[AdaWorldAPI]

Summary

• Researched papermerge/papermerge-core (the real source — ciur/papermerge is a now-empty meta-repo, worth knowing if anyone else goes looking) for UX-facing architectural lessons applicable to ogar-doc-ir, scoped deliberately to exclude infra concerns (search indexing, folders/permissions, the task-queue boundary — those are archive-layer, not this crate's).
• Adds docs/DOCIR-UX-HARVEST.md: two proposals, each a doc.v2-scale design sketch (not a live code change) —
  1. Page identity — DocPage today is addressed only by Vec position, no stable id. Papermerge's actual mechanism (real citations, not README paraphrase): a structural page edit mints a whole new document version and copies forward unaffected pages' already-recognized content by explicit old-id→new-id mapping, never re-running OCR. Proposes PageId + PageOp + CarriedForward.
  2. Document type + field schema — proposes FieldKind/FieldSchema/DocumentType, mirroring papermerge's converged "one JSONB value + typed generated columns" design for custom fields, to validate a document's TypedFields against an expected schema.
• Appends a D-DOCIR-UX-HARVEST entry to docs/DISCOVERY-MAP.md.
• Caught mid-write: origin/main had just landed the document_actions.rs W4 council ruling (no separate ogar-doc crate — persistence ActionDefs live in ogar-vocab::document_actions, executed by tesseract-paperless/paperless-kv). Synced to main first, then corrected every reference in the new doc that assumed a nonexistent ogar-doc crate, and re-verified the two struct-citation line numbers against the post-sync file.

Test plan

• Every citation checked against real papermerge-core source (file:line), not the README
• Every ogar-doc-ir citation re-verified against the current (post-sync) lib.rs line numbers
• Confirmed no stale `ogar-doc` crate references remain after the sync-driven correction
• docs/DISCOVERY-MAP.md diff reviewed — purely additive, one entry appended at the true tail, upstream's own new content intact
• N/A — documentation only, no code changed, no build/test surface affected

🤖 Generated with Claude Code

---

Generated by Claude Code

[coderabbitai]

Important

Draft PR not reviewed

Draft PRs are not automatically reviewed by default.

• Trigger a manual review

To automatically review draft PRs, update your CodeRabbit configuration:

reviews:
  auto_review:
    drafts: true

---

Thanks for using CodeRabbit! It's free for OSS, and your support helps us grow. If you like it, consider giving us a shout-out.

❤️ Share

• X
• Mastodon
• Reddit
• LinkedIn

_{Comment @coderabbitai help to get the list of available commands.}

[cursor]

Bugbot couldn't run - usage limit reached

Bugbot is counted against Cursor usage for this user or team, and this run hit a usage or spend limit.

A user or team admin can review and increase usage limits in the Cursor dashboard.

(requestId: serverGenReqId_a9eb7579-2f42-42e1-ad69-3cfc3d8fdc1d)

[AdaWorldAPI]

Closing — wrong layer. ogar-doc-ir is an IR: perceptual structure, source-agnostic, carries facts not policy. This doc proposed page-editing capabilities and document-type metadata, which is UX, and UX belongs in paperless-rs. The IR would only take a finding like this if it were a render primitive — something the IR must represent so a renderer can draw it — and nothing here is that.

Re-doing it in the right repo, aimed at usability rather than data-model archaeology.

---

Generated by Claude Code