New mode for Harakat

+ Updated letters
+ Updated ligatures

New mode for Harakat to shift Harakat position for bidi get_display

Author: mpcabd

arabic_reshaper/__version__.py

@@ -1 +1 @@
-__version__ = '2.0.11'
+__version__ = '2.0.12'

arabic_reshaper/arabic_reshaper.py

@@ -167,15 +167,24 @@ def reshape(self, text):
         delete_harakat = self.configuration.getboolean('delete_harakat')
         delete_tatweel = self.configuration.getboolean('delete_tatweel')
         support_zwj = self.configuration.getboolean('support_zwj')
+        shift_harakat_position = self.configuration.getboolean(
+            'shift_harakat_position'
+        )
+
         positions_harakat = {}
 
         for letter in text:
             if HARAKAT_RE.match(letter):
                 if not delete_harakat:
                     position = len(output) - 1
+                    if shift_harakat_position:
+                        position -= 1
                     if position not in positions_harakat:
                         positions_harakat[position] = []
-                    positions_harakat[position].append(letter)
+                    if shift_harakat_position:
+                        positions_harakat[position].insert(0, letter)
+                    else:
+                        positions_harakat[position].append(letter)
             elif letter == TATWEEL and delete_tatweel:
                 pass
             elif letter == ZWJ and not support_zwj:

arabic_reshaper/default-config.ini

@@ -6,6 +6,9 @@ language = Arabic
 # Whether to delete the Harakat (Tashkeel) before reshaping or not.
 delete_harakat = yes
 
+# Whether to shift the Harakat (Tashkeel) one position so they appear correctly when string is reversed
+shift_harakat_position = no
+
 # Whether to delete the Tatweel (U+0640) before reshaping or not.
 delete_tatweel = no
 
@@ -189,6 +192,7 @@ ARABIC LIGATURE QAF WITH MEEM WITH HAH = no
 ARABIC LIGATURE QAF WITH MEEM WITH MEEM = no
 ARABIC LIGATURE QAF WITH MEEM WITH YEH = no
 ARABIC LIGATURE QAF WITH YEH = no
+ARABIC LIGATURE QALA USED AS KORANIC STOP SIGN = no
 ARABIC LIGATURE REH WITH SUPERSCRIPT ALEF = no
 ARABIC LIGATURE SAD WITH ALEF MAKSURA = no
 ARABIC LIGATURE SAD WITH HAH = no
@@ -199,6 +203,7 @@ ARABIC LIGATURE SAD WITH MEEM = no
 ARABIC LIGATURE SAD WITH MEEM WITH MEEM = no
 ARABIC LIGATURE SAD WITH REH = no
 ARABIC LIGATURE SAD WITH YEH = no
+ARABIC LIGATURE SALLA USED AS KORANIC STOP SIGN = no
 ARABIC LIGATURE SEEN WITH ALEF MAKSURA = no
 ARABIC LIGATURE SEEN WITH HAH = no
 ARABIC LIGATURE SEEN WITH HAH WITH JEEM = no
@@ -215,18 +220,19 @@ ARABIC LIGATURE SEEN WITH MEEM WITH JEEM = no
 ARABIC LIGATURE SEEN WITH MEEM WITH MEEM = no
 ARABIC LIGATURE SEEN WITH REH = no
 ARABIC LIGATURE SEEN WITH YEH = no
-ARABIC LIGATURE SHADDA WITH DAMMATAN ISOLATED FORM = no
-ARABIC LIGATURE SHADDA WITH KASRATAN ISOLATED FORM = no
-ARABIC LIGATURE SHADDA WITH FATHA ISOLATED FORM = no
+ARABIC LIGATURE SHADDA WITH DAMMA = no
 ARABIC LIGATURE SHADDA WITH DAMMA ISOLATED FORM = no
-ARABIC LIGATURE SHADDA WITH KASRA ISOLATED FORM = no
-ARABIC LIGATURE SHADDA WITH SUPERSCRIPT ALEF ISOLATED FORM = no
-ARABIC LIGATURE SHADDA WITH FATHA MEDIAL FORM = no
 ARABIC LIGATURE SHADDA WITH DAMMA MEDIAL FORM = no
-ARABIC LIGATURE SHADDA WITH KASRA MEDIAL FORM = no
+ARABIC LIGATURE SHADDA WITH DAMMATAN ISOLATED FORM = no
 ARABIC LIGATURE SHADDA WITH FATHA = no
-ARABIC LIGATURE SHADDA WITH DAMMA = no
+ARABIC LIGATURE SHADDA WITH FATHA ISOLATED FORM = no
+ARABIC LIGATURE SHADDA WITH FATHA MEDIAL FORM = no
 ARABIC LIGATURE SHADDA WITH KASRA = no
+ARABIC LIGATURE SHADDA WITH KASRA ISOLATED FORM = no
+ARABIC LIGATURE SHADDA WITH KASRA MEDIAL FORM = no
+ARABIC LIGATURE SHADDA WITH KASRATAN ISOLATED FORM = no
+ARABIC LIGATURE SHADDA WITH SUPERSCRIPT ALEF = no
+ARABIC LIGATURE SHADDA WITH SUPERSCRIPT ALEF ISOLATED FORM = no
 ARABIC LIGATURE SHEEN WITH ALEF MAKSURA = no
 ARABIC LIGATURE SHEEN WITH HAH = no
 ARABIC LIGATURE SHEEN WITH HAH WITH MEEM = no

arabic_reshaper/letters.py

@@ -94,8 +94,8 @@
     '\u0647': ('\uFEE9', '\uFEEB', '\uFEEC', '\uFEEA'),
     # ARABIC LETTER WAW
     '\u0648': ('\uFEED', '', '', '\uFEEE'),
-    # ARABIC LETTER ALEF MAKSURA
-    '\u0649': ('\uFEEF', '', '', '\uFEF0'),
+    # ARABIC LETTER (UIGHUR KAZAKH KIRGHIZ)? ALEF MAKSURA
+    '\u0649': ('\uFEEF', '\uFBE8', '\uFBE9', '\uFEF0'),
     # ARABIC LETTER YEH
     '\u064A': ('\uFEF1', '\uFEF3', '\uFEF4', '\uFEF2'),
     # ARABIC LETTER ALEF WASLA

arabic_reshaper/ligatures.py

@@ -583,6 +583,9 @@
     ('ARABIC LIGATURE QAF WITH YEH', (
         '\u0642\u064A', ('\uFC36', '', '', '\uFC7F'),
     )),
+    ('ARABIC LIGATURE QALA USED AS KORANIC STOP SIGN', (
+        '\u0642\u0644\u06D2', ('\uFDF1', '', '', ''),
+    )),
     ('ARABIC LIGATURE REH WITH SUPERSCRIPT ALEF', (
         '\u0631\u0670', ('\uFC5C', '', '', ''),
     )),
@@ -613,6 +616,9 @@
     ('ARABIC LIGATURE SAD WITH YEH', (
         '\u0635\u064A', ('\uFD06', '', '', '\uFD22'),
     )),
+    ('ARABIC LIGATURE SALLA USED AS KORANIC STOP SIGN', (
+        '\u0635\u0644\u06D2', ('\uFDF0', '', '', ''),
+    )),
     ('ARABIC LIGATURE SEEN WITH ALEF MAKSURA', (
         '\u0633\u0649', ('\uFCFB', '', '', '\uFD17'),
     )),
@@ -688,10 +694,8 @@
 
         ('\uFC62', '\uFC62', '\uFC62', '\uFC62'),
     )),
-    ('ARABIC LIGATURE SHADDA WITH SUPERSCRIPT ALEF ISOLATED FORM', (
-        '(?:\u0670\u0651|\u0651\u0670)',
-
-        ('\uFC63', '\uFC63', '\uFC63', '\uFC63'),
+    ('ARABIC LIGATURE SHADDA WITH SUPERSCRIPT ALEF', (
+        '(?:\u0651\u0670|\u0670\u0651)', ('\uFC63', '', '', ''),
     )),
 
     # There is a special case when they are with Tatweel

arabic_reshaper/tests/test_002_reshaping.py

@@ -177,6 +177,22 @@ def test_reshaping(self):
         _reshaping_test(self)
 
 
+class TestReshapingWithShiftedHarakatWithoutLigatures(unittest.TestCase):
+    def setUp(self):
+        self.reshaper = arabic_reshaper.ArabicReshaper({
+            'delete_harakat': False,
+            'support_ligatures': False,
+            'shift_harakat_position': True,
+        })
+        self.cases = (
+            ('فُعِلَ', 'ُﻓِﻌَﻞ'),
+            ('فُعِّلَ', 'ُﻓِّﻌَﻞ'),
+        )
+
+    def test_reshaping(self):
+        _reshaping_test(self)
+
+
 class TestReshapingSomeLigatures(unittest.TestCase):
     def setUp(self):
         self.reshaper = arabic_reshaper.ArabicReshaper({