fix(files): resolve an export asset by its stored id, rewrite by its spelling

An embed carries two representations and they are not interchangeable: metadata
resolves by the stored id, while the rewrite finds the embed by searching the
document for the spelling it used. Using one for both either drops a
percent-encoded asset or bundles it behind a link still pointing at the API.

Author: waleedlatif1

apps/sim/app/api/files/export/[id]/route.test.ts

@@ -161,6 +161,29 @@ describe('markdown export bundling', () => {
     expect(zip.file('assets/bad.png')).toBeNull()
   })
 
+  /**
+   * The two id representations have to stay distinct: metadata resolves by the stored id, while the
+   * rewrite finds the embed by the spelling the document used. Collapsing them either drops the
+   * asset or bundles it behind a link still pointing at the API.
+   */
+  it('resolves and rewrites an embed whose id is percent-encoded in the document', async () => {
+    embeds('wf%5Fa')
+    assetsResolveTo((id) => (id === 'wf_a' ? assetRecord(id, 1 * MB) : null))
+    mockDownloadFile.mockImplementation(async ({ key }: { key: string }) =>
+      key.endsWith('doc.md')
+        ? Buffer.from('# Doc\n![x](/api/files/view/wf%5Fa)\n')
+        : Buffer.from('png-bytes')
+    )
+
+    const response = await GET(request(), context)
+
+    const zip = await JSZip.loadAsync(Buffer.from(await response.arrayBuffer()))
+    expect(zip.file('assets/wf_a.png')).not.toBeNull()
+    const md = await zip.file('doc.md')?.async('string')
+    expect(md).toContain('./assets/wf_a.png')
+    expect(md).not.toContain('/api/files/view/')
+  })
+
   it('skips an asset the caller cannot read', async () => {
     embeds('secret')
     mockVerifyFileAccess.mockImplementation(async (key: string) => !key.endsWith('secret'))

apps/sim/app/api/files/export/[id]/route.ts

@@ -51,6 +51,19 @@ function safeFilename(name: string): string {
     .replace(/[\r\n\t]/g, '')
 }
 
+/**
+ * The stored id behind an embed's spelling. Ids arrive spelled as the document writes them, because
+ * the rewrite below locates each embed by searching for that spelling; metadata lookup needs the
+ * decoded id instead, so the two representations are kept distinct rather than reconciled.
+ */
+function storedId(spelledId: string): string {
+  try {
+    return decodeURIComponent(spelledId)
+  } catch {
+    return spelledId
+  }
+}
+
 function deduplicatedFilename(preferred: string, existing: Set<string>, imageId: string): string {
   if (!existing.has(preferred)) return preferred
   const ext = path.extname(preferred)
@@ -160,7 +173,7 @@ export const GET = withRouteHandler(
     const assetTargets = (
       await mapWithConcurrency(imageIds, MATERIALIZE_CONCURRENCY, async (imageId) => {
         try {
-          const imgRecord = await getFileMetadataById(imageId)
+          const imgRecord = await getFileMetadataById(storedId(imageId))
           if (!imgRecord) return null
           if (!(await verifyFileAccess(imgRecord.key, userId))) return null
           return { imageId, record: imgRecord }