Commit graph

2 commits

Author SHA1 Message Date
Suphonchai Phoonsawat
0b6ee8cc84 feat: add GET /api/documents/latest returning latest version file with stamped ticket
All checks were successful
Deploy eCourt Note Web to Server / Validate Database Schema (Drizzle) (push) Successful in 4s
Deploy eCourt Note Web to Server / Build Web Docker Image (push) Successful in 1m23s
Deploy eCourt Note Web to Server / Deploy eCourt Note Web to Server (push) Successful in 5s
Deploy eCourt Note Web to Server / Notify Deployment Status (push) Successful in 0s
Server-to-server endpoint for the e-Court backend (x-api-key against
DOCVER_API_KEY or EXTERNAL_API_KEY):
- type=pdf returns the latest version's PDF bytes (404 when absent)
- type=docx stamps fresh ticket custom properties (exp + HmacSha256 via
  addinTicketSecret(), Ed25519 when PRIVATE_KEY_HEX is set) into an
  in-memory copy before returning — MinIO object is never modified
- docId comes from fid inside the ticket_data query param, which must
  include fid/uid/uname so the stamped file stays redeemable by the
  add-in at /api/documents/token (buildStampedTicketData drops null
  values because parseDocverTicket rejects tid:null)
- adds parseTicketDataParam, parseLatestTypeParam, requireDocumentApiKey,
  findLatestVersionForDoc, getVersionObjectBytes + unit tests (320 pass)
- syncs PLAN-DOCVER.md, OpenAPI spec and .env.example

Co-Authored-By: Claude Code <noreply@anthropic.com>
2026-10-31 11:09:42 +07:00
Suphonchai Phoonsawat
cdae1ceedd Refactor document versioning to extract fields from uploaded .docx files on the server side
- Removed the `fields` parameter from the version upload form, as it is now extracted from the uploaded .docx file using the `extractContentControls` function.
- Updated related types and interfaces to reflect the new structure of version fields.
- Modified the OpenAPI specification to document the changes in the versioning process.
- Adjusted tests to ensure compliance with the new behavior, including validation for the absence of the `fields` parameter.
- Implemented a new function `extractDocumentSnapshot` to handle the extraction of content controls and document properties from .docx files.
- Updated metadata and versioning services to align with the new extraction logic.
2026-10-31 10:21:17 +07:00