Add typehints, make custom page quotient breach function private since the intention of outsourcing it from build_image_info is to make it testable seperately

image_prediction/image_extractor/extractors/parsable.py

@@ -5,7 +5,7 @@ import traceback
 from functools import partial, lru_cache
 from itertools import chain, starmap, filterfalse
 from operator import itemgetter, truth
-from typing import List
+from typing import List, Iterable, Iterator
 
 import fitz
 from PIL import Image
@@ -59,7 +59,7 @@ class ParsablePDFImageExtractor(ImageExtractor):
         yield from image_metadata_pairs
 
     @staticmethod
-    def __filter_valid_images(image_metadata_pairs):
+    def __filter_valid_images(image_metadata_pairs: Iterable[ImageMetadataPair]) -> Iterator[ImageMetadataPair]:
         def validate(image: Image.Image, metadata: dict):
             try:
                 # TODO: stand-in heuristic for testing if image is valid => find cleaner solution (RED-5148)

image_prediction/transformer/transformers/response.py

@@ -27,7 +27,7 @@ def build_image_info(data: dict) -> dict:
     quotient = round(compute_geometric_quotient(), 4)
 
     min_image_to_page_quotient_breached = bool(quotient < CONFIG.filters.image_to_page_quotient.min)
-    max_image_to_page_quotient_breached = is_max_image_to_page_quotient_breached(
+    max_image_to_page_quotient_breached = __is_max_image_to_page_quotient_breached(
         quotient, data["classification"]["label"]
     )
     min_image_width_to_height_quotient_breached = bool(
@@ -77,7 +77,7 @@ def build_image_info(data: dict) -> dict:
     return image_info
 
 
-def is_max_image_to_page_quotient_breached(quotient, label):
+def __is_max_image_to_page_quotient_breached(quotient: float, label: str) -> bool:
     default_max_quotient = CONFIG.filters.image_to_page_quotient.max
     customized_entries = CONFIG.filters.image_to_page_quotient.customized.max
     max_quotient = customized_entries.get(label, default_max_quotient)

test/unit_tests/response_transformer_test.py

@@ -1,6 +1,6 @@
 import pytest
 
-from image_prediction.transformer.transformers.response import is_max_image_to_page_quotient_breached
+from image_prediction.transformer.transformers.response import __is_max_image_to_page_quotient_breached
 
 
 @pytest.fixture
@@ -18,4 +18,4 @@ def expected_is_breached(quotient, label):
 @pytest.mark.parametrize("quotient", [0.1, 0.5])
 @pytest.mark.parametrize("label", ["logo", "signature"])
 def test_customized_per_label_ratio_breach(quotient, label, expected_is_breached):
-    assert is_max_image_to_page_quotient_breached(quotient, label) == expected_is_breached
+    assert __is_max_image_to_page_quotient_breached(quotient, label) == expected_is_breached