(converter) Add truncation att the parser step to prevent the converter from spending too much time on exessively large documents

code/processes/crawling-process/model/java/nu/marginalia/model/crawldata/CrawledDocument.java

@@ -59,9 +59,14 @@ public final class CrawledDocument implements SerializableCrawlData {
     }
 
     public Document parseBody() throws IOException {
+        // Prevent stalls from parsing excessively large documents
+
+        byte[] bytes = documentBodyBytes.length > 200_000
+                ? Arrays.copyOf(documentBodyBytes, 200_000) : documentBodyBytes;
+
         return DocumentBodyToString.getParsedData(
                 ContentType.parse(contentType),
-                documentBodyBytes,
+                bytes,
                 url);
     }