Allow no transcription unit test

docs/usage/datasets/extract.md

@@ -28,7 +28,7 @@ If an image download fails for whatever reason, it won't appear in the transcrip
 | `--max-height`                   | Images larger than this height will be resized to this height.                                                                                                                                                                       | `int`           |                                                    |
 | `--keep-spaces`                  | Transcriptions are trimmed by default. Use this flag to disable this behaviour.                                                                                                                                                      | `bool`          | False                                              |
 | `--image-format`                 | Images will be saved under this format.                                                                                                                                                                                              | `str`           | `.jpg`                                             |
-| `--allow-empty`                  | Elements with no transcriptions are skipped by default. This flag disables this behaviour.                                                                                                                                           | `bool`          | False                                |
+| `--allow-empty`                  | Elements with no transcriptions are skipped by default. This flag disables this behaviour.                                                                                                                                           | `bool`          | False                                              |
 
 The `--tokens` argument expects a YAML-formatted file with a specific format. A list of entries with each entry describing a NER entity. The label of the entity is the key to a dict mapping the starting and ending tokens respectively.
 

tests/test_extract.py

@@ -12,7 +12,11 @@ import pytest
 from PIL import Image, ImageChops
 
 from arkindex_export import Element, Transcription
-from dan.datasets.extract.exceptions import NoEndTokenError, UnknownTokenInText
+from dan.datasets.extract.exceptions import (
+    NoEndTokenError,
+    NoTranscriptionError,
+    UnknownTokenInText,
+)
 from dan.datasets.extract.extract import IIIF_FULL_SIZE, ArkindexExtractor
 from dan.datasets.extract.utils import EntityType, insert_token, remove_spaces
 from dan.utils import parse_tokens
@@ -487,3 +491,26 @@ def test_download_image_error(iiif_url, caplog, capsys):
     # Check stdout
     captured = capsys.readouterr()
     assert captured.out == "deadbeef: Image URL must be HTTP(S)\n"
+
+
+@pytest.mark.parametrize("allow_empty", (True, False))
+def test_empty_transcription(allow_empty, mock_database):
+    extractor = ArkindexExtractor(
+        folders=["train", "val", "test"],
+        element_type=["text_line"],
+        parent_element_type="double_page",
+        output=None,
+        entity_separators=None,
+        tokens=None,
+        transcription_worker_version=None,
+        entity_worker_version=None,
+        keep_spaces=False,
+        image_extension=".jpg",
+        allow_empty=allow_empty,
+    )
+    element_no_transcription = Element(id="unknown")
+    if allow_empty:
+        assert extractor.extract_transcription(element_no_transcription) == ""
+    else:
+        with pytest.raises(NoTranscriptionError):
+            extractor.extract_transcription(element_no_transcription)