From 2fb8a3c8dc427a8f6f208c7f23a36b054df6580f Mon Sep 17 00:00:00 2001
From: rafaelmmiller <150964962+rafaelsideguide@users.noreply.github.com>
Date: Tue, 19 Nov 2024 10:04:42 -0300
Subject: [PATCH] fix schema

---
 apps/api/requests.http                           | 16 ++++++++++++++++
 .../scraper/scrapeURL/transformers/llmExtract.ts |  9 +++++++++
 2 files changed, 25 insertions(+)

diff --git a/apps/api/requests.http b/apps/api/requests.http
index 9fe36aa7..0e3b9206 100644
--- a/apps/api/requests.http
+++ b/apps/api/requests.http
@@ -58,4 +58,20 @@ content-type: application/json
 {
   "url": "firecrawl.dev",
   "sitemapOnly": true
+}
+
+### Extract
+# @name extract
+POST {{baseUrl}}/v1/extract HTTP/1.1
+Authorization: Bearer {{$dotenv TEST_API_KEY}}
+content-type: application/json
+
+{
+  "urls": ["firecrawl.dev"],
+  "prompt": "What is the title, description and main product of the page?",
+  "schema": {
+    "title": "string",
+    "description": "string",
+    "mainProduct": "string"
+  }
 }
\ No newline at end of file
diff --git a/apps/api/src/scraper/scrapeURL/transformers/llmExtract.ts b/apps/api/src/scraper/scrapeURL/transformers/llmExtract.ts
index 1a5abd66..ab619423 100644
--- a/apps/api/src/scraper/scrapeURL/transformers/llmExtract.ts
+++ b/apps/api/src/scraper/scrapeURL/transformers/llmExtract.ts
@@ -108,6 +108,15 @@ export async function generateOpenAICompletions(logger: Logger, options: Extract
             required: ["items"],
             additionalProperties: false,
         };
+    } else if (schema && typeof schema === 'object' && !schema.type) {
+      schema = {
+          type: "object",
+          properties: Object.fromEntries(
+              Object.entries(schema).map(([key, value]) => [key, { type: value }])
+          ),
+          required: Object.keys(schema),
+          additionalProperties: false
+      };
     }
 
     schema = normalizeSchema(schema);