feat: Add spellcheck annotate (#1434)

* feat: 🔥 Add annotation spellcheck

* docs: 📝 Add API doc for spellcheck - additional data

* refactor: 🎨 Black

* test: 🧪 Add unittest and fix code

* docs: 📝 Update OpenAPI insights/annotate

doc/references/api.yml

@@ -329,7 +329,7 @@ paths:
         The annotation is an integer that can take 4 values: `0`, `1`, `2`, `-1`. `0` means the insight is incorrect
         (so it won't be applied), `1` means it is correct (so it will be applied) and `-1` means the insight
         won't be returned to the user (_skip_). `2` is used when user submit some data to the annotate endpoint 
-        (for example in some cases of category annotation).
+        (for example in some cases of category annotation or ingredients spellcheck).
         
         We use the voting mecanism system to remember which insight to skip for a user (authenticated or not).
       requestBody:
@@ -344,18 +344,23 @@ paths:
                   description: ID of the insight
                 annotation:
                   type: integer
-                  description: "Annotation of the prediction: 1 to accept the prediction, 0 to refuse it, and -1 for _skip_"
+                  description: "Annotation of the prediction: 1 to accept the prediction, 0 to refuse it, and -1 for _skip_, 2 to accept and add data"
                   enum:
                     - 0
                     - 1
                     - -1
+                    - 2
                 update:
                   type: integer
                   description: "Send the update to Openfoodfacts if `update=1`, don't send the update otherwise. This parameter is useful if the update is performed client-side"
                   default: 1
                   enum:
                     - 0
                     - 1
+                data:
+                  type: object
+                  description: "Additional data provided by the user as key-value pairs"
+
               required:
                 - "insight_id"
                 - "annotation"

robotoff/insights/annotate.py

@@ -19,6 +19,7 @@
     add_label_tag,
     add_packaging,
     add_store,
+    save_ingredients,
     select_rotate_image,
     unselect_image,
     update_emb_codes,
@@ -100,6 +101,12 @@ class AnnotationStatus(Enum):
     description="Open Food Facts update failed",
 )
 
+INVALID_DATA = AnnotationResult(
+    status_code=AnnotationStatus.error_invalid_data.value,
+    status=AnnotationStatus.error_invalid_data.name,
+    description="The data schema is invalid.",
+)
+
 
 class InsightAnnotator(metaclass=abc.ABCMeta):
     @classmethod
@@ -671,6 +678,38 @@ def is_data_required(cls) -> bool:
         return True
 
 
+class IngredientSpellcheckAnnotator(InsightAnnotator):
+    @classmethod
+    def process_annotation(
+        cls,
+        insight: ProductInsight,
+        data: Optional[dict] = None,
+        auth: Optional[OFFAuthentication] = None,
+        is_vote: bool = False,
+    ) -> AnnotationResult:
+        # Possibility for the annotator to change the spellcheck correction if data is provided
+        if data is not None:
+            annotation = data.get("annotation")
+            if not annotation or len(data) > 1:
+                return INVALID_DATA
+            # We add the new annotation to the Insight.
+            json_data = insight.data
+            json_data["annotation"] = annotation
+            insight.data = json_data
+            insight.save()
+
+        ingredient_text = data.get("annotation") if data else insight.data["correction"]
+        save_ingredients(
+            product_id=insight.get_product_id(),
+            ingredient_text=ingredient_text,
+            lang=insight.value_tag,
+            insight_id=insight.id,
+            auth=auth,
+            is_vote=is_vote,
+        )
+        return UPDATED_ANNOTATION_RESULT
+
+
 ANNOTATOR_MAPPING: dict[str, Type] = {
     InsightType.packager_code.name: PackagerCodeAnnotator,
     InsightType.label.name: LabelAnnotator,
@@ -683,6 +722,7 @@ def is_data_required(cls) -> bool:
     InsightType.nutrition_image.name: NutritionImageAnnotator,
     InsightType.nutrition_table_structure.name: NutritionTableStructureAnnotator,
     InsightType.is_upc_image.name: UPCImageAnnotator,
+    InsightType.ingredient_spellcheck.name: IngredientSpellcheckAnnotator,
 }
 
 

tests/integration/insights/test_annotate.py

@@ -1,6 +1,14 @@
+from unittest.mock import Mock
+
 import pytest
 
-from robotoff.insights.annotate import AnnotationResult, CategoryAnnotator
+from robotoff.insights.annotate import (
+    UPDATED_ANNOTATION_RESULT,
+    INVALID_DATA,
+    AnnotationResult,
+    CategoryAnnotator,
+    IngredientSpellcheckAnnotator,
+)
 from robotoff.models import ProductInsight
 
 from ..models_utils import ProductInsightFactory, clean_db
@@ -92,3 +100,63 @@ def test_process_annotation_with_invalid_user_input_data(self, user_data, mocker
             status="error_invalid_data",
             description="`data` is invalid, expected a single `value_tag` string field with the category tag",
         )
+
+
+class TestIngredientSpellcheckAnnotator:
+
+    @pytest.fixture
+    def mock_save_ingredients(self, mocker) -> Mock:
+        return mocker.patch("robotoff.insights.annotate.save_ingredients")
+
+    @pytest.fixture
+    def spellcheck_insight(self):
+        return ProductInsightFactory(
+            type="ingredient_spellcheck",
+            data={
+                "original": "List of ingredient",
+                "correction": "List fo ingredients",
+            },
+        )
+
+    def test_process_annotation(
+        self,
+        mock_save_ingredients: Mock,
+        spellcheck_insight: ProductInsightFactory,
+    ):
+        user_data = {"annotation": "List of ingredients"}
+        annotation_result = IngredientSpellcheckAnnotator.process_annotation(
+            insight=spellcheck_insight,
+            data=user_data,
+        )
+        assert annotation_result == UPDATED_ANNOTATION_RESULT
+        assert "annotation" in spellcheck_insight.data
+        mock_save_ingredients.assert_called()
+
+    @pytest.mark.parametrize(
+        "user_data",
+        [{}, {"annotation": "List of ingredients", "wrong_key": "wrong_item"}],
+    )
+    def test_process_annotation_invalid_data(
+        self,
+        user_data: dict,
+        mock_save_ingredients: Mock,
+        spellcheck_insight: ProductInsightFactory,
+    ):
+        annotation_result = IngredientSpellcheckAnnotator.process_annotation(
+            insight=spellcheck_insight,
+            data=user_data,
+        )
+        assert annotation_result == INVALID_DATA
+        mock_save_ingredients.assert_not_called()
+
+    def test_process_annotate_no_user_data(
+        self,
+        mock_save_ingredients: Mock,
+        spellcheck_insight: ProductInsightFactory,
+    ):
+        annotation_result = IngredientSpellcheckAnnotator.process_annotation(
+            insight=spellcheck_insight,
+        )
+        assert annotation_result == UPDATED_ANNOTATION_RESULT
+        assert "annotation" not in spellcheck_insight.data
+        mock_save_ingredients.assert_called()