asf/scripts/seed-traceability.py

#!/usr/bin/env python3
"""Seed tenant t-demo-003 with 5 artifacts + 5 relationships for a rich Traceability Graph demo.

Endpoint testati via curl:
- POST /api/v1/artifacts
- POST /api/v1/relationships
- GET  /api/v1/traceability/graph
"""
from __future__ import annotations

import json
import subprocess
import sys
from typing import Any

BASE = "http://127.0.0.1:8000/api/v1"
H = [
    "-H", "X-ASF-Tenant-Id: t-demo-003",
    "-H", "X-ASF-User-Id: u-mario",
    "-H", "Content-Type: application/json",
]


def curl(method: str, path: str, body: dict[str, Any] | None = None) -> tuple[int, dict[str, Any] | list | str]:
    cmd = ["curl", "-s", "-m", "15", "-o", "-", "-w", "\n%{http_code}", "-X", method] + H
    if body is not None:
        cmd += ["--data", json.dumps(body, ensure_ascii=False)]
    cmd.append(f"{BASE}{path}")
    r = subprocess.run(cmd, capture_output=True, text=True)
    out_raw = r.stdout.strip() or "{}"
    # Last line is the http code we injected with -w
    if "\n" in out_raw:
        json_part, code_part = out_raw.rsplit("\n", 1)
    else:
        json_part, code_part = "", out_raw
    try:
        code = int(code_part)
    except ValueError:
        code = 0
    try:
        data = json.loads(json_part or "null")
    except Exception as exc:  # pragma: no cover
        data = f"JSON_ERR {exc} RAW={json_part[:200]!r}"
    print(f"  [{method:6s} {path:24s}] HTTP={code:3d}",
          f"type={data.get('type','?') if isinstance(data,dict) else '?'}\tid={str(data.get('id',''))[:10] if isinstance(data,dict) else '?'}",
          file=sys.stderr)
    return code, data


def main() -> int:
    print("# Step 1: list existing artifacts + project_id", file=sys.stderr)
    _, arts = curl("GET", "/artifacts?limit=50")
    assert isinstance(arts, list), f"artifacts list not a list: {arts!r}"
    req_103 = None
    project_id = None
    for a in arts:
        if isinstance(a, dict) and a.get("external_id") == "REQ-103":
            req_103 = a
            project_id = a.get("project_id")
            break
    if not req_103:
        # fallback: first artifact
        req_103 = arts[0] if arts else None
        project_id = req_103.get("project_id") if isinstance(req_103, dict) else None
    print(f"  req_103 id={str(req_103['id'])[:10] if isinstance(req_103,dict) else None} project_id={str(project_id)[:10] if project_id else None}",
          file=sys.stderr)

    def make(type_: str, ext: str, title: str, content: str) -> dict[str, Any]:
        code, data = curl(
            "POST", "/artifacts",
            {
                "type": type_,
                "external_id": ext,
                "title": title,
                "content": content,
                "status": "done",
                "project_id": project_id,
            },
        )
        assert code == 201 and isinstance(data, dict) and "id" in data, f"make {type_} FAILED HTTP={code} {data}"
        return data

    print(file=sys.stderr); print("# Step 2: create 5 extra artifacts", file=sys.stderr)
    task   = make("task",         "TASK-101", "Implementa tenant trust header coerce",       "FastAPI correlation middleware: UUID UUID5 per stringhe t-demo-xxx, no 400 invalid")
    pr     = make("pull_request", "PR#88",    "feat(rbac): tenant trust + policies v1",     "feat(rbac): add trust header middleware + policy service helpers")
    test   = make("test",         "TEST-22",  "cross-tenant reads/writes blocked",         "Pytest suite proving cross-tenant artifact operations return 404/403")
    build  = make("build",        "BUILD-910","asf-core build #910 (main)",                "GitHub Actions build main branch, all checks green.")
    deploy = make("deployment",   "DEP-44",   "asf-core v0.5.0 → prod-west-1",             "ArgoCD sync asf-core to cluster prod-west-1 eu-west")

    print(file=sys.stderr); print("# Step 3: create 5 relationships (chain: DEP → BUILD → PR → TASK → REQ, TEST → PR)", file=sys.stderr)
    rels_pairs = [
        (task,   req_103, "implements"),
        (pr,     task,    "generated_from"),
        (build,  pr,      "generated_from"),
        (test,   pr,      "verifies"),
        (deploy, build,   "deployed_as"),
    ]
    for src, tgt, kind in rels_pairs:
        code, rel = curl(
            "POST", "/relationships",
            {
                "source_artifact_id": src["id"],
                "target_artifact_id": tgt["id"],
                "relationship_type": kind,
                "meta": {"seed": "traceability_demo"},
            },
        )
        assert code < 300 and isinstance(rel, dict) and "id" in rel, f"REL FAIL {kind} -> {rel}"

    print(file=sys.stderr); print("# Step 4: GET traceability graph → counts", file=sys.stderr)
    code, graph = curl("GET", "/traceability/graph?limit=200")
    assert code == 200 and isinstance(graph, dict), f"GRAPH FAIL HTTP={code} {graph}"
    nodes = graph.get("nodes", [])
    edges = graph.get("edges", [])
    print(f"\n=== SEED SUCCESS ===", file=sys.stderr)
    print(f"  graph.nodes = {len(nodes)}   graph.edges = {len(edges)}", file=sys.stderr)
    # print to stdout for bash pipeline capture (JSON):
    print(json.dumps({"nodes": len(nodes), "edges": len(edges),
                      "node_ids": [{"id":str(n["id"])[:10],"type":n.get("type"),"label":n.get("label")} for n in nodes],
                      "edge_labels": [{"s":e["source"][:8],"t":e["target"][:8],"lbl":e.get("label")} for e in edges]}))
    return 0


if __name__ == "__main__":
    raise SystemExit(main())