Bound ChatGPT DOM extraction time

src/capture.ts

@@ -207,6 +207,19 @@ async function waitForCdp(port: number, process: ChildProcess): Promise<void> {
   throw new Error("Chromium CDP endpoint did not become ready.");
 }
 
+async function withTimeout<T>(promise: Promise<T>, timeoutMs: number, label: string): Promise<T> {
+  let timeout: NodeJS.Timeout | undefined;
+  const timeoutPromise = new Promise<never>((_resolve, reject) => {
+    timeout = setTimeout(() => reject(new Error(`${label} timed out after ${timeoutMs}ms.`)), timeoutMs);
+  });
+  try {
+    return await Promise.race([promise, timeoutPromise]);
+  } finally {
+    if (timeout) clearTimeout(timeout);
+    promise.catch(() => undefined);
+  }
+}
+
 async function stopProcess(process: ChildProcess): Promise<void> {
   if (process.exitCode !== null || process.signalCode !== null) return;
   process.kill("SIGTERM");
@@ -274,7 +287,7 @@ async function extractConversation(sourceUrl: string): Promise<ExtractedConversa
   await page.waitForLoadState("networkidle", { timeout: 5_000 }).catch(() => undefined);
   await page.waitForTimeout(1000);
 
-  const extracted = await page.evaluate(() => {
+  const extracted = await withTimeout(page.evaluate(() => {
     function cleanHtml(root: Element): string {
       const clone = root.cloneNode(true) as Element;
       clone.querySelectorAll("script, style, iframe, object, embed, form, button, textarea, input, nav").forEach((node) => node.remove());
@@ -331,7 +344,7 @@ async function extractConversation(sourceUrl: string): Promise<ExtractedConversa
     }
 
     return { title, messages };
-  });
+  }), config.captureTimeoutMs, "Extracting ChatGPT share DOM");
 
   if (!extracted.messages.length) {
     throw new Error("Unsupported ChatGPT share page: no conversation messages were found.");