def split_syllables(letters = []):
""" Returns the syllables in a given word as a list """
## Generic algorithm:
## Each vowel and combination is its own syllable. Consonants and aytham get added to the end of the previous syllable
# ensure that the word is a valid word
TamilWord.validate_word(''.join(letters))
# initialize empty list
syllables = []
# loop through letters in the word
for letter in letters:
# if letter is a vowel or combination, it gets its own syllable
if TamilLetter.is_combination(letter) or TamilLetter.is_vowel(letter):
syllables.append(letter)
# if codepoint is a consonant or aytham, add it to the end of the previously-added codepoint
elif TamilLetter.is_consonant(letter) or TamilLetter.is_aytham(letter):
# ensure that at least one character already exists
if len(syllables) > 0:
syllables[-1] = syllables[-1] + letter
# if the first letter is a consonant (probably b/c it' s a loanword), add it to the beginning of the string
else:
syllables.append(letter)
# if codepoint was neither a vowel, aytham, a pulli or a combination ending, an unexpected error has occurred
else:
raise Exception("Unknown error: The letter \'%s\' in word %s is neither a vowel, consonant, combination or aytham" %(letter, ''.join(letters)))
def get_class(word = u''):
""" Returns the noun class for a given word """
### See flowchart in docs/material/sendhil/noun_classes.png for more details
word = TamilWord(word)
## For noun classes ending in a consonant:
## Ending in ம் -> class 1
## Ending in ல், ன், ய், ள், ண் AND has exactly two graphemes AND first letter is kuril: class 2
## All other consonant-ending nouns -> class 3
if TamilLetter.is_consonant(word[-1]):
if word[-1] == u'ம்':
return 1
elif word[-1] in (u'ல்', u'ன்', u'ய்', u'ள்', u'ண்') and len(word)==2 and TamilLetter.is_kuril(word[0]):
return 2
else:
return 3
## For noun classes not ending in a combination:
## TODO: What if a noun ends in Aytham?
## TODO: Has thiTamilLettertested on one-letter long all-vowel nouns (e.g. ஈ)?
## Ends in இ, ஈ or ஐ -> class 4
## Ends in ஆ, ஏ, ஊ or ஓ -> class 5
## Has exactly two graphemes AND first letter is kuril -> class 6
## Ends in று and penultimate grapheme is not a consonant -> class 7
## Ends in டு and penultimate grapheme is not a consonant -> class 8
## TODO: The above two rules assume the word is at least two graphemes long. Valid assumption?
## All other cases: class 9
else:
_, ending_vowel = TamilLetter.split_combination(word[-1])
if ending_vowel in (u'இ', u'ஈ', u'ஐ'):
return 4
elif ending_vowel in (u'ஆ', u'ஏ', u'ஊ', u'ஓ'):
return 5
elif len(word)==2 and TamilLetter.is_kuril(TamilWord[0]):
return 6
elif word[-1]==u'று' and not TamilLetter.is_consonant(word[-2]):
return 7
elif word[-1]==u'டு' and not TamilLetter.is_consonant(word[-2]):
return 8
else:
return 9
def validate(self):
""" Checks whether the given word is valid """
# simple test: every element of the string has to be a valid Tamil character
for codepoint in self.text:
TamilLetter.validate_letter(codepoint)
#TODO: implement method this more thoroughly
#TODO: check for pulli or combination_ending at the beginning of a word
return True
def direct_object(self):
"""Returns the direct object of a given noun."""
direct_object = TamilNoun(self.word)
# direct objects use the suffix 'ஐ'
suffix = u"ஐ"
# map noun class to its particular connector
CONNECTOR_BY_CLASS = {
1: u"த்த்",
2: direct_object[-1],
3: direct_object[-1],
4: u"ய்",
5: u"வ்",
6: u"வ்",
7: u"ற்ற்",
8: u"ட்ட்",
9: TamilLetter.split_combination(direct_object[-1])[0]
}
noun_class = TamilNoun.get_class(self.word)
if noun_class not in CONNECTOR_BY_CLASS.keys():
raise ValueError("""%s is an invalid noun class for word %s.
Must be between 1 and 9""" % (noun_class, self.word))
connector = TamilWord(CONNECTOR_BY_CLASS.get(noun_class))
# remove last letter + add two-letter connector + suffix
if noun_class in (1, 7, 8):
del direct_object[-1]
direct_object.word += (
connector[0] +
TamilLetter.get_combination(connector[-1], suffix))
# remove last letter + add one-letter connector + suffix
elif noun_class in (3, 9):
del direct_object[-1]
direct_object.word += \
(TamilLetter.get_combination(connector[0], suffix))
# nothing to remove; just add connector + suffix
elif noun_class in (2, 4, 5, 6):
direct_object.word += \
(TamilLetter.get_combination(connector[0], suffix))
return direct_object.word
def direct_object(self):
"""Returns the direct object of a given noun."""
direct_object = TamilNoun(self.word)
# direct objects use the suffix 'ஐ'
suffix = u"ஐ"
# map noun class to its particular connector
CONNECTOR_BY_CLASS = {1: u"த்த்",
2: direct_object[-1],
3: direct_object[-1],
4: u"ய்",
5: u"வ்",
6: u"வ்",
7: u"ற்ற்",
8: u"ட்ட்",
9: TamilLetter.split_combination(direct_object[-1])[0]
}
noun_class = TamilNoun.get_class(self.word)
if noun_class not in CONNECTOR_BY_CLASS.keys():
raise ValueError("""%s is an invalid noun class for word %s.
Must be between 1 and 9""" % (noun_class, self.word))
connector = TamilWord(CONNECTOR_BY_CLASS.get(noun_class))
# remove last letter + add two-letter connector + suffix
if noun_class in (1, 7, 8):
del direct_object[-1]
direct_object.word += (connector[0] +
TamilLetter.get_combination(connector[-1], suffix))
# remove last letter + add one-letter connector + suffix
elif noun_class in (3, 9):
del direct_object[-1]
direct_object.word += \
(TamilLetter.get_combination(connector[0], suffix))
# nothing to remove; just add connector + suffix
elif noun_class in (2, 4, 5, 6):
direct_object.word += \
(TamilLetter.get_combination(connector[0], suffix))
return direct_object.word
def split_letters(word=u''):
""" Returns the graphemes (i.e. the Tamil characters) in a given word as a list """
# ensure that the word is a valid word
TamilWord.validate_word(word)
# list (which will be returned to user)
letters = []
# a tuple of all combination endings and of all அ combinations
combination_endings = TamilLetter.get_combination_endings()
a_combinations = TamilLetter.get_combination_column(u'அ').values()
# loop through
for codepoint in word:
# if codepoint is an அ combination, a vowel, aytham or a space, add it to the list
if codepoint in a_combinations or TamilLetter.is_whitespace(codepoint) or TamilLetter.is_vowel(codepoint) or TamilLetter.is_aytham(codepoint):
letters.append(codepoint)
# if codepoint is a combination ending or a pulli ('்'), add it to the end of the previously-added codepoint
elif codepoint in combination_endings or codepoint==TamilLetter.get_pulli():
# ensure that at least one character already exists
if len(letters) > 0:
letters[-1] = letters[-1] + codepoint
# otherwise raise an Error. validate_word() should catch this, however
else:
raise ValueError("Unknown error: The combination ending %s cannot be the first character of a word" %(codepoint))
# if codepoint was neither a vowel, aytham, a pulli or a combination ending, an unexpected error has occurred
else:
raise ValueError("Unknown error: The codepoint \'%s\' in word %s is neither a vowel, consonant, combination or aytham" %(codepoint, word))
#TODO: Write extensive test cases for this
return letters
def get_class(word=u''):
""" Returns the noun class for a given noun """
### See flowchart (docs/material/sendhil/noun_classes.png) for details
# Convert 'word' from a unicode object to a TamilWord object
word = TamilWord(word)
## For noun classes ending in a consonant:
## Ending in ம் -> class 1
## Ending in ல், ன், ய், ள், ண் AND has exactly two graphemes AND
## first letter is kuril: class 2
## All other consonant-ending nouns -> class 3
if TamilLetter.is_consonant(word[-1]):
if word[-1] == u'ம்':
return 1
elif (word[-1] in (u'ல்', u'ன்', u'ய்', u'ள்', u'ண்') and
len(word) == 2 and TamilLetter.is_kuril(word[0])):
return 2
else:
return 3
## For noun classes not ending in a combination:
## TODO: What if a noun ends in Aytham? Is this possible?
## TODO: Ensure this is tested on single letter nouns (e.g. ஈ)
## Ends in இ, ஈ or ஐ -> class 4
## Ends in ஆ, ஏ, ஊ or ஓ -> class 5
## Has exactly two graphemes AND first letter is kuril -> class 6
## Ends in று and penultimate grapheme is not a consonant -> class 7
## Ends in டு and penultimate grapheme is not a consonant -> class 8
## TODO: Is it ok that rules assume the word is >= two graphemes long?
## All other cases: class 9
else:
_, ending_vowel = TamilLetter.split_combination(word[-1])
if ending_vowel in (u'இ', u'ஈ', u'ஐ'):
return 4
elif ending_vowel in (u'ஆ', u'ஏ', u'ஊ', u'ஓ'):
return 5
elif len(word) == 2 and TamilLetter.is_kuril(TamilWord[0]):
return 6
elif word[-1] == u'று' and not TamilLetter.is_consonant(word[-2]):
return 7
elif word[-1] == u'டு' and not TamilLetter.is_consonant(word[-2]):
return 8
else:
return 9
def get_class(word=u''):
""" Returns the noun class for a given noun """
### See flowchart (docs/material/sendhil/noun_classes.png) for details
# Convert 'word' from a unicode object to a TamilWord object
word = TamilWord(word)
## For noun classes ending in a consonant:
## Ending in ம் -> class 1
## Ending in ல், ன், ய், ள், ண் AND has exactly two graphemes AND
## first letter is kuril: class 2
## All other consonant-ending nouns -> class 3
if TamilLetter.is_consonant(word[-1]):
if word[-1] == u'ம்':
return 1
elif (word[-1] in (u'ல்', u'ன்', u'ய்', u'ள்', u'ண்')
and len(word) == 2 and TamilLetter.is_kuril(word[0])):
return 2
else:
return 3
## For noun classes not ending in a combination:
## TODO: What if a noun ends in Aytham? Is this possible?
## TODO: Ensure this is tested on single letter nouns (e.g. ஈ)
## Ends in இ, ஈ or ஐ -> class 4
## Ends in ஆ, ஏ, ஊ or ஓ -> class 5
## Has exactly two graphemes AND first letter is kuril -> class 6
## Ends in று and penultimate grapheme is not a consonant -> class 7
## Ends in டு and penultimate grapheme is not a consonant -> class 8
## TODO: Is it ok that rules assume the word is >= two graphemes long?
## All other cases: class 9
else:
_, ending_vowel = TamilLetter.split_combination(word[-1])
if ending_vowel in (u'இ', u'ஈ', u'ஐ'):
return 4
elif ending_vowel in (u'ஆ', u'ஏ', u'ஊ', u'ஓ'):
return 5
elif len(word) == 2 and TamilLetter.is_kuril(word[0]):
return 6
elif word[-1] == u'று' and not TamilLetter.is_consonant(word[-2]):
return 7
elif word[-1] == u'டு' and not TamilLetter.is_consonant(word[-2]):
return 8
else:
return 9