Merge pull request #1 from explosion/add-tests

Adding unit tests

.github/workflows/tests.yml

@@ -0,0 +1,52 @@
+name: Unit Tests
+
+on:
+  pull_request:
+  push:
+    branches:
+      - main
+
+jobs:
+  setup:
+    runs-on: ubuntu-latest
+    steps:
+      - uses: actions/checkout@v3
+      - name: Set up Python 3.9
+        uses: actions/setup-python@v4
+        with:
+          python-version: "3.10"
+          cache: "pip" # caching pip dependencies
+
+      - name: Setup ffmpeg
+        uses: FedericoCarboni/setup-ffmpeg@v2
+        id: setup-ffmpeg
+
+      - name: Check out Prodigy
+        uses: actions/checkout@v3
+        with:
+          repository: explosion/prodigy
+          ref: v1.14.0
+          path: ./prodigy
+          ssh-key: ${{ secrets.GHA_PRODIGY_READ }}
+
+      - name: Install Prodigy
+        run: |
+          ls -la
+          pip install ./prodigy
+          python -m prodigy --help
+
+      - name: Install dependencies
+        run: |
+          pip install --upgrade pip
+          pip install -e .
+          pip install ruff pytest
+      
+      - name: Run ruff
+        if: always()
+        shell: bash
+        run: python -m ruff prodigy_whisper tests
+
+      - name: Run pytest
+        if: always()
+        shell: bash
+        run: python -m pytest tests

README.md

@@ -10,6 +10,8 @@ You can install this plugin via `pip`.
 pip install "prodigy-whisper @ git+https://github.com/explosion/prodigy-whisper"
 ```
 
+You'll also want to make sure that `ffmpeg` is installed before running these recipes. 
+
 To learn more about this plugin, you can check the [Prodigy docs](https://prodi.gy/docs/plugins/#whisper).
 
 ## Issues? 

prodigy_whisper/__init__.py

@@ -123,7 +123,6 @@ def add_transcript_input_metadata(stream: StreamType) -> StreamType:
             "show_audio_timeline": segment,
             "show_audio_minimap": not segment,
             "audio_bar_width": 1,
-            "javascript": "console.log('this works yo');"
         },
     }
 

pyproject.toml

@@ -0,0 +1,2 @@
+[tool.ruff]
+line-length = 120

tests/test_smoke.py

@@ -0,0 +1,16 @@
+from prodigy_whisper import whisper_audio_transcribe
+
+
+def test_smoke_no_segment(tmpdir):
+    # Make sure we can train without errors
+    components = whisper_audio_transcribe("xxx", "audio", model="tiny.en")
+    assert "transcript" in next(components["stream"])
+
+
+def test_smoke_segment(tmpdir):
+    # Make sure we can train without errors
+    components = whisper_audio_transcribe("xxx", "audio", model="tiny.en", segment=True)
+    for ex in components["stream"]:
+        assert "transcript" in ex
+        assert ex["meta"]["path"] == "audio/talking.mp3"
+        assert isinstance(ex["meta"]["segment_id"], int)