[WIP] Monadic refactoring

image_prediction/image_extractor/extractors/parsable.py

@@ -55,15 +55,12 @@ class ParsablePDFImageExtractor(ImageExtractor):
 
         maybe_image_metadata_pairs = map(partial(metadatum_to_image_metadata_pair, self.doc), metadata)
         image_metadata_pairs = [pair.value for pair in maybe_image_metadata_pairs if pair.is_just()]
-        print(image_metadata_pairs)
         clear_caches()
 
         #  TODO: In the future, consider to introduce an image validator as a pipeline component rather than doing the
         #   validation here. Invalid images can then be split into a different stream and joined with the intact images
         #   again for the formatting step.
         image_metadata_pairs = self.__filter_valid_images(image_metadata_pairs)
-        image_metadata_pairs = list(image_metadata_pairs)
-        print(image_metadata_pairs)
 
         image_metadata_pairs = stitch_pairs(list(image_metadata_pairs), tolerance=self.tolerance)
 
@@ -136,7 +133,8 @@ def xref_to_maybe_image(doc, xref) -> Maybe:
 
 
 def make_maybe_image_metadata_pair(image: Maybe, metadata: Maybe):
-    return Just(image.bind(curry(2, make_image_metadata_pair))).amap(metadata)
+    f = image.bind(lambda img: Just(lambda mdt: ImageMetadataPair(img, mdt)))
+    return f.amap(metadata)
 
 
 def make_image_metadata_pair(image: Image.Image, metadatum: dict) -> Just:

test/unit_tests/image_extractor_test.py

@@ -98,4 +98,4 @@ def test_bad_xref_handling(bad_xref_pdf, dvc_test_data):
     with pytest.raises(BadXref):
         extract_image(doc, xref)
 
-    assert xref_to_maybe_image(doc, xref) is None
+    assert xref_to_maybe_image(doc, xref).is_nothing()