From 257556f42483f3d1fe3296b4a66351c8fdfc8587 Mon Sep 17 00:00:00 2001
From: Devin AI <158243242+devin-ai-integration[bot]@users.noreply.github.com>
Date: Tue, 8 Apr 2025 23:21:25 +0000
Subject: [PATCH] Fix map endpoint including non-existent URLs in results

Co-Authored-By: Nicolas Camara <nick@sideguide.dev>
---
 apps/api/src/__tests__/snips/map.test.ts | 19 +++++++++++++++++++
 apps/api/src/controllers/v1/map.ts       |  4 +++-
 2 files changed, 22 insertions(+), 1 deletion(-)

diff --git a/apps/api/src/__tests__/snips/map.test.ts b/apps/api/src/__tests__/snips/map.test.ts
index 8c9ffe05..069f5417 100644
--- a/apps/api/src/__tests__/snips/map.test.ts
+++ b/apps/api/src/__tests__/snips/map.test.ts
@@ -31,4 +31,23 @@ describe("Map tests", () => {
     expect(response.body.success).toBe(true);
     expect(response.body.links.some(x => x.match(/^https:\/\/www\.hfea\.gov\.uk\/choose-a-clinic\/clinic-search\/results\/?\?options=\d+$/))).toBe(true);
   }, 60000);
+
+  it.concurrent("does not include non-existent child URLs in parent results", async () => {
+    const nonExistentPath = "/non-existent-path-" + Date.now(); // Use timestamp to ensure uniqueness
+    const childUrl = "https://firecrawl.dev" + nonExistentPath;
+    
+    await map({
+      url: childUrl,
+      useMock: "map-non-existent-child",
+    });
+    
+    const parentResponse = await map({
+      url: "https://firecrawl.dev",
+      useMock: "map-parent-after-child",
+    });
+    
+    expect(parentResponse.statusCode).toBe(200);
+    expect(parentResponse.body.success).toBe(true);
+    expect(parentResponse.body.links.some(link => link.includes(nonExistentPath))).toBe(false);
+  }, 30000);
 });
diff --git a/apps/api/src/controllers/v1/map.ts b/apps/api/src/controllers/v1/map.ts
index 49890d90..9a946c68 100644
--- a/apps/api/src/controllers/v1/map.ts
+++ b/apps/api/src/controllers/v1/map.ts
@@ -161,7 +161,9 @@ export async function getMapResults({
       );
       allResults = await Promise.all(pagePromises);
 
-      await redis.set(cacheKey, JSON.stringify(allResults), "EX", 48 * 60 * 60); // Cache for 48 hours
+      if (allResults.flat().filter(result => result !== null && result !== undefined).length > 0) {
+        await redis.set(cacheKey, JSON.stringify(allResults), "EX", 48 * 60 * 60); // Cache for 48 hours
+      }
     }
 
     // Parallelize sitemap index query with search results