def spell_time(self, time, relative):
"""\
Convert a time expression into words (assuming accusative).
:param time: The 24hr numerical time value in a string, e.g. '8:05'
:param relative: If true, time is interpreted as relative, i.e. \
0:15 will generate '15 minutes' and not '0 hours and \
15 minutes'.
:return: Czech time string with all numerals written out as words
"""
if ':' not in time: # 'now' and similar
return time
hours, mins = map(int, time.split(':'))
time_str = []
if not (relative and hours == 0):
hr_id = 'hodin' + self.HR_ENDING.get(hours, '')
hours = word_for_number(hours, 'F4')
time_str.extend((hours, hr_id))
if mins == 0 and (not relative or hours != 0):
return ' '.join(time_str)
if time_str:
time_str.append('a')
min_id = 'minut' + self.HR_ENDING.get(mins, self.HR_ENDING_DEFAULT)
mins = word_for_number(mins, 'F4')
return ' '.join(time_str + [mins, min_id])
def spell_time(self, time, relative):
"""\
Convert a time expression into words (assuming accusative).
:param time: The 24hr numerical time value in a string, e.g. '8:05'
:param relative: If true, time is interpreted as relative, i.e. \
0:15 will generate '15 minutes' and not '0 hours and \
15 minutes'.
:return: Czech time string with all numerals written out as words
"""
if ':' not in time: # 'now' and similar
return time
hours, mins = map(int, time.split(':'))
time_str = []
if not (relative and hours == 0):
hr_id = 'hodin' + self.HR_ENDING.get(hours, '')
hours = word_for_number(hours, 'F4')
time_str.extend((hours, hr_id))
if mins == 0 and (not relative or hours != 0):
return ' '.join(time_str)
if time_str:
time_str.append('a')
min_id = 'minut' + self.HR_ENDING.get(mins, self.HR_ENDING_DEFAULT)
mins = word_for_number(mins, 'F4')
return ' '.join(time_str + [mins, min_id])
def expand_numbers(stop_name):
"""Spell out all numbers that appear as separate tokens in the word (separated by spaces)."""
tokens = []
for token in stop_name.split(' '):
if NUM_TOKEN.match(token):
try:
num_word = word_for_number(int(token), 'F1')
if token.startswith('0') and len(token) > 1:
tokens.append('nula')
tokens.append(num_word)
except:
tokens.append(token)
else:
tokens.append(token)
return ' '.join(tokens)
def expand_numbers(stop_name):
"""Spell out all numbers that appear as separate tokens in the word (separated by spaces)."""
tokens = []
for token in stop_name.split(" "):
if NUM_TOKEN.match(token):
try:
num_word = word_for_number(int(token), "F1")
if token.startswith("0") and len(token) > 1:
tokens.append("nula")
tokens.append(num_word)
except:
tokens.append(token)
else:
tokens.append(token)
return " ".join(tokens)
def spell_temperature(self, value, interval):
"""Convert a temperature expression into words (assuming nominative).
:param value: Temperature value (whole number in degrees as string), \
e.g. '1' or '-10'.
:param interval: Boolean indicating whether to treat this as a start \
of an interval, i.e. omit the degrees word.
:return: Czech temperature expression as string
"""
ret = ''
value = int(value)
if value < 0:
ret += 'mínus '
value = abs(value)
ret += word_for_number(value, 'M1')
if not interval:
ret += ' stup' + self.DEG_ENDING.get(value, self.DEG_ENDING_DEFAULT)
return ret
def spell_temperature(self, value, interval):
"""Convert a temperature expression into words (assuming nominative).
:param value: Temperature value (whole number in degrees as string), \
e.g. '1' or '-10'.
:param interval: Boolean indicating whether to treat this as a start \
of an interval, i.e. omit the degrees word.
:return: Czech temperature expression as string
"""
ret = ''
value = int(value)
if value < 0:
ret += 'mínus '
value = abs(value)
ret += word_for_number(value, 'M1')
if not interval:
ret += ' stup' + self.DEG_ENDING.get(value, self.DEG_ENDING_DEFAULT)
return ret
def __init__(self, *args, **kwargs):
super(PTICSSLUPreprocessing, self).__init__(*args, **kwargs)
num_norms = []
for num in xrange(60):
num_norms.append(([unicode(num)], [word_for_number(num, 'F1')]))
self.text_normalization_mapping += num_norms
self.text_normalization_mapping += [
(['ve'], ['v']),
(['ke'], ['k']),
(['ku'], ['k']),
(['ze'], ['z']),
# (['se'], ['s']), # do not use this, FJ
(['barandov'], ['barrandov']),
(['litňanská'], ['letňanská']),
(['ípé', 'pa', 'pavlova'], ['i', 'p', 'pavlova']),
(['í', 'pé', 'pa', 'pavlova'], ['i', 'p', 'pavlova']),
(['čaplinovo'], ['chaplinovo']),
(['čaplinova'], ['chaplinova']),
(['zologická'], ['zoologická']),
]
def __init__(self, *args, **kwargs):
super(PTICSSLUPreprocessing, self).__init__(*args, **kwargs)
num_norms = []
for num in xrange(60):
num_norms.append(([unicode(num)], [word_for_number(num, 'F1')]))
self.text_normalization_mapping += num_norms
self.text_normalization_mapping += [
(['ve'], ['v']),
(['ke'], ['k']),
(['ku'], ['k']),
(['ze'], ['z']),
# (['se'], ['s']), # do not use this, FJ
(['barandov'], ['barrandov']),
(['litňanská'], ['letňanská']),
(['ípé', 'pa', 'pavlova'], ['i', 'p', 'pavlova']),
(['í', 'pé', 'pa', 'pavlova'], ['i', 'p', 'pavlova']),
(['čaplinovo'], ['chaplinovo']),
(['čaplinova'], ['chaplinova']),
(['zologická'], ['zoologická']),
]
