v2.1.0 (#44)

* initial work

* fix CI

* Work done

* Fix README

Author: mpcabd

.gitignore

@@ -24,6 +24,7 @@ wheels/
 *.egg-info/
 .installed.cfg
 *.egg
+.tox/
 
 # Installer logs
 pip-log.txt

.travis.yml

@@ -1,12 +1,19 @@
-# https://travis-ci.org/mpcabd/python-arabic-reshaper
-dist: xenial
 language: python
-python:
-  - "2.7"
-  - "3.5"
-  - "3.6"
-  - "3.7"
 install:
-  - "pip install -e ."
-script:
-  - "python setup.py test"
+  - pip install tox
+  - pip install -e .
+matrix:
+  include:
+    - python: 2.7
+      env:
+       - TOX_ENV=py27
+    - python: 3.6
+      env:
+        - TOX_ENV=py36
+    - python: 3.7
+      env:
+        - TOX_ENV=py37
+    - python: 3.8
+      env:
+        - TOX_ENV=py38
+script: tox -e $TOX_ENV

README.md

@@ -187,6 +187,45 @@ define an environment variable with the name
 to the configuration file. This way the reshape function will pick it
 automatically, and you won't have to change your old code.
 
+## Settings based on a TrueType® font
+
+If you intend to render the text in a TrueType® font, you can tell the library
+to generate its configuration by reading the font file to figure out what's
+supported in the font and what's not.
+
+To use this feature you need to install the library with an extra option
+(not necessary when you install it with conda):
+
+    pip install --upgrade arabic-reshaper[with-fonttools]
+
+Then you can use the reshaper like this:
+
+```python
+import arabic_reshaper
+
+reshaper = arabic_reshaper.ArabicReshaper(
+    arabic_reshaper.config_for_true_type_font(
+        '/path/to/true-type-font.ttf',
+        arabic_reshaper.ENABLE_ALL_LIGATURES
+    )
+)
+```
+
+This will parse the font file, and figure out what ligatures it supports and enable them,
+as well as whether it has isolated forms or `use_unshaped_instead_of_isolated` should be
+enabled.
+
+The second parameter to `config_for_true_type_font` can be one of
+
+- `ENABLE_NO_LIGATURES`
+- `ENABLE_SENTENCES_LIGATURES`
+- `ENABLE_WORDS_LIGATURES`
+- `ENABLE_LETTERS_LIGATURES`
+- `ENABLE_ALL_LIGATURES` (default)
+
+which controls what ligatures to look for, depending on your usage,
+see [default-config.ini](default-config.ini) to know what ligatures are there.
+
 ## Tashkeel/Harakat issue
 
 [Harakat or Tashkeel](http://en.wikipedia.org/wiki/Arabic_diacritics#Tashkil_.28marks_used_as_phonetic_guides.29)
@@ -209,6 +248,10 @@ https://github.com/mpcabd/python-arabic-reshaper/tarball/master
 
 ## Version History
 
+### 2.1.0
+
+* Added support for settings based on a TrueType® font
+
 ### 2.0.14
 
 * New option `use_unshaped_instead_of_isolated` to get around some fonts missing the isolated form for letters.
@@ -290,8 +333,8 @@ to Python.
 ## Contact
 
 Abdullah Diab (mpcabd)
-Email: 	mpcabd@gmail.com
-Blog:	http://mpcabd.xyz
+Email:  mpcabd@gmail.com
+Blog:   http://mpcabd.xyz
 
 For more info visit my blog
 [post here](http://mpcabd.xyz/python-arabic-text-reshaper/)

arabic_reshaper/__init__.py

@@ -1,6 +1,12 @@
 import os
 
 from .arabic_reshaper import reshape, default_reshaper, ArabicReshaper
+from .reshaper_config import (config_for_true_type_font,
+                              ENABLE_NO_LIGATURES,
+                              ENABLE_SENTENCES_LIGATURES,
+                              ENABLE_WORDS_LIGATURES,
+                              ENABLE_LETTERS_LIGATURES,
+                              ENABLE_ALL_LIGATURES)
 
 
 exec(open(os.path.join(os.path.dirname(__file__), '__version__.py')).read())

arabic_reshaper/__version__.py

@@ -1 +1 @@
-__version__ = '2.0.15'
+__version__ = '2.1.0'

arabic_reshaper/arabic_reshaper.py

@@ -7,30 +7,14 @@
 # Email: mpcabd@gmail.com
 # Website: http://mpcabd.xyz
 
-# Ported and tweaked from Java to Python, from Better Arabic Reshaper
-# [https://github.com/agawish/Better-Arabic-Reshaper/]
-
-# Usage:
-# Install python-bidi [https://github.com/MeirKriheli/python-bidi], can be
-# installed from pip `pip install python-bidi`.
-
-# import arabic_reshaper
-# from bidi.algorithm import get_display
-# reshaped_text = arabic_reshaper.reshape('اللغة العربية رائعة')
-# bidi_text = get_display(reshaped_text)
-# Now you can pass `bidi_text` to any function that handles
-# displaying/printing of the text, like writing it to PIL Image or passing it
-# to a PDF generating method.
 from __future__ import unicode_literals
 
 import re
-import os
 
-from configparser import ConfigParser
 from itertools import repeat
-from pkg_resources import resource_filename
 
 from .ligatures import LIGATURES
+from .reshaper_config import auto_config
 from .letters import (UNSHAPED, ISOLATED, TATWEEL, ZWJ, LETTERS_ARABIC,
                       LETTERS_ARABIC_V2, LETTERS_KURDISH, FINAL,
                       INITIAL, MEDIAL, connects_with_letters_before_and_after,
@@ -73,72 +57,19 @@ class ArabicReshaper(object):
     See the default configuration file :file:`default-config.ini` for details
     on how to configure your reshaper.
     """
+
     def __init__(self, configuration=None, configuration_file=None):
         super(ArabicReshaper, self).__init__()
 
-        configuration_files = [
-            resource_filename(__name__, 'default-config.ini')
-        ]
-
-        if not os.path.exists(configuration_files[0]):
-            raise Exception(
-                ('Default configuration file {} not found,' +
-                 ' check the module installation.').format(
-                     configuration_files[0],
-                 )
-            )
-
-        loaded_from_envvar = False
-
-        if not configuration_file:
-            configuration_file = os.getenv(
-                'PYTHON_ARABIC_RESHAPER_CONFIGURATION_FILE'
-            )
-            if configuration_file:
-                loaded_from_envvar = True
-
-        if configuration_file:
-            if not os.path.exists(configuration_file):
-                raise Exception(
-                    'Configuration file {} not found{}.'.format(
-                        configuration_file,
-                        loaded_from_envvar and (
-                            ' it is set in your environment variable ' +
-                            'PYTHON_ARABIC_RESHAPER_CONFIGURATION_FILE'
-                        ) or ''
-                    )
-                )
-            configuration_files.append(configuration_file)
-
-        configuration_parser = ConfigParser()
-        configuration_parser.read(
-            configuration_files
-        )
-
-        if configuration:
-            configuration_parser.read_dict({
-                'ArabicReshaper': configuration
-            })
-
-        if 'ArabicReshaper' not in configuration_parser:
-            raise ValueError(
-                'Invalid configuration: '
-                'A section with the name ArabicReshaper was not found'
-            )
-
-        configuration = configuration_parser['ArabicReshaper']
-        self.configuration = configuration
+        self.configuration = auto_config(configuration, configuration_file)
         self.language = self.configuration.get('language')
 
-        
         if self.language == 'ArabicV2':
             self.letters = LETTERS_ARABIC_V2
         elif self.language == 'Kurdish':
             self.letters = LETTERS_KURDISH
         else:
             self.letters = LETTERS_ARABIC
-           
-        
 
     @property
     def _ligatures_re(self):
@@ -215,15 +146,15 @@ def reshape(self, text):
                 previous_letter = output[-1]
                 if previous_letter[FORM] == NOT_SUPPORTED:
                     output.append((letter, isolated_form))
-                elif not connects_with_letter_before(letter,self.letters):
+                elif not connects_with_letter_before(letter, self.letters):
                     output.append((letter, isolated_form))
                 elif not connects_with_letter_after(
-                        previous_letter[LETTER],self.letters):
+                        previous_letter[LETTER], self.letters):
                     output.append((letter, isolated_form))
                 elif (previous_letter[FORM] == FINAL and not
                       connects_with_letters_before_and_after(
-                          previous_letter[LETTER],self.letters
-                      )):
+                          previous_letter[LETTER], self.letters
+                )):
                     output.append((letter, isolated_form))
                 elif previous_letter[FORM] == isolated_form:
                     output[-1] = (

arabic_reshaper/ligatures.py

@@ -20,9 +20,9 @@
 # This way we make sure we replace the longest ligatures first
 
 from __future__ import unicode_literals
+from itertools import chain
 
-LIGATURES = (
-    # Sentences
+SENTENCES_LIGATURES = (
     ('ARABIC LIGATURE BISMILLAH AR-RAHMAN AR-RAHEEM', (
         '\u0628\u0633\u0645\u0020'
         '\u0627\u0644\u0644\u0647\u0020'
@@ -44,8 +44,9 @@
 
         ('\uFDFA', '', '', '')
     )),
+)
 
-    # Words
+WORDS_LIGATURES = (
     ('ARABIC LIGATURE ALLAH', (
         '\u0627\u0644\u0644\u0647', ('\uFDF2', '', '', ''),
     )),
@@ -73,8 +74,9 @@
     ('RIAL SIGN', (
         '\u0631[\u06CC\u064A]\u0627\u0644', ('\uFDFC', '', '', ''),
     )),
+)
 
-    # Letters
+LETTERS_LIGATURES = (
     ('ARABIC LIGATURE AIN WITH ALEF MAKSURA', (
         '\u0639\u0649', ('\uFCF7', '', '', '\uFD13'),
     )),
@@ -927,3 +929,5 @@
         '\u0638\u0645', ('\uFC28', '\uFCB9', '\uFD3B', ''),
     )),
 )
+
+LIGATURES = tuple(chain(SENTENCES_LIGATURES, WORDS_LIGATURES, LETTERS_LIGATURES))