('T3', 'U0', ['0'], [])
f.add_arc('T3', 'U1', ['1'], [kFRENCH_AND+' '+kFRENCH_TRANS[1]])
f.add_arc('T3', 'T:(U2-U9)_COMMON', [], [])
f.add_arc('T4', 'U0', ['0'], [])
f.add_arc('T4', 'U1', ['1'], [kFRENCH_AND+' '+kFRENCH_TRANS[1]])
f.add_arc('T4', 'T:(U2-U9)_COMMON', [], [])
f.add_arc('T5', 'U0', ['0'], [])
f.add_arc('T5', 'U1', ['1'], [kFRENCH_AND+' '+kFRENCH_TRANS[1]])
f.add_arc('T5', 'T:(U2-U9)_COMMON', [], [])
f.add_arc('T6', 'U0', ['0'], [])
f.add_arc('T6', 'U1', ['1'], [kFRENCH_AND+' '+kFRENCH_TRANS[1]])
f.add_arc('T6', 'T:(U2-U9)_COMMON', [], [])
for i in range(10):
if i == 1: opList = [ kFRENCH_AND + ' '+kFRENCH_TRANS[i+10] ]
elif i in [7,8,9]: opList = [ kFRENCH_TRANS[10]+' '+kFRENCH_TRANS[i] ]
else: opList = [ kFRENCH_TRANS[i+10] ]
f.add_arc('T7', 'U'+str(10+i), [str(i)], opList )
#=========================================================================== weird behavior on epsilon, backup above
# f.add_arc('T7', 'U10', ['0'], [kFRENCH_TRANS[10]])
# f.add_arc('T7', 'U11', ['1'], [kFRENCH_AND+' '+kFRENCH_TRANS[11]])
# f.add_arc('T7', 'T:(U11-U19)_COMMON', [], [])
#===========================================================================
f.add_arc('T8', 'U0', ['0'], [])
f.add_arc('T8', 'U1', ['1'], [kFRENCH_TRANS[1]])
f.add_arc('T8', 'T:(U2-U9)_COMMON', [], [])
f.add_arc('T9', 'U10', ['0'], [kFRENCH_TRANS[10]])
f.add_arc('T9', 'T:(U11-U19)_COMMON', [], [])
# Prepare 10s - arcs to exit to 1s - Major Mapping END
# prepare Common: 1s
for i in range(1,10):
if i!=1: f.add_arc('T:(U2-U9)_COMMON', 'U'+str(i), [str(i)], [ kFRENCH_TRANS[i]])
if i+10 not in [17,18,19]: f.add_arc('T:(U11-U19)_COMMON', 'U'+str(10+i), [str(i)], [ kFRENCH_TRANS[i+10]])
else: f.add_arc('T:(U11-U19)_COMMON', 'U'+str(10+i), [str(i)], [ kFRENCH_TRANS[10]+' '+kFRENCH_TRANS[i]])
return f
def add_zero_padding():
# Now, the third fst - the zero-padding fst
f3 = FST('soundex-padzero')
# Indicate initial and final states
f3.add_state('0')
f3.add_state('1')
f3.add_state('1a')
f3.add_state('1b')
f3.add_state('1c')
f3.add_state('2')
f3.add_state('3a')
f3.add_state('3b')
f3.add_state('4')
f3.initial_state = '0'
f3.set_final('4')
# Add the arcs
for letter in string.letters:
f3.add_arc('0', '1', (letter), (letter))
for n in range(10):
f3.add_arc('1', '1a', (str(n)), (str(n)))
f3.add_arc('1a', '1b', (str(n)), (str(n)))
f3.add_arc('1b', '4', (str(n)), (str(n)))
f3.add_arc('1', '2', (str(n)), (str(n)))
f3.add_arc('1', '3a', (str(n)), (str(n)))
f3.add_arc('3a', '3b', (str(n)), (str(n)))
f3.add_arc('0', '1', (), ())
f3.add_arc('1', '4', (), ('000'))
f3.add_arc('2', '4', (), ('00'))
f3.add_arc('3b', '4', (), ('0'))
return f3
def french_count():
f = FST('french')
f.add_state('start')
f.add_state('2')
f.add_state('3')
f.add_state('4')
f.add_state('5')
f.add_state('6')
f.add_state('7')
f.add_state('8')
f.add_state('9')
f.add_state('10')
f.add_state('11')
f.add_state('12')
f.add_state('13')
f.add_state('14')
f.add_state('15')
f.add_state('16')
# f.add_state('17')
# f.add_state('18')
# f.add_state('19')
# f.add_state('20')
# f.add_state('21')
# f.add_state('22')
f.initial_state = 'start'
f.set_final('4')
f.set_final('6')
f.set_final('7')
f.set_final('9')
# f.set_final('13')
# f.set_final('14')
# f.set_final('15')
f.set_final('12')
f.set_final('16')
# f.set_final('17')
# f.set_final('18')
# f.set_final('20')
# f.set_final('21')
# f.set_final('22')
#takes care of single digit
f.add_arc('start', '2', '0', ())
f.add_arc('2', '3', '0', ())
for digit in range(0, 10):
f.add_arc('3', '4', [str(digit)], [kFRENCH_TRANS[digit]])
#Two-digits -in range (10,16)
f.add_arc('2', '5', '1', ())
for digit in range(0, 7):
f.add_arc('5', '6', [str(digit)], [kFRENCH_TRANS[digit + 10]])
#takes care of 17,18,19
for digit in range(7, 10):
f.add_arc('5', '7', [str(digit)],
[kFRENCH_TRANS[10]] + [kFRENCH_TRANS[digit]])
#takes care of 20
for digit in range(2, 7):
f.add_arc('2', '8', [str(digit)], [kFRENCH_TRANS[digit * 10]])
f.add_arc('8', '9', '0', ())
#takes care of 21-29(Takes care of 20 to 60)
for digit in range(1, 10):
if digit == 1:
f.add_arc('8', '4', [str(digit)],
[kFRENCH_AND] + [kFRENCH_TRANS[digit]])
else:
f.add_arc('8', '4', [str(digit)], [kFRENCH_TRANS[digit]])
#takes care of 70-79
f.add_arc('2', '10', '7', [kFRENCH_TRANS[60]])
for digit in range(0, 7):
if digit == 1:
f.add_arc('10', '6', [str(digit)],
[kFRENCH_AND] + [kFRENCH_TRANS[digit + 10]])
else:
f.add_arc('10', '6', [str(digit)], [kFRENCH_TRANS[digit + 10]])
for digit in range(7, 10):
f.add_arc('10', '7', [str(digit)],
[kFRENCH_TRANS[10]] + [kFRENCH_TRANS[digit]])
#takes care of 80-89
f.add_arc('2', '11', '8', [kFRENCH_TRANS[4]] + [kFRENCH_TRANS[20]])
#takes care of 80
f.add_arc('11', '12', '0', ())
for digit in range(1, 10):
f.add_arc('11', '4', [str(digit)], [kFRENCH_TRANS[digit]])
#takes care of 90-99
f.add_arc('2', '13', '9', [kFRENCH_TRANS[4]] + [kFRENCH_TRANS[20]])
for digit in range(0, 7):
f.add_arc('13', '6', [str(digit)], [kFRENCH_TRANS[digit + 10]])
for digit in range(7, 10):
f.add_arc('13', '7', [str(digit)],
[kFRENCH_TRANS[10]] + [kFRENCH_TRANS[digit]])
#takes care of 100
f.add_arc('start', '14', '1', ['cent'])
f.add_arc('14', '15', '0', ())
f.add_arc('15', '16', '0', ())
#takes care of 101-109
for digit in range(1, 10):
f.add_arc('15', '4', [str(digit)], [kFRENCH_TRANS[digit]])
#takes care of 110-119
f.add_arc('14', '5', '1', ())
#takes care of 120-169
for digit in range(2, 7):
f.add_arc('14', '8', [str(digit)], [kFRENCH_TRANS[digit * 10]])
#takes care of 170-179
f.add_arc('14', '10', '7', [kFRENCH_TRANS[60]])
#takes care of 180-189
f.add_arc('14', '11', '8', [kFRENCH_TRANS[4]] + [kFRENCH_TRANS[20]])
#takes care of 190-199
f.add_arc('14', '13', '9', [kFRENCH_TRANS[4]] + [kFRENCH_TRANS[20]])
#takes care of 200-999
for digit in range(2, 10):
f.add_arc('start', '14', [str(digit)],
[kFRENCH_TRANS[digit]] + ['cent'])
return f
def letters_to_numbers():
"""
Returns an FST that converts letters to numbers as specified by
the soundex algorithm
"""
# Let's define our first FST
f1 = FST('soundex-generate')
# Indicate that '1' is the initial state
f1.add_state('start')
f1.add_state('next')
f1.add_state('next0')
f1.add_state('next1')
f1.add_state('next2')
f1.add_state('next3')
f1.add_state('next4')
f1.add_state('next5')
f1.add_state('next6')
f1.initial_state = 'start'
# Set all the final states
f1.set_final('next0')
f1.set_final('next1')
f1.set_final('next2')
f1.set_final('next3')
f1.set_final('next4')
f1.set_final('next5')
f1.set_final('next6')
group0 = [
'a', 'e', 'h', 'i', 'o', 'u', 'w', 'y', 'A', 'E', 'H', 'I', 'O', 'U',
'W', 'Y'
]
group1 = ['b', 'f', 'p', 'v', 'B', 'F', 'P', 'V']
group2 = [
'c', 'g', 'j', 'k', 'q', 's', 'x', 'z', 'C', 'G', 'J', 'K', 'Q', 'S',
'X', 'Z'
]
group3 = ['d', 't', 'D', 'T']
group4 = ['l', 'L']
group5 = ['m', 'n', 'M', 'N']
group6 = ['r', 'R']
# Add the rest of the arcs
for letter in string.letters:
if (letter in group0):
f1.add_arc('start', 'next0', (letter), (letter))
f1.add_arc('next', 'next0', (letter), ())
f1.add_arc('next0', 'next0', (letter), ())
f1.add_arc('next1', 'next0', (letter), ())
f1.add_arc('next2', 'next0', (letter), ())
f1.add_arc('next3', 'next0', (letter), ())
f1.add_arc('next4', 'next0', (letter), ())
f1.add_arc('next5', 'next0', (letter), ())
f1.add_arc('next6', 'next0', (letter), ())
elif (letter in group1):
f1.add_arc('start', 'next1', (letter), (letter))
f1.add_arc('next', 'next1', (letter), ('1'))
f1.add_arc('next1', 'next1', (letter), ())
f1.add_arc('next0', 'next1', (letter), ('1'))
f1.add_arc('next2', 'next1', (letter), ('1'))
f1.add_arc('next3', 'next1', (letter), ('1'))
f1.add_arc('next4', 'next1', (letter), ('1'))
f1.add_arc('next5', 'next1', (letter), ('1'))
f1.add_arc('next6', 'next1', (letter), ('1'))
elif (letter in group2):
f1.add_arc('start', 'next2', (letter), (letter))
f1.add_arc('next', 'next2', (letter), ('2'))
f1.add_arc('next2', 'next2', (letter), ())
f1.add_arc('next0', 'next2', (letter), ('2'))
f1.add_arc('next1', 'next2', (letter), ('2'))
f1.add_arc('next3', 'next2', (letter), ('2'))
f1.add_arc('next4', 'next2', (letter), ('2'))
f1.add_arc('next5', 'next2', (letter), ('2'))
f1.add_arc('next6', 'next2', (letter), ('2'))
elif (letter in group3):
f1.add_arc('start', 'next3', (letter), (letter))
f1.add_arc('next', 'next3', (letter), ('3'))
f1.add_arc('next3', 'next3', (letter), ())
f1.add_arc('next0', 'next3', (letter), ('3'))
f1.add_arc('next1', 'next3', (letter), ('3'))
f1.add_arc('next2', 'next3', (letter), ('3'))
f1.add_arc('next4', 'next3', (letter), ('3'))
f1.add_arc('next5', 'next3', (letter), ('3'))
f1.add_arc('next6', 'next3', (letter), ('3'))
elif (letter in group4):
f1.add_arc('start', 'next4', (letter), (letter))
f1.add_arc('next', 'next4', (letter), ('4'))
f1.add_arc('next4', 'next4', (letter), ())
f1.add_arc('next0', 'next4', (letter), ('4'))
f1.add_arc('next1', 'next4', (letter), ('4'))
f1.add_arc('next2', 'next4', (letter), ('4'))
f1.add_arc('next3', 'next4', (letter), ('4'))
f1.add_arc('next5', 'next4', (letter), ('4'))
f1.add_arc('next6', 'next4', (letter), ('4'))
elif (letter in group5):
f1.add_arc('start', 'next5', (letter), (letter))
f1.add_arc('next', 'next5', (letter), ('5'))
f1.add_arc('next5', 'next5', (letter), ())
f1.add_arc('next0', 'next5', (letter), ('5'))
f1.add_arc('next1', 'next5', (letter), ('5'))
f1.add_arc('next2', 'next5', (letter), ('5'))
f1.add_arc('next3', 'next5', (letter), ('5'))
f1.add_arc('next4', 'next5', (letter), ('5'))
f1.add_arc('next6', 'next5', (letter), ('5'))
elif (letter in group6):
f1.add_arc('start', 'next6', (letter), (letter))
f1.add_arc('next', 'next6', (letter), ('6'))
f1.add_arc('next6', 'next6', (letter), ())
f1.add_arc('next0', 'next6', (letter), ('6'))
f1.add_arc('next1', 'next6', (letter), ('6'))
f1.add_arc('next2', 'next6', (letter), ('6'))
f1.add_arc('next3', 'next6', (letter), ('6'))
f1.add_arc('next4', 'next6', (letter), ('6'))
f1.add_arc('next5', 'next6', (letter), ('6'))
return f1
def letters_to_numbers():
"""
Returns an FST that converts letters to numbers as specified by
the soundex algorithm
"""
# Let's define our first FST
f1 = FST('soundex-generate')
# Indicate that '1' is the initial state
f1.add_state('1')
f1.add_state('2')
f1.add_state('3')
f1.add_state('4')
f1.add_state('5')
f1.add_state('6')
f1.add_state('7')
f1.add_state('8')
f1.initial_state = '1'
# Set all the final states
f1.set_final('2')
f1.set_final('3')
f1.set_final('4')
f1.set_final('5')
f1.set_final('6')
f1.set_final('7')
f1.set_final('8')
set1 = {
'a': 1,
'e': 1,
'i': 1,
'o': 1,
'u': 1,
'h': 1,
'w': 1,
'y': 1,
'A': 1,
'E': 1,
'I': 1,
'O': 1,
'U': 1,
'H': 1,
'W': 1,
'Y': 1
}
set2 = {'b': 1, 'f': 1, 'p': 1, 'v': 1, 'B': 1, 'F': 1, 'P': 1, 'V': 1}
set3 = {
'c': 1,
'g': 1,
'j': 1,
'k': 1,
'q': 1,
's': 1,
'x': 1,
'z': 1,
'C': 1,
'G': 1,
'J': 1,
'K': 1,
'Q': 1,
'S': 1,
'X': 1,
'Z': 1
}
set4 = {'d': 1, 't': 1, 'D': 1, 'T': 1}
set5 = {'l': 1, 'L': 1}
set6 = {'m': 1, 'n': 1, 'M': 1, 'N': 1}
set7 = {'r': 1, 'R': 1}
# Add the rest of the arcs
for letter in string.ascii_letters:
if letter in set1:
f1.add_arc('1', '2', (letter), (letter))
f1.add_arc('2', '2', (letter), ())
f1.add_arc('3', '2', (letter), ())
f1.add_arc('4', '2', (letter), ())
f1.add_arc('5', '2', (letter), ())
f1.add_arc('6', '2', (letter), ())
f1.add_arc('7', '2', (letter), ())
f1.add_arc('8', '2', (letter), ())
elif letter in set2:
f1.add_arc('1', '3', (letter), (letter))
f1.add_arc('2', '3', (letter), ('1'))
f1.add_arc('3', '3', (letter), ())
f1.add_arc('4', '3', (letter), ('1'))
f1.add_arc('5', '3', (letter), ('1'))
f1.add_arc('6', '3', (letter), ('1'))
f1.add_arc('7', '3', (letter), ('1'))
f1.add_arc('8', '3', (letter), ('1'))
elif letter in set3:
f1.add_arc('1', '4', (letter), (letter))
f1.add_arc('2', '4', (letter), ('2'))
f1.add_arc('3', '4', (letter), ('2'))
f1.add_arc('4', '4', (letter), ())
f1.add_arc('5', '4', (letter), ('2'))
f1.add_arc('6', '4', (letter), ('2'))
f1.add_arc('7', '4', (letter), ('2'))
f1.add_arc('8', '4', (letter), ('2'))
elif letter in set4:
f1.add_arc('1', '5', (letter), (letter))
f1.add_arc('2', '5', (letter), ('3'))
f1.add_arc('3', '5', (letter), ('3'))
f1.add_arc('4', '5', (letter), ('3'))
f1.add_arc('5', '5', (letter), ())
f1.add_arc('6', '5', (letter), ('3'))
f1.add_arc('7', '5', (letter), ('3'))
f1.add_arc('8', '5', (letter), ('3'))
elif letter in set5:
f1.add_arc('1', '6', (letter), (letter))
f1.add_arc('2', '6', (letter), ('4'))
f1.add_arc('3', '6', (letter), ('4'))
f1.add_arc('4', '6', (letter), ('4'))
f1.add_arc('5', '6', (letter), ('4'))
f1.add_arc('6', '6', (letter), ())
f1.add_arc('7', '6', (letter), ('4'))
f1.add_arc('8', '6', (letter), ('4'))
elif letter in set6:
f1.add_arc('1', '7', (letter), (letter))
f1.add_arc('2', '7', (letter), ('5'))
f1.add_arc('3', '7', (letter), ('5'))
f1.add_arc('4', '7', (letter), ('5'))
f1.add_arc('5', '7', (letter), ('5'))
f1.add_arc('6', '7', (letter), ('5'))
f1.add_arc('8', '7', (letter), ('5'))
elif letter in set7:
f1.add_arc('1', '8', (letter), (letter))
f1.add_arc('2', '8', (letter), ('6'))
f1.add_arc('3', '8', (letter), ('6'))
f1.add_arc('4', '8', (letter), ('6'))
f1.add_arc('5', '8', (letter), ('6'))
f1.add_arc('6', '8', (letter), ('6'))
f1.add_arc('7', '8', (letter), ('6'))
f1.add_arc('8', '8', (letter), ())
return f1
def add_zero_padding():
# Now, the third fst - the zero-padding fst
f3 = FST('soundex-padzero')
f3.add_state('1')
f3.add_state('1a')
f3.add_state('1b')
f3.add_state('2')
f3.initial_state = '1'
# The soundex string will either need no padding in which case its final
# state is 1, or it will need 1 to 3 zeros and have final state 2
f3.set_final('1')
f3.set_final('2')
f3.add_arc('1', '2', (), ('000'))
f3.add_arc('1a', '2', (), ('00'))
f3.add_arc('1b', '2', (), ('0'))
for letter in string.letters:
f3.add_arc('1', '1', (letter), (letter))
for number in xrange(10):
f3.add_arc('1', '1a', (str(number)), (str(number)))
for number in xrange(10):
f3.add_arc('1a', '1b', (str(number)), (str(number)))
for number in xrange(10):
f3.add_arc('1b', '2', (str(number)), (str(number)))
return f3
def letters_to_numbers():
"""
Returns an FST that converts letters to numbers as specified by
the soundex algorithm
"""
# Let's define our first FST
f1 = FST('soundex-generate')
letter_groups = [['b','f','p','v','B','F','P','V'],['c','C', 'g','G','J', 'j', 'K','k','Q', 'q','S', 's','X', 'x', 'Z','z'],['d','D','T','t'],['L','l'],['M','N','m','n'],['R','r']]
vowels = ['a','e','i','o','u','w','y','h','A','E','I','O','U','W','Y','H']
states_num = len(letter_groups)
f1.add_state('start')
f1.add_state('vowels')
f1.set_final('vowels')
for i in range(states_num) :
f1.add_state(i)
f1.set_final(i)
f1.initial_state = 'start'
# Add the rest of the arcs
# f1.add_arc('vowels','start',(),())
for letter in string.ascii_letters:
if letter in vowels :
f1.add_arc('start','vowels',(letter),(letter)) #first char is vowel
f1.add_arc('vowels','vowels',(letter),()) #ignoring consecutive vowels iin start
for i in range(states_num) :
f1.add_arc(i,'vowels',(letter),())
else :
for conso_state in range(states_num):
if letter in letter_groups[conso_state] :
f1.add_arc('start',conso_state,(letter),(letter))
f1.add_arc('vowels',conso_state,(letter),(str(conso_state+1)[0]))
f1.add_arc(conso_state,conso_state,(letter),())
for other_conso_state in range(states_num):
if other_conso_state != conso_state :
f1.add_arc(other_conso_state,conso_state,(letter),(str(conso_state+1)[0]))
return f1
def letters_to_numbers():
"""
Returns an FST that converts letters to numbers as specified by
the soundex algorithm
"""
# Let's define our first FST
f1 = FST('soundex-generate')
char_removal = [
'a', 'e', 'h', 'i', 'o', 'u', 'w', 'y', 'A', 'E', 'I', 'O', 'U', 'W',
'Y'
]
char_replace1 = ['b', 'f', 'p', 'v', 'B', 'F', 'P', 'V']
char_replace2 = [
'c', 'g', 'j', 'k', 'q', 's', 'x', 'z', 'C', 'G', 'J', 'K', 'Q', 'S',
'X', 'Z'
]
char_replace3 = ['d', 't', 'D', 'T']
char_replace4 = ['l', 'L']
char_replace5 = ['m', 'n', 'M', 'N']
char_replace6 = ['r', 'R']
#indicate all the states
f1.add_state('start')
f1.add_state('next')
f1.add_state('rp1')
f1.add_state('rp2')
f1.add_state('rp3')
f1.add_state('rp4')
f1.add_state('rp5')
f1.add_state('rp6')
#indicate the initial and final state
f1.initial_state = 'start'
f1.set_final('next')
f1.set_final('rp1')
f1.set_final('rp2')
f1.set_final('rp3')
f1.set_final('rp4')
f1.set_final('rp5')
f1.set_final('rp6')
#Add the arcs
for letter in char_removal:
f1.add_arc('start', 'next', (letter), (letter))
f1.add_arc('next', 'next', (letter), ())
for letter in char_replace1:
f1.add_arc('start', 'rp1', (letter), (letter))
f1.add_arc('next', 'rp1', (letter), ('1'))
f1.add_arc('rp1', 'rp1', (letter), ())
for letter in char_removal:
f1.add_arc('rp1', 'rp1', (letter), ())
for letter in char_replace2:
f1.add_arc('rp1', 'rp2', (letter), ('2'))
for letter in char_replace3:
f1.add_arc('rp1', 'rp3', (letter), ('3'))
for letter in char_replace4:
f1.add_arc('rp1', 'rp4', (letter), ('4'))
for letter in char_replace5:
f1.add_arc('rp1', 'rp5', (letter), ('5'))
for letter in char_replace6:
f1.add_arc('rp1', 'rp6', (letter), ('6'))
##end of rp1 to all rp6 connections
for letter in char_replace2:
f1.add_arc('start', 'rp2', (letter), (letter))
f1.add_arc('next', 'rp2', (letter), ('2'))
f1.add_arc('rp2', 'rp2', (letter), ())
for letter in char_removal:
f1.add_arc('rp2', 'rp2', (letter), ())
for letter in char_replace1:
f1.add_arc('rp2', 'rp1', (letter), ('1'))
for letter in char_replace3:
f1.add_arc('rp2', 'rp3', (letter), ('3'))
for letter in char_replace4:
f1.add_arc('rp2', 'rp4', (letter), ('4'))
for letter in char_replace5:
f1.add_arc('rp2', 'rp5', (letter), ('5'))
for letter in char_replace6:
f1.add_arc('rp2', 'rp6', (letter), ('6'))
##end of rp2 to all rp6 connections
for letter in char_replace3:
f1.add_arc('start', 'rp3', (letter), (letter))
f1.add_arc('next', 'rp3', (letter), ('3'))
f1.add_arc('rp3', 'rp3', (letter), ())
for letter in char_removal:
f1.add_arc('rp3', 'rp3', (letter), ())
for letter in char_replace1:
f1.add_arc('rp3', 'rp1', (letter), ('1'))
for letter in char_replace2:
f1.add_arc('rp3', 'rp2', (letter), ('2'))
for letter in char_replace4:
f1.add_arc('rp3', 'rp4', (letter), ('4'))
for letter in char_replace5:
f1.add_arc('rp3', 'rp5', (letter), ('5'))
for letter in char_replace6:
f1.add_arc('rp3', 'rp6', (letter), ('6'))
##end of rp3 to all rp6 connections
for letter in char_replace4:
f1.add_arc('start', 'rp4', (letter), (letter))
f1.add_arc('next', 'rp4', (letter), ('4'))
f1.add_arc('rp4', 'rp4', (letter), ())
for letter in char_removal:
f1.add_arc('rp4', 'rp4', (letter), ())
for letter in char_replace1:
f1.add_arc('rp4', 'rp1', (letter), ('1'))
for letter in char_replace2:
f1.add_arc('rp4', 'rp2', (letter), ('2'))
for letter in char_replace3:
f1.add_arc('rp4', 'rp3', (letter), ('3'))
for letter in char_replace5:
f1.add_arc('rp4', 'rp5', (letter), ('5'))
for letter in char_replace6:
f1.add_arc('rp4', 'rp6', (letter), ('6'))
##end of rp4 to all rp6 connections
for letter in char_replace5:
f1.add_arc('start', 'rp5', (letter), (letter))
f1.add_arc('next', 'rp5', (letter), ('5'))
f1.add_arc('rp5', 'rp5', (letter), ())
for letter in char_removal:
f1.add_arc('rp5', 'rp5', (letter), ())
for letter in char_replace1:
f1.add_arc('rp5', 'rp1', (letter), ('1'))
for letter in char_replace2:
f1.add_arc('rp5', 'rp2', (letter), ('2'))
for letter in char_replace3:
f1.add_arc('rp5', 'rp3', (letter), ('3'))
for letter in char_replace4:
f1.add_arc('rp5', 'rp4', (letter), ('4'))
for letter in char_replace6:
f1.add_arc('rp5', 'rp6', (letter), ('6'))
##end of rp5 to all rp6 connections
for letter in char_replace6:
f1.add_arc('start', 'rp6', (letter), (letter))
f1.add_arc('next', 'rp6', (letter), ('6'))
f1.add_arc('rp6', 'rp6', (letter), ())
for letter in char_removal:
f1.add_arc('rp6', 'rp6', (letter), ())
for letter in char_replace1:
f1.add_arc('rp6', 'rp1', (letter), ('1'))
for letter in char_replace2:
f1.add_arc('rp6', 'rp2', (letter), ('2'))
for letter in char_replace3:
f1.add_arc('rp6', 'rp3', (letter), ('3'))
for letter in char_replace4:
f1.add_arc('rp6', 'rp4', (letter), ('4'))
for letter in char_replace5:
f1.add_arc('rp6', 'rp5', (letter), ('5'))
##end of rp6 to all rp6 connections
return f1
def letters_to_numbers():
"""
Returns an FST that converts letters to numbers as specified by
the soundex algorithm
"""
# Let's define our first FST
f1 = FST('soundex-generate')
# Indicate that '1' is the initial state
f1.add_state('start')
f1.add_state('next')
f1.initial_state = 'start'
#create 0,1,2,3,4,5,6 states for all classes
for x in range(0,7):
f1.add_state(str(x))
f1.set_final(str(x))
list_0 = ['a','e','h','i','o','u','w','y']
list_1 = ['b','f','p','v']
list_2 = ['c','g','j','k','q','s','x','z']
list_3 = ['d','t']
list_4 = ['l']
list_5 = ['m','n']
list_6 = ['r']
all_lists = [list_0,list_1,list_2,list_3,list_4,list_5,list_6]
# Set all the final states
for index,item in enumerate(all_lists):
for letter in item:
f1.add_arc('start',str(index),(letter),(letter))
f1.add_arc('start',str(index),(letter.upper()),(letter.upper()))
for x in range(0,7):
for index,item in enumerate(all_lists):
for letter in item:
if x == index:
f1.add_arc(str(x),str(index),(letter),())
f1.add_arc(str(x),str(index),(letter.upper()),())
elif index == 0:
f1.add_arc(str(x),str(index),(letter),())
f1.add_arc(str(x),str(index),(letter.upper()),())
else:
f1.add_arc(str(x),str(index),(letter),(str(index)))
f1.add_arc(str(x),str(index),(letter.upper()),(str(index)))
return f1
def add_zero_padding():
# Now, the third fst - the zero-padding fst
f3 = FST('soundex-padzero')
f3.add_state('1')
f3.add_state('1a')
f3.add_state('1b')
f3.add_state('2')
f3.initial_state = '1'
f3.set_final('2')
for letter in string.letters:
f3.add_arc('1', '1', (letter), (letter))
f3.add_arc('1', '1a', (), ('0'))
for number in xrange(10):
f3.add_arc('1', '1a', (str(number)), (str(number)))
f3.add_arc('1a', '1b', (), ('0'))
for number in xrange(10):
f3.add_arc('1a', '1b', (str(number)), (str(number)))
f3.add_arc('1b', '2', (), ('0'))
for number in xrange(10):
f3.add_arc('1b', '2', (str(number)), (str(number)))
return f3
def add_zero_padding():
# Now, the third fst - the zero-padding fst
f3 = FST('soundex-padzero')
f3.add_state('start')
f3.initial_state = 'start'
for x in range(4):
f3.add_state(str(x))
f3.set_final(str(3))
# Add the arcs
f3.add_arc(str(0),str(1),(''),('0'))
f3.add_arc('start','1',(''),('0'))
f3.add_arc(str(1),str(2),(''),('0'))
f3.add_arc(str(2),str(3),(''),('0'))
for letter in string.letters:
f3.add_arc('start', '0', (letter), (letter))
for n in range(10):
f3.add_arc('start','1',(str(n)),(str(n)))
for x in range(3):
for n in range(10):
f3.add_arc(str(x), str(x+1), (str(n)), (str(n)))
for n in range(10):
f3.add_arc(str(3),str(3),(str(n)),())
'''
f3.add_state('1')
f3.add_state('1a')
f3.add_state('1b')
f3.add_state('2')
f3.initial_state = '1'
f3.set_final('2')
for letter in string.letters:
f3.add_arc('1', '1', (letter), (letter))
for number in xrange(10):
f3.add_arc('1', '1', (str(number)), (str(number)))
f3.add_arc('1', '1a', (), ('0'))
f3.add_arc('1a', '1b', (), ('0'))
f3.add_arc('1b', '2', (), ('0'))
'''
return f3
def generate(self, analysis):
"""Generate the morphologically correct word
e.g.
p = Parser()
analysis = ['p','a','n','i','c','+past form']
p.generate(analysis)
---> 'panicked'
"""
# Let's define our first FST
f1 = FST('morphology-generate')
# Indicate initial and final states
f1.add_state('start')
f1.add_state('vowel')
f1.add_state('consonant')
f1.add_state('c')
f1.add_state('form_1')
f1.add_state('form_2')
f1.initial_state = 'start'
f1.set_final('form_1')
f1.set_final('form_2')
# Generate
vowels = 'aeiou'
for vowel in vowels:
f1.add_arc('start', 'vowel', vowel, vowel)
f1.add_arc('vowel', 'vowel', vowel, vowel)
f1.add_arc('consonant', 'vowel', vowel, vowel)
f1.add_arc('c', 'vowel', vowel, vowel)
for letter in string.ascii_lowercase:
if letter in vowels:
continue
if not letter == 'c':
f1.add_arc('vowel', 'consonant', letter, letter)
f1.add_arc('start', 'consonant', letter, letter)
f1.add_arc('consonant', 'consonant', letter, letter)
f1.add_arc('c', 'consonant', letter, letter)
f1.add_arc('vowel', 'c', 'c', 'c')
f1.add_arc('c', 'form_1', '+past form', 'ked')
f1.add_arc('c', 'form_1', '+present participle form', 'king')
f1.add_arc('consonant', 'form_2', '+past form', 'ed')
f1.add_arc('consonant', 'form_2', '+present participle form', 'ing')
output = f1.transduce(analysis)[0]
return "".join(output)
def letters_to_numbers():
"""
Returns an FST that converts letters to numbers as specified by
the soundex algorithm
"""
vowels = 'aeiouwhyAEIOUWHY'
q1 = 'BFPVbfpv'
q2 = 'CGJKQSXZcgjkqsxz'
q3 = 'dtDT'
q4 = 'lL'
q5 = 'mnMN'
q6 = 'rR'
# Let's define our first FST
f1 = FST('soundex-generate')
# Indicate that '1' is the initial state
'''f1.add_state('start')
f1.add_state('next')
f1.initial_state = 'start'
# Set all the final states
f1.set_final('next')
# Add the rest of the arcs
for letter in string.ascii_lowercase:
f1.add_arc('start', 'next', (letter), (letter))
f1.add_arc('next', 'next', (letter), ('0'))
return f1'''
f1.add_state('0')
f1.add_state('1')
f1.add_state('v')
f1.add_state('q1')
f1.add_state('q2')
f1.add_state('q3')
f1.add_state('q4')
f1.add_state('q5')
f1.add_state('q6')
f1.initial_state = '0'
f1.set_final('1')
f1.set_final('v')
f1.set_final('q1')
f1.set_final('q2')
f1.set_final('q3')
f1.set_final('q4')
f1.set_final('q5')
f1.set_final('q6')
for letter in string.ascii_letters:
f1.add_arc('0', '1', (letter), (letter))
if letter in vowels:
f1.add_arc('1', 'v', (letter), ())
f1.add_arc('v', 'v', (letter), ())
f1.add_arc('q1', 'v', (letter), ())
f1.add_arc('q2', 'v', (letter), ())
f1.add_arc('q3', 'v', (letter), ())
f1.add_arc('q4', 'v', (letter), ())
f1.add_arc('q5', 'v', (letter), ())
f1.add_arc('q6', 'v', (letter), ())
if letter in q1:
f1.add_arc('1', 'q1', (letter), ('1'))
f1.add_arc('v', 'q1', (letter), ('1'))
f1.add_arc('q1','q1', (letter), ())
f1.add_arc('q2','q1', (letter), ('1'))
f1.add_arc('q3', 'q1', (letter), ('1'))
f1.add_arc('q4', 'q1', (letter), ('1'))
f1.add_arc('q5', 'q1', (letter), ('1'))
f1.add_arc('q6', 'q1', (letter), ('1'))
if letter in q2:
f1.add_arc('1', 'q2', (letter), ('2'))
f1.add_arc('v', 'q2', (letter), ('2'))
f1.add_arc('q1','q2', (letter), ('2'))
f1.add_arc('q2','q2', (letter), ())
f1.add_arc('q3', 'q2', (letter), ('2'))
f1.add_arc('q4', 'q2', (letter), ('2'))
f1.add_arc('q5', 'q2', (letter), ('2'))
f1.add_arc('q6', 'q2', (letter), ('2'))
if letter in q3:
f1.add_arc('1', 'q3', (letter), ('3'))
f1.add_arc('v', 'q3', (letter), ('3'))
f1.add_arc('q1','q3', (letter), ('3'))
f1.add_arc('q2','q3', (letter), ('3'))
f1.add_arc('q3', 'q3', (letter), ())
f1.add_arc('q4', 'q3', (letter), ('3'))
f1.add_arc('q5', 'q3', (letter),('3'))
f1.add_arc('q6', 'q3', (letter), ('3'))
if letter in q4:
f1.add_arc('1', 'q4', (letter), ('4'))
f1.add_arc('v', 'q4', (letter), ('4'))
f1.add_arc('q1','q4', (letter), ('4'))
f1.add_arc('q2','q4', (letter), ('4'))
f1.add_arc('q3', 'q4', (letter), ('4'))
f1.add_arc('q4', 'q4', (letter), ())
f1.add_arc('q5', 'q4', (letter), ('4'))
f1.add_arc('q6', 'q4', (letter), ('4'))
if letter in q5:
f1.add_arc('1', 'q5', (letter), ('5'))
f1.add_arc('v', 'q5', (letter), ('5'))
f1.add_arc('q1','q5', (letter), ('5'))
f1.add_arc('q2','q5', (letter), ('5'))
f1.add_arc('q3','q5', (letter), ('5'))
f1.add_arc('q4','q5', (letter), ('5'))
f1.add_arc('q5', 'q5', (letter), ())
f1.add_arc('q6', 'q5', (letter), ('5'))
if letter in q6:
f1.add_arc('1', 'q6', (letter), ('6'))
f1.add_arc('v', 'q6', (letter), ('6'))
f1.add_arc('q1','q6', (letter), ('6'))
f1.add_arc('q2','q6', (letter), ('6'))
f1.add_arc('q3', 'q6', (letter), ('6'))
f1.add_arc('q4', 'q6', (letter), ('6'))
f1.add_arc('q5', 'q6', (letter), ('6'))
f1.add_arc('q6', 'q6', (letter), ())
return f1
if __name__ == '__main__':
f1 = FST('test-generate')
# Indicate that '1' is the initial state
f1.add_state('start')
f1.add_state('next')
f1.initial_state = 'start'
# Set all the final states
f1.set_final('next')
# Add the rest of the arcs
for letter in ['A', 'B', 'C', 'D']:
f1.add_arc('start', 'next', letter, '1')
f1.add_arc('next', 'next', letter, '0')
f2 = FST('test-generate')
f2.add_state('start')
f2.add_state('next')
f2.initial_state = 'start'
f2.set_final('next')
f2.add_arc('start', 'next', '1', 'a')
f2.add_arc('start', 'next', '1', 'an')
f2.add_arc('next', 'next', '0', 'b')
output = compose(tuple('BAD'), f1, f2)
print output
for o in output:
def french_count():
f = FST('french')
f.add_state('start')
f.add_state('final')
f.add_state('0XX')
f.add_state('00X')
f.add_state('01X')
f.add_state('XX')
f.add_state('7X')
f.add_state('8X')
f.add_state('9X')
f.add_state('NXX')
f.add_state('N0X')
f.initial_state = 'start'
f.set_final('final')
# single digit numbers
for ii in xrange(10):
if ii == 0:
f.add_arc('start', '0XX', [str(ii)], ())
f.add_arc('0XX', '00X', [str(ii)], ())
f.add_arc('00X', 'final', [str(ii)], [kFRENCH_TRANS[ii]])
# 10-19
f.add_arc('0XX', '01X', [str(1)], ())
for ii in xrange(10):
if ii < 7:
f.add_arc('01X', 'final', [str(ii)], [kFRENCH_TRANS[ii+10]])
else:
f.add_arc('01X', 'final', [str(ii)], [kFRENCH_TRANS[10],kFRENCH_TRANS[ii]])
# 20-69
for ii in xrange(2,7):
f.add_arc('0XX', 'XX', [str(ii)], [kFRENCH_TRANS[ii*10]])
# dont add anything that ends in 0...e.g. 20, 30, 40, 50, 60
f.add_arc('XX', 'final', [str(0)], ())
# add "and" to numbers if number is 21,31,41,51,61
f.add_arc('XX', 'final', [str(1)], [kFRENCH_AND,kFRENCH_TRANS[1]])
for ii in xrange(2,10):
f.add_arc('XX', 'final', [str(ii)], [kFRENCH_TRANS[ii]])
# 70s
f.add_arc('0XX', '7X', [str(7)], [kFRENCH_TRANS[60]])
for ii in xrange(10):
if ii == 0:
f.add_arc('7X', 'final', [str(ii)], [kFRENCH_TRANS[10]])
elif ii == 1:
f.add_arc('7X', 'final', [str(ii)], [kFRENCH_AND,kFRENCH_TRANS[11]])
elif 2 <= ii <= 6:
f.add_arc('7X', 'final', [str(ii)], [kFRENCH_TRANS[ii+10]])
else:
f.add_arc('7X', 'final', [str(ii)], [kFRENCH_TRANS[10],kFRENCH_TRANS[ii]])
# 80s
f.add_arc('0XX', '8X', [str(8)], [kFRENCH_TRANS[4],kFRENCH_TRANS[20]])
for ii in xrange(10):
if ii == 0:
f.add_arc('8X', 'final', [str(ii)], ())
else:
f.add_arc('8X', 'final', [str(ii)], [kFRENCH_TRANS[ii]])
# 90s
f.add_arc('0XX', '9X', [str(9)], [kFRENCH_TRANS[4],kFRENCH_TRANS[20]])
for ii in xrange(10):
if ii < 7:
f.add_arc('9X', 'final', [str(ii)], [kFRENCH_TRANS[ii+10]])
else:
f.add_arc('9X', 'final', [str(ii)], [kFRENCH_TRANS[10],kFRENCH_TRANS[ii]])
# X00-X09
for ii in xrange(1,10):
if ii == 1:
f.add_arc('start', 'NXX', [str(ii)], [kFRENCH_TRANS[100]])
else:
f.add_arc('start', 'NXX', [str(ii)], [kFRENCH_TRANS[ii],kFRENCH_TRANS[100]])
f.add_arc('NXX', 'N0X', [str(0)], ())
for ii in xrange(10):
if ii == 0:
f.add_arc('N0X', 'final', [str(0)], ())
else:
f.add_arc('N0X', 'final', [str(ii)], [kFRENCH_TRANS[ii]])
# X10-X19
f.add_arc('NXX', '01X', [str(1)], ())
# X20-X69
for ii in xrange(2,7):
f.add_arc('NXX', 'XX', [str(ii)], [kFRENCH_TRANS[ii*10]])
# X70s, X80s, X90s
f.add_arc('NXX', '7X', [str(7)], [kFRENCH_TRANS[60]])
f.add_arc('NXX', '8X', [str(8)], [kFRENCH_TRANS[4],kFRENCH_TRANS[20]])
f.add_arc('NXX', '9X', [str(9)], [kFRENCH_TRANS[4],kFRENCH_TRANS[20]])
return f
def french_count():
f = FST('french')
f.add_state('0')
f.add_state('1')
f.add_state('2')
f.add_state('3')
f.add_state('4')
f.add_state('5')
f.add_state('6')
f.add_state('7')
f.add_state('8')
f.add_state('9')
f.add_state('10')
f.add_state('11')
f.add_state('12')
f.add_state('13')
f.add_state('14')
f.add_state('15')
f.add_state('16')
f.add_state('17')
f.add_state('18')
f.add_state('19')
f.add_state('20')
f.add_state('21')
f.add_state('22')
f.add_state('23')
f.add_state('24')
f.add_state('25')
f.initial_state = '0'
f.set_final('1')
f.set_final('3')
f.set_final('6')
f.set_final('7')
f.set_final('8')
f.set_final('9')
f.set_final('11')
f.set_final('13')
f.set_final('14')
f.set_final('18')
f.set_final('20')
zero = [0]
one = [1]
two_to_six = [2,3,4,5,6]
one_to_six = [1,2,3,4,5,6]
seven = [7]
seven_eight_nine = [7,8,9]
eight = [8]
nine = [9]
singles_all = [1,2,3,4,5,6,7,8,9]
singles = [2,3,4,5,6,7,8,9]
tens = [20,30,40,50]
# Edge from initial to final, if preceding zero in input
for i in zero:
# f.add_arc('0','9', str(i), [kFRENCH_TRANS[i]])
f.add_arc('0','0', str(i), ())
f.add_arc('4','6', str(i), ())
f.add_arc('5','8', str(i), ())
f.add_arc('0','9', str(i), [kFRENCH_TRANS[i]])
f.add_arc('10','11', str(i), [kFRENCH_TRANS[i+10]])
f.add_arc('12','13', str(i), [kFRENCH_TRANS[20]])
f.add_arc('16','18', str(i), [kFRENCH_TRANS[20],kFRENCH_TRANS[10]])
f.add_arc('17','19', str(i), ())
f.add_arc('19','9', str(i), ())
for i in one:
f.add_arc('0','2', str(i), ())
f.add_arc('17','2', str(i), ())
f.add_arc('0','17', str(i), [kFRENCH_TRANS[100]])
f.add_arc('0','5', str(i), [kFRENCH_TRANS[i*10]])
f.add_arc('17','5', str(i), [kFRENCH_TRANS[i*10]])
f.add_arc('4','7', str(i), [kFRENCH_AND, kFRENCH_TRANS[i]])
f.add_arc('10','11', str(i), [kFRENCH_AND, kFRENCH_TRANS[i+10]])
f.add_arc('12','14', str(i), [kFRENCH_TRANS[20], kFRENCH_AND, kFRENCH_TRANS[i]])
f.add_arc('16','20', str(i), [kFRENCH_TRANS[20], kFRENCH_AND, kFRENCH_TRANS[i+10]])
for i in one_to_six:
f.add_arc('2','3', str(i), [kFRENCH_TRANS[i+10]])
for i in two_to_six:
f.add_arc('0','4', str(i), [kFRENCH_TRANS[i*10]])
f.add_arc('17','4', str(i), [kFRENCH_TRANS[i*10]])
f.add_arc('10','11', str(i), [kFRENCH_TRANS[i+10]])
f.add_arc('16','20', str(i), [kFRENCH_TRANS[20],kFRENCH_TRANS[i+10]])
for i in singles:
f.add_arc('4','7', str(i), [kFRENCH_TRANS[i]])
f.add_arc('0','17', str(i), [kFRENCH_TRANS[i],kFRENCH_TRANS[100]])
f.add_arc('12','14', str(i), [kFRENCH_TRANS[20], kFRENCH_TRANS[i]])
for i in singles_all:
f.add_arc('0','1', str(i), [kFRENCH_TRANS[i]])
f.add_arc('19','1', str(i), [kFRENCH_TRANS[i]])
for i in seven_eight_nine:
f.add_arc('5','8', str(i), [kFRENCH_TRANS[i]])
f.add_arc('10','11', str(i), [kFRENCH_TRANS[10], kFRENCH_TRANS[i]])
f.add_arc('16','20', str(i), [kFRENCH_TRANS[20], kFRENCH_TRANS[10], kFRENCH_TRANS[i]])
for i in seven:
f.add_arc('0','10',str(i), [kFRENCH_TRANS[60]])
f.add_arc('17','10',str(i), [kFRENCH_TRANS[60]])
for i in eight:
f.add_arc('0','12',str(i), [kFRENCH_TRANS[4]])
f.add_arc('17','12',str(i), [kFRENCH_TRANS[4]])
for i in nine:
f.add_arc('0','16',str(i), [kFRENCH_TRANS[4]])
f.add_arc('17','16',str(i), [kFRENCH_TRANS[4]])
return f
return output_list
if __name__ == '__main__':
f1 = FST('test-generate')
# Indicate that '1' is the initial state
f1.add_state('start')
f1.add_state('next')
f1.initial_state = 'start'
# Set all the final states
f1.set_final('next')
# Add the rest of the arcs
for letter in ['A','B','C','D']:
f1.add_arc('start', 'next', letter, '1')
f1.add_arc('next', 'next', letter, '0')
f2 = FST('test-generate')
f2.add_state('start')
f2.add_state('next')
f2.initial_state = 'start'
f2.set_final('next')
f2.add_arc('start', 'next', '1', 'a')
f2.add_arc('start', 'next', '1', 'an')
f2.add_arc('next', 'next', '0', 'b')
output = compose(tuple('BAD'), f1, f2)
print output
for o in output:
def french_count():
f = FST('french')
f.add_state('start')
f.initial_state = 'start'
f.add_state('1stzero')
f.add_state('tens')
f.add_state('seventeen')
f.add_state('final_seventeen')
f.add_state('eighteen')
f.add_state('final_eighteen')
f.add_state('nineteen')
f.add_state('final_nineteen')
f.add_state('zero')
f.add_state('ones')
f.add_state('20-69')
f.add_state('70-ten')
f.add_state('80s')
f.add_state('90s')
f.add_state('100s')
f.add_state('et')
f.add_state('10-et')
f.add_state('et-un')
f.add_state('et-onze')
f.set_final('zero')
f.set_final('ones')
f.set_final('tens')
f.set_final('final_seventeen')
f.set_final('final_eighteen')
f.set_final('final_nineteen')
f.set_final('20-69')
f.set_final('70-ten')
f.set_final('80s')
f.set_final('90s')
f.set_final('et-un')
f.set_final('et-onze')
# 100 - 999
f.add_arc('start', '1stzero', '1', [kFRENCH_TRANS[100]])
for i in range(2, 10):
f.add_arc('start', '100s', str(i), [kFRENCH_TRANS[i]])
f.add_arc('100s', '1stzero', (), [kFRENCH_TRANS[100]])
# 0 - 9
f.add_arc('start', '1stzero', '0', [])
f.add_arc('1stzero', 'ones', '0', [])
for ii in range(1, 10):
f.add_arc('ones', 'ones', str(ii), [kFRENCH_TRANS[ii]])
f.add_arc('ones', 'ones', '0', [])
# for i in range(10):
# f.add_arc('ten-6', 'ten-6', str(i), kFRENCH_TRANS[(i+10])
# 10 - 16
f.add_arc('1stzero', 'tens', '1', [])
f.add_arc('tens', 'tens', '0', [kFRENCH_TRANS[10]])
f.add_arc('tens', 'tens', '1', [kFRENCH_TRANS[11]])
f.add_arc('tens', 'tens', '2', [kFRENCH_TRANS[12]])
f.add_arc('tens', 'tens', '3', [kFRENCH_TRANS[13]])
f.add_arc('tens', 'tens', '4', [kFRENCH_TRANS[14]])
f.add_arc('tens', 'tens', '5', [kFRENCH_TRANS[15]])
f.add_arc('tens', 'tens', '6', [kFRENCH_TRANS[16]])
f.add_arc('tens', 'seventeen', '7', [kFRENCH_TRANS[10]])
f.add_arc('seventeen', 'final_seventeen', (), [kFRENCH_TRANS[7]])
f.add_arc('tens', 'eighteen', '8', [kFRENCH_TRANS[10]])
f.add_arc('eighteen', 'final_eighteen', (), [kFRENCH_TRANS[8]])
f.add_arc('tens', 'nineteen', '9', [kFRENCH_TRANS[10]])
f.add_arc('nineteen', 'final_nineteen', (), [kFRENCH_TRANS[9]])
# 20 - 69
f.add_arc('1stzero', '20-69', '2', [kFRENCH_TRANS[20]])
f.add_arc('1stzero', '20-69', '3', [kFRENCH_TRANS[30]])
f.add_arc('1stzero', '20-69', '4', [kFRENCH_TRANS[40]])
f.add_arc('1stzero', '20-69', '5', [kFRENCH_TRANS[50]])
f.add_arc('1stzero', '20-69', '6', [kFRENCH_TRANS[60]])
# special cases:
for i in range(2, 10):
f.add_arc('20-69', '20-69', str(i), [kFRENCH_TRANS[i]])
# handles 20, 30 ... 60
for i in range(20, 60, 10):
f.add_arc('20-69', '20-69', '0', [])
# handles 21, 31, ... 61
f.add_arc('20-69', 'et', '1', [kFRENCH_AND])
f.add_arc('et', 'et-un', (),[kFRENCH_TRANS[1]])
# 70 - 79
f.add_arc('1stzero', '70-ten', '7', [kFRENCH_TRANS[60]])
f.add_arc('70-ten', '70-ten', '0', [kFRENCH_TRANS[10]])
# handle 71 here
f.add_arc('70-ten', '10-et', '1', [kFRENCH_AND])
f.add_arc('10-et', 'et-onze', (),[kFRENCH_TRANS[11]])
f.add_arc('70-ten', '70-ten', '2', [kFRENCH_TRANS[12]])
f.add_arc('70-ten', '70-ten', '3', [kFRENCH_TRANS[13]])
f.add_arc('70-ten', '70-ten', '4', [kFRENCH_TRANS[14]])
f.add_arc('70-ten', '70-ten', '5', [kFRENCH_TRANS[15]])
f.add_arc('70-ten', '70-ten', '6', [kFRENCH_TRANS[16]])
f.add_arc('70-ten', 'seventeen', '7', [kFRENCH_TRANS[10]])
f.add_arc('seventeen', 'final_seventeen', (), [kFRENCH_TRANS[7]])
f.add_arc('70-ten', 'eighteen', '8', [kFRENCH_TRANS[10]])
f.add_arc('eighteen', 'final_eighteen', (), [kFRENCH_TRANS[8]])
f.add_arc('70-ten', 'nineteen', '9', [kFRENCH_TRANS[10]])
f.add_arc('nineteen', 'final_nineteen', (), [kFRENCH_TRANS[9]])
# 80 - 89
f.add_arc('1stzero', '80s', '8', [kFRENCH_TRANS[4]])
f.add_arc('80s', 'ones', (), [kFRENCH_TRANS[20]])
f.add_arc('80s', '80s', '0', [kFRENCH_TRANS[20]])
# 90 - 99
f.add_arc('1stzero', '90s', '9', [kFRENCH_TRANS[4]])
f.add_arc('90s', 'tens', (), [kFRENCH_TRANS[20]])
return f
def letters_to_numbers():
"""
Returns an FST that converts letters to numbers as specified by
the soundex algorithm
"""
# Let's define our first FST
f1 = FST('soundex-generate')
aeoy = ['a','e','h','i','o','u','w','y']
one = ['b','f','p','v']
two = ['c','g','j','k','q','s','x','z']
three = ['d','t']
four = ['l']
five = ['m','n']
six = ['r']
# Indicate that '1' is the initial state
f1.add_state('initial')
f1.add_state('0')
f1.add_state('1')
f1.add_state('2')
f1.add_state('3')
f1.add_state('4')
f1.add_state('5')
f1.add_state('6')
f1.initial_state = 'initial'
# Set all the final states
f1.set_final('0')
f1.set_final('1')
f1.set_final('2')
f1.set_final('3')
f1.set_final('4')
f1.set_final('5')
f1.set_final('6')
# Add the rest of the arcs
for letter in string.ascii_letters:
f1.add_arc('initial','0',(letter),(letter))
if letter in aeoy:
f1.add_arc('0','0', (letter), ())
f1.add_arc('1','0', (letter), ())
f1.add_arc('2','0', (letter), ())
f1.add_arc('3','0', (letter), ())
f1.add_arc('4','0', (letter), ())
f1.add_arc('5','0', (letter), ())
f1.add_arc('6','0', (letter), ())
else:
if letter in one:
f1.add_arc('0','1', (letter), '1')
f1.add_arc('2','1', (letter), '1')
f1.add_arc('3','1', (letter), '1')
f1.add_arc('4','1', (letter), '1')
f1.add_arc('5','1', (letter), '1')
f1.add_arc('6','1', (letter), '1')
f1.add_arc('1','0', (letter), ())
if letter in two:
f1.add_arc('0','2', (letter), '2')
f1.add_arc('1','2', (letter), '2')
f1.add_arc('3','2', (letter), '2')
f1.add_arc('4','2', (letter), '2')
f1.add_arc('5','2', (letter), '2')
f1.add_arc('6','2', (letter), '2')
f1.add_arc('2','0', (letter), ())
if letter in three:
f1.add_arc('0','3', (letter), '3')
f1.add_arc('1','3', (letter), '3')
f1.add_arc('2','3', (letter), '3')
f1.add_arc('4','3', (letter), '3')
f1.add_arc('5','3', (letter), '3')
f1.add_arc('6','3', (letter), '3')
f1.add_arc('3','0', (letter), ())
if letter in four:
f1.add_arc('0','4', (letter), '4')
f1.add_arc('1','4', (letter), '4')
f1.add_arc('2','4', (letter), '4')
f1.add_arc('3','4', (letter), '4')
f1.add_arc('5','4', (letter), '4')
f1.add_arc('6','4', (letter), '4')
f1.add_arc('4','0', (letter), ())
if letter in five:
f1.add_arc('0','5', (letter), '5')
f1.add_arc('1','5', (letter), '5')
f1.add_arc('2','5', (letter), '5')
f1.add_arc('3','5', (letter), '5')
f1.add_arc('4','5', (letter), '5')
f1.add_arc('6','5', (letter), '5')
f1.add_arc('5','0', (letter), ())
if letter in six:
f1.add_arc('0','6', (letter), '6')
f1.add_arc('1','6', (letter), '6')
f1.add_arc('2','6', (letter), '6')
f1.add_arc('3','6', (letter), '6')
f1.add_arc('4','6', (letter), '6')
f1.add_arc('5','6', (letter), '6')
f1.add_arc('6','0', (letter), ())
return f1
def letters_to_numbers():
"""
Returns an FST that converts letters to numbers as specified by
the soundex algorithm
"""
# Let's define our first FST
f1 = FST('soundex-generate')
# Indicate that '1' is the initial state
f1.add_state('start')
f1.add_state('1')
f1.add_state('2')
f1.add_state('3')
f1.add_state('4')
f1.add_state('5')
f1.add_state('6')
f1.add_state('v')
f1.initial_state = 'start'
# Set all the final states
f1.set_final('1')
f1.set_final('2')
f1.set_final('3')
f1.set_final('4')
f1.set_final('5')
f1.set_final('6')
f1.set_final('v')
# Add the rest of the arcs
#have loops for each of these lists
list1 = ['b', 'f', 'p', 'v', 'B', 'F', 'P', 'V']
list2 = [
'c', 'g', 'j', 'k', 'q', 's', 'x', 'z', 'C', 'G', 'J', 'K', 'Q', 'S',
'X', 'Z'
]
list3 = ['d', 't', 'D', 'T']
list4 = ['l', 'L']
list5 = ['m', 'n', 'M', 'N']
list6 = ['r', 'R']
listv = [
'a', 'e', 'i', 'o', 'u', 'h', 'w', 'y', 'A', 'E', 'I', 'O', 'U', 'H',
'W', 'Y'
]
for letter in list1:
f1.add_arc('start', '1', (letter), (letter))
f1.add_arc('1', '1', (letter), ())
f1.add_arc('2', '1', (letter), '1')
f1.add_arc('3', '1', (letter), '1')
f1.add_arc('4', '1', (letter), '1')
f1.add_arc('5', '1', (letter), '1')
f1.add_arc('6', '1', (letter), '1')
f1.add_arc('v', '1', (letter), '1')
for letter in list2:
f1.add_arc('start', '2', (letter), (letter))
f1.add_arc('1', '2', (letter), '2')
f1.add_arc('2', '2', (letter), ())
f1.add_arc('3', '2', (letter), '2')
f1.add_arc('4', '2', (letter), '2')
f1.add_arc('5', '2', (letter), '2')
f1.add_arc('6', '2', (letter), '2')
f1.add_arc('v', '2', (letter), '2')
for letter in list3:
f1.add_arc('start', '3', (letter), (letter))
f1.add_arc('1', '3', (letter), '3')
f1.add_arc('2', '3', (letter), '3')
f1.add_arc('3', '3', (letter), ())
f1.add_arc('4', '3', (letter), '3')
f1.add_arc('5', '3', (letter), '3')
f1.add_arc('6', '3', (letter), '3')
f1.add_arc('v', '3', (letter), '3')
for letter in list3:
f1.add_arc('start', '3', (letter), (letter))
f1.add_arc('1', '3', (letter), '3')
f1.add_arc('2', '3', (letter), '3')
f1.add_arc('3', '3', (letter), ())
f1.add_arc('4', '3', (letter), '3')
f1.add_arc('5', '3', (letter), '3')
f1.add_arc('6', '3', (letter), '3')
f1.add_arc('v', '3', (letter), '3')
for letter in list4:
f1.add_arc('start', '4', (letter), (letter))
f1.add_arc('1', '4', (letter), '4')
f1.add_arc('2', '4', (letter), '4')
f1.add_arc('3', '4', (letter), '4')
f1.add_arc('4', '4', (letter), ())
f1.add_arc('5', '4', (letter), '4')
f1.add_arc('6', '4', (letter), '4')
f1.add_arc('v', '4', (letter), '4')
for letter in list5:
f1.add_arc('start', '5', (letter), (letter))
f1.add_arc('1', '5', (letter), '5')
f1.add_arc('2', '5', (letter), '5')
f1.add_arc('3', '5', (letter), '5')
f1.add_arc('4', '5', (letter), '5')
f1.add_arc('5', '5', (letter), ())
f1.add_arc('6', '5', (letter), '5')
f1.add_arc('v', '5', (letter), '5')
for letter in list6:
f1.add_arc('start', '6', (letter), (letter))
f1.add_arc('1', '6', (letter), '6')
f1.add_arc('2', '6', (letter), '6')
f1.add_arc('3', '6', (letter), '6')
f1.add_arc('4', '6', (letter), '6')
f1.add_arc('5', '6', (letter), '6')
f1.add_arc('6', '6', (letter), ())
f1.add_arc('v', '6', (letter), '6')
for letter in listv:
f1.add_arc('start', 'v', (letter), (letter))
f1.add_arc('1', 'v', (letter), ())
f1.add_arc('2', 'v', (letter), ())
f1.add_arc('3', 'v', (letter), ())
f1.add_arc('4', 'v', (letter), ())
f1.add_arc('5', 'v', (letter), ())
f1.add_arc('6', 'v', (letter), ())
f1.add_arc('v', 'v', (letter), ())
return f1
def letters_to_numbers():
"""
Returns an FST that converts letters to numbers as specified by
the soundex algorithm
"""
# Let's define our first FST
f1 = FST('soundex-generate')
# Add all states
f1.add_state('0')
f1.add_state('1')
f1.add_state('2')
f1.add_state('3')
f1.add_state('4')
f1.add_state('5')
f1.add_state('6')
f1.add_state('7')
f1.add_state('2a')
f1.add_state('3a')
f1.add_state('4a')
f1.add_state('5a')
f1.add_state('6a')
f1.add_state('7a')
# Indicate that '0' is the initial state
f1.initial_state = '0'
# Set all the final states
f1.set_final('1')
f1.set_final('2')
f1.set_final('3')
f1.set_final('4')
f1.set_final('5')
f1.set_final('6')
f1.set_final('7')
f1.set_final('2a')
f1.set_final('3a')
f1.set_final('4a')
f1.set_final('5a')
f1.set_final('6a')
f1.set_final('7a')
# Define all lists
remove_letters = [
'a', 'e', 'h', 'i', 'o', 'u', 'w', 'y', 'A', 'E', 'H', 'I', 'O', 'U',
'W', 'Y'
]
g1 = ['b', 'f', 'p', 'v', 'B', 'F', 'P', 'V']
g2 = [
'c', 'g', 'j', 'k', 'q', 's', 'x', 'z', 'C', 'G', 'J', 'K', 'Q', 'S',
'X', 'Z'
]
g3 = ['d', 't', 'D', 'T']
g4 = ['l', 'L']
g5 = ['m', 'n', 'M', 'N']
g6 = ['r', 'R']
# Add the rest of the arcs
for letter in string.ascii_letters:
# Retain the first character
if letter in g1:
f1.add_arc('0', '2a', (letter), (letter))
f1.add_arc('2a', '2', (letter), ())
f1.add_arc('3a', '2', (letter), ('1'))
f1.add_arc('4a', '2', (letter), ('1'))
f1.add_arc('5a', '2', (letter), ('1'))
f1.add_arc('6a', '2', (letter), ('1'))
f1.add_arc('7a', '2', (letter), ('1'))
f1.add_arc('1', '2', (letter), ('1'))
f1.add_arc('2', '2', (letter), ())
f1.add_arc('3', '2', (letter), ('1'))
f1.add_arc('4', '2', (letter), ('1'))
f1.add_arc('5', '2', (letter), ('1'))
f1.add_arc('6', '2', (letter), ('1'))
f1.add_arc('7', '2', (letter), ('1'))
if letter in g2:
f1.add_arc('0', '3a', (letter), (letter))
f1.add_arc('3a', '3', (letter), ())
f1.add_arc('2a', '3', (letter), ('2'))
f1.add_arc('4a', '3', (letter), ('2'))
f1.add_arc('5a', '3', (letter), ('2'))
f1.add_arc('6a', '3', (letter), ('2'))
f1.add_arc('7a', '3', (letter), ('2'))
f1.add_arc('1', '3', (letter), ('2'))
f1.add_arc('3', '3', (letter), ())
f1.add_arc('2', '3', (letter), ('2'))
f1.add_arc('4', '3', (letter), ('2'))
f1.add_arc('5', '3', (letter), ('2'))
f1.add_arc('6', '3', (letter), ('2'))
f1.add_arc('7', '3', (letter), ('2'))
if letter in g3:
f1.add_arc('0', '4a', (letter), (letter))
f1.add_arc('4a', '4', (letter), ())
f1.add_arc('2a', '4', (letter), ('3'))
f1.add_arc('3a', '4', (letter), ('3'))
f1.add_arc('5a', '4', (letter), ('3'))
f1.add_arc('6a', '4', (letter), ('3'))
f1.add_arc('7a', '4', (letter), ('3'))
f1.add_arc('1', '4', (letter), ('3'))
f1.add_arc('4', '4', (letter), ())
f1.add_arc('2', '4', (letter), ('3'))
f1.add_arc('3', '4', (letter), ('3'))
f1.add_arc('5', '4', (letter), ('3'))
f1.add_arc('6', '4', (letter), ('3'))
f1.add_arc('7', '4', (letter), ('3'))
if letter in g4:
f1.add_arc('0', '5a', (letter), (letter))
f1.add_arc('5a', '5', (letter), ())
f1.add_arc('2a', '5', (letter), ('4'))
f1.add_arc('3a', '5', (letter), ('4'))
f1.add_arc('4a', '5', (letter), ('4'))
f1.add_arc('6a', '5', (letter), ('4'))
f1.add_arc('7a', '5', (letter), ('4'))
f1.add_arc('1', '5', (letter), ('4'))
f1.add_arc('5', '5', (letter), ())
f1.add_arc('2', '5', (letter), ('4'))
f1.add_arc('3', '5', (letter), ('4'))
f1.add_arc('4', '5', (letter), ('4'))
f1.add_arc('6', '5', (letter), ('4'))
f1.add_arc('7', '5', (letter), ('4'))
if letter in g5:
f1.add_arc('0', '6a', (letter), (letter))
f1.add_arc('6a', '6', (letter), ())
f1.add_arc('2a', '6', (letter), ('5'))
f1.add_arc('3a', '6', (letter), ('5'))
f1.add_arc('4a', '6', (letter), ('5'))
f1.add_arc('5a', '6', (letter), ('5'))
f1.add_arc('7a', '6', (letter), ('5'))
f1.add_arc('1', '6', (letter), ('5'))
f1.add_arc('6', '6', (letter), ())
f1.add_arc('2', '6', (letter), ('5'))
f1.add_arc('3', '6', (letter), ('5'))
f1.add_arc('4', '6', (letter), ('5'))
f1.add_arc('5', '6', (letter), ('5'))
f1.add_arc('7', '6', (letter), ('5'))
if letter in g6:
f1.add_arc('0', '7a', (letter), (letter))
f1.add_arc('7a', '7', (letter), ())
f1.add_arc('2a', '7', (letter), ('6'))
f1.add_arc('3a', '7', (letter), ('6'))
f1.add_arc('4a', '7', (letter), ('6'))
f1.add_arc('5a', '7', (letter), ('6'))
f1.add_arc('6a', '7', (letter), ('6'))
f1.add_arc('1', '7', (letter), ('6'))
f1.add_arc('7', '7', (letter), ())
f1.add_arc('2', '7', (letter), ('6'))
f1.add_arc('3', '7', (letter), ('6'))
f1.add_arc('4', '7', (letter), ('6'))
f1.add_arc('5', '7', (letter), ('6'))
f1.add_arc('6', '7', (letter), ('6'))
# Remove letters
if letter in remove_letters:
f1.add_arc('0', '1', (letter), (letter))
f1.add_arc('1', '1', (letter), ())
f1.add_arc('2a', '1', (letter), ())
f1.add_arc('3a', '1', (letter), ())
f1.add_arc('4a', '1', (letter), ())
f1.add_arc('5a', '1', (letter), ())
f1.add_arc('6a', '1', (letter), ())
f1.add_arc('7a', '1', (letter), ())
f1.add_arc('2', '1', (letter), ())
f1.add_arc('3', '1', (letter), ())
f1.add_arc('4', '1', (letter), ())
f1.add_arc('5', '1', (letter), ())
f1.add_arc('6', '1', (letter), ())
f1.add_arc('7', '1', (letter), ())
return f1
def letters_to_numbers():
"""
Returns an FST that converts letters to numbers as specified by
the soundex algorithm
"""
# Let's define our first FST
f1 = FST('soundex-generate')
# Indicate that '1' is the initial state
states = ['q1', 'q2', 'n1', 'n2', 'n3', 'n4', 'n5', 'n6']
for state in states:
f1.add_state(state)
f1.initial_state = 'q1'
# Set all the final states
for state in ['q2', 'n1', 'n2', 'n3', 'n4', 'n5', 'n6']:
f1.set_final(state)
# Add the rest of the arcs
for letter in string.ascii_lowercase:
f1.add_arc('q1', 'q2', (letter), (letter))
if letter in set('aehiouwy'):
for state in ['q2', 'n1', 'n2', 'n3', 'n4', 'n5', 'n6']:
f1.add_arc(state, state, (letter), ())
else:
if letter in set('bfpv'):
for state in ['q2', 'n2', 'n3', 'n4', 'n5', 'n6']:
f1.add_arc(state, 'n1', (letter), ('1'))
f1.add_arc('n1', 'n1', (letter), ())
elif letter in set('cgjkqsxz'):
for state in ['q2', 'n1', 'n3', 'n4', 'n5', 'n6']:
f1.add_arc(state, 'n2', (letter), ('2'))
f1.add_arc('n2', 'n2', (letter), ())
elif letter in set('dt'):
for state in ['q2', 'n1', 'n2', 'n4', 'n5', 'n6']:
f1.add_arc(state, 'n3', (letter), ('3'))
f1.add_arc('n3', 'n3', (letter), ())
elif letter in set('l'):
for state in ['q2', 'n1', 'n2', 'n3', 'n5', 'n6']:
f1.add_arc(state, 'n4', (letter), ('4'))
f1.add_arc('n4', 'n4', (letter), ())
elif letter in set('mn'):
for state in ['q2', 'n1', 'n2', 'n3', 'n4', 'n6']:
f1.add_arc(state, 'n5', (letter), ('5'))
f1.add_arc('n5', 'n5', (letter), ())
elif letter in set('r'):
for state in ['q2', 'n1', 'n2', 'n3', 'n4', 'n5']:
f1.add_arc(state, 'n6', (letter), ('6'))
f1.add_arc('n6', 'n6', (letter), ())
return f1
def french_count():
f = FST('french')
f.add_state('1')
f.add_state('2')
f.add_state('3')
f.add_state('4')
f.add_state('5')
f.add_state('6')
f.add_state('7')
f.add_state('8')
f.add_state('9')
f.add_state('10')
f.initial_state = '1'
f.set_final('4')
#hundred's place
for i in range(1):
f.add_arc('1','2',[str(i)],())
for i in range(1,2):
f.add_arc('1','9',[str(i)],[kFRENCH_TRANS[100]])
for i in range(2,10):
f.add_arc('1','9',[str(i)],[kFRENCH_TRANS[i]]+[kFRENCH_TRANS[100]])
#ten's place when hundred's place was 0
for i in range (0,10):
if i==0:
f.add_arc('2','3',[str(i)],())
if i==1:
f.add_arc('2','5',[str(i)],())
if i>1 and i<7:
f.add_arc('2','6',[str(i)],[kFRENCH_TRANS[i*10]])
if i==7:
f.add_arc('2','7',[str(i)],[kFRENCH_TRANS[60]])
if i==8:
f.add_arc('2','8',[str(i)],[kFRENCH_TRANS[4]]+ [kFRENCH_TRANS[20]])
if i==9:
f.add_arc('2','5',[str(i)],[kFRENCH_TRANS[4]]+ [kFRENCH_TRANS[20]])
#ten's place when hundred's place was 1-9
for i in range (0,10):
if i==0:
f.add_arc('9','10',[str(i)],())
if i==1:
f.add_arc('9','5',[str(i)],())
if i>1 and i<7:
f.add_arc('9','6',[str(i)],[kFRENCH_TRANS[i*10]])
if i==7:
f.add_arc('9','7',[str(i)],[kFRENCH_TRANS[60]])
if i==8:
f.add_arc('9','8',[str(i)],[kFRENCH_TRANS[4]]+ [kFRENCH_TRANS[20]])
if i==9:
f.add_arc('9','5',[str(i)],[kFRENCH_TRANS[4]]+ [kFRENCH_TRANS[20]])
#one's place
#state 10-->4
for ii in xrange(0,10):
if ii==0:
f.add_arc('10', '4', [str(ii)], ())
else:
f.add_arc('10', '4', [str(ii)], [kFRENCH_TRANS[ii]])
#state 3-->4
for ii in xrange(0,10):
f.add_arc('3', '4', [str(ii)], [kFRENCH_TRANS[ii]])
#state 5-->4
for i in range(0,10):
if i<7:
f.add_arc('5','4',[str(i)],[kFRENCH_TRANS[10+i]])
else:
f.add_arc('5','4',[str(i)],[kFRENCH_TRANS[10]]+[kFRENCH_TRANS[i]])
#state 6-->4
for i in range(0,10):
if i==0:
f.add_arc('6','4',[str(i)],())
if i==1:
f.add_arc('6','4',[str(i)],[kFRENCH_AND]+[kFRENCH_TRANS[i]])
if i>1:
f.add_arc('6','4',[str(i)],[kFRENCH_TRANS[i]])
#state 7-->4
for i in range(0,10):
if i==1:
f.add_arc('7','4',[str(i)],[kFRENCH_AND]+[kFRENCH_TRANS[10+i]])
elif i>1 and i<7:
f.add_arc('7','4',[str(i)],[kFRENCH_TRANS[10+i]])
elif i==0:
f.add_arc('7','4',[str(i)],[kFRENCH_TRANS[10+i]])
else:
f.add_arc('7','4',[str(i)],[kFRENCH_TRANS[10]]+[kFRENCH_TRANS[i]])
#state 8-->4
for ii in xrange(0,10):
if ii==0:
f.add_arc('8', '4', [str(ii)], ())
else:
f.add_arc('8', '4', [str(ii)], [kFRENCH_TRANS[ii]])
return f
def french_count():
f = FST('french')
f.add_state('start')
# one number and two trailing unknowns
f.add_state('n**')
# exception from state n**
f.add_state('n**+')
# two numbers and one trailing unknown
f.add_state('nn*')
# zero and two uknown digits trailing and so on
f.add_state('0**')
f.add_state('00*')
f.add_state('00n')
f.add_state('0n*')
f.add_state('0n*+')
f.add_state('0nn')
f.add_state('n00')
f.add_state('nnn')
f.add_state('nnn*')
f.add_state('*et*')
# vegasimal counting for 7 in ((0/n)n*)
f.add_state('0n*Vega7+')
f.add_state('0n*Vega7')
f.add_state('0nnVega7')
# vegasimal counting for 8 in ((0/n)n*)
f.add_state('0n*Vega8')
f.add_state('0n*Vega8+')
f.add_state('0nnVega8')
# vegasimal counting for 9 in ((0/n)n*)
f.add_state('0n*Vega9')
f.add_state('0n*Vega9+')
f.add_state('0n*Vega9++')
f.add_state('0nnVega9')
# set final states
f.set_final('00n')
f.set_final('0nn')
f.set_final('nnn')
f.set_final('n00')
f.set_final('0nnVega7')
f.set_final('0nnVega8')
f.set_final('0nnVega9')
# initial state
f.initial_state = 'start'
# remove initial zeroes
f.add_arc('start', '0**', '0', ())
f.add_arc('0**', '00*', '0', ())
for ii in xrange(10):
#from '0n*Vega8' to '0nnVega8
if ii != 0:
f.add_arc('0n*Vega8+', '0nnVega8', str(ii), [kFRENCH_TRANS[ii]])
elif ii == 0:
f.add_arc('0n*Vega8+', '0nnVega8', str(ii), ())
#from '0n*Vega7' to '0nnVega7' 7-9
if ii == 0 or ii == 7 or ii ==8 or ii == 9:
f.add_arc('0n*Vega7', '0n*Vega7+', (), [kFRENCH_TRANS[10]])
f.add_arc('0n*Vega7+', '0n*Vega7+', str(ii), [kFRENCH_TRANS[ii]])
#
f.add_arc('0n*Vega9+', '0n*Vega9++', (), [kFRENCH_TRANS[10]])
f.add_arc('0n*Vega9++', '0nnVega9', str(ii), [kFRENCH_TRANS[ii]])
if ii == 0:
f.add_arc('0n*Vega7+', '0nnVega7', '0', ())
f.add_arc('0n*Vega9++', '0nnVega9', '0', ())
elif ii == 7 or ii == 8 or ii == 9:
f.add_arc('0n*Vega7+', '0nnVega7', str(ii), [kFRENCH_TRANS[ii]])
#from '0n*Vega' to '0nnVega' 2-6
if ii == 2 or ii == 3 or ii ==4 or ii == 5 or ii == 6:
f.add_arc('0n*Vega7', '0nnVega7', str(ii), [kFRENCH_TRANS[ii+10]])
f.add_arc('0n*Vega9+', '0nnVega9', str(ii), [kFRENCH_TRANS[ii+10]])
if ii == 1:
f.add_arc('0**','0n*', str(ii), [kFRENCH_TRANS[10]])
f.add_arc('n**','0n*', str(ii), [kFRENCH_TRANS[10]])
f.add_arc('0n*Vega7', '0n*Vega7+', str(ii), [kFRENCH_AND])
f.add_arc('0n*Vega7+', '0nnVega7', str(ii), [kFRENCH_TRANS[ii+10]])
f.add_arc('0n*Vega9+', '0nnVega9', str(ii), [kFRENCH_TRANS[ii+10]])
#from '00*' to '00n'
f.add_arc('00*', '00n', str(ii), [kFRENCH_TRANS[ii]])
#from '*n*' to '*nn' 2-9
if ii != 0 and ii !=9:
f.add_arc('0n*','0nn', str(ii+1), [kFRENCH_TRANS[ii+1]])
f.add_arc('0n*+','0nn', str(ii), [kFRENCH_TRANS[ii]])
#from 'start' to 'nnn' 200,300,...,900
if ii != 0 and ii !=1:
f.add_arc('start','n**+', str(ii), [kFRENCH_TRANS[ii]])
f.add_arc('n**+', 'n**', (), [kFRENCH_TRANS[100]])
#from 'n**' to 'n0*' 0
if ii == 0:
f.add_arc('n**', 'n00', '00', ())
if ii == 1:
f.add_arc('start', 'n**', '1', [kFRENCH_TRANS[100]])
#from '*n*' to '*et*' 1
f.add_arc('0n*','*et*', '1', [kFRENCH_AND])
#from '*et*' to '*nn' 1
f.add_arc('*et*','0nn', (), [kFRENCH_TRANS[1]])
#from '0**' to '*nn' 10-16
for ii in xrange(10,17):
f.add_arc('0**','0nn', str(ii), [kFRENCH_TRANS[ii]])
f.add_arc('n**','0nn', str(ii), [kFRENCH_TRANS[ii]])
#from '0**' to '*nn' 20-60
for ii in xrange(2,7):
f.add_arc('0**', '0nn', str(ii*10), [kFRENCH_TRANS[ii*10]])
f.add_arc('n**', '0nn', str(ii*10), [kFRENCH_TRANS[ii*10]])
#from '0**', to *n*
f.add_arc('0**','0n*', str(ii), [kFRENCH_TRANS[ii*10]])
#from 'n**' to '0n*'
f.add_arc('n**', '0n*+', str(ii), [kFRENCH_TRANS[ii*10]])
for ii in xrange(7,10):
if ii == 7:
f.add_arc('0**', '0n*Vega7', str(ii), [kFRENCH_TRANS[60]])
f.add_arc('n**', '0n*Vega7', str(ii), [kFRENCH_TRANS[60]])
elif ii == 8:
f.add_arc('0**', '0n*Vega8', str(ii), [kFRENCH_TRANS[4]])
f.add_arc('n**', '0n*Vega8', str(ii), [kFRENCH_TRANS[4]])
f.add_arc('0n*Vega8', '0n*Vega8+', (), [kFRENCH_TRANS[20]])
elif ii == 9:
f.add_arc('0**', '0n*Vega9', str(ii), [kFRENCH_TRANS[4]])
f.add_arc('n**', '0n*Vega9', str(ii), [kFRENCH_TRANS[4]])
f.add_arc('0n*Vega9', '0n*Vega9+', (), [kFRENCH_TRANS[20]])
f.add_arc('n**', '0n*+', '0', ())
return f
def french_count():
f = FST('french')
f.add_state('start')
#states
f.add_state('dig1zero')
f.add_state('dig2zero')
f.add_state('f_dig3')
f.add_state('dig2_one')
f.add_state('state5')
f.add_state('state6')
f.add_state('dig2_two')
f.add_state('dig2_three')
f.add_state('dig2_four')
f.add_state('dig2_five')
f.add_state('dig2_six')
f.add_state('dig2_sev')
f.add_state('dig2_eig')
f.add_state('dig2_nine')
f.add_state('dig1_nzero')
f.add_state('sec_last')
f.add_state('last')
f.add_state('p1')
f.add_state('p2')
f.add_state('p3')
f.initial_state = 'start'
f.set_final('start')
f.set_final('f_dig3')
f.set_final('state5')
f.set_final('state6')
f.set_final('dig2_two')
f.set_final('dig2_three')
f.set_final('dig2_four')
f.set_final('dig2_five')
f.set_final('dig2_six')
f.set_final('dig2_sev')
f.set_final('dig2_eig')
f.set_final('dig2_nine')
f.set_final('last')
f.set_final('p1')
f.set_final('p2')
f.set_final('p3')
f.set_final('sec_last')
# case for 09X
f.add_arc('dig1zero', 'dig2_nine', '9', ())
f.add_arc('dig2_nine', 'dig2_nine', '0',
[kFRENCH_TRANS[4]] + [kFRENCH_TRANS[20]] + [kFRENCH_TRANS[10]])
for k in range(1, 7):
f.add_arc('dig2_nine', 'dig2_nine', str(k), [kFRENCH_TRANS[4]] +
[kFRENCH_TRANS[20]] + [kFRENCH_TRANS[k + 10]])
for k in range(7, 10):
f.add_arc('dig2_nine', 'dig2_nine', str(k),
[kFRENCH_TRANS[4]] + [kFRENCH_TRANS[20]] +
[kFRENCH_TRANS[10]] + [kFRENCH_TRANS[k]])
# case for 08X
f.add_arc('dig1zero', 'dig2_eig', '8', ())
f.add_arc('dig2_eig', 'dig2_eig', '0',
[kFRENCH_TRANS[4]] + [kFRENCH_TRANS[20]])
for j in range(1, 10):
f.add_arc('dig2_eig', 'dig2_eig', str(j), [kFRENCH_TRANS[4]] +
[kFRENCH_TRANS[20]] + [kFRENCH_TRANS[j]])
# case for 07X
f.add_arc('dig1zero', 'dig2_sev', '7', ())
f.add_arc('dig2_sev', 'dig2_sev', '0',
[kFRENCH_TRANS[60]] + [kFRENCH_TRANS[10]])
f.add_arc('dig2_sev', 'dig2_sev', '1',
[kFRENCH_TRANS[60]] + [kFRENCH_AND] + [kFRENCH_TRANS[11]])
for k in range(2, 7):
f.add_arc('dig2_sev', 'dig2_sev', str(k),
[kFRENCH_TRANS[60]] + [kFRENCH_TRANS[k + 10]])
for k in range(7, 10):
f.add_arc('dig2_sev', 'dig2_sev', str(k), [kFRENCH_TRANS[60]] +
[kFRENCH_TRANS[10]] + [kFRENCH_TRANS[k]])
#00X case in french
f.add_arc('start', 'dig1zero', '0', ())
f.add_arc('dig1zero', 'dig2zero', '0', ())
for ii in range(10):
f.add_arc('dig2zero', 'f_dig3', [str(ii)], [kFRENCH_TRANS[ii]])
#case for 02X
f.add_arc('dig1zero', 'dig2_two', '2', ())
f.add_arc('dig2_two', 'dig2_two', '0', [kFRENCH_TRANS[20]])
f.add_arc('dig2_two', 'dig2_two', '1',
[kFRENCH_TRANS[20]] + [kFRENCH_AND] + [kFRENCH_TRANS[1]])
for j in range(2, 10):
f.add_arc('dig2_two', 'dig2_two', str(j),
[kFRENCH_TRANS[20]] + [kFRENCH_TRANS[j]])
#01X case
f.add_arc('dig1zero', 'dig2_one', '1', ())
for j in range(7):
f.add_arc('dig2_one', 'state5', [str(j)], [kFRENCH_TRANS[j + 10]])
for j in range(7, 10):
f.add_arc('dig2_one', 'state6', [str(j)],
[kFRENCH_TRANS[10]] + [kFRENCH_TRANS[j]])
# case for 04X
f.add_arc('dig1zero', 'dig2_four', '4', ())
f.add_arc('dig2_four', 'dig2_four', '0', [kFRENCH_TRANS[40]])
f.add_arc('dig2_four', 'dig2_four', '1',
[kFRENCH_TRANS[40]] + [kFRENCH_AND] + [kFRENCH_TRANS[1]])
for j in range(2, 10):
f.add_arc('dig2_four', 'dig2_four', str(j),
[kFRENCH_TRANS[40]] + [kFRENCH_TRANS[j]])
# case for 03X
f.add_arc('dig1zero', 'dig2_three', '3', ())
f.add_arc('dig2_three', 'dig2_three', '0', [kFRENCH_TRANS[30]])
f.add_arc('dig2_three', 'dig2_three', '1',
[kFRENCH_TRANS[30]] + [kFRENCH_AND] + [kFRENCH_TRANS[1]])
for j in range(2, 10):
f.add_arc('dig2_three', 'dig2_three', str(j),
[kFRENCH_TRANS[30]] + [kFRENCH_TRANS[j]])
# case for 05X
f.add_arc('dig1zero', 'dig2_five', '5', ())
f.add_arc('dig2_five', 'dig2_five', '0', [kFRENCH_TRANS[50]])
f.add_arc('dig2_five', 'dig2_five', '1',
[kFRENCH_TRANS[50]] + [kFRENCH_AND] + [kFRENCH_TRANS[1]])
for k in range(2, 10):
f.add_arc('dig2_five', 'dig2_five', str(k),
[kFRENCH_TRANS[50]] + [kFRENCH_TRANS[k]])
# case for 06X
f.add_arc('dig1zero', 'dig2_six', '6', ())
f.add_arc('dig2_six', 'dig2_six', '0', [kFRENCH_TRANS[60]])
f.add_arc('dig2_six', 'dig2_six', '1',
[kFRENCH_TRANS[60]] + [kFRENCH_AND] + [kFRENCH_TRANS[1]])
for k in range(2, 10):
f.add_arc('dig2_six', 'dig2_six', str(k),
[kFRENCH_TRANS[60]] + [kFRENCH_TRANS[k]])
f.add_arc('start', 'dig1_nzero', '1', [kFRENCH_TRANS[100]])
for j in range(2, 10):
f.add_arc('start', 'dig1_nzero', str(j),
[kFRENCH_TRANS[j]] + [kFRENCH_TRANS[100]])
for i in range(1, 10):
f.add_arc('sec_last', 'sec_last', str(i), [kFRENCH_TRANS[i]])
f.add_arc('dig1_nzero', 'dig2_six', '6', ())
f.add_arc('dig1_nzero', 'dig2_sev', '7', ())
f.add_arc('dig1_nzero', 'dig2_eig', '8', ())
f.add_arc('dig1_nzero', 'dig2_nine', '9', ())
f.add_arc('dig1_nzero', 'sec_last', '0', ())
f.add_arc('dig1_nzero', 'dig2_one', '1', ())
f.add_arc('dig1_nzero', 'dig2_two', '2', ())
f.add_arc('dig1_nzero', 'dig2_three', '3', ())
f.add_arc('dig1_nzero', 'dig2_four', '4', ())
f.add_arc('dig1_nzero', 'dig2_five', '5', ())
f.add_arc('sec_last', 'last', '0', ())
return f
