def __init__(self): self.custom_dictionary = CustomDictionary()
elif o == "--input":
input_file = a
else:
print "Unhandled Option\n"
logger.critical("Unhandled Option")
usage()
sys.exit(2)
if not input_file:
print "No input file set, use --input option."
usage()
logger.critical("No input file set")
sys.exit(2)
try:
raw_tex = open(input_file, "r").read()
except IOError as e:
print "Can't read input file:" + str(e) + "\n"
logger.critical("Can't read input file : " + str(e))
sys.exit(2)
custom_dictionary = CustomDictionary(no_dictionary_update=no_dictionary_update)
haiku_list = find_haiku_in_tex(raw_tex, custom_dictionary)
custom_dictionary.save_dict()
logger.info("Found the following haiku: " + str(haiku_list))
if len(haiku_list)==0:
print "Found no Haiku, sorry :("
else:
print "Found the following Haiku:"
for haiku in haiku_list:
print haiku
class HaikuFinder(object):
def __init__(self):
self.custom_dictionary = CustomDictionary()
def save_dict(self):
self.custom_dictionary.save_dict();
def find_haiku_in_text(self, raw_text):
haiku_found = []
#Split at double line breaks
paragraphs = raw_text.split("\n\n")
#ignore single line breaks, tabs - replace with spaces
#remove all non-alphanumeric characters/non-punctuation characters
paragraphs = [nonalphanumeric_pattern.sub(' ',p).strip() for p in paragraphs]
for paragraph in paragraphs:
clauses = split_at_punctuation(paragraph)
if len(clauses) < 3:
continue
clauses_with_syllable_count = self.count_syllables_in_clauses(clauses)
haiku_found.extend(self.find_haiku_in_clauses(clauses_with_syllable_count))
return haiku_found
def find_haiku_in_clauses(self, clauses):
haiku_found = []
for i in range(0, len(clauses)-2):
if [clauses[i][2], clauses[i+1][2], clauses[i+2][2]] == [5,7,5]:
haiku_found.append(''.join(clauses[i][0] + clauses[i][1] + " " \
+ clauses[i+1][0] + clauses[i+1][1] + " " \
+ clauses[i+2][0] + clauses[i+2][1]))
return haiku_found
#Returns None if word isn't recognised
def count_syllables_in_clause(self, clause):
return reduce(sum_syllables, [self.number_of_syllables(word) for word in clause.split()])
def count_syllables_in_clauses(self, clauses_with_punctuation):
return [(clause, ending_punctuation, self.count_syllables_in_clause(clause))
for (clause, ending_punctuation) in clauses_with_punctuation]
#returns None if word isn't recognised
def number_of_syllables(self, word):
if word == "":
return 0
#If the word is hypenated then use the sum of the word on each side of the dash
if "-" in word:
return reduce(sum_syllables, [self.number_of_syllables(w) for w in word.split("-")])
is_vowel_sound = lambda syllable: isdigit(syllable[-1])
try:
#cmudict actually returns a list of phonetics, so by default choose first length
return len([1 for syllable in cmudictionary[word.lower()][0] if is_vowel_sound(syllable) ])
except KeyError as e:
num = self.custom_dictionary.number_of_syllables(word)
if num is None:
return None
return num
