Bug fix. (#191)

app/api/api_v1/routers/search.py

@@ -76,7 +76,7 @@ def _search_request(
                 data_access_search_response,
                 limit=search_body.limit,
                 offset=search_body.offset,
-            )
+            ).increment_pages()
         else:
             return _OPENSEARCH_CONNECTION.query_families(
                 search_request_body=search_body,
@@ -85,7 +85,7 @@ def _search_request(
                     db
                 ),
                 preference="default_search_preference",
-            )
+            ).increment_pages()
 
 
 @search_router.post("/searches")

app/api/api_v1/schemas/search.py

@@ -172,23 +172,25 @@ class SearchResponse(BaseModel):
 
     families: Sequence[SearchResponseFamily]
 
-    @validator("families", always=True)
-    @classmethod
-    def increment_pages(cls, value):
+    def increment_pages(self):
         """PDF page numbers must be incremented from our 0-indexed values."""
-        for family in value:
-            for family_document in family.family_documents:
-                for index, passage_match in enumerate(
+        for family_index, family in enumerate(self.families):
+            for family_document_index, family_document in enumerate(
+                family.family_documents
+            ):
+                for passage_match_index, passage_match in enumerate(
                     family_document.document_passage_matches
                 ):
                     if (
                         passage_match.text_block_page
                         or passage_match.text_block_page == 0
                     ):
-                        family_document.document_passage_matches[
-                            index
-                        ].text_block_page += 1
-        return value
+                        self.families[family_index].family_documents[
+                            family_document_index
+                        ].document_passage_matches[
+                            passage_match_index
+                        ].text_block_page += 1  # type: ignore
+        return self
 
 
 Top5FamilyList = conlist(SearchResponseFamily, max_items=5)

tests/unit/app/schemas/test_schemas.py

@@ -1,7 +1,13 @@
 import pytest
+from datetime import datetime
 
 from app.api.api_v1.schemas.document import FamilyDocumentResponse
-from app.api.api_v1.schemas.search import SearchResponseFamilyDocument
+from app.api.api_v1.schemas.search import (
+    SearchResponseDocumentPassage,
+    SearchResponseFamilyDocument,
+    SearchResponseFamily,
+    SearchResponse,
+)
 
 CLIMATE_LAWS_DOMAIN_PATHS = [
     "climate-laws.org",
@@ -98,3 +104,94 @@ def test_non_climate_laws_source_url_left_in_document(source_domain_path, scheme
         document_role=None,
     )
     assert document_response.source_url == given_url
+
+
+def test_search_response() -> None:
+    """
+    Test that instantiating Search Response objects is done correctly.
+
+    Particularly testing of the validators.
+    """
+    search_response = SearchResponse(
+        hits=1,
+        query_time_ms=1,
+        total_time_ms=1,
+        families=[
+            SearchResponseFamily(
+                family_slug="example_slug",
+                family_name="Example Family",
+                family_description="This is an example family",
+                family_category="Example Category",
+                family_date=str(
+                    datetime.now()
+                ),  # You can replace this with an actual date string
+                family_last_updated_date=str(
+                    datetime.now()
+                ),  # You can replace this with an actual date string
+                family_source="Example Source",
+                family_geography="Example Geography",
+                family_metadata={"key1": "value1", "key2": "value2"},
+                family_title_match=True,
+                family_description_match=False,
+                family_documents=[
+                    SearchResponseFamilyDocument(
+                        document_title="Document Title",
+                        document_slug="Document Slug",
+                        document_type="Executive",
+                        document_source_url="https://cdn.example.com/file.pdf",
+                        document_url=None,
+                        document_content_type="application/pdf",
+                        document_passage_matches=[
+                            SearchResponseDocumentPassage(
+                                text="Example",
+                                text_block_id="p_0_b_0",
+                                text_block_page=0,
+                                text_block_coords=None,
+                            ),
+                            SearchResponseDocumentPassage(
+                                text="Example",
+                                text_block_id="p_1_b_0",
+                                text_block_page=1,
+                                text_block_coords=None,
+                            ),
+                            SearchResponseDocumentPassage(
+                                text="Example",
+                                text_block_id="p_1_b_2",
+                                text_block_page=1,
+                                text_block_coords=None,
+                            ),
+                        ],
+                    )
+                ],
+            )
+        ],
+    )
+
+    first_document_initial_pages = [
+        page.text_block_page
+        for page in search_response.families[0]
+        .family_documents[0]
+        .document_passage_matches
+    ]
+
+    search_response_incremented = search_response.increment_pages()
+
+    first_document_incremented_pages = [
+        page.text_block_page
+        for page in search_response_incremented.families[0]
+        .family_documents[0]
+        .document_passage_matches
+    ]
+
+    assert len(first_document_initial_pages) == len(first_document_incremented_pages)
+
+    assert first_document_initial_pages != first_document_incremented_pages
+
+    expected_pages = []
+    for page in first_document_initial_pages:
+        if page is None:
+            expected_pages.append(page)
+        else:
+            expected_pages.append(page + 1)
+
+    assert expected_pages == first_document_incremented_pages