Toxic PDFs crashing your RAG pipeline? đź’Ą
OpenCrawling now runs Apache Tika 4.1.0 with PipesForkParser process isolation!
• Isolated child JVMs
• 30s watchdog timeouts
• Bounded heaps
• Zero-Trust Zip Slip defense
@pjlucidi.bsky.social
Explore
@pjlucidi.bsky.social