Fixes for #4, #5, #6

.github/workflows/release.yml

@@ -1,99 +1,71 @@
+name: CI
+
 on:
   push:
-    # Sequence of patterns matched against refs/tags
-    tags:
-      - 'v*' # Push events to matching v*, i.e. v1.0, v20.15.10
+    branches: [master]
+  pull_request:
+  release:
+    types: [published]
+  workflow_dispatch:
 
-name: Release
+concurrency:
+  group: ${{ github.workflow }}-${{ github.ref }}
+  cancel-in-progress: ${{ github.event_name == 'pull_request' }}
 
 jobs:
-  build:
-    name: Creating Windows and OSX wheels
-
+  build_wheels:
+    name: Wheels on ${{ matrix.os }}
     runs-on: ${{ matrix.os }}
     strategy:
       fail-fast: false
       matrix:
-        os: [macos-latest, windows-latest]
-        python-version: [3.6, 3.7, 3.8, 3.9]
+        os: [ubuntu-latest, ubuntu-24.04-arm, windows-latest, macos-14]
 
     steps:
-    - uses: actions/checkout@v2
-
-    - name: Set up Python ${{ matrix.python-version }}
-      uses: actions/setup-python@v2
-      with:
-        python-version: ${{ matrix.python-version }}
+      - uses: actions/checkout@v7
 
-    - name: Install python dependencies
-      run: |
-        python -m pip install --upgrade pip
-        python -m pip install --upgrade setuptools
-        pip install wheel twine
-        pip install -e '.'
+      - name: Build and test wheels
+        uses: pypa/cibuildwheel@v4.1.0
 
-    - name: Publishing
-      env:
-        TWINE_USERNAME: ${{ secrets.PYPI_USERNAME }}
-        TWINE_PASSWORD: ${{ secrets.PYPI_PASSWORD }}
-      run: |
-        python setup.py bdist_wheel
-        twine upload dist/*
-
-  build_manylinux2010:
-    name: Creating manylinux releases
+      - uses: actions/upload-artifact@v7
+        with:
+          name: wheels-${{ matrix.os }}
+          path: ./wheelhouse/*.whl
 
+  build_sdist:
+    name: Source distribution
     runs-on: ubuntu-latest
-
     steps:
-    - uses: actions/checkout@v2
+      - uses: actions/checkout@v7
 
-    - name: Build (manylinux2014)
-      uses: RalfG/python-wheels-manylinux-build@v0.3-manylinux2014_x86_64
-      with:
-        python-versions: 'cp36-cp36m cp37-cp37m cp38-cp38 cp39-cp39'
+      - name: Install uv
+        uses: astral-sh/setup-uv@v8.3.2
+        with:
+          enable-cache: true
 
-    - name: Set up Python
-      uses: actions/setup-python@v1
-      with:
-        python-version: 3.8
+      - name: Build sdist
+        run: uv build --sdist
 
-    - name: Install python dependencies
-      run: |
-        python -m pip install --upgrade pip
-        python -m pip install --upgrade setuptools
-        pip install wheel twine
-
-    - name: Publishing
-      env:
-        TWINE_USERNAME: ${{ secrets.PYPI_USERNAME }}
-        TWINE_PASSWORD: ${{ secrets.PYPI_PASSWORD }}
-      run: |
-        twine upload dist/*-manylinux*.whl
-
-  sdist:
-    name: Creating source release
+      - uses: actions/upload-artifact@v7
+        with:
+          name: sdist
+          path: ./dist/*.tar.gz
 
+  publish:
+    name: Publish to PyPI
+    needs: [build_wheels, build_sdist]
     runs-on: ubuntu-latest
-
+    if: github.event_name == 'release' && github.repository == 'TkTech/mutf8'
+    environment:
+      name: pypi
+      url: https://pypi.org/p/mutf8
+    permissions:
+      id-token: write
     steps:
-    - uses: actions/checkout@v2
-
-    - name: Set up Python
-      uses: actions/setup-python@v1
-      with:
-        python-version: 3.8
-
-    - name: Install python dependencies
-      run: |
-        python -m pip install --upgrade pip
-        python -m pip install --upgrade setuptools
-        pip install wheel twine
+      - uses: actions/download-artifact@v8
+        with:
+          path: dist
+          merge-multiple: true
 
-    - name: Publishing
-      env:
-        TWINE_USERNAME: ${{ secrets.PYPI_USERNAME }}
-        TWINE_PASSWORD: ${{ secrets.PYPI_PASSWORD }}
-      run: |
-        python setup.py sdist
-        twine upload dist/*
+      - name: Publish
+        uses: pypa/gh-action-pypi-publish@release/v1

.gitignore

@@ -1,9 +1,14 @@
 *.pyc
 *.pyd
 venv
+.venv/
 .eggs
 cython_debug/*
 build/
 _build/
+dist/
+wheelhouse/
 *.egg-info
 *.so
+uv.lock
+*.idea

README.md

@@ -1,10 +1,9 @@
-![Tests](https://github.com/TkTech/mutf8/workflows/Tests/badge.svg?branch=master)
+![CI](https://github.com/TkTech/mutf8/actions/workflows/release.yml/badge.svg)
 
 # mutf-8
 
 This package contains simple pure-python as well as C encoders and decoders for
-the MUTF-8 character encoding. In most cases, you can also parse the even-rarer
-CESU-8.
+the MUTF-8 character encoding.
 
 These days, you'll most likely encounter MUTF-8 when working on files or
 protocols related to the JVM. Strings in a Java `.class` file are encoded using
@@ -14,26 +13,15 @@ serializer.
 This library was extracted from [Lawu][], a Python library for working with JVM
 class files.
 
-## 🎉 Installation
+## Installation
 
+Prebuilt binary packages are available for most platforms and Python versions.
 Install the package from PyPi:
 
 ```
 pip install mutf8
 ```
 
-Binary wheels are available for the following:
-
-|                  | py3.6 | py3.7 | py3.8 | py3.9 |
-| ---------------- | ----- | ----- | ----- | ----- |
-| OS X (x86_64)    | y     | y     | y     | y     |
-| Windows (x86_64) | y     | y     | y     | y     |
-| Linux (x86_64)   | y     | y     | y     | y     |
-
-If binary wheels are not available, it will attempt to build the C extension
-from source with any C99 compiler. If it could not build, it will fall back
-to a pure-python version.
-
 ## Usage
 
 Encoding and decoding is simple:
@@ -48,23 +36,48 @@ bytes = encode_modified_utf8(unicode)
 This module *does not* register itself globally as a codec, since importing
 should be side-effect-free.
 
-## 📈 Benchmarks
+## Benchmarks
 
-The C extension is significantly faster - often 20x to 40x faster.
+The C extension is dramatically faster than the pure-Python fallback — from
+~10× on tiny strings (where per-call overhead dominates) to several hundred
+times on realistic payloads, and over 1000× when encoding large ASCII.
 
 <!-- BENCHMARK START -->
 
 ### MUTF-8 Decoding
-| Name                         |   Min (μs) |   Max (μs) |   StdDev |           Ops |
-|------------------------------|------------|------------|----------|---------------|
-| cmutf8-decode_modified_utf8  |    0.00009 |    0.00080 |  0.00000 | 9957678.56358 |
-| pymutf8-decode_modified_utf8 |    0.00190 |    0.06040 |  0.00000 |  450455.96019 |
+
+| Input | C | Python | Speedup |
+|-------|---|--------|--------:|
+| ASCII (16 B) | 81.5 ± 26 ns | 1.04 ± 0.18 µs | 13× |
+| ASCII (1 KB) | 185 ± 38 ns | 65.7 ± 8.5 µs | 356× |
+| ASCII (64 KB) | 5.58 ± 0.66 µs | 4.35 ± 0.23 ms | 780× |
+| CJK (16 B) | 127 ± 15 ns | 1.33 ± 0.15 µs | 10× |
+| CJK (1 KB) | 681 ± 117 ns | 99.3 ± 9.4 µs | 146× |
+| CJK (64 KB) | 31.8 ± 3.3 µs | 6.28 ± 0.063 ms | 198× |
+| Emoji (16 B) | 100 ± 21 ns | 882 ± 122 ns | 9× |
+| Emoji (1 KB) | 538 ± 119 ns | 79.6 ± 6.9 µs | 148× |
+| Emoji (64 KB) | 26.2 ± 4.1 µs | 5.32 ± 0.13 ms | 203× |
+| Mixed (16 B) | 118 ± 27 ns | 2.17 ± 0.16 µs | 18× |
+| Mixed (1 KB) | 860 ± 162 ns | 76.7 ± 9.2 µs | 89× |
+| Mixed (64 KB) | 38 ± 5 µs | 5.3 ± 0.48 ms | 139× |
 
 ### MUTF-8 Encoding
-| Name                         |   Min (μs) |   Max (μs) |   StdDev |            Ops |
-|------------------------------|------------|------------|----------|----------------|
-| cmutf8-encode_modified_utf8  |    0.00008 |    0.00151 |  0.00000 | 11897361.05101 |
-| pymutf8-encode_modified_utf8 |    0.00180 |    0.16650 |  0.00000 |   474390.98091 |
+
+| Input | C | Python | Speedup |
+|-------|---|--------|--------:|
+| ASCII (16 B) | 70.5 ± 5.8 ns | 1.23 ± 0.15 µs | 18× |
+| ASCII (1 KB) | 109 ± 16 ns | 49 ± 6.2 µs | 449× |
+| ASCII (64 KB) | 1.55 ± 0.27 µs | 2.93 ± 0.19 ms | 1889× |
+| CJK (16 B) | 81.4 ± 12 ns | 2.02 ± 0.26 µs | 25× |
+| CJK (1 KB) | 624 ± 127 ns | 108 ± 16 µs | 173× |
+| CJK (64 KB) | 32.5 ± 2.9 µs | 6.8 ± 0.24 ms | 209× |
+| Emoji (16 B) | 73 ± 10 ns | 1.17 ± 0.12 µs | 16× |
+| Emoji (1 KB) | 621 ± 76 ns | 67 ± 8.5 µs | 108× |
+| Emoji (64 KB) | 33.3 ± 6 µs | 3.98 ± 0.12 ms | 119× |
+| Mixed (16 B) | 108 ± 97 ns | 2.48 ± 0.35 µs | 23× |
+| Mixed (1 KB) | 1.39 ± 0.29 µs | 63.3 ± 3.5 µs | 45× |
+| Mixed (64 KB) | 81.2 ± 14 µs | 3.91 ± 0.093 ms | 48× |
+
 <!-- BENCHMARK END -->
 
 ## C Extension

mutf8/__init__.py

@@ -10,6 +10,8 @@
     for MUTF-8/CESU-8 into the python core.
 """
 
+__version__ = "1.0.6"
+
 
 try:
     from mutf8.cmutf8 import decode_modified_utf8, encode_modified_utf8