[COSM] removal of PYTHON3 in utf8.py (#259)

* [COSM] removal of python3 version checks in  utf8.py

* Version Checks PYTHON3 has been removed

* Encode2Unicode.py PYTHON3 Version handling is successfully passed all unit tests

examples/solpattiyal.py

@@ -19,10 +19,6 @@
 import tamil
 from transliterate import *
 
-PYTHON3 = sys.version[0] > "2"
-if not PYTHON3:
-    sys.stdout = codecs.getwriter("utf-8")(sys.stdout)
-
 
 # use generators for better memory footprint -- 04/04/15
 class WordFrequency(object):

examples/tamilwordgrid.py

@@ -22,9 +22,6 @@
 import codecs
 from math import sqrt
 
-PYTHON3 = sys.version > "3"
-
-
 # Vertical / Horizontal Word Grids
 class Solver:
     def __init__(self, wordgrid):
@@ -140,10 +137,9 @@ def precompute(self):
         )
         self.grid_size = 3 + int(self.max_word_len)
         # sort words in order
-        if PYTHON3:
-            self.words = sorted(self.words, key=len)
-        else:
-            self.words.sort(cmp=WordGrid.sorter)
+        self.words = sorted(self.words, key=len)
+        #else:
+        #    self.words.sort(cmp=WordGrid.sorter)
         # prepare a random grid of dim [#words x #max-word-length]
         # len(self.words)
         for itr_r in range(self.grid_size):

examples/wordlist.py

@@ -11,7 +11,7 @@
 import tamil
 
 LINUX = not hasattr(sys, "getwindowsversion")
-PYTHON3 = sys.version > "3"
+
 
 
 # compute word intersection graph of the a wordlist

examples/wordxsec.py

@@ -8,9 +8,6 @@
 
 import tamil
 
-PYTHON3 = version > "3"
-
-
 # compute word intersection graph of the a wordlist
 # optimized for using the symmetry in computation but not space
 class WordXSec:

solthiruthi/Ezhimai.py

@@ -12,9 +12,6 @@
 from . import WordSpeller
 from . import resources
 
-PYTHON3 = sys.version > "3"
-
-
 class PattiyalThiruthi(WordSpeller.ISpeller):
     def __init__(self, option):
         """ spell checker based on whitelist agarathi """
@@ -37,9 +34,8 @@ def loadWordFile(filename):
         # words will be loaded from the file into the Trie structure
         with codecs.open(filename, "r", "utf-8") as fp:
             data = map(lambda word: word.strip(), fp.readlines())
-        if PYTHON3:
-            return frozenset(data)
-        return set(data)
+        return frozenset(data)
+        #return set(data)
 
 
 if __name__ == "__main__":

solthiruthi/datastore.py

@@ -11,7 +11,7 @@
 
 from tamil import utf8
 
-PYTHON3 = sys.version[0] == "3"
+
 
 
 class Queue(list):

solthiruthi/dictionary.py

@@ -11,7 +11,7 @@
 from . import datastore
 from . import resources
 
-PYTHON3 = sys.version[0] == "3"
+
 
 
 # specify dictionary interface without specifying storage

solthiruthi/heuristics.py

@@ -9,9 +9,6 @@
 
 from tamil import utf8
 
-PYTHON3 = sys.version[0] == "3"
-
-
 def get_letters(word):
     if isinstance(word, list):
         chars = word

solthiruthi/scoring.py

@@ -12,10 +12,7 @@
 
 from tamil import utf8
 from . import resources
-
-PYTHON3 = sys.version[0] == "3"
-if PYTHON3:
-    from functools import reduce
+from functools import reduce
 
 
 class NGStats:

solthiruthi/solthiruthi.py

@@ -10,7 +10,7 @@
 
 # from pprint import pprint
 
-PYTHON3 = sys.version[0] == "3"
+
 
 
 class Solthiruthi:

spell/spell.py

@@ -36,9 +36,6 @@
 
 # Make Bi-Lingual dictionary
 
-PYTHON3 = sys.version_info[0] == 3
-assert PYTHON3, "சொல்திருத்தி செயலி பைத்தான் 3-இல் மற்றுமே இயங்கும்!"
-
 _DEBUG = False
 
 

tamil/date.py

@@ -7,9 +7,6 @@
 from datetime import datetime as datetime_cpy
 from typing import Type
 
-PYTHON3 = sys.version > "3"
-assert PYTHON3, "This module requires Python 3"
-
 TA_WEEKDAYS_SHORT = [
     "திங்கள்",
     "செவ்வாய்",

tamil/numeral.py

@@ -6,8 +6,6 @@
 import re
 import sys
 
-PYTHON3 = sys.version > "3"
-assert PYTHON3, "Python3 or larger required for this module"
 SPACE = re.compile("\s+")
 
 

tamil/tscii2utf8.py

@@ -7,9 +7,6 @@
 
 import tamil
 
-PYTHON3 = version[0] > "2"
-
-
 def usage():
     return u"tscii2utf8.py <filename-1> <filename-2> ... "
 
@@ -23,10 +20,9 @@ def usage():
         try:
             with codecs.open(fname, "r", "utf-8") as fileHandle:
                 output = tamil.tscii.convert_to_unicode(fileHandle.read())
-                if PYTHON3:
-                    print(output)
-                else:
-                    print(output.encode("utf-8"))
+                print(output)
+#                else:
+#                    print(output.encode("utf-8"))
         except Exception as fileOrConvException:
             print(
                 u"tscii2utf8 error - file %s could not be processed due to - %s"

tamil/txt2unicode/encode2unicode.py

@@ -20,17 +20,7 @@
 # <http://www.gnu.org/licenses/>.                                            #
 #                                                                            #
 ##############################################################################
-from sys import version
-
-PYTHON3 = version > "3"
-del version
-
-try:
-    # python 2
-    from .orddic import OrderedDict
-except ImportError as ime:
-    # python 3
-    from collections import OrderedDict
+from collections import OrderedDict
 
 from .encode2utf8 import (
     anjal2utf8,
@@ -281,8 +271,9 @@ def _get_unique_ch(text, all_common_encodes):
     special_chars = [".", ",", ";", ":", "", " ", "\r", "\t", "=", "\n"]
     for line in text:
         for word in line.split(" "):
-            if not PYTHON3:
-                word = word.decode("utf-8")
+        #this place is little trickier how can we change it for python 3 ?
+        #   if not PYTHON3:
+        #       word = word.decode("utf-8")
             for ch in all_common_encodes:
                 if ch in word:
                     word = word.replace(ch, "")
@@ -330,8 +321,9 @@ def _get_unique_common_encodes():
     _all_common_encodes_single_char_ = set([])
 
     for name, encode in _all_encodes_.items():
+    # This place has python3 unicode handling?
         encode_utf8 = set(
-            [PYTHON3 and ch or ch.decode("utf-8") for ch in encode.keys()]
+            [ ch for ch in encode.keys()]
         )
         _all_unicode_encodes_[name] = encode_utf8
     _all_unique_encodes_full_ = _all_unicode_encodes_.copy()

tamil/utf8.py

@@ -15,8 +15,6 @@
 from copy import copy
 from sys import version
 
-PYTHON3 = version > "3"
-assert PYTHON3, "PYTHON3 required to operate Open-Tamil library"
 import functools
 
 ## constants

tests/opentamiltests/__init__.py

@@ -18,6 +18,7 @@
 WINDOWS = sys.platform.find("win") != -1
 LINUX = not WINDOWS
 
+
 import tamil
 
 import transliterate

tests/solthiruthi_suffixremoval.py

@@ -41,7 +41,6 @@ def test_basic_plural_stripper(self):
         for w, x in zip(words_list, expected):
             rval = obj.removeSuffix(w)
             self.assertTrue(rval[1])
-            # if not PYTHON3: print(utf8.get_letters(w),u'->',rval[1])
             self.assertEqual(rval[0], x)
         return
 

webapp/opentamilapp/tamilwordgrid.py

@@ -11,7 +11,7 @@
 
 import tamil
 
-PYTHON3 = sys.version > "3"
+
 
 
 # Vertical / Horizontal Word Grids
@@ -129,10 +129,10 @@ def precompute(self):
         )
         self.grid_size = 3 + int(self.max_word_len)
         # sort words in order
-        if PYTHON3:
-            self.words = sorted(self.words, key=len)
-        else:
-            self.words.sort(cmp=WordGrid.sorter)
+
+        self.words = sorted(self.words, key=len)
+        #else:
+        #    self.words.sort(cmp=WordGrid.sorter)
         # prepare a random grid of dim [#words x #max-word-length]
         # len(self.words)
         for itr_r in range(self.grid_size):