40 lines
1.4 KiB
Markdown
40 lines
1.4 KiB
Markdown
# Phase 0 – Generic Platform Foundation
|
||
|
||
This package introduces the new platform namespace `at.procon.dip` without breaking the existing
|
||
TED runtime under `at.procon.ted`.
|
||
|
||
## New project identity
|
||
- Project name: **Procon Document Intelligence Platform**
|
||
- Short name: **DIP**
|
||
- Base namespace: `at.procon.dip`
|
||
|
||
## Intent
|
||
Phase 0 is intentionally light-weight. It defines the canonical vocabulary and SPI contracts that
|
||
later phases will implement incrementally:
|
||
- generic document root model
|
||
- optional owner tenant + explicit document visibility/access model
|
||
- ingestion adapters
|
||
- type detection
|
||
- extraction
|
||
- text normalization
|
||
- vectorization provider abstraction
|
||
- generic search scope
|
||
|
||
## Access model
|
||
Documents are no longer assumed to be always tenant-owned.
|
||
|
||
Examples:
|
||
- public TED notice -> `ownerTenant = null`, `visibility = PUBLIC`
|
||
- tenant-owned private document -> `ownerTenant = tenantA`, `visibility = TENANT`
|
||
|
||
This keeps ownership and access semantics separate from the beginning of the generalized model.
|
||
|
||
## Compatibility
|
||
The new Spring Boot entry point is `at.procon.dip.DocumentIntelligencePlatformApplication` and it
|
||
explicitly scans the legacy TED packages so the current runtime remains operational while future
|
||
phases migrate modules gradually.
|
||
|
||
|
||
## Phase 1 note
|
||
The additive `DOC` schema and generic persistence services are introduced in `README_PHASE1.md`.
|