release_identity.py

from __future__ import annotations

import hashlib
import json
import os
from pathlib import Path
from typing import Any


BEHAVIOR_PATHS = (
    "answering.py",
    "engine.py",
    "clause_retrieval.py",
    "normative_runtime.py",
    "normative_status.py",
    "release_identity.py",
    "source_router.py",
    "proof_bundle.py",
    "hybrid_retrieval.py",
    "query_understanding.py",
    "clarification.py",
    "conversation_intent.py",
    "sources.py",
    "llm.py",
    "decision_runtime.py",
    "normative_metadata.py",
    "prompts.py",
    "config.py",
    "ui_gradio.py",
    "institution_onboarding.py",
    "normative_contract.py",
    "ontology.py",
    "rdfox_pilot.py",
    "requirements.txt",
    "data/registry/document_semantic_profiles.json",
    "data/registry/normative_decision_contracts.json",
    "data/registry/normative_relations.json",
    "data/registry/legal_documents.json",
    "data/mckf/mckf_shapes.ttl",
)


def compute_release_manifest(root: Path, corpus_build_id: str) -> dict[str, Any]:
    artifacts: dict[str, dict[str, Any]] = {}
    for relative in BEHAVIOR_PATHS:
        path = root / relative
        if not path.exists():
            continue
        payload = path.read_bytes()
        artifacts[relative] = {
            "sha256": hashlib.sha256(payload).hexdigest(),
            "bytes": len(payload),
        }
    models = {
        "provider": os.getenv("LLM_PROVIDER", "auto"),
        "hf": os.getenv("HF_MODEL") or os.getenv("GEN_MODEL", "Qwen/Qwen2.5-7B-Instruct"),
        "groq": os.getenv("GROQ_MODEL", "openai/gpt-oss-120b"),
        "openai": os.getenv("OPENAI_MODEL", "gpt-4o-mini"),
    }
    material = json.dumps(
        {"corpus_build_id": corpus_build_id, "artifacts": artifacts, "models": models},
        ensure_ascii=False,
        sort_keys=True,
        separators=(",", ":"),
    )
    release_id = "mitranlil-" + hashlib.sha256(material.encode("utf-8")).hexdigest()[:20]
    return {
        "schema": "MCKF-RuntimeRelease-v1.0",
        "release_id": release_id,
        "corpus_build_id": corpus_build_id,
        "behavior_artifacts": artifacts,
        "configured_models": models,
    }