fix(knowledge): prevent documentId fallthrough and use byte-count limit

- Use if/else so filename lookup only runs when no documentId is provided,
  preventing stale IDs from silently replacing unrelated documents
- Check utf8 byte length instead of character count for 1MB size limit,
  correctly handling multi-byte characters (CJK, emoji)

Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>

Author: waleedlatif1

apps/sim/app/api/knowledge/[id]/documents/upsert/route.ts

@@ -103,9 +103,7 @@ export async function POST(req: NextRequest, { params }: { params: Promise<{ id:
       if (existingDoc.length > 0) {
         existingDocumentId = existingDoc[0].id
       }
-    }
-
-    if (!existingDocumentId) {
+    } else {
       const docsByFilename = await db
         .select({ id: document.id })
         .from(document)

apps/sim/tools/knowledge/upsert_document.ts

@@ -77,12 +77,12 @@ export const knowledgeUpsertDocumentTool: ToolConfig<
       if (!textContent || textContent.length < 1) {
         throw new Error('Document content cannot be empty')
       }
-      if (textContent.length > 1000000) {
-        throw new Error('Document content exceeds maximum size of 1MB')
-      }
-
       const utf8Bytes = new TextEncoder().encode(textContent)
       const contentBytes = utf8Bytes.length
+
+      if (contentBytes > 1_000_000) {
+        throw new Error('Document content exceeds maximum size of 1MB')
+      }
       let base64Content: string
       if (typeof Buffer !== 'undefined') {
         base64Content = Buffer.from(textContent, 'utf8').toString('base64')