From 192d056bef828addd811607a85c6852b4c04231a Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Gerg=C5=91=20M=C3=B3ricz?= <mo.geryy@gmail.com>
Date: Mon, 19 May 2025 21:36:15 +0200
Subject: [PATCH] feat(scrapeURL/pdf/mu): add timeout and created_at (#1570)

---
 apps/api/src/scraper/scrapeURL/engines/pdf/index.ts | 8 +++++++-
 1 file changed, 7 insertions(+), 1 deletion(-)

diff --git a/apps/api/src/scraper/scrapeURL/engines/pdf/index.ts b/apps/api/src/scraper/scrapeURL/engines/pdf/index.ts
index 29a07a5d..733b06df 100644
--- a/apps/api/src/scraper/scrapeURL/engines/pdf/index.ts
+++ b/apps/api/src/scraper/scrapeURL/engines/pdf/index.ts
@@ -27,6 +27,8 @@ async function scrapePDFWithRunPodMU(
     tempFilePath,
   });
 
+  const preCacheCheckStartTime = Date.now();
+
   try {
     const cachedResult = await getPdfResultFromCache(base64Content);
     
@@ -54,6 +56,8 @@ async function scrapePDFWithRunPodMU(
       input: {
         file_content: base64Content,
         filename: path.basename(tempFilePath) + ".pdf",
+        timeout: timeToRun ? timeToRun - (Date.now() - preCacheCheckStartTime) : undefined,
+        created_at: Date.now(),
       },
     },
     logger: meta.logger.child({
@@ -103,6 +107,8 @@ export async function scrapePDF(
   meta: Meta,
   timeToRun: number | undefined,
 ): Promise<EngineScrapeResult> {
+  const startTime = Date.now();
+  
   if (!meta.options.parsePDF) {
     if (meta.pdfPrefetch !== undefined && meta.pdfPrefetch !== null) {
       const content = (await readFile(meta.pdfPrefetch.filePath)).toString("base64");
@@ -167,7 +173,7 @@ export async function scrapePDF(
           }),
         },
         tempFilePath,
-        timeToRun,
+        timeToRun ? (timeToRun - (Date.now() - startTime)) : undefined,
         base64Content,
       );
     } catch (error) {