SPIKE: Perform OCR/Text Recognition on All Document Uploads

[cholly75]

As a Court, so that I can quickly search and scan within a document, I need all uploaded PDF files to be OCR'd and/or scanned for text.

We can currently scrape text-ready PDFs using PDF.js

We do not currently have the ability to OCR PDF files that only consist of a flattened/image layer.

Pre-Conditions

Acceptance Criteria

• Non-order/opinion document text should not be incorporated into search indices

Notes

• Do not interfere with uploading mechanism - can/should do post-processing?
• Batch? On demand?
• Where will we store the data?
• Some pre-work has been done on this --- see here: Search: Practitioner-filed Brief Documents #8745 (comment)

Tasks

Test Cases

Definition of Done (Updated 2026-01-28)

Product Owner

• Acceptance criteria have been met and validated on the Court's test environment.
• Associated test cases defined in TestRail have been updated if necessary.
• Successful test run is performed in TestRail.
• User guides are updated if necessary.

UX

• All new functionality has been verified to work with keyboard navigation and screen reader software.
• UI should be touch optimized and responsive for external users.

Engineering

• Automated test scripts have been written, including visual tests for newly added PDFs.
• Successful test run is performed in TestRail.
• New screens have been added to cypress accessibility axe.
• Interactors should validate entities before calling persistence methods.
• Types have been added to all added and updated functions.
• Code refactored for clarity and to remove any known technical debt.
• Acceptance criteria for the story has been met.
• If there are special deployment instructions, they have been added to the CHANGES.md file and the PR description.
• Code that resides in the shared folder that only runs on the API or browser has been moved to either /web-client or /web-api.