PRD Implementation

[AzvenusK]

Problem

Need to implement the DPP in line with PRD_DPP.md and the following two condition:

• As a Data‑Responsible Manager, we want to see a list of available DPP schemas by sector so we can understand which ones apply to our products.
• As a software developer for a hypothetical manufacturing company, we want tooling to automate DPP creation for this company's product line.

Action-plan of Implementation

Goal

Make schemas discoverable by sector and DPP type, provide a developer‑friendly DPP creation pipeline, and complete domain mapping heuristics.

---

1. Schema metadata and sector listing

• Add config/schemas.yml with, per schema: #5
• Extend SchemaRegistry to load config/schemas.yml catalog #6
• Add sector‑facing APIs in SchemaRegistry: #7

---

2. Developer‑friendly DPP creation pipeline

• Create adapters.py (or similar) with input adapters: #8
• Implement concrete SchemaMapper subclasses: #9
• Wire mappers into register_default_mappers(): #10
• Add high‑level facade api.py: #11
• Export create_dpp from __init__.py for easy developer access.

---

3. Mapping Solution for Ontology

• Mapping solution complete.

---

4. Turn PRD “starter kits” into code

• Add config/starter_kits.yml:
  • Each entry includes: name, sector, input_schema, output_schema, and mapping/config files to use (aligned with PRD starter kits).
• Extend api.py with helper functions:
  • list_starter_kits() -> List[Dict[str, Any]]
  • run_starter_kit(kit_name: str, files: List[Path]) -> dict
  • run_starter_kit should:
    • Look up kit configuration.
    • Call create_dpp with appropriate schemas, adapters, and config.
• (Later) Extend CLI to expose:
  • dpp list-schemas --sector battery
  • dpp list-starter-kits
  • dpp run-starter-kit battery_isa95 --input ./data.

---

5. Minimal sequencing / milestones

• Short term:
  • Fix ECLASS domain mapping (Section 3).
  • Implement ECLASSMapper and register it.
  • Implement create_dpp for input_schema="ECLASS" and output_schema="ECLASS" or “Generic DPP (ECLASS JSON‑LD)”.
• Next:
  • Implement schema metadata (schemas.yml) and list_schemas_by_sector for the manager persona.
  • Implement one concrete starter kit (e.g. Generic DPP from ECLASS) and wire it into run_starter_kit.
• Later:
  • Add ISA‑95 adapters and ISA95Mapper.
  • Add Battery DPP starter kit and richer validation paths.

This gives a vertical slice where managers can discover schemas by sector and developers can generate a DPP for an ECLASS‑based product line with minimal glue. Which of these sections (1–5) do you want to turn into an actual PR first?