feat(examples): add self-test example
CI / changes (pull_request) Successful in 7s
CI / client (pull_request) Skipped
CI / server (pull_request) Successful in 1m7s
CI / image (pull_request) Skipped

- add examples/ with a ready-to-import self-test: environment, credential,
  snippets and a scenario that creates, runs and checks another scenario
  through the liqa REST API and UI, then cleans up
- add test/examples.spec.ts: imports every example and parses its code
- link the examples from README, docs/scenario.md and CLAUDE.md

Closes #6
This commit is contained in:
ars9 committed 2026-10-02 15:39:02 +03:00
1 parent 28dfce246a
commit c1924060eb
7 files changed
+433

No files matched your search

+280
View File
@@ -0,0 +1,280 @@
# liqa self-test example: a scenario that tests the liqa instance it runs on.
#
# Import this file (Import page in the UI, POST /api/v1/import with
# Content-Type: application/yaml, or the import_yaml MCP tool), then run the
# "liqa self-test" scenario. See README.md next to this file.
#
# Every item has a fixed id, so importing the file again updates the same
# resources instead of creating duplicates.
---
# Environment: where the liqa under test is reachable from the liqa server
kind: environment
id: 418eacb0-ab16-4ad6-ab50-c0274fdc49f1
name: liqa-self-test
description: >-
The liqa instance the self-test scenario runs against. Steps run inside the
liqa server (the Docker container), so the compose deployment published at
http://localhost:13000 is reached at its in-container port 3000.
data:
# REST API base (without /api/v1)
liqa_url: http://localhost:3000
# Web UI base; use http://localhost:5173 with the Vite dev server
ui_url: http://localhost:3000
# Name prefix of the temporary resources the self-test creates and deletes
temp_prefix: "[self-test] "
# Value the target scenario reads from its own (temporary) environment
greeting: Hello from the liqa self-test
---
# Credential: test data handed over to the temporary target scenario
kind: credential
id: b0c9ccab-6743-4ac1-9017-d99e0f5d7474
name: liqa-self-test-fixture
data: '{"username":"self-test-user","token":"not-a-real-secret-0123456789"}'
---
# Snippet: liqa-api
kind: snippet
id: f9d2612c-266c-411a-9483-5d860599f208
version: 1
alias: liqa-api
title: Call the liqa REST API
description: >-
`context.runSnippet("liqa-api", method, path, body?)` sends a request to
`liqa_url` + `/api/v1` + path and returns the parsed JSON response (`null`
for an empty body). Throws on a non-2xx status.
code: |
const [method, path, body] = args;
const url = context.getEnv("liqa_url") + "/api/v1" + path;
const response = await context.page.request.fetch(url, {
method,
data: body,
// POST /scenarios/:id/run/:runId/wait blocks until the run finishes
timeout: 120_000,
failOnStatusCode: false,
});
const text = await response.text();
if (!response.ok()) {
throw new Error(`${method} ${path} returned ${response.status()}: ${text}`);
}
return text ? JSON.parse(text) : null;
---
# Snippet: liqa-run-and-wait
kind: snippet
id: b36c1075-2700-421f-a2e7-ab1a80adb7e7
version: 1
alias: liqa-run-and-wait
title: Run a scenario and wait for it
description: >-
`context.runSnippet("liqa-run-and-wait", scenarioId, environmentId)` starts
a run of the scenario's latest version and returns the finished run (with
its step runs and logs). Uses the liqa-api snippet.
code: |
const [scenarioId, environmentId] = args;
const run = await context.runSnippet(
"liqa-api",
"POST",
`/scenarios/${scenarioId}/run`,
{ environmentId },
);
context.log(`Started run ${run.id}, waiting for it to finish`);
return context.runSnippet(
"liqa-api",
"POST",
`/scenarios/${scenarioId}/run/${run.id}/wait`,
);
---
# Snippet: self-test-cleanup
kind: snippet
id: f99e3b17-d28d-4a80-8ae4-f7e8252e0be5
version: 1
alias: self-test-cleanup
title: Delete temporary self-test resources
description: >-
Deletes every scenario, credential and environment whose name starts with
the `temp_prefix` environment value, and returns how many were deleted.
Uses the liqa-api snippet.
code: |
const prefix = context.getEnv("temp_prefix");
let deleted = 0;
// Scenarios first: they reference the credentials and environments
for (const path of ["/scenarios", "/credentials", "/environments"]) {
const page = await context.runSnippet("liqa-api", "GET", `${path}?limit=1000`);
for (const item of page.data.filter((i) => i.name.startsWith(prefix))) {
await context.runSnippet("liqa-api", "DELETE", `${path}/${item.id}`);
context.log(`Deleted ${path.slice(1, -1)} "${item.name}"`);
deleted += 1;
}
}
return deleted;
---
# Scenario: liqa self-test
kind: scenario
id: 571b2894-dacc-4428-b459-4e3fefac5908
version: 1
name: liqa self-test
description: |
Tests the liqa instance it runs on, through its REST API and web UI:
1. checks that the instance is up and removes leftovers of earlier runs
2. creates a temporary environment, credential and two-step target scenario
3. runs the target scenario, waits for it and checks its step outputs and logs
4. opens the finished run in the web UI
5. deletes the temporary resources
Uses the `liqa-api`, `liqa-run-and-wait` and `self-test-cleanup` snippets and
the `fixture` credential. See `examples/self-test/README.md`.
environmentId: 418eacb0-ab16-4ad6-ab50-c0274fdc49f1
timeoutSeconds: 300
credentials:
- credentialId: b0c9ccab-6743-4ac1-9017-d99e0f5d7474
alias: fixture
steps:
- title: Check the instance is up
execCode: |
const response = await context.page.request.get(
context.getEnv("liqa_url") + "/health",
);
const health = await response.json();
context.assert(() => response.ok(), "Health endpoint responds", {
status: response.status(),
});
context.assert(() => health.status === "ok", "Health status is ok", health);
// A failed earlier run may have left its temporary resources behind
const deleted = await context.runSnippet("self-test-cleanup");
if (deleted > 0) {
context.warn(`Deleted ${deleted} leftover resource(s) of an earlier run`);
}
return health;
- title: Create the target scenario
execCode: |
const prefix = context.getEnv("temp_prefix");
const api = (method, path, body) =>
context.runSnippet("liqa-api", method, path, body);
const environment = await api("POST", "/environments", {
name: `${prefix}environment`,
data: { greeting: context.getEnv("greeting") },
});
// Hand the fixture credential over to the target scenario
const fixture = context.getCredential("fixture");
const credential = await api("POST", "/credentials", {
name: `${prefix}credential`,
data: JSON.stringify(fixture),
});
const scenario = await api("POST", "/scenarios", {
name: `${prefix}target scenario`,
description: "Temporary scenario created by the liqa self-test",
environmentId: environment.id,
timeoutSeconds: 60,
});
await api("POST", `/scenarios/${scenario.id}/credentials`, {
credentialId: credential.id,
alias: "secret",
});
await api("POST", `/scenarios/${scenario.id}/steps`, {
title: "Read the environment and credential",
execCode: `
const secret = context.getCredential("secret");
context.assert(() => typeof secret.token === "string", "Credential has a token");
// Return the token length, not the token: step outputs are not secret
return { greeting: context.getEnv("greeting"), tokenLength: secret.token.length };
`,
});
await api("POST", `/scenarios/${scenario.id}/steps`, {
title: "Render the greeting",
execCode: `
const { greeting } = await context.getStepOutput(-1);
await context.page.setContent("<h1>" + greeting + "</h1>");
await expect(context.page.locator("h1")).toHaveText(greeting);
return { heading: await context.page.locator("h1").textContent() };
`,
});
const created = await api("GET", `/scenarios/${scenario.id}`);
context.assert(
() => created.steps.length === 2,
"Target scenario has 2 steps",
{ steps: created.steps.length },
);
context.log(`Created target scenario ${scenario.id}`);
return {
scenarioId: scenario.id,
scenarioName: scenario.name,
environmentId: environment.id,
credentialId: credential.id,
};
- title: Run the target scenario
timeoutSeconds: 150
execCode: |
const { scenarioId, environmentId } = await context.getStepOutput(-1);
const run = await context.runSnippet(
"liqa-run-and-wait",
scenarioId,
environmentId,
);
context.assert(() => run.status === "pass", "Target run passed", {
status: run.status,
errors: run.logs.filter((l) => l.level === "error").map((l) => l.message),
});
context.assert(
() => run.stepRuns.every((s) => s.status === "pass"),
"All target steps passed",
run.stepRuns.map((s) => ({ order: s.order, status: s.status })),
);
const [read, render] = run.stepRuns.map((s) => JSON.parse(s.output));
const greeting = context.getEnv("greeting");
context.assert(
() => read.greeting === greeting,
"Target read the greeting from its environment",
read,
);
context.assert(
() => read.tokenLength === context.getCredential("fixture").token.length,
"Target read the token from its credential",
);
context.assert(
() => render.heading === greeting,
"Target rendered the greeting",
render,
);
context.assert(
() =>
run.logs.some(
(l) =>
l.context === "assertion:pass" &&
l.message === "Credential has a token",
),
"Target logged its passed assertion",
);
return { runId: run.id };
- title: Check the run in the UI
execCode: |
const { scenarioId, scenarioName } = await context.getStepOutput(2);
const { runId } = await context.getStepOutput(3);
const page = context.page;
await page.goto(
`${context.getEnv("ui_url")}/scenarios/${scenarioId}/runs/${runId}`,
);
await expect(page.getByText(scenarioName).first()).toBeVisible();
await expect(page.getByText("Credential has a token").first()).toBeVisible();
context.log(`Run page checked: ${page.url()}`);
return { url: page.url() };
- title: Clean up
execCode: |
const { scenarioId } = await context.getStepOutput(2);
const deleted = await context.runSnippet("self-test-cleanup");
context.assert(() => deleted === 3, "Deleted the 3 temporary resources", {
deleted,
});
const response = await context.page.request.get(
`${context.getEnv("liqa_url")}/api/v1/scenarios/${scenarioId}`,
);
context.assert(
() => response.status() === 404,
"Target scenario is gone",
{ status: response.status() },
);
return { deleted };