Allow extracting elements holding transcriptions without entities

dan/datasets/extract/arkindex.py

@@ -116,6 +116,9 @@ class ArkindexExtractor:
             supported_types=list(self.tokens),
         )
 
+        if not entities.count():
+            return transcription.text.strip()
+
         return self.translate(
             entities_to_xml(
                 transcription.text, entities, entity_separators=self.entity_separators

tests/test_extract.py

@@ -425,6 +425,33 @@ def test_empty_transcription(allow_empty, mock_database):
             extractor.extract_transcription(element_no_transcription)
 
 
+def test_extract_transcription_no_tokens(mock_database):
+    extractor = ArkindexExtractor(
+        element_type=["text_line"],
+        entity_separators=None,
+        # No tokens provided to the extractor
+        tokens=None,
+    )
+    element = Element.get_by_id("test-page_1-line_1")
+    assert extractor.extract_transcription(element) == "Coupez  Bouis  7.12.14"
+
+
+def test_extract_transcription_without_entities(mock_database):
+    extractor = ArkindexExtractor(
+        element_type=["text_line"],
+        entity_separators=None,
+        tokens=EXTRACTION_DATA_PATH / "tokens.yml",
+    )
+    element = Element.get_by_id("test-page_1-line_1")
+    # Deleting all entities on the element transcriptions while leaving the transcriptions intact
+    TranscriptionEntity.delete().where(
+        TranscriptionEntity.transcription.in_(
+            Transcription.select().where(Transcription.element == element)
+        )
+    ).execute()
+    assert extractor.extract_transcription(element) == "Coupez  Bouis  7.12.14"
+
+
 @pytest.mark.parametrize(
     "nestation, xml_output, separators",
     (