runbook-tutorials/rtut/__main__.py
tegwick 8a5cbc1c29
All checks were successful
CI Smoke / host-smoke (push) Successful in 0s
CI Smoke / container-smoke (push) Successful in 2s
Work Records / validate (push) Successful in 12s
Run outcomes, score, explicit exercise recording, and step captures (RBT-WP-0004-T13)
Co-Authored-By: Claude Sonnet 5.5 <noreply@anthropic.com>

Assistant: claude-code
Assistant-Model: sonnet
Assistant-Process: 295952@bnt-lap001
Assistant-Session: e93f64ad-516c-46eb-9666-aad8d300c477
2026-09-29 11:15:58 +02:00

180 lines
7.7 KiB
Python

"""CLI: python -m rtut {list,validate,show,render} ..."""
from __future__ import annotations
import argparse
import sys
from pathlib import Path
from . import pack as P
def _packs(root: str) -> list[P.Pack]:
if Path(root).is_file():
return [P.load(Path(root))]
return [P.load(p) for p in P.discover(Path(root))]
def _run(pk, a) -> int:
from . import run as R
try:
if a.action == "new":
r = R.create(pk, a.operator, dict(kv.split("=", 1) for kv in a.param))
print(r.data["run_id"])
return 0
if a.action == "list":
print("\n".join(R.list_runs(pk)))
return 0
r = R.load(pk, a.run_id)
if a.action == "status":
for sid, st in r.data["steps"].items():
v = st.get("verification", {})
print(f"{st['state']:8} {sid} {'verified' if v.get('passed') else ''}")
return 0
if a.action == "cmd":
print(R.command_for(r, a.step)); return 0
if a.action == "start":
R.start(r, a.step); print(R.command_for(r, a.step) or ""); return 0
if a.action == "verify":
res = R.verify(r, a.step, confirm=a.yes, manual_ok=a.passed or None)
print("passed" if res["passed"] else "FAILED"); return 0 if res["passed"] else 1
if a.action == "done":
R.complete(r, a.step, dict(kv.split("=", 1) for kv in a.evidence)); return 0
if a.action == "fail":
R.fail(r, a.step, a.note); return 0
if a.action == "retry":
R.retry(r, a.step); return 0
if a.action == "skip":
R.skip(r, a.step, a.note); return 0
if a.action == "finalize":
R.finalize(r, a.outcome or "", a.note)
if r.data["outcome"] == "succeeded" and a.apply:
print("recorded:", R.apply_exercise(r))
print(r.data["outcome"]); return 0
if a.action == "capture":
R.add_capture(r, a.step, sys.stdin.read(), a.command); print("captured"); return 0
if a.action == "feedback":
R.add_feedback(r, a.rating or "", a.note); print("recorded"); return 0
old, new = R.exercise_update(r)
import difflib
print("".join(difflib.unified_diff(old.splitlines(True), new.splitlines(True), "pack.yaml", "pack.yaml")))
if a.apply:
pk.path.write_text(new); print("written")
return 0
except (R.RunError, KeyError, ValueError) as e:
print(f"error: {e}", file=sys.stderr)
return 1
def main(argv: list[str] | None = None) -> int:
ap = argparse.ArgumentParser(prog="rtut")
sub = ap.add_subparsers(dest="cmd", required=True)
for name in ("list", "validate"):
sp = sub.add_parser(name)
sp.add_argument("root", nargs="?", default=".")
sp = sub.add_parser("run", help="record a run of a native pack")
sp.add_argument("action", choices=["new", "list", "status", "cmd", "start", "verify", "done", "fail", "retry", "skip", "finish", "feedback", "finalize", "capture"])
sp.add_argument("manifest")
sp.add_argument("run_id", nargs="?")
sp.add_argument("step", nargs="?")
sp.add_argument("--operator", "-o", default="")
sp.add_argument("--param", "-p", action="append", default=[], help="name=value")
sp.add_argument("--evidence", "-e", action="append", default=[], help="name=value (non-secret)")
sp.add_argument("--note", default="")
sp.add_argument("--outcome", choices=["succeeded", "failed", "abandoned"], help="finalize: how the run ended")
sp.add_argument("--command", default="", help="capture: the command you actually ran; output is read from stdin")
sp.add_argument("--rating", choices=["good", "neutral", "poor"], help="feedback: how the runbook felt")
sp.add_argument("--yes", action="store_true", help="confirm running a verify command")
sp.add_argument("--pass", dest="passed", action="store_true", help="manual check: operator confirms it passed")
sp.add_argument("--apply", action="store_true", help="finish: write the exercise-status change")
sp = sub.add_parser("ui", help="serve the local UI on 127.0.0.1")
sp.add_argument("roots", nargs="*", default=["."])
sp.add_argument("--port", type=int, default=8877)
sp = sub.add_parser("score", help="how often each runbook was tried and how it went")
sp.add_argument("root", nargs="?", default=".")
sp = sub.add_parser("feedback", help="list the latest feedback of every run of a pack")
sp.add_argument("manifest")
sp = sub.add_parser("launch"); sp.add_argument("manifest")
sp.add_argument("--timeout", type=float, default=30.0)
sp = sub.add_parser("show"); sp.add_argument("manifest")
sp = sub.add_parser("render"); sp.add_argument("manifest"); sp.add_argument("step")
sp.add_argument("--param", "-p", action="append", default=[], help="name=value")
a = ap.parse_args(argv)
if a.cmd == "ui":
from . import ui
srv = ui.serve([Path(r) for r in a.roots], a.port)
print(f"http://127.0.0.1:{a.port} (Ctrl-C stops)")
try:
srv.serve_forever()
except KeyboardInterrupt:
pass
return 0
if a.cmd == "score":
from . import run as R
for p in _packs(a.root):
if p.ok and p.data.get("engine") == "native" and p.spec == "runbook-pack/v0.1":
c = R.scorecard(p)
print(f"{p.id}\ttried {c['tried']}\tsucceeded {c['succeeded']}\tfailed {c['failed']}\tabandoned {c['abandoned']}\topen {c['open']}\t[{p.data['exercise_status']}]")
return 0
if a.cmd in ("list", "validate"):
packs = _packs(a.root)
if not packs:
print("no packs found", file=sys.stderr)
return 1
bad = 0
for p in packs:
d = p.data
if a.cmd == "list":
print(f"{p.id}\t{d.get('engine')}\t{d.get('exercise_status')}\t{d.get('title')}")
else:
print(f"{'ok ' if p.ok else 'FAIL'} {p.id}")
for e in p.errors:
print(f" - {e}")
bad += not p.ok
return 1 if bad and a.cmd == "validate" else 0
pk = P.load(Path(a.manifest))
if not pk.ok:
print("\n".join(pk.errors), file=sys.stderr)
return 1
if a.cmd == "feedback":
from . import run as R
for f in R.all_feedback(pk):
print(f"{f['at']} {R.RATINGS[f['rating']]} {f['operator']} {f['run_id']} {f['comment']}")
return 0
if a.cmd == "run":
return _run(pk, a)
if a.cmd == "launch":
from . import legacy
try:
proc = legacy.launch(pk, a.timeout)
except (ValueError, TimeoutError) as e:
print(f"error: {e}", file=sys.stderr)
return 1
print(f"running: {pk.data['legacy'].get('url', '(no url)')} (Ctrl-C stops it)")
try:
return proc.wait()
except KeyboardInterrupt:
proc.terminate()
return 0
if a.cmd == "show":
d = pk.data
print(f"{d['title']} [{d['exercise_status']}] engine={d['engine']}")
for i, s in enumerate(d.get("steps", []), 1):
print(f"{i}. [{s['owner']}] {s['title']} ({s.get('risk', 'read-only')})")
print(f" done when: {s['verify']['done_when']}")
return 0
values = dict(kv.split("=", 1) for kv in a.param)
step = next((s for s in pk.data.get("steps", []) if s["id"] == a.step), None)
if not step or "command" not in step:
print("no such step or step has no command", file=sys.stderr)
return 1
try:
print(P.render(step["command"], values, pk))
except (KeyError, ValueError) as e:
print(f"error: {e}", file=sys.stderr)
return 1
return 0
if __name__ == "__main__":
sys.exit(main())