def apply_T1(word):
'''There is a syllable boundary in front of every CV-sequence.'''
T1 = ' T1'
WORD = _split_consonants_and_vowels(word)
CONTINUE_VV = 0
CONTINUE_VVV = 0
for i, v in enumerate(WORD):
if i == 0 and is_consonant(v[0][0]):
continue
elif is_consonant(v[0]) and i + 1 != len(WORD):
WORD[i] = v[:-1] + '.' + v[-1]
elif is_vowel(v[0]):
if len(v) > 2:
CONTINUE_VVV += 1
elif len(v) > 1:
CONTINUE_VV += 1
word = ''.join(WORD)
return word, CONTINUE_VV, CONTINUE_VVV, T1
def apply_T4(word):
'''An agglutination diphthong that ends in /u, y/ usually contains a
syllable boundary when -C# or -CCV follow, e.g., [lau.ka.us],
[va.ka.ut.taa].'''
T4 = ''
WORD = word.split('.')
for i, v in enumerate(WORD):
# i % 2 != 0 prevents this rule from applying to first, third, etc.
# syllables, which receive stress (WSP)
if is_consonant(v[-1]) and i % 2 != 0:
if i + 1 == len(WORD) or is_consonant(WORD[i + 1][0]):
if contains_Vu_diphthong(v):
I = v.rfind('u')
WORD[i] = v[:I] + '.' + v[I:]
T4 = ' T4'
elif contains_Vy_diphthong(v):
I = v.rfind('y')
WORD[i] = v[:I] + '.' + v[I:]
T4 = ' T4'
word = '.'.join(WORD)
return word, T4
def apply_T4(word): # OPTIMIZE
'''An agglutination diphthong that ends in /u, y/ usually contains a
syllable boundary when -C# or -CCV follow, e.g., [lau.ka.us],
[va.ka.ut.taa].'''
WORD = _split_consonants_and_vowels(word)
for k, v in WORD.iteritems():
if len(v) == 2 and v.endswith(('u', 'y')):
if WORD.get(k + 2, 0):
if not WORD.get(k + 3, 0):
if len(WORD[k + 2]) == 1 and is_consonant(WORD[k + 2]):
WORD[k] = v[0] + '.' + v[1]
elif len(WORD[k + 1]) == 1 and WORD.get(k + 3, 0):
if is_consonant(WORD[k + 3][0]):
WORD[k] = v[0] + '.' + v[1]
elif len(WORD[k + 2]) == 2:
WORD[k] = v[0] + '.' + v[1]
word = _compile_dict_into_word(WORD)
return word
def apply_T4(word): # TODO: toggle variation
'''An agglutination diphthong that ends in /u, y/ optionally contains a
syllable boundary when -C# or -CCV follow, e.g., [lau.ka.us],
[va.ka.ut.taa].'''
WORD = word.split('.')
PARTS = [[] for part in range(len(WORD))]
for i, v in enumerate(WORD):
# i % 2 != 0 prevents this rule from applying to first, third, etc.
# syllables, which receive stress (WSP)
if is_consonant(v[-1]) and i % 2 != 0:
if i + 1 == len(WORD) or is_consonant(WORD[i + 1][0]):
vv = u_y_final_diphthongs(v)
if vv:
I = vv.start(1) + 1
PARTS[i].append(v[:I] + '.' + v[I:])
# include original form (non-application of rule)
PARTS[i].append(v)
WORDS = [w for w in product(*PARTS)]
for WORD in WORDS:
WORD = '.'.join(WORD)
RULE = ' T4' if word != WORD else ''
yield WORD, RULE
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):
'''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]
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):
WORD[i] = v[:-1] + '.' + v[-1]
WORD = ''.join(WORD)
RULE = ' T1' if word != WORD else ''
return WORD, RULE
def apply_T4(word):
'''An agglutination diphthong that ends in /u, y/ usually contains a
syllable boundary when -C# or -CCV follow, e.g., [lau.ka.us],
[va.ka.ut.taa].'''
WORD = word.split('.')
for i, v in enumerate(WORD):
# i % 2 != 0 prevents this rule from applying to first, third, etc.
# syllables, which receive stress (WSP)
if is_consonant(v[-1]) and i % 2 != 0:
if i + 1 == len(WORD) or is_consonant(WORD[i + 1][0]):
vv = u_y_final_diphthongs(v)
if vv and not is_long(vv.group(1)):
I = vv.start(1) + 1
WORD[i] = v[:I] + '.' + v[I:]
WORD = '.'.join(WORD)
RULE = ' T4' if word != WORD else ''
return WORD, RULE
def apply_T1(word):
'''There is a syllable boundary in front of every CV-sequence.'''
WORD = _split_consonants_and_vowels(word)
for k, v in WORD.iteritems():
if k == 1 and is_consonantal_onset(v):
WORD[k] = '.' + v
elif is_consonant(v[0]) and WORD.get(k + 1, 0):
WORD[k] = v[:-1] + '.' + v[-1]
word = _compile_dict_into_word(WORD)
return word
def apply_T5(word): # BROKEN
'''If a (V)VVV-sequence contains a VV-sequence that could be an /i/-final
diphthong, there is a syllable boundary between it and the third vowel,
e.g., [raa.ois.sa], [huo.uim.me], [la.eis.sa], [sel.vi.äi.si], [tai.an],
[säi.e], [oi.om.me].'''
T5 = ''
WORD = word.split('.')
for i, v in enumerate(WORD):
if contains_VVV(v) and any(i for i in i_DIPHTHONGS if i in v):
I = v.rfind('i') - 1 or 2
I = I + 2 if is_consonant(v[I - 1]) else I
WORD[i] = v[:I] + '.' + v[I:]
T5 = ' T5'
word = '.'.join(WORD)
return word, T5
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