From d49f62fb5641ea123fcbba8d9eb41bcfeef74f1e Mon Sep 17 00:00:00 2001
From: Nicolas <nicolascamara29@gmail.com>
Date: Wed, 20 Nov 2024 11:50:14 -0800
Subject: [PATCH] Nick: extract fixes

---
 apps/api/src/__tests__/e2e_extract/index.test.ts          | 3 ++-
 apps/api/src/scraper/scrapeURL/transformers/llmExtract.ts | 4 +++-
 2 files changed, 5 insertions(+), 2 deletions(-)

diff --git a/apps/api/src/__tests__/e2e_extract/index.test.ts b/apps/api/src/__tests__/e2e_extract/index.test.ts
index 31f14f4a..346958da 100644
--- a/apps/api/src/__tests__/e2e_extract/index.test.ts
+++ b/apps/api/src/__tests__/e2e_extract/index.test.ts
@@ -86,7 +86,8 @@ describe("E2E Tests for Extract API Routes", () => {
           type: "array",
           items: {
             type: "string"
-          }
+          },
+          required: ["items"]
         },
       });
     expect(response.statusCode).toBe(200);
diff --git a/apps/api/src/scraper/scrapeURL/transformers/llmExtract.ts b/apps/api/src/scraper/scrapeURL/transformers/llmExtract.ts
index 7c2a6696..41d76779 100644
--- a/apps/api/src/scraper/scrapeURL/transformers/llmExtract.ts
+++ b/apps/api/src/scraper/scrapeURL/transformers/llmExtract.ts
@@ -165,7 +165,9 @@ export async function generateOpenAICompletions(logger: Logger, options: Extract
         }
     }
 
-    if (options.schema && options.schema.type === "array") {
+    // If the users actually wants the items object, they can specify it as 'required' in the schema
+    // otherwise, we just return the items array
+    if (options.schema && options.schema.type === "array" && !schema?.required?.includes("items")) {
         extract = extract?.items;
     }
     return { extract, warning };