def apply_T1(word):
'''There is a syllable boundary in front of every CV-sequence.'''
# split consonants and vowels: 'balloon' -> ['b', 'a', 'll', 'oo', 'n']
WORD = [w for w in re.split('([ieAyOauo]+)', word) if w]
count = 0
for i, v in enumerate(WORD):
if i == 0 and is_consonant(v[0]):
continue
elif is_consonant(v[0]) and i + 1 != len(WORD):
if is_cluster(v): # WSP
if count % 2 == 0:
WORD[i] = v[0] + '.' + v[1:] # CC > C.C, CCC > C.CC
else:
WORD[i] = '.' + v # CC > .CC, CCC > .CCC
# elif is_sonorant(v[0]) and is_cluster(v[1:]): # NEW
# if count % 2 == 0:
# WORD[i] = v[0:2] + '.' + v[2:]
# else:
# WORD[i] = v[0] + '.' + v[1:]
else:
WORD[i] = v[:-1] + '.' + v[-1] # CC > C.C, CCC > CC.C
count += 1
WORD = ''.join(WORD)
RULE = ' T1' if word != WORD else ''
return WORD, RULE
def apply_T1(word, T1E=True):
# split consonants and vowels: 'balloon' -> ['b', 'a', 'll', 'oo', 'n']
WORD = [w for w in re.split('([ieAyOauo]+)', word) if w]
# these are to keep track of which sub-rules are applying
A, B, C, D, E, F, G = '', '', '', '', '', '', ''
# a count divisible by 2 indicates an even syllable
count = 1
for i, v in enumerate(WORD):
# T1B
# If there is a consonant cluster word-initially, the entire cluster
# forms the onset of the first syllable:
# CCV > #CCV
if i == 0 and is_consonant(v[0]):
B = 'b'
elif is_consonant(v[0]):
count += 1
# True if the current syllable is unstressed, else False
unstressed = count % 2 == 0
# T1C
# If there is a consonant cluster word-finally, the entire cluster
# forms the coda of the final syllable:
# VCC# > VCC#
if i + 1 == len(WORD):
C = 'c'
# T1D
# If there is a bare "Finnish" consonant cluster word-medially and
# the previous syllable receives stress, the first consonant of the
# cluster forms the coda of the previous syllable (to create a
# heavy syllable); otherwise, the whole cluster forms the onset of
# the current syllable (thisis the /kr/ rule):
# 'VCCV > 'VC.CV, VCCV > V.CCV
elif is_cluster(v):
D = 'd'
WORD[i] = v[0] + '.' + v[1:] if unstressed else '.' + v
elif is_cluster(v[1:]):
# T1E (optional)
# If there is a word-medial "Finnish" consonant cluster that is
# preceded by a sonorant consonant, if the previous syllable
# receives stress, the sonorant consonant and the first
# consonant of the cluster form the coda of the previous
# syllable, and the remainder of the cluster forms the onset of
# the current syllable:
# 'VlCC > VlC.C
if T1E and is_sonorant(v[0]) and unstressed:
E = 'e'
WORD[i] = v[:2] + '.' + v[2:]
# T1F
# If there is a word-medial "Finnish" cluster that follows a
# consonant, that first consonant forms the coda of the
# previous syllable, and the cluster forms the onset of the
# current syllable:
# VCkr > VC.kr
else:
F = 'f'
WORD[i] = v[0] + '.' + v[1:]
# T1A
# There is a syllable boundary in front of every CV sequence:
# VCV > V.CV, CCV > C.CV
else:
WORD[i] = v[:-1] + '.' + v[-1]
A = 'a'
WORD = ''.join(WORD)
RULE = ' T1' + A + B + C + D + E + F + G if word != WORD else ''
return WORD, RULE