mirror of
https://github.com/fabro-sh/fabro.git
synced 2026-10-16 04:33:16 +00:00
## Summary
Replaces the single `queued` pre-execution state with explicit `pending`
and `runnable` states, and wires approve/deny actions for
parent-generated child runs that require human approval before they can
execute. This diff covers the web UI and OpenAPI spec layers of that
change.
## What changed
**Run status model**
- `queued` is removed from all TypeScript types, display maps, column
routing, and tests.
- `pending` (awaiting approval) and `runnable` (eligible for the
scheduler) replace it as distinct board columns and `RunStatus` variants
with their own labels and colors (`runnable` gets cyan; `pending` stays
muted).
**Approval actions**
- New `approveRun` / `denyRun` API calls in `run-actions.ts` invoke the
new `POST /runs/{id}/approve` and `POST /runs/{id}/deny` endpoints.
- `canApprove` predicate requires both `status.kind === "pending"` and
`lifecycle.approval?.state === "pending"` — a run whose status is
pending but has no approval record does not expose the action.
- `useApproveRun` / `useDenyRun` mutations in `mutations.ts` follow the
same pattern as `useCancelRun`.
- `ActionsMenu` in `run-detail.tsx` gains Approve (lifecycle group) and
Deny (destructive group) menu items.
**Board and event plumbing**
- `columnForStatus` now routes `pending → pending column` and `runnable
→ runnable column`; `submitted` stays in the pending column.
- `BOARD_STATUS_EVENTS` and `RUN_SUMMARY_EVENTS` replace `run.queued`
with `run.start_requested`, `run.pending`, `run.approved`, `run.denied`,
and `run.runnable`.
- The `pending` column is hidden when empty (same behaviour the old
`queued` column had).
**Waterfall phases (`run-phases.ts`)**
- `queued` phase is removed; `pending` and `runnable` phases are added
in order.
- The submitted phase closes at `run.start_requested` rather than
`run.queued`.
- Each phase derives its timestamps from its own event rather than a
single `firstTs` lookup, making multi-phase pre-execution timelines
accurate.
**OpenAPI spec**
- `POST /api/v1/runs/{id}/approve` and `POST /api/v1/runs/{id}/deny`
endpoints added with 200/404/409 responses.
- `startRun` description updated to describe the pending/runnable
branching behaviour.
- `cancelRun` description updated to reference `pending`/`runnable`
instead of `queued`.
### Plan Summary
- **Task 3** (OpenAPI schema additions for approve/deny endpoints) —
complete in this diff.
- **Task 6** (Web UI surfaces: board columns, run-detail actions,
waterfall phases, event subscriptions) — complete in this diff.
- **Task 7** (doc cleanup: references to `queued` replaced in plans,
brainstorms, and QA docs) — complete in this diff.
### Fabro Details
<details>
<summary>Ran 9 stages in 127m 37s for $104.98</summary>
| Stage | Duration | Cost | Retries |
|---|---|---|---|
| start | 0s | – | 0 |
| toolchain | 2s | – | 0 |
| preflight_compile | 2m 15s | – | 0 |
| preflight_lint | 2m 29s | – | 0 |
| implement | 92m 10s | $91.53 | 0 |
| simplify_opus | 18m 35s | $10.65 | 0 |
| simplify_gpt | 7m 36s | $2.81 | 0 |
| verify | 3m 42s | – | 0 |
| fmt | 3s | – | 0 |
| **Total** | **127m 37s** | **$104.98** | **0** |
</details>
<details>
<summary>Ran <code>ImplementPlan.fabro</code> (12 nodes and 15
edges)</summary>
```dot
digraph ImplementPlan {
graph [
goal="Implement and simplify",
model_stylesheet="
* { model: claude-opus-4-7; }
"
]
rankdir=LR
start [shape=Mdiamond, label="Start"]
exit [shape=Msquare, label="Exit"]
toolchain [label="Toolchain", shape=parallelogram, script="command -v cargo >/dev/null || { curl --proto '=https' --tlsv1.2 -sSf https://sh.rustup.rs | sh -s -- -y && sudo ln -sf $HOME/.cargo/bin/* /usr/local/bin/; }; cargo --version 2>&1", max_retries=0]
preflight_compile [label="Preflight Compile", shape=parallelogram, script="cargo check -q --workspace 2>&1", max_retries=0]
preflight_lint [label="Preflight Lint", shape=parallelogram, script="cargo +nightly-2026-04-14 clippy -q --workspace --all-targets -- -D warnings 2>&1", max_retries=0]
fix_lints [label="Fix Lints", prompt="The preflight lint step failed. Read the build output from context and fix all clippy lint warnings.", max_visits=3]
implement [label="Implement", prompt="Read the plan file referenced in the goal and implement every step. Make all the code changes described in the plan. Use red/green TDD.", model="gpt-55", reasoning_effort="xhigh"]
simplify_opus [label="Simplify (Opus)", prompt="@prompts/simplify.md"]
simplify_gpt [label="Simplify (GPT-55)", prompt="@prompts/simplify.md", model="gpt-55"]
verify [label="Verify", shape=parallelogram, script="cargo +nightly-2026-04-14 clippy -q --workspace --all-targets -- -D warnings 2>&1 && cargo nextest run --cargo-quiet --workspace --status-level fail 2>&1 && cargo dev docs refresh 2>&1 && cargo dev docs check 2>&1", goal_gate=true, retry_target="fixup"]
fixup [label="Fixup", prompt="The verify step failed. Read the build output from context and fix all clippy lint warnings, test failures, and generated docs errors.", max_visits=3]
fmt [label="Format", shape=parallelogram, script="cargo +nightly-2026-04-14 fmt --all 2>&1", max_retries=0]
start -> toolchain
toolchain -> preflight_compile [condition="outcome=succeeded"]
toolchain -> exit
preflight_compile -> preflight_lint [condition="outcome=succeeded"]
preflight_compile -> exit
preflight_lint -> implement [condition="outcome=succeeded"]
preflight_lint -> fix_lints
fix_lints -> preflight_lint
implement -> simplify_opus -> simplify_gpt -> verify
verify -> fmt [condition="outcome=succeeded"]
verify -> fixup
fixup -> verify
fmt -> exit
}
```
</details>
⚒️ Generated with [Fabro](https://fabro.sh)
---------
Co-authored-by: Fabro <noreply@fabro.sh>
Co-authored-by: fabro <fabro@anthropic.com>
Co-authored-by: Bryan Helmkamp <bryan@brynary.com>
398 lines
14 KiB
Rust
398 lines
14 KiB
Rust
#![expect(
|
|
clippy::disallowed_methods,
|
|
reason = "integration tests stage fixtures with sync std::fs; test infrastructure, not Tokio-hot path"
|
|
)]
|
|
|
|
use std::path::PathBuf;
|
|
|
|
use axum::body::Body;
|
|
use axum::http::{Request, StatusCode};
|
|
use fabro_config::Storage;
|
|
use fabro_types::RunId;
|
|
use fabro_types::settings::interp::InterpString;
|
|
use tempfile::tempdir;
|
|
use tower::ServiceExt;
|
|
|
|
use crate::helpers::{
|
|
MINIMAL_DOT, POLL_ATTEMPTS, POLL_INTERVAL, TestAppSettings, api, checked_response,
|
|
minimal_manifest_json, minimal_manifest_json_with_dry_run, response_json, response_status,
|
|
test_app_state_with_options, test_app_with_scheduler, test_settings, wait_for_run_status,
|
|
};
|
|
|
|
const HUMAN_GATE_DOT: &str = r#"digraph GateTest {
|
|
graph [goal="Test gate"]
|
|
start [shape=Mdiamond]
|
|
exit [shape=Msquare]
|
|
work [shape=box, prompt="Do work"]
|
|
gate [shape=hexagon, type="human", label="Approve?"]
|
|
done [shape=box, prompt="Finish"]
|
|
revise [shape=box, prompt="Revise"]
|
|
|
|
start -> work -> gate
|
|
gate -> done [label="[A] Approve"]
|
|
gate -> revise [label="[R] Revise"]
|
|
done -> exit
|
|
revise -> gate
|
|
}"#;
|
|
|
|
fn temp_storage_settings() -> (tempfile::TempDir, TestAppSettings, PathBuf) {
|
|
let temp = tempdir().expect("tempdir should create");
|
|
let mut settings = test_settings();
|
|
let storage_dir = temp.path().join("storage");
|
|
settings.server_settings.server.storage.root =
|
|
InterpString::parse(&storage_dir.to_string_lossy());
|
|
(temp, settings, storage_dir)
|
|
}
|
|
|
|
async fn create_run(app: &axum::Router, manifest: serde_json::Value) -> String {
|
|
let request = Request::builder()
|
|
.method("POST")
|
|
.uri(api("/runs"))
|
|
.header("content-type", "application/json")
|
|
.body(Body::from(
|
|
serde_json::to_vec(&manifest).expect("manifest fixture should serialize"),
|
|
))
|
|
.expect("create-run request should build");
|
|
let response = app.clone().oneshot(request).await.unwrap();
|
|
let body = response_json(response, StatusCode::CREATED, "POST /api/v1/runs").await;
|
|
body["id"]
|
|
.as_str()
|
|
.expect("create-run response should include an id")
|
|
.to_string()
|
|
}
|
|
|
|
async fn start_run(app: &axum::Router, run_id: &str) {
|
|
let request = Request::builder()
|
|
.method("POST")
|
|
.uri(api(&format!("/runs/{run_id}/start")))
|
|
.body(Body::empty())
|
|
.expect("start-run request should build");
|
|
let response = app.clone().oneshot(request).await.unwrap();
|
|
response_status(
|
|
response,
|
|
StatusCode::OK,
|
|
format!("POST /api/v1/runs/{run_id}/start"),
|
|
)
|
|
.await;
|
|
}
|
|
|
|
async fn wait_for_question(app: &axum::Router, run_id: &str) -> serde_json::Value {
|
|
for _ in 0..POLL_ATTEMPTS {
|
|
let request = Request::builder()
|
|
.method("GET")
|
|
.uri(api(&format!("/runs/{run_id}/questions")))
|
|
.body(Body::empty())
|
|
.expect("questions request should build");
|
|
let response = app.clone().oneshot(request).await.unwrap();
|
|
let body = response_json(
|
|
response,
|
|
StatusCode::OK,
|
|
format!("GET /api/v1/runs/{run_id}/questions"),
|
|
)
|
|
.await;
|
|
if let Some(question) = body["data"].as_array().and_then(|items| items.first()) {
|
|
return question.clone();
|
|
}
|
|
tokio::time::sleep(POLL_INTERVAL).await;
|
|
}
|
|
panic!("question should have appeared for {run_id}");
|
|
}
|
|
|
|
async fn load_questions(app: &axum::Router, run_id: &str) -> serde_json::Value {
|
|
let request = Request::builder()
|
|
.method("GET")
|
|
.uri(api(&format!("/runs/{run_id}/questions")))
|
|
.body(Body::empty())
|
|
.expect("questions request should build");
|
|
let response = app.clone().oneshot(request).await.unwrap();
|
|
response_json(
|
|
response,
|
|
StatusCode::OK,
|
|
format!("GET /api/v1/runs/{run_id}/questions"),
|
|
)
|
|
.await
|
|
}
|
|
|
|
#[tokio::test]
|
|
async fn get_system_info_returns_runtime_fields() {
|
|
let (_temp, settings, expected_storage_dir) = temp_storage_settings();
|
|
let configured_server_url = settings.server_settings.server.web.url.as_source();
|
|
let app =
|
|
fabro_server::test_support::build_test_router(test_app_state_with_options(settings, 5));
|
|
|
|
let request = Request::builder()
|
|
.method("GET")
|
|
.uri(api("/system/info"))
|
|
.body(Body::empty())
|
|
.unwrap();
|
|
let response = app.oneshot(request).await.unwrap();
|
|
|
|
let body = response_json(response, StatusCode::OK, "GET /api/v1/system/info").await;
|
|
assert!(body["version"].as_str().is_some());
|
|
assert_eq!(body["server_url"], configured_server_url);
|
|
assert_eq!(body["storage_engine"], "slatedb");
|
|
assert_eq!(
|
|
body["storage_dir"],
|
|
expected_storage_dir.display().to_string()
|
|
);
|
|
assert_eq!(body["runs"]["total"], 0);
|
|
assert_eq!(body["runs"]["active"], 0);
|
|
assert!(body["uptime_secs"].as_i64().is_some());
|
|
assert!(
|
|
body.get("features").is_none(),
|
|
"system info should not include a features field"
|
|
);
|
|
}
|
|
|
|
#[tokio::test]
|
|
async fn get_system_resources_returns_server_visible_metrics() {
|
|
let (_temp, settings, expected_storage_dir) = temp_storage_settings();
|
|
let app =
|
|
fabro_server::test_support::build_test_router(test_app_state_with_options(settings, 5));
|
|
|
|
let request = Request::builder()
|
|
.method("GET")
|
|
.uri(api("/system/resources"))
|
|
.body(Body::empty())
|
|
.unwrap();
|
|
let response = app.oneshot(request).await.unwrap();
|
|
|
|
let body = response_json(response, StatusCode::OK, "GET /api/v1/system/resources").await;
|
|
let sampled_at = body["sampled_at"]
|
|
.as_str()
|
|
.expect("sampled_at should be an RFC3339 timestamp");
|
|
chrono::DateTime::parse_from_rfc3339(sampled_at).expect("sampled_at should parse as RFC3339");
|
|
|
|
assert_eq!(body["cpu"]["supported"], true);
|
|
assert_eq!(body["cpu"]["scope"], "server_environment");
|
|
assert!(
|
|
body["cpu"]["sample_window_ms"].is_null()
|
|
|| body["cpu"]["sample_window_ms"]
|
|
.as_i64()
|
|
.is_some_and(|value| value >= 0),
|
|
"sample window should be null until a delta sample is available or nonnegative: {body}"
|
|
);
|
|
assert!(
|
|
body["cpu"]["logical_cpus"].as_i64().unwrap_or_default() > 0,
|
|
"logical CPU count should be positive"
|
|
);
|
|
assert_percent_if_present(&body["cpu"]["usage_percent"]);
|
|
|
|
assert_eq!(body["memory"]["supported"], true);
|
|
assert!(matches!(
|
|
body["memory"]["scope"].as_str(),
|
|
Some("host" | "cgroup")
|
|
));
|
|
assert_nonnegative_i64(&body["memory"]["total_bytes"], "memory.total_bytes");
|
|
assert_nonnegative_i64(&body["memory"]["used_bytes"], "memory.used_bytes");
|
|
assert_nonnegative_i64(&body["memory"]["available_bytes"], "memory.available_bytes");
|
|
assert_percent_if_present(&body["memory"]["used_percent"]);
|
|
|
|
assert_eq!(body["disk"]["supported"], true);
|
|
assert_eq!(body["disk"]["scope"], "storage_filesystem");
|
|
assert_eq!(
|
|
body["disk"]["storage_path"],
|
|
expected_storage_dir.display().to_string()
|
|
);
|
|
assert!(body["disk"]["mount_point"].as_str().is_some());
|
|
assert_nonnegative_i64(&body["disk"]["total_bytes"], "disk.total_bytes");
|
|
assert_nonnegative_i64(&body["disk"]["used_bytes"], "disk.used_bytes");
|
|
assert_nonnegative_i64(&body["disk"]["available_bytes"], "disk.available_bytes");
|
|
assert_percent_if_present(&body["disk"]["used_percent"]);
|
|
assert_nonnegative_i64(
|
|
&body["disk"]["fabro_managed_bytes"],
|
|
"disk.fabro_managed_bytes",
|
|
);
|
|
assert_nonnegative_i64(
|
|
&body["disk"]["fabro_reclaimable_bytes"],
|
|
"disk.fabro_reclaimable_bytes",
|
|
);
|
|
assert!(
|
|
body["disk"]["fabro_managed_bytes"]
|
|
.as_i64()
|
|
.unwrap_or_default()
|
|
>= body["disk"]["fabro_reclaimable_bytes"]
|
|
.as_i64()
|
|
.unwrap_or_default(),
|
|
"managed bytes must include everything reclaimable: {body}"
|
|
);
|
|
assert!(
|
|
body["notes"]
|
|
.as_array()
|
|
.is_some_and(std::vec::Vec::is_empty),
|
|
"initial test app resources should not need notes: {body}"
|
|
);
|
|
}
|
|
|
|
fn assert_nonnegative_i64(value: &serde_json::Value, name: &str) {
|
|
assert!(
|
|
value.as_i64().is_some_and(|value| value >= 0),
|
|
"{name} should be a nonnegative integer: {value:?}"
|
|
);
|
|
}
|
|
|
|
fn assert_percent_if_present(value: &serde_json::Value) {
|
|
if value.is_null() {
|
|
return;
|
|
}
|
|
let percent = value
|
|
.as_f64()
|
|
.expect("percent should be numeric when present");
|
|
assert!(
|
|
(0.0..=100.0).contains(&percent),
|
|
"percent should be between 0 and 100: {percent}"
|
|
);
|
|
}
|
|
|
|
#[tokio::test(flavor = "multi_thread", worker_threads = 2)]
|
|
async fn test_app_state_with_options_respects_max_concurrent_runs() {
|
|
let app = test_app_with_scheduler(test_app_state_with_options(test_settings(), 1));
|
|
|
|
let first_run = create_run(&app, minimal_manifest_json(HUMAN_GATE_DOT)).await;
|
|
let second_run = create_run(&app, minimal_manifest_json(HUMAN_GATE_DOT)).await;
|
|
|
|
start_run(&app, &first_run).await;
|
|
start_run(&app, &second_run).await;
|
|
|
|
let question = wait_for_question(&app, &first_run).await;
|
|
assert_eq!(question["stage"], "gate");
|
|
|
|
tokio::time::sleep(POLL_INTERVAL * 5).await;
|
|
|
|
let second_questions = load_questions(&app, &second_run).await;
|
|
assert!(
|
|
second_questions["data"]
|
|
.as_array()
|
|
.is_some_and(std::vec::Vec::is_empty),
|
|
"second run should still be waiting for scheduler capacity while the first waits at the human gate: {second_questions}"
|
|
);
|
|
}
|
|
|
|
#[tokio::test(flavor = "multi_thread", worker_threads = 2)]
|
|
async fn get_system_disk_usage_returns_summary_and_verbose_rows() {
|
|
let (_temp, settings, storage_dir) = temp_storage_settings();
|
|
let app = test_app_with_scheduler(test_app_state_with_options(settings, 5));
|
|
|
|
let run_id = create_run(&app, minimal_manifest_json_with_dry_run(MINIMAL_DOT)).await;
|
|
start_run(&app, &run_id).await;
|
|
let status = wait_for_run_status(&app, &run_id, &["succeeded", "failed"]).await;
|
|
assert_eq!(status, "succeeded");
|
|
|
|
let logs_dir = storage_dir.join("logs");
|
|
std::fs::create_dir_all(&logs_dir).unwrap();
|
|
std::fs::write(logs_dir.join("server.log"), b"log line\n").unwrap();
|
|
|
|
let request = Request::builder()
|
|
.method("GET")
|
|
.uri(api("/system/df?verbose=true"))
|
|
.body(Body::empty())
|
|
.unwrap();
|
|
let response = app.oneshot(request).await.unwrap();
|
|
|
|
let body = response_json(
|
|
response,
|
|
StatusCode::OK,
|
|
"GET /api/v1/system/df?verbose=true",
|
|
)
|
|
.await;
|
|
let summary = body["summary"].as_array().expect("summary array");
|
|
assert!(summary.iter().any(|row| row["type"] == "other"));
|
|
|
|
let row_size = |type_: &str| {
|
|
summary
|
|
.iter()
|
|
.find(|row| row["type"] == type_)
|
|
.and_then(|row| row["size_bytes"].as_i64())
|
|
.unwrap_or_default()
|
|
};
|
|
let total_size = body["total_size_bytes"].as_i64().unwrap_or_default();
|
|
assert!(total_size > 0);
|
|
assert_eq!(
|
|
row_size("runs") + row_size("logs") + row_size("other"),
|
|
total_size,
|
|
);
|
|
assert!(
|
|
body["runs"]
|
|
.as_array()
|
|
.is_some_and(|runs| runs.iter().any(|entry| entry["run_id"] == run_id))
|
|
);
|
|
}
|
|
|
|
#[tokio::test(flavor = "multi_thread", worker_threads = 2)]
|
|
async fn prune_runs_supports_dry_run_and_deletion() {
|
|
let (_temp, settings, storage_dir) = temp_storage_settings();
|
|
let app = test_app_with_scheduler(test_app_state_with_options(settings, 5));
|
|
|
|
let run_id = create_run(&app, minimal_manifest_json_with_dry_run(MINIMAL_DOT)).await;
|
|
start_run(&app, &run_id).await;
|
|
let status = wait_for_run_status(&app, &run_id, &["succeeded", "failed"]).await;
|
|
assert_eq!(status, "succeeded");
|
|
|
|
let run_id_parsed: RunId = run_id.parse().unwrap();
|
|
let run_dir = Storage::new(&storage_dir)
|
|
.run_scratch(&run_id_parsed)
|
|
.root()
|
|
.to_path_buf();
|
|
assert!(run_dir.exists());
|
|
|
|
let dry_run_request = Request::builder()
|
|
.method("POST")
|
|
.uri(api("/system/prune/runs"))
|
|
.header("content-type", "application/json")
|
|
.body(Body::from(r#"{"before":"9999"}"#))
|
|
.unwrap();
|
|
let dry_run_response = app.clone().oneshot(dry_run_request).await.unwrap();
|
|
let dry_run_body = response_json(
|
|
dry_run_response,
|
|
StatusCode::OK,
|
|
"POST /api/v1/system/prune/runs",
|
|
)
|
|
.await;
|
|
assert_eq!(dry_run_body["dry_run"], true);
|
|
assert_eq!(dry_run_body["total_count"], 1);
|
|
assert_eq!(dry_run_body["runs"][0]["run_id"], run_id);
|
|
assert!(run_dir.exists());
|
|
|
|
let delete_request = Request::builder()
|
|
.method("POST")
|
|
.uri(api("/system/prune/runs"))
|
|
.header("content-type", "application/json")
|
|
.body(Body::from(r#"{"dry_run":false,"before":"9999"}"#))
|
|
.unwrap();
|
|
let delete_response = app.clone().oneshot(delete_request).await.unwrap();
|
|
let delete_body = response_json(
|
|
delete_response,
|
|
StatusCode::OK,
|
|
"POST /api/v1/system/prune/runs",
|
|
)
|
|
.await;
|
|
assert_eq!(delete_body["dry_run"], false);
|
|
assert_eq!(delete_body["deleted_count"], 1);
|
|
assert!(!run_dir.exists());
|
|
}
|
|
|
|
#[tokio::test(flavor = "multi_thread", worker_threads = 2)]
|
|
async fn attach_events_returns_sse_stream() {
|
|
let (_temp, settings, _storage_dir) = temp_storage_settings();
|
|
let app = test_app_with_scheduler(test_app_state_with_options(settings, 5));
|
|
let run_id = RunId::new();
|
|
|
|
let request = Request::builder()
|
|
.method("GET")
|
|
.uri(api(&format!("/attach?run_id={run_id}")))
|
|
.body(Body::empty())
|
|
.unwrap();
|
|
let response = checked_response(
|
|
app.clone().oneshot(request).await.unwrap(),
|
|
StatusCode::OK,
|
|
format!("GET /api/v1/attach?run_id={run_id}"),
|
|
)
|
|
.await;
|
|
let content_type = response
|
|
.headers()
|
|
.get("content-type")
|
|
.expect("content-type should be present")
|
|
.to_str()
|
|
.unwrap();
|
|
assert!(content_type.contains("text/event-stream"));
|
|
}
|