Modernise packaging and clean up Python 2 residue

.github/workflows/ci.yml

@@ -0,0 +1,32 @@
+name: CI
+
+on:
+  push:
+    branches: [master]
+  pull_request:
+    branches: [master]
+
+jobs:
+  test:
+    runs-on: ubuntu-latest
+    strategy:
+      fail-fast: false
+      matrix:
+        python-version: ["3.10", "3.11", "3.12", "3.13", "3.14"]
+
+    steps:
+      - uses: actions/checkout@v4
+
+      - name: Install uv
+        uses: astral-sh/setup-uv@v5
+        with:
+          enable-cache: true
+
+      - name: Set up Python ${{ matrix.python-version }}
+        run: uv python install ${{ matrix.python-version }}
+
+      - name: Install dependencies
+        run: uv sync --group dev
+
+      - name: Run tests
+        run: uv run pytest --cov=arabic_reshaper --cov-report=term-missing

.github/workflows/release.yml

@@ -0,0 +1,33 @@
+name: Release
+
+on:
+  release:
+    types: [published]
+
+jobs:
+  test:
+    runs-on: ubuntu-latest
+    strategy:
+      fail-fast: true
+      matrix:
+        python-version: ["3.10", "3.11", "3.12", "3.13", "3.14"]
+    steps:
+      - uses: actions/checkout@v4
+      - uses: astral-sh/setup-uv@v5
+        with:
+          enable-cache: true
+      - run: uv python install ${{ matrix.python-version }}
+      - run: uv sync --group dev
+      - run: uv run pytest
+
+  publish:
+    needs: test
+    runs-on: ubuntu-latest
+    environment: pypi
+    permissions:
+      id-token: write
+    steps:
+      - uses: actions/checkout@v4
+      - uses: astral-sh/setup-uv@v5
+      - run: uv build
+      - uses: pypa/gh-action-pypi-publish@release/v1

.gitignore

@@ -47,3 +47,6 @@ venv/
 # ignore
 .ignore/
 .DS_Store
+
+# local test fonts — not committed
+fonts/

README.md

@@ -1,6 +1,6 @@
 ## Python Arabic Reshaper
 
-[![Build Status](https://app.travis-ci.com/mpcabd/python-arabic-reshaper.svg?branch=master)](https://app.travis-ci.com/mpcabd/python-arabic-reshaper)
+[![CI](https://github.com/mpcabd/python-arabic-reshaper/actions/workflows/ci.yml/badge.svg)](https://github.com/mpcabd/python-arabic-reshaper/actions/workflows/ci.yml)
 
 Reconstruct Arabic sentences to be used in applications that don't support
 Arabic script.
@@ -248,6 +248,14 @@ https://github.com/mpcabd/python-arabic-reshaper/tarball/master
 
 ## Version History
 
+### 3.0.1
+
+* Modernised packaging: migrated from `setuptools` to `hatchling` via `pyproject.toml`
+* Dropped support for Python < 3.10
+* Switched CI from Travis CI to GitHub Actions
+* Added type hints and improved test coverage for font configuration
+* Removed obsolete Anaconda publishing scripts
+
 ### 3.0.0
 * Stop supporting Python 2.7
 * Remove dependency on `future`. See #88.

arabic_reshaper/__init__.py

@@ -1,7 +1,6 @@
-import os
-
 from .arabic_reshaper import reshape, default_reshaper, ArabicReshaper
 from .reshaper_config import (config_for_true_type_font,
+                              ArabicReshaperConfigurationError,
                               ENABLE_NO_LIGATURES,
                               ENABLE_SENTENCES_LIGATURES,
                               ENABLE_WORDS_LIGATURES,

arabic_reshaper/arabic_reshaper.py

@@ -9,6 +9,7 @@
 
 import re
 
+from functools import cached_property
 from itertools import repeat
 
 from .ligatures import LIGATURES
@@ -35,7 +36,7 @@
 )
 
 
-class ArabicReshaper(object):
+class ArabicReshaper:
     """
     A class for Arabic reshaper, it allows for fine-tune configuration over the
     API.
@@ -56,7 +57,7 @@ class ArabicReshaper(object):
     """
 
     def __init__(self, configuration=None, configuration_file=None):
-        super(ArabicReshaper, self).__init__()
+        super().__init__()
 
         self.configuration = auto_config(configuration, configuration_file)
         self.language = self.configuration.get('language')
@@ -68,33 +69,25 @@ def __init__(self, configuration=None, configuration_file=None):
         else:
             self.letters = LETTERS_ARABIC
 
-    @property
+    @cached_property
     def _ligatures_re(self):
-        if not hasattr(self, '__ligatures_re'):
-            patterns = []
-            re_group_index_to_ligature_forms = {}
-            index = 0
-            FORMS = 1
-            MATCH = 0
-            for ligature_record in LIGATURES:
-                ligature, replacement = ligature_record
-                if not self.configuration.getboolean(ligature):
-                    continue
-                re_group_index_to_ligature_forms[index] = replacement[FORMS]
-                patterns.append('({})'.format(replacement[MATCH]))
-                index += 1
-            self._re_group_index_to_ligature_forms = (
-                re_group_index_to_ligature_forms
-            )
-            self.__ligatures_re = re.compile('|'.join(patterns), re.UNICODE)
-        return self.__ligatures_re
+        patterns = []
+        self._re_group_index_to_ligature_forms = {}
+        index = 0
+        FORMS = 1
+        MATCH = 0
+        for ligature, replacement in LIGATURES:
+            if not self.configuration.getboolean(ligature):
+                continue
+            self._re_group_index_to_ligature_forms[index] = replacement[FORMS]
+            patterns.append(f'({replacement[MATCH]})')
+            index += 1
+        return re.compile('|'.join(patterns), re.UNICODE)
 
     def _get_ligature_forms_from_re_group_index(self, group_index):
-        if not hasattr(self, '_re_group_index_to_ligature_forms'):
-            return self._ligatures_re
         return self._re_group_index_to_ligature_forms[group_index]
 
-    def reshape(self, text):
+    def reshape(self, text: str) -> str:
         if not text:
             return ''
 
@@ -217,6 +210,8 @@ def reshape(self, text):
                 if not forms[ligature_form]:
                     continue
                 output[a] = (forms[ligature_form], NOT_SUPPORTED)
+                # Pad the replaced positions with empty sentinels so that
+                # Harakat position indices remain aligned with the output list.
                 output[a+1:b] = repeat(('', NOT_SUPPORTED), b - 1 - a)
 
         result = []

arabic_reshaper/letters.py

@@ -506,22 +506,22 @@
     ZWJ: (ZWJ, ZWJ, ZWJ, ZWJ),
 }
 
-def connects_with_letter_before(letter,LETTERS):
-    if letter not in LETTERS:
+def connects_with_letter_before(letter, letters):
+    if letter not in letters:
         return False
-    forms = LETTERS[letter]
+    forms = letters[letter]
     return forms[FINAL] or forms[MEDIAL]
 
 
-def connects_with_letter_after(letter,LETTERS):
-    if letter not in LETTERS:
+def connects_with_letter_after(letter, letters):
+    if letter not in letters:
         return False
-    forms = LETTERS[letter]
+    forms = letters[letter]
     return forms[INITIAL] or forms[MEDIAL]
 
 
-def connects_with_letters_before_and_after(letter,LETTERS):
-    if letter not in LETTERS:
+def connects_with_letters_before_and_after(letter, letters):
+    if letter not in letters:
         return False
-    forms = LETTERS[letter]
+    forms = letters[letter]
     return forms[MEDIAL]

arabic_reshaper/ligatures.py

@@ -71,6 +71,7 @@
         '\u0648\u0633\u0644\u0645', ('\uFDF8', '', '', ''),
     )),
     ('RIAL SIGN', (
+        # Regex (not a literal string): matches both Farsi YEH (U+06CC) and Arabic YEH (U+064A).
         '\u0631[\u06CC\u064A]\u0627\u0644', ('\uFDFC', '', '', ''),
     )),
 )

arabic_reshaper/reshaper_config.py

@@ -22,6 +22,10 @@
 except ImportError:
     with_font_config = False
 
+class ArabicReshaperConfigurationError(ValueError):
+    """Raised when the reshaper configuration is invalid or missing."""
+
+
 ENABLE_NO_LIGATURES = 0b000
 ENABLE_SENTENCES_LIGATURES = 0b001
 ENABLE_WORDS_LIGATURES = 0b010
@@ -361,7 +365,10 @@
 }
 
 
-def auto_config(configuration=None, configuration_file=None):
+def auto_config(
+    configuration: dict | None = None,
+    configuration_file: str | None = None,
+):
     loaded_from_envvar = False
 
     configuration_parser = ConfigParser()
@@ -378,14 +385,12 @@ def auto_config(configuration=None, configuration_file=None):
 
     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 ''
-                )
+            env_note = (
+                ' (set via PYTHON_ARABIC_RESHAPER_CONFIGURATION_FILE)'
+                if loaded_from_envvar else ''
+            )
+            raise ArabicReshaperConfigurationError(
+                f'Configuration file not found: {configuration_file}{env_note}'
             )
         configuration_parser.read((configuration_file,))
 
@@ -395,22 +400,40 @@ def auto_config(configuration=None, configuration_file=None):
         })
 
     if 'ArabicReshaper' not in configuration_parser:
-        raise ValueError(
+        raise ArabicReshaperConfigurationError(
             'Invalid configuration: '
             'A section with the name ArabicReshaper was not found'
         )
 
     return configuration_parser['ArabicReshaper']
 
 
-def config_for_true_type_font(font_file_path,
-                              ligatures_config=ENABLE_ALL_LIGATURES):
+def config_for_true_type_font(font_file_path: str, ligatures_config: int = ENABLE_ALL_LIGATURES) -> dict:
+    """Return a reshaper configuration dict tuned to the capabilities of a TrueType font.
+
+    Inspects the font's cmap table to determine which positional Arabic letter
+    forms are present, and checks each ligature glyph so that only ligatures
+    the font actually supports are enabled.
+
+    Args:
+        font_file_path: Path to the .ttf/.otf font file.
+        ligatures_config: Bitmask of ENABLE_*_LIGATURES flags controlling which
+            ligature categories to probe. Defaults to ENABLE_ALL_LIGATURES.
+
+    Returns:
+        A configuration dict suitable for passing to ArabicReshaper().
+
+    Raises:
+        ImportError: If fonttools is not installed.
+        ArabicReshaperConfigurationError: If the font path is invalid.
+    """
     if not with_font_config:
-        raise Exception('fonttools not installed, ' +
-                        'install it then rerun this.\n' +
-                        '$ pip install arabic-teshaper[with-fonttools]')
+        raise ImportError(
+            'fonttools is not installed. '
+            'Install it with: pip install arabic-reshaper[with-fonttools]'
+        )
     if not font_file_path or not os.path.exists(font_file_path):
-        raise Exception('Invalid path to font file')
+        raise ArabicReshaperConfigurationError(f'Invalid path to font file: {font_file_path}')
     ttfont = TTFont(font_file_path)
     has_isolated = True
     for k, v in LETTERS_ARABIC.items():

arabic_reshaper/tests/test_003_reshaping.py

@@ -28,8 +28,6 @@ def setUp(self):
 
 
         )
-        print(self.cases[0][0])
-
     def test_reshaping(self):
         _reshaping_test(self)