Files
archie/ui/test/attempt-config.test.js
T
sam 7414713d61 feat(webui): add the per-run task detail page
A per-task page at #/tasks/{id} with tabs for the attempt's stages and log, changed
files, configuration and a raw debug view. Reachable from the task list, deep
linkable via ?tab= and ?attempt=, and reload safe. Existing URLs (#/tasks?task=N
and #/tasks?status=...) are unchanged.

- A stage rail per attempt: name, duration, terminal status and error text, with
  unattributed events surfaced and the agent's own verdict shown separately.
- No pass/fail badge and no exit code anywhere: the backend has neither, and a stage
  that returned nil after producing a poor diff is a success. Status word, duration
  and error text are shown instead.
- The log pane filters by attempt, stage and level, states the coverage caveat for
  stage filters (agent and tool output records no stage), and keeps the three read
  outcomes distinct: no log reader in this deployment, no file for this attempt, and
  no entries matching the filter.
- Retry is labelled as a new run and its confirm says the previous attempt's commits
  are discarded, because it hard-resets the branch onto the base.
- An absent capture reads "not captured for this run" rather than implying nothing
  changed, and no elapsed-run figure is shown since only per-stage durations exist.
- The events cache is invalidated on retry (it previously kept the ended attempt's
  events) and the log cache is keyed by attempt, so two attempts cannot be conflated.
- ui/dist is rebuilt and committed: the gate runs the node tests but never rebuilds
  the bundle, so a stale one would otherwise ship behind a green gate.
- docs/prds/task-run-detail.md records the surface and its limits, and
  docs/architecture/observability.md describes it alongside the read endpoints.

Both new vocabularies are pinned by tests on each side (config schema, change
statuses) so a one-sided rename fails a test instead of silently degrading the page.
2026-09-18 18:33:03 +10:00

112 lines
4.6 KiB
JavaScript

import { test } from "node:test";
import assert from "node:assert/strict";
import { render, cleanup } from "@testing-library/preact";
import { AttemptConfig, CONFIG_SCHEMA, selectConfigEvent } from "../src/tasks/attempt-config.jsx";
const EVENTS = [
{
kind: "config_captured",
attempt: 1,
at: "2026-09-18T07:00:00.100Z",
stage: "",
data: { schema: CONFIG_SCHEMA, document: { BotUser: "archie-attempt-one", Models: { main: "openai/gpt" } } },
},
{
kind: "config_captured",
attempt: 2,
at: "2026-09-18T07:05:00.100Z",
stage: "",
data: { schema: CONFIG_SCHEMA, document: { BotUser: "archie-attempt-two", Models: { main: "openai/gpt" } } },
},
];
function mount(props) {
const { container, unmount } = render(<AttemptConfig {...props} />);
return { root: container, unmount };
}
// Selecting by kind alone would show the newest attempt's configuration on
// every attempt -- the merge-attempts bug the attempt column exists to prevent.
test("the configuration is selected by kind and attempt together", () => {
const { root, unmount } = mount({ events: EVENTS, attempt: 1 });
assert.match(root.textContent, /archie-attempt-one/);
assert.doesNotMatch(root.textContent, /archie-attempt-two/);
assert.match(root.textContent, /attempt 1's effective task-runtime configuration/);
unmount();
});
test("an attempt with no config event of its own says so, never another attempt's", () => {
const { root, unmount } = mount({ events: EVENTS, attempt: 3 });
assert.equal(root.querySelector(".empty-title").textContent, "Not captured for this run");
assert.doesNotMatch(root.textContent, /archie-attempt-one|archie-attempt-two/);
assert.match(root.textContent, /Nothing here says the run used the defaults/);
unmount();
});
test("the document is passed through verbatim, pretty-printed", () => {
const { root, unmount } = mount({ events: EVENTS, attempt: 2 });
const json = root.querySelector(".run-json").textContent;
assert.match(json, /"BotUser": "archie-attempt-two"/);
assert.match(json, /"Models"/);
unmount();
});
// The page must not imply this document is the whole configuration: it is the
// non-secret task-runtime subset.
test("the panel states the scope of what was captured", () => {
const { root, unmount } = mount({ events: EVENTS, attempt: 1 });
assert.match(root.textContent, /non-secret subset/);
assert.match(root.textContent, /not the dashboard configuration view/);
assert.match(root.textContent, /providers, repositories, identities, credentials and lock state/);
unmount();
});
test("an unrecognised schema is named and its payload is still shown", () => {
const events = [
{ kind: "config_captured", attempt: 2, at: "2026-09-18T07:05:00.100Z", data: { schema: "archie/task-config@9", document: { future: true } } },
];
const { root, unmount } = mount({ events, attempt: 2 });
assert.match(root.textContent, /unknown schema \(archie\/task-config@9\)/);
assert.match(root.querySelector(".run-json").textContent, /"future": true/);
unmount();
});
test("a config event with no data at all renders the raw payload rather than guessing", () => {
const events = [{ kind: "config_captured", attempt: 2, at: "2026-09-18T07:05:00.100Z" }];
const { root, unmount } = mount({ events, attempt: 2 });
assert.match(root.textContent, /unknown schema/);
assert.ok(root.querySelector(".run-json"));
unmount();
});
test("loading and failure states are distinct, and the failure offers a retry", () => {
const loading = mount({ events: undefined, attempt: 1 });
assert.match(loading.root.textContent, /Loading this task's events/);
loading.unmount();
const failed = mount({ events: null, attempt: 1, onRetry: () => {} });
assert.match(failed.root.textContent, /Could not load this task's events/);
assert.equal(failed.root.querySelector("button").textContent, "Retry");
failed.unmount();
});
test("selectConfigEvent ignores other kinds and other attempts", () => {
const events = [
{ kind: "stage_finish", attempt: 2, data: {} },
...EVENTS,
];
assert.equal(selectConfigEvent(events, 2).attempt, 2);
assert.equal(selectConfigEvent(events, 3), null);
assert.equal(selectConfigEvent(undefined, 1), null);
});
// CONFIG_SCHEMA mirrors events.ConfigCapturedSchema (internal/events/events.go),
// which the Go suite pins by the same literal. Pinning both sides is what makes
// a one-sided rename a failing test rather than every attempt rendering as
// "unknown schema".
test("the config schema is the one the producer stamps", () => {
assert.equal(CONFIG_SCHEMA, "archie/task-config@1");
});
test.after(() => cleanup());