Braille for Tamil #223

 - Initial commit provides table of Barathi Braille including English/Tamil
   letter support.
 - Appropriate handling of Vowel Consonants and Pulli are made in validation
   with data from IIT-M Acharya website, and other references.
 - More work to be done in rendering the Braille cells to images.
 - Analysis of printed braille image to text can be a future project.

setup.py

@@ -18,7 +18,7 @@
       url='https://github.com/Ezhil-Language-Foundation/open-tamil',
       packages=['tamil','transliterate','ngram','tamil.txt2ipa','tamil.txt2unicode',
       'tamil.utils', 'solthiruthi','spell','tamilstemmer','tamilmorse','tamilsandhi',
-      'valai','kural','tamiltts'],
+      'valai','kural','tamiltts','tabraille'],
       package_dir={'solthiruthi': 'solthiruthi','tamilmorse':'tamilmorse',
       'tamilsandhi':'tamil-sandhi-checker/tamilsandhi'},
       package_data={'solthiruthi': ['data/*.txt'],'tamilmorse':['data/*.json'],

tabraille/__init__.py

@@ -0,0 +1,4 @@
+# This Python file uses the following encoding: utf-8
+# (C) 2020 Muthiah Annamalai
+# This file is part of open-tamil project
+from .table import BrailleCell, table, map_to_braille

tabraille/table.py

@@ -0,0 +1,149 @@
+# This Python file uses the following encoding: utf-8
+# (C) 2020 முத்து அண்ணாமலை
+# This file is part of open-tamil project
+import copy as _copy
+from tamil.utf8 import get_letters, pulli_symbols
+from tamil.utf8 import grantha_agaram_letters, grantha_mei_letters
+from tamil.utf8 import uyir_letters, ayudha_letter, splitMeiUyir
+
+# Barathi Braille வழியில் உருவான தமிழ் பிரெயில்
+# Ref: https://en.wikipedia.org/wiki/Tamil_Braille
+class BrailleCell:
+    """
+    Describe 3x2 (3-row, 2-col) Braille cell.
+    positions are 1-6 from Left->Right, Top->Bot.
+    """
+    def __init__(self,letter,pos):
+        self.letter=letter
+        self.pos = _copy.copy(pos)
+        self.cells = [[False,False],[False,False],[False,False]]
+        assert len(pos) < 6,"6 cell positions can only be specified."
+        if len(pos) > 0:
+            assert (max(pos) < 7) and (min(pos) > 0), "positions are [1-6] only"
+        for _pos in pos:
+            x,y= (_pos-1)%2,(_pos-1)//2
+            self.cells[y][x]=True
+        #print(self,self.letter)
+
+    def gethash(self):
+        return "".join(["%d"%x for  x in self.pos])
+
+    def __str__(self):
+        result = ''
+        for i in range(3):
+            for j in range(2):
+                if self.cells[i][j]:
+                    result += '*'
+                else:
+                    result += '_'
+            result += '\n'
+        return result
+
+# எ.கா.
+# =>வணக்கம்
+#=> "⠧⠼⠅⠰⠅⠍⠰"
+
+# Ref: https://en.wikipedia.org/wiki/Tamil_Braille
+# Generate Braille cells from Tamil Text
+table = {}
+table['அ']=BrailleCell('அ',[1])
+table['ஆ']=BrailleCell('ஆ',[2,4,5])
+table['இ']=BrailleCell('இ',[2,3])
+table['ஈ']=BrailleCell('ஈ',[4,5])
+table['உ']=BrailleCell('உ',[1,5,6])
+table['ஊ']=BrailleCell('ஊ',[1,3,4,6])
+table['எ']=BrailleCell('எ',[3,6])
+table['ஏ']=BrailleCell('ஏ',[1,4])
+table['ஐ']=BrailleCell('ஐ',[2,5])
+table['ஒ']=BrailleCell('ஒ',[1,2,5,6])
+table['ஓ']=BrailleCell('ஓ',[1,4,5])
+table['ஔ']=BrailleCell('ஔ',[2,3,6])
+table['க']=BrailleCell('க',[1,5])
+table['ங']=BrailleCell('ங',[2,5,6])
+table['ச']=BrailleCell('ச',[1,2])
+table['ஜ']=BrailleCell('ஜ',[2,3,4])
+table['ஞ']=BrailleCell('ஞ',[3,4])
+table['ட']=BrailleCell('ட',[2,3,4,5,6])
+table['ண']=BrailleCell('ண',[2,4,5,6])
+table['த']=BrailleCell('த',[2,3,4,5])
+table['ந']=BrailleCell('ந',[1,2,4,5])
+table['ப']=BrailleCell('ப',[1,2,3,5])
+table['ம']=BrailleCell('ம',[1,2,5])
+table['ய']=BrailleCell('ய',[1,2,4,5,6])
+table['ர']=BrailleCell('ர',[1,3,4,5])
+table['ல']=BrailleCell('ல',[1,3,5])
+table['ள']=BrailleCell('ள',[2,4,6])
+table['வ']=BrailleCell('வ',[1,3,5,6])
+table['ஶ']=BrailleCell('ஶ',[1,2,6])
+table['ஷ']=BrailleCell('ஷ',[1,2,3,5,6])
+table['ஸ']=BrailleCell('ஸ',[2,3,5])
+table['ஹ']=BrailleCell('ஹ',[1,3,4])
+table['க்ஷ']=BrailleCell('க்ஷ',[1,2,3,4,5])
+table['ற']=BrailleCell('ற',[1,2,3,4,6])
+table['ழ']=BrailleCell('ழ',[1,3,4,5,6])
+table['ன']=BrailleCell('ன',[4,6])
+table['்']=BrailleCell('்',[2])
+table['ஃ']=BrailleCell('ஃ',[6])
+
+# Ref: http://www.acharya.gen.in:8080/disabilities/br_tut.php#tabular
+# ஆங்கில் பிரெயில் எழுத்துக்களும் தமிழ் உயிரெழுத்துக்களுடன்
+# ஒன்றாகவே குறியிடப்பட்டுள்ளன.
+table['a']=table['அ']
+table['e']=table['ஏ']
+table['i']=table['இ']
+table['o']=table['ஓ']
+table['u']=table['உ']
+# ஆங்கில மெய்/consonant களுக்கு மற்ற தனிப்பட்ட
+# குறியீடுகள் உள்ளன.
+table['b']=BrailleCell('b',[1,3])
+table['c']=BrailleCell('c',[1,2])#sa
+table['d']=BrailleCell('d',[1,2,4])
+table['f']=BrailleCell('f',[1,2,3])
+table['g']=BrailleCell('g',[1,2,3,4])
+table['h']=BrailleCell('h',[1,3,4]) #ha
+table['j']=BrailleCell('j',[2,3,4]) #ja
+table['k']=BrailleCell('k',[1,5]) #ka
+table['l']=BrailleCell('l',[1,3,5])#la
+table['m']=BrailleCell('m',[1,2,5])#ma
+table['n']=BrailleCell('n',[1,2,4,5])#Na
+table['p']=BrailleCell('p',[1,2,3,5])#pa
+table['q']=BrailleCell('q',[1,2,3,4,5])#ksha
+table['r']=BrailleCell('r',[1,3,4,5])#chinna R
+table['s']=BrailleCell('s',[2,3,5])#ஸ
+table['t']=BrailleCell('t',[2,3,4,5])#த
+table['v']=BrailleCell('v',[1,3,5,6])#வ
+table['w']=BrailleCell('w',[2,3,4,6])
+table['x']=BrailleCell('x',[1,2,5,6])#ஒ
+table['y']=BrailleCell('y',[1,2,4,5,6])#ய
+table['z']=BrailleCell('z',[1,4,5,6])
+
+table[',']=BrailleCell(',',[3])
+table[';']=BrailleCell(';',[3,5])
+table[':']=BrailleCell(':',[3,4])
+table['.']=BrailleCell('.',[3,4,6])
+table['!']=BrailleCell('!',[3,4,5])
+table['(']=BrailleCell('(',[3,4,5,6])
+table[')']=BrailleCell(')',[3,4,5,6])
+table['\'']=BrailleCell('\'',[3,5,6])
+table['"']=BrailleCell('"',[4,5,6])
+table[' ']=BrailleCell(' ',[])
+assert len(table) == 74,"Barathi Braille for Tamil is strictly 74  entries."
+
+def map_to_braille(tamil_string):
+    result = []
+    for letter in get_letters(tamil_string):
+        if letter in grantha_mei_letters:
+            pos = grantha_mei_letters.index(letter)
+            agaram = grantha_agaram_letters[pos]
+            result.append(table[agaram])
+            result.append(table[pulli_symbols[0]])
+        elif letter in uyir_letters or letter == ayudha_letter:
+            result.append(table[letter])
+        else:
+            lMei,lUyir = splitMeiUyir(letter)
+            pos = grantha_mei_letters.index(lMei)
+            agaram = grantha_agaram_letters[pos]
+            result.append(table[agaram])
+            if lUyir != 'அ':
+                result.append(table[lUyir])
+    return result

tamil/utf8.py

@@ -577,8 +577,8 @@ def splitMeiUyir(uyirmei_char):
     This function split uyirmei compound character into mei + uyir characters
     and returns in tuple.
 
-    Input : It must be unicode tamil char.
-
+    Input : It must be unicode tamil char, V=vowel, C=char, VC=compound VC.
+    Output: (mei,uyir) if it is VC, otherwise return input if it is V, or C.
     Written By : Arulalan.T
     Date : 22.09.2014
 

tests/braille_tests.py

@@ -0,0 +1,32 @@
+# -*- coding: utf-8 -*-
+# This file is part of Open-Tamil package unittests
+# (C) 2018 Muthu Annamalai
+
+# setup the paths
+from opentamiltests import *
+import tabraille
+
+import string
+class BrailleTests(unittest.TestCase):
+    def test_braille(self):
+        self.assertEqual(len(tabraille.table),74,'74 elements expected in table')
+
+    def test_atoz(self):
+        for letter in string.ascii_lowercase:
+            self.assertTrue(tabraille.table[letter])
+
+    def test_braille_mapping(self):
+        expected = [tabraille.table['ந'],
+        tabraille.table['ஆ'],
+        tabraille.table['ன'],
+        tabraille.table['்']]
+        self.assertListEqual(tabraille.map_to_braille('நான்'),expected)
+
+    def test_braille_mapping2(self):
+        word = 'கப்பல்'
+        letters = ['க','ப','்','ப','ல','்']
+        expected=[tabraille.table[l] for l in letters]
+        self.assertEqual(tabraille.map_to_braille(word),expected)
+
+if __name__ == "__main__":
+    unittest.main()