first commit

tests/__init__.py

@@ -0,0 +1 @@
+# Package de tests unitaires

tests/conftest.py

@@ -0,0 +1,8 @@
+import sys
+from unittest.mock import MagicMock
+
+# Mock global pour éviter le besoin d'installer des paquets lourds en local pour les tests unitaires
+sys.modules['easyocr'] = MagicMock()
+sys.modules['transformers'] = MagicMock()
+sys.modules['torch'] = MagicMock()
+sys.modules['torchvision'] = MagicMock()

tests/test_organizer.py

@@ -0,0 +1,83 @@
+import pytest
+from pathlib import Path
+from unittest.mock import MagicMock, patch
+from tweet_classifier.organizer import ImageThemeOrganizer
+
+@pytest.fixture
+def temp_workspace(tmp_path):
+    """
+    Crée une structure temporaire pour les tests.
+    """
+    input_dir = tmp_path / "inputs"
+    input_dir.mkdir()
+    
+    # Créer quelques fichiers d'images factices
+    (input_dir / "tweet1.png").write_text("fake image content 1")
+    (input_dir / "tweet2.jpg").write_text("fake image content 2")
+    (input_dir / "not_an_image.txt").write_text("some text")
+    
+    return input_dir
+
+@patch("tweet_classifier.organizer.CSVDatabaseManager")
+@patch("tweet_classifier.ocr.easyocr.Reader")
+@patch("tweet_classifier.classifier.pipeline")
+def test_scan_new_files(mock_pipeline, mock_reader, mock_db_class, temp_workspace):
+    """
+    Vérifie que l'organisateur scanne les fichiers et les ajoute à la base de données.
+    """
+    mock_db = MagicMock()
+    mock_db_class.return_value = mock_db
+    
+    organizer = ImageThemeOrganizer(input_dir=temp_workspace)
+    
+    organizer.scan_new_files()
+    
+    # Vérifier que add_files a été appelé avec les bonnes images
+    assert mock_db.add_files.called
+    added_files = mock_db.add_files.call_args[0][0]
+    assert len(added_files) == 2
+    assert any(img.name == "tweet1.png" for img in added_files)
+    assert any(img.name == "tweet2.jpg" for img in added_files)
+
+@patch("tweet_classifier.organizer.CSVDatabaseManager")
+@patch("tweet_classifier.ocr.easyocr.Reader")
+@patch("tweet_classifier.classifier.pipeline")
+def test_process_record_success(mock_pipeline, mock_reader, mock_db_class, temp_workspace):
+    """
+    Vérifie le traitement réussi d'un enregistrement DB.
+    """
+    mock_db = MagicMock()
+    mock_db_class.return_value = mock_db
+    
+    mock_reader_instance = MagicMock()
+    mock_reader.return_value = mock_reader_instance
+    mock_reader_instance.readtext.return_value = [
+        (None, "Ceci est un tweet agressif", None)
+    ]
+    
+    mock_pipeline_instance = MagicMock()
+    mock_pipeline.return_value = mock_pipeline_instance
+    mock_pipeline_instance.return_value = {
+        "labels": ["Insulte"],
+        "scores": [0.9]
+    }
+    
+    organizer = ImageThemeOrganizer(input_dir=temp_workspace)
+    
+    image_path = temp_workspace / "tweet1.png"
+    record = {
+        "filename": "tweet1.png",
+        "filepath": str(image_path)
+    }
+    
+    organizer._process_record(record)
+
+    # Vérifier le déplacement
+    dest_path = temp_workspace / "ok" / "Insulte" / "tweet1.png"
+    assert dest_path.exists()
+
+    # Vérifier la mise à jour DB : le nouvel emplacement doit être enregistré
+    mock_db.update_file_status.assert_called_with(
+        "tweet1.png", 'processed', "Ceci est un tweet agressif", "Insulte", 0.9,
+        new_filepath=str(dest_path)
+    )

tests/test_web_generator.py

@@ -0,0 +1,19 @@
+import pytest
+from pathlib import Path
+from tweet_classifier.web_generator import WebReportGenerator
+
+def test_web_generator_finds_csv_in_output_dir(tmp_path):
+    output_dir = tmp_path / "ok"
+    output_dir.mkdir()
+    
+    # Create a CSV in output_dir
+    csv_file = output_dir / "my_tweets.csv"
+    csv_file.write_text("filepath,status,detected_category,confidence,ocr_text\n/path/to/img,processed,Cat,0.9,Text")
+    
+    # Instantiate with a non-existent path but pointing to the output_dir
+    # The WebReportGenerator will now check in output_dir
+    generator = WebReportGenerator(csv_path=Path("my_tweets.csv"), output_dir=output_dir)
+    
+    # This should now point to output_dir / "my_tweets.csv"
+    assert generator.csv_path.exists()
+    assert generator.csv_path == csv_file