WIP: pytest and GitHub Actions

.github/workflows/pypi-publish.yml

@@ -0,0 +1,31 @@
+name: Upload Python Package
+
+on:
+  release:
+    types: [published]
+
+permissions:
+  contents: read
+
+jobs:
+  deploy:
+
+    runs-on: ubuntu-latest
+
+    steps:
+    - uses: actions/checkout@v4
+    - name: Set up Python
+      uses: actions/setup-python@v3
+      with:
+        python-version: '3.x'
+    - name: Install dependencies
+      run: |
+        python -m pip install --upgrade pip
+        pip install build
+    - name: Build package
+      run: python -m build
+    - name: Publish package
+      uses: pypa/gh-action-pypi-publish@27b31702a0e7fc50959f5ad993c78deac1bdfc29
+      with:
+        user: __token__
+        password: ${{ secrets.PYPI_API_TOKEN }}

.github/workflows/pytest.yml

@@ -0,0 +1,27 @@
+name: Test with pytest
+
+on:
+  push:
+  #   branches: [ "main" ]
+  # pull_request:
+  #   branches: [ "main" ]
+
+permissions:
+  contents: read
+
+jobs:
+  test:
+    runs-on: ubuntu-latest
+    steps:
+    - uses: actions/checkout@v4
+    - name: Set up Python
+      uses: actions/setup-python@v3
+      with:
+        python-version: '3.x'
+    - name: Install dependencies
+      run: |
+        python -m pip install --upgrade pip
+        pip install pytest pytest-cov
+        pip install .
+    - name: Test with pytest
+      run: pytest --cov=src

tests/test_fingerprint.py

@@ -0,0 +1,36 @@
+import pytest
+
+from pybitshred.fingerprint import Fingerprint, jaccard_distance
+
+def test_jaccard_distance():
+    # Test case 1: Empty fingerprints
+    fp_a = Fingerprint(bytearray(), 0)
+    fp_b = Fingerprint(bytearray(), 0)
+    with pytest.raises(ZeroDivisionError):
+        jaccard_distance(fp_a, fp_b)
+
+    # Test case 2: Fingerprint with all bits set to 0
+    fp_a = Fingerprint(bytearray(b"\x00\x00\x00\x00"), 0)
+    fp_b = Fingerprint(bytearray(b"\x00\x00\x00\x00"), 0)
+    with pytest.raises(ZeroDivisionError):
+        jaccard_distance(fp_a, fp_b)
+
+    # Test case 3: Identical fingerprints with all bits set to 0
+    fp_a = Fingerprint(bytearray(b"\xFF\xFF\xFF\xFF"), 32)
+    fp_b = Fingerprint(bytearray(b"\xFF\xFF\xFF\xFF"), 32)
+    assert jaccard_distance(fp_a, fp_b) == pytest.approx(1.0)
+
+    # Test case 4: Identical fingerprint with some bits set to 1
+    fp_a = Fingerprint(bytearray(b"\x0F\x00\xFF\x55"), 16)
+    fp_b = Fingerprint(bytearray(b"\x0F\x00\xFF\x55"), 16)
+    assert jaccard_distance(fp_a, fp_b) == pytest.approx(1.0)
+
+    # Test case 5: Fingerprint with different bit vectors
+    fp_a = Fingerprint(bytearray(b"\xAA\x55\xAA\x55"), 16)
+    fp_b = Fingerprint(bytearray(b"\x55\xAA\x55\xAA"), 16)
+    assert jaccard_distance(fp_a, fp_b) == pytest.approx(0.0)
+
+    # Test case 6: Fingerprint with different bit vectors
+    fp_a = Fingerprint(bytearray(b"\xAA\xBB\xCC\xDD"), 20)
+    fp_b = Fingerprint(bytearray(b"\xCC\xDD\xEE\xFF"), 30)
+    assert jaccard_distance(fp_a, fp_b) == pytest.approx(0.470588)

tests/test_utils.py

@@ -0,0 +1,65 @@
+from pybitshred.utils import bit_count, bit_vector_set, djb2_hash
+
+
+def test_djb2_hash():
+    # Test case 1: Empty data
+    assert djb2_hash(b"") == 5381
+
+    # Test case 2: Single byte data
+    assert djb2_hash(b"A") == 177638
+
+    # Test case 3: Multiple byte data
+    assert djb2_hash(b"Hello, World!") == 2531426958
+
+    # Test case 4: Same data should produce same hash
+    assert djb2_hash(b"Hello, World!") == djb2_hash(b"Hello, World!")
+
+    # Test case 5: Different data should produce different hash
+    assert djb2_hash(b"Hello, World!") != djb2_hash(b"Hello, GitHub Copilot!")
+
+
+def test_bit_vector_set():
+    # Test case 1: Setting a bit at the beginning of the vector
+    vector = bytearray(b"\x00\x00\x00\x00")
+    bit_vector_set(vector, 0)
+    assert vector == bytearray(b"\x01\x00\x00\x00")
+
+    # Test case 2: Setting a bit in the middle of the vector
+    vector = bytearray(b"\x00\x00\x00\x00")
+    bit_vector_set(vector, 12)
+    assert vector == bytearray(b"\x00\x10\x00\x00")
+
+    # Test case 3: Setting a bit at the end of the vector
+    vector = bytearray(b"\x00\x00\x00\x00")
+    bit_vector_set(vector, 31)
+    assert vector == bytearray(b"\x00\x00\x00\x80")
+
+    # Test case 4: Setting multiple bits
+    vector = bytearray(b"\x00\x00\x00\x00")
+    bit_vector_set(vector, 0)
+    bit_vector_set(vector, 12)
+    bit_vector_set(vector, 31)
+    assert vector == bytearray(b"\x01\x10\x00\x80")
+
+    # Test case 5: Setting a bit that is already set
+    vector = bytearray(b"\x00\x00\x00\x00")
+    bit_vector_set(vector, 3)
+    bit_vector_set(vector, 3)
+    assert vector == bytearray(b"\x08\x00\x00\x00")
+
+
+def test_bit_count():
+    # Test case 1: Empty fingerprint
+    assert bit_count(bytearray()) == 0
+
+    # Test case 2: Fingerprint with all bits set to 0
+    assert bit_count(bytearray(b"\x00\x00\x00\x00")) == 0
+
+    # Test case 3: Fingerprint with all bits set to 1
+    assert bit_count(bytearray(b"\xFF\xFF\xFF\xFF")) == 32
+
+    # Test case 4: Fingerprint with some bits set to 1
+    assert bit_count(bytearray(b"\x0F\x00\xFF\x55")) == 16
+
+    # Test case 5: Fingerprint with alternating bits set to 1 and 0
+    assert bit_count(bytearray(b"\xAA\x55\xAA\x55")) == 16