CONCD-599 unit test for the Asset.get_ocr_transcript() method

LibraryOfCongress · Apr 18, 2024 · 6d0ec9f · 6d0ec9f
1 parent 61cc224
commit 6d0ec9f
Show file tree

Hide file tree

Showing 2 changed files with 8 additions and 0 deletions.
diff --git a/concordia/tests/data/test-european.jpg b/concordia/tests/data/test-european.jpg
diff --git a/concordia/tests/test_models.py b/concordia/tests/test_models.py
@@ -1,7 +1,9 @@
 from datetime import timedelta
 
+import pytesseract
 from django.test import TestCase
 from django.utils import timezone
+from PIL import Image
 
 from concordia.models import Campaign, Transcription, UserProfileActivity
 from concordia.utils import get_anonymous_user
@@ -20,6 +22,12 @@ def setUp(self):
             reviewed_by=anon,
         )
 
+    def get_ocr_transcript(self):
+        image = Image.open("concordia/tests/data/test-european.jpg")
+        phrase = "marrón rápido salta sobre el perro"
+        self.assertFalse(phrase in pytesseract.image_to_string(image))
+        self.assertTrue(phrase in pytesseract.image_to_string(image, lang="spa"))
+
     def test_get_contributor_count(self):
         self.assertEqual(self.asset.get_contributor_count(), 2)