Skip to content

Improve analyzing large datarooms #16

Description

@mpstaton

Issue #1: Large Dataroom Performance - PDF Page Count Extraction (2025-11-30)

Status: Open - Needs Fix
Severity: High
Discovered: 2025-11-30 during Star-Catcher dataroom analysis

Problem Description

When processing large datarooms (100+ files, especially with many PDFs), the analyzer becomes extremely slow during the initial scan phase. The root cause is that _get_page_count() is called synchronously for every PDF file during inventory building.

Observed behavior with Star-Catcher dataroom:

  • 218 total files (152 PDFs, 10 Excel files)
  • Scanner opens every PDF with pypdf.PdfReader just to count pages
  • Malformed PDFs trigger hundreds of "Ignoring wrong pointing object" warnings
  • Each PDF open/parse adds latency
  • Total scan time: Several minutes for just the inventory phase

Metadata

Metadata

Assignees

Labels

bugSomething isn't working

Projects

Status
Ready

Relationships

None yet

Development

No branches or pull requests

Issue actions