diff --git a/knip.json b/knip.json index b5ff3c2e5d..8c0f71f3af 100644 --- a/knip.json +++ b/knip.json @@ -22,7 +22,7 @@ "ignoreDependencies": ["cordis"] }, "packages/support/acp-snapshot": { - "entry": ["tests/**/*.spec.ts"], + "entry": ["tests/**/*.spec.ts", "tests/fixtures/fake-acp-agent.ts"], "project": ["src/**/*.ts", "tests/**/*.ts"], "ignoreDependencies": ["cordis"] }, diff --git a/packages/support/acp-snapshot/src/harness.ts b/packages/support/acp-snapshot/src/harness.ts index a46f752d5c..652e60a469 100644 --- a/packages/support/acp-snapshot/src/harness.ts +++ b/packages/support/acp-snapshot/src/harness.ts @@ -224,7 +224,12 @@ export async function runScenario(input: InputScript, opts: RunOptions): Promise sessionUpdate(params: SessionNotification): Promise { for (let i = updateWaiters.length - 1; i >= 0; i--) { const waiter = updateWaiters[i] - if (waiter !== undefined && waiter.match(params.update)) { + // The index is always in-bounds (i only decreases; splice removes at + // i, so lower entries stay valid); the guard satisfies + // noUncheckedIndexedAccess. + /* v8 ignore next 1 -- unreachable in-bounds guard, see above */ + if (waiter === undefined) continue + if (waiter.match(params.update)) { updateWaiters.splice(i, 1) waiter.resolve() } @@ -351,6 +356,10 @@ async function runStep( /** Resolve once the child process exits (any code/signal). */ function waitForExit(child: ChildProcessWithoutNullStreams): Promise { + // Race guard: both call sites run within one synchronous frame of + // stdin.end()/kill(), so the exit event cannot have been delivered yet; + // kept for any future caller that awaits in between. + /* v8 ignore next 1 -- unreachable race guard, see above */ if (child.exitCode !== null || child.signalCode !== null) return Promise.resolve() return new Promise(resolve => child.once('exit', () => { resolve() })) } diff --git a/packages/support/acp-snapshot/src/suite.ts b/packages/support/acp-snapshot/src/suite.ts index 48d6692258..14a4df54da 100644 --- a/packages/support/acp-snapshot/src/suite.ts +++ b/packages/support/acp-snapshot/src/suite.ts @@ -101,8 +101,14 @@ export interface SnapshotSuiteOptions { mode: 'replay' | 'record' } -/** The sibling child-fixture paths for a scenario (`session.1.jsonl` …). */ -function childFixturePaths(dir: string, childSessions: number): string[] { +/** + * The sibling child-fixture paths for a scenario (`session.1.jsonl` …). + * + * @param dir The scenario's snapshots directory (`/`). + * @param childSessions How many subagent child sessions the scenario records. + * @returns One path per child, 1-based, in fixture order. + */ +export function childFixturePaths(dir: string, childSessions: number): string[] { return Array.from({ length: childSessions }, (_, i) => join(dir, `session.${i + 1}.jsonl`)) } @@ -118,8 +124,11 @@ function childFixturePaths(dir: string, childSessions: number): string[] { * is an idempotent no-op. A header with no `cwd` falls back to a sentinel that * cannot occur in a log (NOT `''`, which `String.split` would match on every * character boundary and corrupt the output). + * + * @param fixture The committed `session.jsonl` content. + * @returns The fixture's own volatile values, ready for {@link normalizeSessionLog}. */ -function fixtureContext(fixture: string): NormalizeContext { +export function fixtureContext(fixture: string): NormalizeContext { const firstLine = fixture.split('\n').find(line => line.trim().length > 0) ?? '{}' const header = JSON.parse(firstLine) as { id?: unknown; cwd?: unknown } return { @@ -134,8 +143,12 @@ function fixtureContext(fixture: string): NormalizeContext { * ({@link normalizeSessionLog}) so headers harvested from different runs — * each embedding its own temp cwd in the composed prompt — compare on equal * footing. + * + * @param rawLog The session `.jsonl` content to extract headers from. + * @param ctx The volatile values of the run that produced it. + * @returns The normalized `data.header` payloads, in log order. */ -function normalizedHeaders(rawLog: string, ctx: NormalizeContext): unknown[] { +export function normalizedHeaders(rawLog: string, ctx: NormalizeContext): unknown[] { return normalizeSessionLog(rawLog, ctx) .split('\n') .filter(line => line.trim().length > 0) @@ -144,8 +157,13 @@ function normalizedHeaders(rawLog: string, ctx: NormalizeContext): unknown[] { .map(record => record.data?.header) } -/** Count the `request/header-delta` events in a session JSONL. */ -function headerDeltaCount(rawLog: string): number { +/** + * Count the `request/header-delta` events in a session JSONL. + * + * @param rawLog The session `.jsonl` content. + * @returns How many `request/header-delta` events the log carries. + */ +export function headerDeltaCount(rawLog: string): number { return rawLog.split('\n') .filter(line => line.trim().length > 0) .filter(line => (JSON.parse(line) as { type?: unknown }).type === 'request/header-delta') diff --git a/packages/support/acp-snapshot/tests/fixtures/fake-acp-agent.ts b/packages/support/acp-snapshot/tests/fixtures/fake-acp-agent.ts new file mode 100644 index 0000000000..cf41412046 --- /dev/null +++ b/packages/support/acp-snapshot/tests/fixtures/fake-acp-agent.ts @@ -0,0 +1,232 @@ +/** + * Scripted fake ACP agent bin for `dsh-acp-snapshot`'s unit specs. Speaks + * newline-delimited JSON-RPC on stdio like the real `dsh-acp-agent` bin, but + * every behavior — how prompts settle, whether session/new rejects, which + * session logs get persisted, what filesystem noise to leave — comes from a + * `behavior.json` sitting NEXT to the `$DSH_SNAPSHOT_FILE` fixture, so a spec + * scripts a whole subprocess run from data. The specs launch it through the + * REAL `runScenario` spawn path (tsx loader, temp cwd, env plumbing), so the + * harness plumbing is exercised for real; only the agent behind the protocol + * is scripted. + * + * The specs (not the golden tier) own this bin: it asserts nothing, echoes + * observable facts into `session/update` text chunks (env probe, permission + * outcome, seeded-workspace listing) for the spec to read off `rawStdout`, and + * exits 0 on stdin EOF after writing the scripted logs — mirroring the real + * bin's dispose-flush-exit shape. + */ + +import { mkdirSync, readFileSync, rmSync, writeFileSync } from 'node:fs' +import { readdirSync } from 'node:fs' +import { dirname, join } from 'node:path' +import { randomUUID } from 'node:crypto' +import { createInterface } from 'node:readline' + +/** One scripted session log: a file path under the sessions root plus its JSONL lines. */ +interface ScriptedLog { + /** Path relative to `$DSH_SNAPSHOT_SESSIONS_ROOT`, e.g. `bucket/a.jsonl` (an empty dir segment is invalid). */ + file: string + /** + * The JSONL records. String templates `{{CWD}}` and `{{SID}}` are replaced + * with the run's real cwd and the ACP session id this bin issued, so a + * written log carries genuine volatile values for the normalizers to scrub. + */ + lines: unknown[] +} + +/** The whole scripted behavior for one run. Every field defaults to the least surprising choice. */ +interface Behavior { + /** Reject every `session/new` (exercises the expect-error step without extra dirs). */ + rejectNewSession?: boolean + /** Reject `session/new` only when `additionalDirectories` is non-empty (the real bridge's rule). */ + rejectExtraDirs?: boolean + /** How `session/prompt` settles: a clean response, a JSON-RPC error, or a hang until `session/cancel`. */ + prompt?: 'respond' | 'error' | 'hang-until-cancel' + /** Before responding to a prompt, send a `session/request_permission` request and echo its outcome as a chunk. */ + permissionProbe?: boolean + /** Echo the `DSH_SNAPSHOT_*` env the harness set as a chunk (spec-side env-plumbing assertions). */ + echoEnv?: boolean + /** Echo the sorted cwd listing as a chunk (spec-side workspace-seeding assertions). */ + echoWorkspace?: boolean + /** Write a line to stderr on boot (spec-side stderr-capture assertions). */ + stderrNote?: string + /** Session logs to persist on stdin EOF. */ + logs?: ScriptedLog[] + /** Leave a stray FILE directly under the sessions root (harvest must skip it). */ + strayRootFile?: boolean + /** Leave a stray non-`.jsonl` file inside a bucket (harvest must skip it). */ + strayBucketFile?: boolean + /** Delete the sessions root entirely (harvest must yield no logs). */ + deleteSessionsRoot?: boolean +} + +const sessionsRoot = process.env.DSH_SNAPSHOT_SESSIONS_ROOT ?? '' +const fixtureFile = process.env.DSH_SNAPSHOT_FILE ?? '' +const behavior: Behavior = fixtureFile === '' + ? {} + : JSON.parse(readFileSync(join(dirname(fixtureFile), 'behavior.json'), 'utf8')) as Behavior + +if (behavior.stderrNote !== undefined) process.stderr.write(`${behavior.stderrNote}\n`) + +let nextOutboundId = 1000 +let sessionId = '' +/** + * The cwd the client passed to `session/new` — used verbatim for `{{CWD}}` + * substitution, mirroring the real bin (whose persisted header carries the + * session cwd as given, NOT `process.cwd()`, which the OS realpaths — on + * macOS `/var/folders/…` vs `/private/var/folders/…`). + */ +let sessionCwd = '' +/** The parked prompt request id while `hang-until-cancel` waits for the cancel notification. */ +let parkedPromptId: number | string | null = null +/** Resolvers for permission-probe responses, keyed by outbound request id. */ +const pendingPermission = new Map void>() + +function send(frame: Record): void { + process.stdout.write(`${JSON.stringify({ jsonrpc: '2.0', ...frame })}\n`) +} + +function respond(id: number | string, result: unknown): void { + send({ id, result }) +} + +function respondError(id: number | string, message: string): void { + send({ id, error: { code: -32603, message } }) +} + +function chunk(text: string): void { + send({ + method: 'session/update', + params: { sessionId, update: { sessionUpdate: 'agent_message_chunk', content: { type: 'text', text } } }, + }) +} + +/** Substitute the `{{CWD}}`/`{{SID}}` templates through a scripted log record. */ +function instantiate(value: unknown): unknown { + if (typeof value === 'string') return value.split('{{CWD}}').join(sessionCwd).split('{{SID}}').join(sessionId) + if (Array.isArray(value)) return value.map(instantiate) + if (value !== null && typeof value === 'object') { + const out: Record = {} + for (const [k, v] of Object.entries(value)) out[k] = instantiate(v) + return out + } + return value +} + +async function handlePrompt(id: number | string): Promise { + if ((behavior.prompt ?? 'respond') === 'hang-until-cancel') { + // A thought chunk BEFORE any message chunk: a promptAndCancel waiter + // watches for agent_message_chunk, so this exercises its non-matching + // update path while the waiter is armed. + send({ + method: 'session/update', + params: { sessionId, update: { sessionUpdate: 'agent_thought_chunk', content: { type: 'text', text: 'mulling' } } }, + }) + } + chunk('thinking about it') + if (behavior.echoEnv === true) { + chunk(`env:${JSON.stringify({ + mode: process.env.DSH_SNAPSHOT, + override: process.env.DSH_SNAPSHOT_OVERRIDE ?? null, + childFiles: process.env.DSH_SNAPSHOT_CHILD_FILES ?? null, + })}`) + } + if (behavior.echoWorkspace === true) { + chunk(`workspace:${readdirSync(process.cwd()).sort().join(',')}`) + } + if (behavior.permissionProbe === true) { + const requestId = nextOutboundId++ + const outcome = await new Promise((resolve) => { + pendingPermission.set(requestId, resolve) + send({ + id: requestId, + method: 'session/request_permission', + params: { + sessionId, + toolCall: { toolCallId: 'call_fake_1', title: 'fake tool', kind: 'execute', status: 'pending' }, + options: [ + { optionId: 'opt-allow', name: 'Allow once', kind: 'allow_once' }, + { optionId: 'opt-reject', name: 'Reject once', kind: 'reject_once' }, + ], + }, + }) + }) + chunk(`permission:${JSON.stringify(outcome)}`) + } + switch (behavior.prompt ?? 'respond') { + case 'respond': + respond(id, { stopReason: 'end_turn' }) + return + case 'error': + respondError(id, 'model exploded') + return + case 'hang-until-cancel': + parkedPromptId = id + return + } +} + +function handleFrame(frame: Record): void { + const id = frame.id as number | string | undefined + const method = frame.method as string | undefined + const params = (frame.params ?? {}) as Record + // A response to one of OUR outbound requests (the permission probe). + if (method === undefined && id !== undefined && typeof id === 'number' && pendingPermission.has(id)) { + const resolve = pendingPermission.get(id) as (outcome: unknown) => void + pendingPermission.delete(id) + resolve((frame.result as { outcome?: unknown } | undefined)?.outcome ?? null) + return + } + switch (method) { + case 'initialize': + respond(id as number | string, { protocolVersion: 1, agentCapabilities: { loadSession: false } }) + return + case 'session/new': { + const extra = params.additionalDirectories as unknown[] | undefined + if (behavior.rejectNewSession === true || (behavior.rejectExtraDirs === true && extra !== undefined && extra.length > 0)) { + respondError(id as number | string, 'unsupported workspace scope') + return + } + sessionId = randomUUID() + sessionCwd = typeof params.cwd === 'string' ? params.cwd : process.cwd() + respond(id as number | string, { sessionId }) + return + } + case 'session/prompt': + void handlePrompt(id as number | string) + return + case 'session/cancel': + if (parkedPromptId !== null) { + const parked = parkedPromptId + parkedPromptId = null + respond(parked, { stopReason: 'cancelled' }) + } + return + default: + // Unknown method: a notification is ignored; a request gets an error so + // the SDK never waits forever on a frame this fake doesn't model. + if (id !== undefined) respondError(id, `unhandled method ${String(method)}`) + } +} + +function flushLogsAndExit(): void { + for (const log of behavior.logs ?? []) { + const target = join(sessionsRoot, log.file) + mkdirSync(dirname(target), { recursive: true }) + writeFileSync(target, log.lines.map(l => JSON.stringify(instantiate(l))).join('\n') + '\n') + } + if (behavior.strayRootFile === true) writeFileSync(join(sessionsRoot, 'stray.txt'), 'not a bucket\n') + if (behavior.strayBucketFile === true) { + mkdirSync(join(sessionsRoot, 'bucket-noise'), { recursive: true }) + writeFileSync(join(sessionsRoot, 'bucket-noise', 'notes.txt'), 'not a session log\n') + } + if (behavior.deleteSessionsRoot === true) rmSync(sessionsRoot, { recursive: true, force: true }) + process.exit(0) +} + +const rl = createInterface({ input: process.stdin }) +rl.on('line', (line) => { + if (line.trim().length === 0) return + handleFrame(JSON.parse(line) as Record) +}) +rl.on('close', () => { flushLogsAndExit() }) diff --git a/packages/support/acp-snapshot/tests/fixtures/record-suite/rec-child/behavior.json b/packages/support/acp-snapshot/tests/fixtures/record-suite/rec-child/behavior.json new file mode 100644 index 0000000000..d44a3a9698 --- /dev/null +++ b/packages/support/acp-snapshot/tests/fixtures/record-suite/rec-child/behavior.json @@ -0,0 +1,13 @@ +{ + "prompt": "respond", + "logs": [ + { "file": "b/parent.jsonl", "lines": [ + { "type": "session", "id": "{{SID}}", "createdAt": 700, "cwd": "{{CWD}}" }, + { "type": "request/header", "seq": 0, "time": 3, "data": { "header": { "config": { "model": "fake" }, "system": "SYS PROMPT", "tools": [{ "name": "t1", "description": "D1", "parameters": { "type": "object" } }] }, "reason": "initial" } } + ]}, + { "file": "b/child.jsonl", "lines": [ + { "type": "session", "id": "abababab-cdcd-4efe-8ada-badabadabada", "createdAt": 800, "cwd": "{{CWD}}", "parentSession": "{{SID}}" }, + { "type": "request/header", "seq": 0, "time": 2, "data": { "header": { "config": { "model": "fake" }, "system": "SYS PROMPT", "tools": [{ "name": "t1", "description": "D1", "parameters": { "type": "object" } }] }, "reason": "initial" } } + ]} + ] +} diff --git a/packages/support/acp-snapshot/tests/fixtures/record-suite/rec-child/input.json b/packages/support/acp-snapshot/tests/fixtures/record-suite/rec-child/input.json new file mode 100644 index 0000000000..6d3e49b830 --- /dev/null +++ b/packages/support/acp-snapshot/tests/fixtures/record-suite/rec-child/input.json @@ -0,0 +1 @@ +{ "steps": [{ "op": "initialize" }, { "op": "newSession" }, { "op": "prompt", "text": "rec child" }] } diff --git a/packages/support/acp-snapshot/tests/fixtures/record-suite/rec-child/session.1.jsonl b/packages/support/acp-snapshot/tests/fixtures/record-suite/rec-child/session.1.jsonl new file mode 100644 index 0000000000..1caf2610b3 --- /dev/null +++ b/packages/support/acp-snapshot/tests/fixtures/record-suite/rec-child/session.1.jsonl @@ -0,0 +1,2 @@ +{"type":"session","id":"abababab-cdcd-4efe-8ada-badabadabada","createdAt":800,"cwd":"/var/folders/2g/b32ct0qn1d728l_v6tdkjytr0000gn/T/acp-snap-cwd-KBQJbW","parentSession":"f6fa7fcf-dd9c-4b39-8815-b25ddcebfd88"} +{"type":"request/header","seq":0,"time":2,"data":{"header":{"config":{"model":"fake"},"system":"{{system}}","tools":"{{tools}}"},"reason":"initial"}} diff --git a/packages/support/acp-snapshot/tests/fixtures/record-suite/rec-child/session.jsonl b/packages/support/acp-snapshot/tests/fixtures/record-suite/rec-child/session.jsonl new file mode 100644 index 0000000000..a2beac360d --- /dev/null +++ b/packages/support/acp-snapshot/tests/fixtures/record-suite/rec-child/session.jsonl @@ -0,0 +1,2 @@ +{"type":"session","id":"f6fa7fcf-dd9c-4b39-8815-b25ddcebfd88","createdAt":700,"cwd":"/var/folders/2g/b32ct0qn1d728l_v6tdkjytr0000gn/T/acp-snap-cwd-KBQJbW"} +{"type":"request/header","seq":0,"time":3,"data":{"header":{"config":{"model":"fake"},"system":"{{system}}","tools":"{{tools}}"},"reason":"initial"}} diff --git a/packages/support/acp-snapshot/tests/fixtures/record-suite/rec-child/stdout.golden.jsonl b/packages/support/acp-snapshot/tests/fixtures/record-suite/rec-child/stdout.golden.jsonl new file mode 100644 index 0000000000..f173b45b77 --- /dev/null +++ b/packages/support/acp-snapshot/tests/fixtures/record-suite/rec-child/stdout.golden.jsonl @@ -0,0 +1,4 @@ +{"jsonrpc":"2.0","id":1,"result":{"protocolVersion":1,"agentCapabilities":{"loadSession":false}}} +{"jsonrpc":"2.0","id":2,"result":{"sessionId":"{{sessionId}}"}} +{"jsonrpc":"2.0","method":"session/update","params":{"sessionId":"{{sessionId}}","update":{"sessionUpdate":"agent_message_chunk","content":{"type":"text","text":"thinking about it"}}}} +{"jsonrpc":"2.0","id":3,"result":{"stopReason":"end_turn"}} diff --git a/packages/support/acp-snapshot/tests/fixtures/record-suite/rec-pin/behavior.json b/packages/support/acp-snapshot/tests/fixtures/record-suite/rec-pin/behavior.json new file mode 100644 index 0000000000..a24e30d80a --- /dev/null +++ b/packages/support/acp-snapshot/tests/fixtures/record-suite/rec-pin/behavior.json @@ -0,0 +1,10 @@ +{ + "prompt": "respond", + "logs": [{ + "file": "b/main.jsonl", + "lines": [ + { "type": "session", "id": "{{SID}}", "createdAt": 600, "cwd": "{{CWD}}" }, + { "type": "request/header", "seq": 0, "time": 4, "data": { "header": { "config": { "model": "fake" }, "system": "SYS PROMPT", "tools": [{ "name": "t1", "description": "D1", "parameters": { "type": "object" } }] }, "reason": "initial" } } + ] + }] +} diff --git a/packages/support/acp-snapshot/tests/fixtures/record-suite/rec-pin/input.json b/packages/support/acp-snapshot/tests/fixtures/record-suite/rec-pin/input.json new file mode 100644 index 0000000000..9573d20b27 --- /dev/null +++ b/packages/support/acp-snapshot/tests/fixtures/record-suite/rec-pin/input.json @@ -0,0 +1 @@ +{ "steps": [{ "op": "initialize" }, { "op": "newSession" }, { "op": "prompt", "text": "rec pin" }] } diff --git a/packages/support/acp-snapshot/tests/fixtures/record-suite/rec-pin/session.jsonl b/packages/support/acp-snapshot/tests/fixtures/record-suite/rec-pin/session.jsonl new file mode 100644 index 0000000000..109a192083 --- /dev/null +++ b/packages/support/acp-snapshot/tests/fixtures/record-suite/rec-pin/session.jsonl @@ -0,0 +1,2 @@ +{"type":"session","id":"ccdc749f-56f3-4267-9750-598b5c60b7b2","createdAt":600,"cwd":"/var/folders/2g/b32ct0qn1d728l_v6tdkjytr0000gn/T/acp-snap-cwd-nOQ4Gy"} +{"type":"request/header","seq":0,"time":4,"data":{"header":{"config":{"model":"fake"},"system":"SYS PROMPT","tools":[{"name":"t1","description":"D1","parameters":{"type":"object"}}]},"reason":"initial"}} diff --git a/packages/support/acp-snapshot/tests/fixtures/record-suite/rec-pin/stdout.golden.jsonl b/packages/support/acp-snapshot/tests/fixtures/record-suite/rec-pin/stdout.golden.jsonl new file mode 100644 index 0000000000..f173b45b77 --- /dev/null +++ b/packages/support/acp-snapshot/tests/fixtures/record-suite/rec-pin/stdout.golden.jsonl @@ -0,0 +1,4 @@ +{"jsonrpc":"2.0","id":1,"result":{"protocolVersion":1,"agentCapabilities":{"loadSession":false}}} +{"jsonrpc":"2.0","id":2,"result":{"sessionId":"{{sessionId}}"}} +{"jsonrpc":"2.0","method":"session/update","params":{"sessionId":"{{sessionId}}","update":{"sessionUpdate":"agent_message_chunk","content":{"type":"text","text":"thinking about it"}}}} +{"jsonrpc":"2.0","id":3,"result":{"stopReason":"end_turn"}} diff --git a/packages/support/acp-snapshot/tests/fixtures/record-suite/rec-skip/behavior.json b/packages/support/acp-snapshot/tests/fixtures/record-suite/rec-skip/behavior.json new file mode 100644 index 0000000000..0967ef424b --- /dev/null +++ b/packages/support/acp-snapshot/tests/fixtures/record-suite/rec-skip/behavior.json @@ -0,0 +1 @@ +{} diff --git a/packages/support/acp-snapshot/tests/fixtures/record-suite/rec-skip/input.json b/packages/support/acp-snapshot/tests/fixtures/record-suite/rec-skip/input.json new file mode 100644 index 0000000000..d1e94c22eb --- /dev/null +++ b/packages/support/acp-snapshot/tests/fixtures/record-suite/rec-skip/input.json @@ -0,0 +1 @@ +{ "steps": [{ "op": "initialize" }] } diff --git a/packages/support/acp-snapshot/tests/fixtures/record-suite/rec-skip/replay.override.json b/packages/support/acp-snapshot/tests/fixtures/record-suite/rec-skip/replay.override.json new file mode 100644 index 0000000000..8ed00c0651 --- /dev/null +++ b/packages/support/acp-snapshot/tests/fixtures/record-suite/rec-skip/replay.override.json @@ -0,0 +1 @@ +[{ "kind": "hang" }] diff --git a/packages/support/acp-snapshot/tests/fixtures/record-suite/rec-skip/session.jsonl b/packages/support/acp-snapshot/tests/fixtures/record-suite/rec-skip/session.jsonl new file mode 100644 index 0000000000..104f2a0df2 --- /dev/null +++ b/packages/support/acp-snapshot/tests/fixtures/record-suite/rec-skip/session.jsonl @@ -0,0 +1 @@ +{"type":"session","id":"{{sessionId}}","createdAt":0,"cwd":"{{cwd}}"} diff --git a/packages/support/acp-snapshot/tests/fixtures/record-suite/rec-skip/stdout.golden.jsonl b/packages/support/acp-snapshot/tests/fixtures/record-suite/rec-skip/stdout.golden.jsonl new file mode 100644 index 0000000000..d6a1d2b232 --- /dev/null +++ b/packages/support/acp-snapshot/tests/fixtures/record-suite/rec-skip/stdout.golden.jsonl @@ -0,0 +1 @@ +{"jsonrpc":"2.0","id":1,"result":{"protocolVersion":1,"agentCapabilities":{"loadSession":false}}} diff --git a/packages/support/acp-snapshot/tests/fixtures/suite/authored-error/behavior.json b/packages/support/acp-snapshot/tests/fixtures/suite/authored-error/behavior.json new file mode 100644 index 0000000000..808d9672b9 --- /dev/null +++ b/packages/support/acp-snapshot/tests/fixtures/suite/authored-error/behavior.json @@ -0,0 +1,10 @@ +{ + "prompt": "error", + "logs": [{ + "file": "b/main.jsonl", + "lines": [ + { "type": "session", "id": "{{SID}}", "createdAt": 500, "cwd": "{{CWD}}" }, + { "type": "turn/end", "seq": 1, "time": 9, "data": { "error": "model exploded" } } + ] + }] +} diff --git a/packages/support/acp-snapshot/tests/fixtures/suite/authored-error/input.json b/packages/support/acp-snapshot/tests/fixtures/suite/authored-error/input.json new file mode 100644 index 0000000000..c281971465 --- /dev/null +++ b/packages/support/acp-snapshot/tests/fixtures/suite/authored-error/input.json @@ -0,0 +1 @@ +{ "steps": [{ "op": "initialize" }, { "op": "newSession" }, { "op": "promptExpectError", "text": "boom" }] } diff --git a/packages/support/acp-snapshot/tests/fixtures/suite/authored-error/replay.override.json b/packages/support/acp-snapshot/tests/fixtures/suite/authored-error/replay.override.json new file mode 100644 index 0000000000..e868115f35 --- /dev/null +++ b/packages/support/acp-snapshot/tests/fixtures/suite/authored-error/replay.override.json @@ -0,0 +1 @@ +[{ "kind": "throw", "chunks": [], "message": "model exploded", "code": "PROVIDER" }] diff --git a/packages/support/acp-snapshot/tests/fixtures/suite/authored-error/session.jsonl b/packages/support/acp-snapshot/tests/fixtures/suite/authored-error/session.jsonl new file mode 100644 index 0000000000..36991a214e --- /dev/null +++ b/packages/support/acp-snapshot/tests/fixtures/suite/authored-error/session.jsonl @@ -0,0 +1,2 @@ +{"type":"session","id":"44444444-3333-4222-8111-000000000000","createdAt":17,"cwd":"/rec/authored-cwd"} +{"type":"turn/end","seq":1,"time":17,"data":{"error":"model exploded"}} diff --git a/packages/support/acp-snapshot/tests/fixtures/suite/authored-error/stdout.golden.jsonl b/packages/support/acp-snapshot/tests/fixtures/suite/authored-error/stdout.golden.jsonl new file mode 100644 index 0000000000..2a1d69bd93 --- /dev/null +++ b/packages/support/acp-snapshot/tests/fixtures/suite/authored-error/stdout.golden.jsonl @@ -0,0 +1,4 @@ +{"jsonrpc":"2.0","id":1,"result":{"protocolVersion":1,"agentCapabilities":{"loadSession":false}}} +{"jsonrpc":"2.0","id":2,"result":{"sessionId":"{{sessionId}}"}} +{"jsonrpc":"2.0","method":"session/update","params":{"sessionId":"{{sessionId}}","update":{"sessionUpdate":"agent_message_chunk","content":{"type":"text","text":"thinking about it"}}}} +{"jsonrpc":"2.0","id":3,"error":{"code":-32603,"message":"model exploded"}} diff --git a/packages/support/acp-snapshot/tests/fixtures/suite/blocked-log/behavior.json b/packages/support/acp-snapshot/tests/fixtures/suite/blocked-log/behavior.json new file mode 100644 index 0000000000..e0a438297d --- /dev/null +++ b/packages/support/acp-snapshot/tests/fixtures/suite/blocked-log/behavior.json @@ -0,0 +1,10 @@ +{ + "prompt": "error", + "logs": [{ + "file": "b/main.jsonl", + "lines": [ + { "type": "session", "id": "{{SID}}", "createdAt": 400, "cwd": "{{CWD}}" }, + { "type": "hook/result", "seq": 1, "time": 8, "data": { "decision": "block", "durationMs": 37 } } + ] + }] +} diff --git a/packages/support/acp-snapshot/tests/fixtures/suite/blocked-log/input.json b/packages/support/acp-snapshot/tests/fixtures/suite/blocked-log/input.json new file mode 100644 index 0000000000..0a711dca3c --- /dev/null +++ b/packages/support/acp-snapshot/tests/fixtures/suite/blocked-log/input.json @@ -0,0 +1 @@ +{ "steps": [{ "op": "initialize" }, { "op": "newSession" }, { "op": "promptExpectError", "text": "blocked" }] } diff --git a/packages/support/acp-snapshot/tests/fixtures/suite/blocked-log/session.jsonl b/packages/support/acp-snapshot/tests/fixtures/suite/blocked-log/session.jsonl new file mode 100644 index 0000000000..6d8474812d --- /dev/null +++ b/packages/support/acp-snapshot/tests/fixtures/suite/blocked-log/session.jsonl @@ -0,0 +1,2 @@ +{"type":"session","id":"99999999-8888-4777-8666-555555555555","createdAt":13,"cwd":"/rec/blocked-cwd"} +{"type":"hook/result","seq":1,"time":13,"data":{"decision":"block","durationMs":99}} diff --git a/packages/support/acp-snapshot/tests/fixtures/suite/blocked-log/stdout.golden.jsonl b/packages/support/acp-snapshot/tests/fixtures/suite/blocked-log/stdout.golden.jsonl new file mode 100644 index 0000000000..2a1d69bd93 --- /dev/null +++ b/packages/support/acp-snapshot/tests/fixtures/suite/blocked-log/stdout.golden.jsonl @@ -0,0 +1,4 @@ +{"jsonrpc":"2.0","id":1,"result":{"protocolVersion":1,"agentCapabilities":{"loadSession":false}}} +{"jsonrpc":"2.0","id":2,"result":{"sessionId":"{{sessionId}}"}} +{"jsonrpc":"2.0","method":"session/update","params":{"sessionId":"{{sessionId}}","update":{"sessionUpdate":"agent_message_chunk","content":{"type":"text","text":"thinking about it"}}}} +{"jsonrpc":"2.0","id":3,"error":{"code":-32603,"message":"model exploded"}} diff --git a/packages/support/acp-snapshot/tests/fixtures/suite/no-model/behavior.json b/packages/support/acp-snapshot/tests/fixtures/suite/no-model/behavior.json new file mode 100644 index 0000000000..0967ef424b --- /dev/null +++ b/packages/support/acp-snapshot/tests/fixtures/suite/no-model/behavior.json @@ -0,0 +1 @@ +{} diff --git a/packages/support/acp-snapshot/tests/fixtures/suite/no-model/input.json b/packages/support/acp-snapshot/tests/fixtures/suite/no-model/input.json new file mode 100644 index 0000000000..d1e94c22eb --- /dev/null +++ b/packages/support/acp-snapshot/tests/fixtures/suite/no-model/input.json @@ -0,0 +1 @@ +{ "steps": [{ "op": "initialize" }] } diff --git a/packages/support/acp-snapshot/tests/fixtures/suite/no-model/session.jsonl b/packages/support/acp-snapshot/tests/fixtures/suite/no-model/session.jsonl new file mode 100644 index 0000000000..104f2a0df2 --- /dev/null +++ b/packages/support/acp-snapshot/tests/fixtures/suite/no-model/session.jsonl @@ -0,0 +1 @@ +{"type":"session","id":"{{sessionId}}","createdAt":0,"cwd":"{{cwd}}"} diff --git a/packages/support/acp-snapshot/tests/fixtures/suite/no-model/stdout.golden.jsonl b/packages/support/acp-snapshot/tests/fixtures/suite/no-model/stdout.golden.jsonl new file mode 100644 index 0000000000..d6a1d2b232 --- /dev/null +++ b/packages/support/acp-snapshot/tests/fixtures/suite/no-model/stdout.golden.jsonl @@ -0,0 +1 @@ +{"jsonrpc":"2.0","id":1,"result":{"protocolVersion":1,"agentCapabilities":{"loadSession":false}}} diff --git a/packages/support/acp-snapshot/tests/fixtures/suite/pin-turn/behavior.json b/packages/support/acp-snapshot/tests/fixtures/suite/pin-turn/behavior.json new file mode 100644 index 0000000000..422e0a17e6 --- /dev/null +++ b/packages/support/acp-snapshot/tests/fixtures/suite/pin-turn/behavior.json @@ -0,0 +1,11 @@ +{ + "prompt": "respond", + "logs": [{ + "file": "b/main.jsonl", + "lines": [ + { "type": "session", "id": "{{SID}}", "createdAt": 100, "cwd": "{{CWD}}" }, + { "type": "request/header", "seq": 0, "time": 100, "data": { "header": { "config": { "model": "fake" }, "system": "SYS PROMPT", "tools": [{ "name": "t1", "description": "D1", "parameters": { "type": "object" } }] }, "reason": "initial" } }, + { "type": "turn/start", "seq": 1, "time": 100, "data": { "turn": 1 } } + ] + }] +} diff --git a/packages/support/acp-snapshot/tests/fixtures/suite/pin-turn/input.json b/packages/support/acp-snapshot/tests/fixtures/suite/pin-turn/input.json new file mode 100644 index 0000000000..b9e2d9bbc5 --- /dev/null +++ b/packages/support/acp-snapshot/tests/fixtures/suite/pin-turn/input.json @@ -0,0 +1 @@ +{ "steps": [{ "op": "initialize" }, { "op": "newSession" }, { "op": "prompt", "text": "pin" }] } diff --git a/packages/support/acp-snapshot/tests/fixtures/suite/pin-turn/session.jsonl b/packages/support/acp-snapshot/tests/fixtures/suite/pin-turn/session.jsonl new file mode 100644 index 0000000000..87bf09c839 --- /dev/null +++ b/packages/support/acp-snapshot/tests/fixtures/suite/pin-turn/session.jsonl @@ -0,0 +1,3 @@ +{"type":"session","id":"12121212-3434-4545-8686-787878787878","createdAt":7,"cwd":"/rec/pin-cwd"} +{"type":"request/header","seq":0,"time":7,"data":{"header":{"config":{"model":"fake"},"system":"SYS PROMPT","tools":[{"name":"t1","description":"D1","parameters":{"type":"object"}}]},"reason":"initial"}} +{"type":"turn/start","seq":1,"time":7,"data":{"turn":1}} diff --git a/packages/support/acp-snapshot/tests/fixtures/suite/pin-turn/stdout.golden.jsonl b/packages/support/acp-snapshot/tests/fixtures/suite/pin-turn/stdout.golden.jsonl new file mode 100644 index 0000000000..f173b45b77 --- /dev/null +++ b/packages/support/acp-snapshot/tests/fixtures/suite/pin-turn/stdout.golden.jsonl @@ -0,0 +1,4 @@ +{"jsonrpc":"2.0","id":1,"result":{"protocolVersion":1,"agentCapabilities":{"loadSession":false}}} +{"jsonrpc":"2.0","id":2,"result":{"sessionId":"{{sessionId}}"}} +{"jsonrpc":"2.0","method":"session/update","params":{"sessionId":"{{sessionId}}","update":{"sessionUpdate":"agent_message_chunk","content":{"type":"text","text":"thinking about it"}}}} +{"jsonrpc":"2.0","id":3,"result":{"stopReason":"end_turn"}} diff --git a/packages/support/acp-snapshot/tests/fixtures/suite/plain-turn/behavior.json b/packages/support/acp-snapshot/tests/fixtures/suite/plain-turn/behavior.json new file mode 100644 index 0000000000..d5cbbf9d28 --- /dev/null +++ b/packages/support/acp-snapshot/tests/fixtures/suite/plain-turn/behavior.json @@ -0,0 +1,15 @@ +{ + "prompt": "respond", + "echoWorkspace": true, + "logs": [ + { "file": "b/parent.jsonl", "lines": [ + { "type": "session", "id": "{{SID}}", "createdAt": 200, "cwd": "{{CWD}}" }, + { "type": "request/header", "seq": 0, "time": 5, "data": { "header": { "config": { "model": "fake" }, "system": "SYS PROMPT", "tools": [{ "name": "t1", "description": "D1", "parameters": { "type": "object" } }] }, "reason": "initial" } }, + { "type": "assistant/chunk", "seq": 1, "time": 5, "data": { "turn": 1, "step": 1, "chunk": { "type": "text-delta", "index": 0, "text": "hi" } } } + ]}, + { "file": "b/child.jsonl", "lines": [ + { "type": "session", "id": "eeeeeeee-1111-4222-8333-444444444444", "createdAt": 300, "cwd": "{{CWD}}", "parentSession": "{{SID}}" }, + { "type": "request/header", "seq": 0, "time": 6, "data": { "header": { "config": { "model": "fake" }, "system": "SYS PROMPT", "tools": [{ "name": "t1", "description": "D1", "parameters": { "type": "object" } }] }, "reason": "initial" } } + ]} + ] +} diff --git a/packages/support/acp-snapshot/tests/fixtures/suite/plain-turn/input.json b/packages/support/acp-snapshot/tests/fixtures/suite/plain-turn/input.json new file mode 100644 index 0000000000..60b9e363b5 --- /dev/null +++ b/packages/support/acp-snapshot/tests/fixtures/suite/plain-turn/input.json @@ -0,0 +1 @@ +{ "steps": [{ "op": "initialize" }, { "op": "newSession" }, { "op": "prompt", "text": "plain" }] } diff --git a/packages/support/acp-snapshot/tests/fixtures/suite/plain-turn/session.1.jsonl b/packages/support/acp-snapshot/tests/fixtures/suite/plain-turn/session.1.jsonl new file mode 100644 index 0000000000..a844f891fc --- /dev/null +++ b/packages/support/acp-snapshot/tests/fixtures/suite/plain-turn/session.1.jsonl @@ -0,0 +1,2 @@ +{"type":"session","id":"eeeeeeee-1111-4222-8333-444444444444","createdAt":12,"cwd":"/rec/plain-cwd","parentSession":"56565656-7878-4989-8a9a-9b9b9b9b9b9b"} +{"type":"request/header","seq":0,"time":12,"data":{"header":{"config":{"model":"fake"},"system":"{{system}}","tools":"{{tools}}"},"reason":"initial"}} diff --git a/packages/support/acp-snapshot/tests/fixtures/suite/plain-turn/session.jsonl b/packages/support/acp-snapshot/tests/fixtures/suite/plain-turn/session.jsonl new file mode 100644 index 0000000000..744998f959 --- /dev/null +++ b/packages/support/acp-snapshot/tests/fixtures/suite/plain-turn/session.jsonl @@ -0,0 +1,3 @@ +{"type":"session","id":"56565656-7878-4989-8a9a-9b9b9b9b9b9b","createdAt":11,"cwd":"/rec/plain-cwd"} +{"type":"request/header","seq":0,"time":11,"data":{"header":{"config":{"model":"fake"},"system":"{{system}}","tools":"{{tools}}"},"reason":"initial"}} +{"type":"assistant/chunk","seq":1,"time":11,"data":{"turn":1,"step":1,"chunk":{"type":"text-delta","index":0,"text":"hi"}}} diff --git a/packages/support/acp-snapshot/tests/fixtures/suite/plain-turn/stdout.golden.jsonl b/packages/support/acp-snapshot/tests/fixtures/suite/plain-turn/stdout.golden.jsonl new file mode 100644 index 0000000000..d0242ae39f --- /dev/null +++ b/packages/support/acp-snapshot/tests/fixtures/suite/plain-turn/stdout.golden.jsonl @@ -0,0 +1,5 @@ +{"jsonrpc":"2.0","id":1,"result":{"protocolVersion":1,"agentCapabilities":{"loadSession":false}}} +{"jsonrpc":"2.0","id":2,"result":{"sessionId":"{{sessionId}}"}} +{"jsonrpc":"2.0","method":"session/update","params":{"sessionId":"{{sessionId}}","update":{"sessionUpdate":"agent_message_chunk","content":{"type":"text","text":"thinking about it"}}}} +{"jsonrpc":"2.0","method":"session/update","params":{"sessionId":"{{sessionId}}","update":{"sessionUpdate":"agent_message_chunk","content":{"type":"text","text":"workspace:seed.txt"}}}} +{"jsonrpc":"2.0","id":3,"result":{"stopReason":"end_turn"}} diff --git a/packages/support/acp-snapshot/tests/fixtures/suite/plain-turn/workspace/seed.txt b/packages/support/acp-snapshot/tests/fixtures/suite/plain-turn/workspace/seed.txt new file mode 100644 index 0000000000..c19e887d68 --- /dev/null +++ b/packages/support/acp-snapshot/tests/fixtures/suite/plain-turn/workspace/seed.txt @@ -0,0 +1 @@ +seeded diff --git a/packages/support/acp-snapshot/tests/harness.spec.ts b/packages/support/acp-snapshot/tests/harness.spec.ts new file mode 100644 index 0000000000..3884e36095 --- /dev/null +++ b/packages/support/acp-snapshot/tests/harness.spec.ts @@ -0,0 +1,233 @@ +import { mkdtemp, rm, writeFile } from 'node:fs/promises' +import { tmpdir } from 'node:os' +import { delimiter, join } from 'node:path' +import { fileURLToPath } from 'node:url' +import { afterAll, describe, expect, it } from 'vitest' +import { runScenario, type AgentUnderTest, type InputStep } from '../src/harness.ts' + +/** + * Unit tests for the subprocess harness, driven through the REAL spawn path + * (tsx loader, temp cwd, env plumbing) against the scripted fake ACP bin in + * ./fixtures/fake-acp-agent.ts. Each case writes a `behavior.json` next to a + * throwaway fixture path; the fake bin echoes observable facts (env, seeded + * workspace, permission outcomes) into `agent_message_chunk` text, so the + * assertions read plain `rawStdout`. + */ + +const AGENT: AgentUnderTest = { + binScript: fileURLToPath(new URL('./fixtures/fake-acp-agent.ts', import.meta.url)), + // The fake bin ignores its config argv; any real path documents the shape. + configPath: fileURLToPath(new URL('./fixtures/fake-acp-agent.ts', import.meta.url)), + tsconfigPath: fileURLToPath(new URL('../../../../tsconfig.json', import.meta.url)), +} + +/** Temp scenario dirs to drop after the suite. */ +const tempDirs: string[] = [] +afterAll(async () => { + for (const dir of tempDirs) await rm(dir, { recursive: true, force: true }) +}) + +/** Write a behavior.json into a fresh temp dir; return the sibling fixture path the harness points the bin at. */ +async function scenario(behavior: object): Promise<{ dir: string; fixtureFile: string }> { + const dir = await mkdtemp(join(tmpdir(), 'acp-snap-spec-')) + tempDirs.push(dir) + await writeFile(join(dir, 'behavior.json'), JSON.stringify(behavior)) + return { dir, fixtureFile: join(dir, 'session.jsonl') } +} + +const boot: InputStep[] = [{ op: 'initialize' }, { op: 'newSession' }] + +describe('runScenario', () => { + it('drives a full turn: initialize (terminal caps), session, prompt, permission stub, harvest', { timeout: 20_000 }, async () => { + const { fixtureFile } = await scenario({ + permissionProbe: true, + logs: [{ + file: 'bucket/main.jsonl', + lines: [ + { type: 'session', id: '{{SID}}', createdAt: 42, cwd: '{{CWD}}' }, + { type: 'turn/start', seq: 1, time: 9, data: { turn: 1 } }, + ], + }], + }) + const result = await runScenario( + { steps: [{ op: 'initialize', terminalOutput: true }, { op: 'newSession' }, { op: 'prompt', text: 'go' }] }, + { agent: AGENT, mode: 'replay', fixtureFile }, + ) + expect(result.sessionId).toBeDefined() + // The harness's client answers a permission request with `cancelled`; the + // fake bin echoes the outcome it received back as a chunk. + expect(result.rawStdout).toContain('permission:{\\"outcome\\":\\"cancelled\\"}') + expect(result.sessionLogs).toHaveLength(1) + expect(result.sessionLogs[0]?.id).toBe(result.sessionId) + expect(result.sessionLogs[0]?.createdAt).toBe(42) + expect(result.sessionLogs[0]?.content).toContain('turn/start') + // The harvested log embeds the run's REAL temp cwd (template-substituted). + expect(result.sessionLogs[0]?.content).toContain(result.cwd) + }) + + it('forwards override/child fixture paths into the child env and captures stderr', { timeout: 20_000 }, async () => { + const { dir, fixtureFile } = await scenario({ echoEnv: true, stderrNote: 'fake bin booted' }) + const childFiles = [join(dir, 'session.1.jsonl'), join(dir, 'session.2.jsonl')] + const result = await runScenario( + { steps: [...boot, { op: 'prompt', text: 'env?' }] }, + { + agent: AGENT, + mode: 'replay', + fixtureFile, + overrideFile: join(dir, 'replay.override.json'), + childFiles, + // A workspaceDir that does not exist is skipped, not an error. + workspaceDir: join(dir, 'no-such-workspace'), + }, + ) + expect(result.stderr).toContain('fake bin booted') + expect(result.rawStdout).toContain('replay.override.json') + // Child paths ride one env var, joined with the platform delimiter. + expect(result.rawStdout).toContain(JSON.stringify(childFiles.join(delimiter)).slice(1, -1)) + }) + + it('seeds the workspace dir into the temp cwd before the run', { timeout: 20_000 }, async () => { + const { dir, fixtureFile } = await scenario({ echoWorkspace: true }) + const workspaceDir = join(dir, 'workspace') + await writeFile(join(dir, 'behavior.json'), JSON.stringify({ echoWorkspace: true })) + const { mkdir } = await import('node:fs/promises') + await mkdir(workspaceDir, { recursive: true }) + await writeFile(join(workspaceDir, 'seeded.txt'), 'hello') + const result = await runScenario( + { steps: [...boot, { op: 'prompt', text: 'ls' }] }, + { agent: AGENT, mode: 'replay', fixtureFile, workspaceDir }, + ) + expect(result.rawStdout).toContain('workspace:seeded.txt') + }) + + it('promptAndCancel waits for the streamed chunk, cancels, and settles the prompt', { timeout: 20_000 }, async () => { + const { fixtureFile } = await scenario({ prompt: 'hang-until-cancel' }) + const result = await runScenario( + { steps: [...boot, { op: 'promptAndCancel', text: 'hang' }] }, + { agent: AGENT, mode: 'replay', fixtureFile }, + ) + expect(result.rawStdout).toContain('"stopReason":"cancelled"') + // The streamed chunk deterministically precedes the cancelled response. + expect(result.rawStdout.indexOf('thinking about it')).toBeLessThan(result.rawStdout.indexOf('cancelled')) + }) + + it('promptExpectError swallows a model-error response as the expected outcome', { timeout: 20_000 }, async () => { + const { fixtureFile } = await scenario({ prompt: 'error' }) + const result = await runScenario( + { steps: [...boot, { op: 'promptExpectError', text: 'boom' }] }, + { agent: AGENT, mode: 'replay', fixtureFile }, + ) + expect(result.rawStdout).toContain('model exploded') + }) + + it('promptExpectError throws when the prompt unexpectedly succeeds (and teardown kills the live child)', { timeout: 20_000 }, async () => { + const { fixtureFile } = await scenario({ prompt: 'respond' }) + await expect(runScenario( + { steps: [...boot, { op: 'promptExpectError', text: 'fine' }] }, + { agent: AGENT, mode: 'replay', fixtureFile }, + )).rejects.toThrow(/expected the prompt to fail/) + }) + + it('newSessionExpectError swallows the rejection, with and without extra dirs', { timeout: 20_000 }, async () => { + const { fixtureFile } = await scenario({ rejectExtraDirs: true }) + const result = await runScenario( + { steps: [{ op: 'initialize' }, { op: 'newSessionExpectError', additionalDirectories: ['/elsewhere'] }] }, + { agent: AGENT, mode: 'replay', fixtureFile }, + ) + // No session was created, so no id and no logs. + expect(result.sessionId).toBeUndefined() + expect(result.sessionLogs).toHaveLength(0) + + const rejectAll = await scenario({ rejectNewSession: true }) + const second = await runScenario( + { steps: [{ op: 'initialize' }, { op: 'newSessionExpectError' }] }, + { agent: AGENT, mode: 'replay', fixtureFile: rejectAll.fixtureFile }, + ) + expect(second.rawStdout).toContain('unsupported workspace scope') + }) + + it('newSessionExpectError throws when session/new unexpectedly succeeds', { timeout: 20_000 }, async () => { + const { fixtureFile } = await scenario({}) + await expect(runScenario( + { steps: [{ op: 'initialize' }, { op: 'newSessionExpectError' }] }, + { agent: AGENT, mode: 'replay', fixtureFile }, + )).rejects.toThrow(/expected session\/new to be rejected/) + }) + + it('a plain cancel step is forwarded (and ignored by an idle agent)', { timeout: 20_000 }, async () => { + const { fixtureFile } = await scenario({}) + const result = await runScenario( + { steps: [...boot, { op: 'cancel' }] }, + { agent: AGENT, mode: 'replay', fixtureFile }, + ) + expect(result.sessionId).toBeDefined() + }) + + it.each([ + [{ op: 'prompt', text: 'x' }, /prompt before newSession/], + [{ op: 'promptExpectError', text: 'x' }, /promptExpectError before newSession/], + [{ op: 'promptAndCancel', text: 'x' }, /promptAndCancel before newSession/], + [{ op: 'cancel' }, /cancel before newSession/], + ] as [InputStep, RegExp][])('rejects %j before newSession', { timeout: 20_000 }, async (step, message) => { + const { fixtureFile } = await scenario({}) + await expect(runScenario( + { steps: [{ op: 'initialize' }, step] }, + { agent: AGENT, mode: 'replay', fixtureFile }, + )).rejects.toThrow(message) + }) + + it('rejects an unknown input op', { timeout: 20_000 }, async () => { + const { fixtureFile } = await scenario({}) + const bogus = { op: 'reticulate' } as unknown as InputStep + await expect(runScenario( + { steps: [bogus] }, + { agent: AGENT, mode: 'replay', fixtureFile }, + )).rejects.toThrow(/unknown input op/) + }) + + it('harvests all logs primary-first, children by createdAt then id, skipping filesystem noise', { timeout: 20_000 }, async () => { + const { fixtureFile } = await scenario({ + strayRootFile: true, + strayBucketFile: true, + logs: [ + // File names chosen so readdir feeds the sort children-first AND + // parent-in-the-middle: the comparator then sees a parent on both + // sides of a pair, plus the same-createdAt (localeCompare) tiebreak. + { file: 'b1/aa-child-c.jsonl', lines: [{ type: 'session', id: 'cccccccc-0000-4000-8000-000000000000', createdAt: 500, parentSession: '{{SID}}' }] }, + { file: 'b1/bb-parent.jsonl', lines: [{ type: 'session', id: '{{SID}}', createdAt: 900 }] }, + { file: 'b1/cc-child-a.jsonl', lines: [{ type: 'session', id: 'aaaaaaaa-0000-4000-8000-000000000000', createdAt: 500, parentSession: '{{SID}}' }] }, + // Missing id/createdAt fall back to ''/0; earliest child by createdAt. + { file: 'b2/orphan-fields.jsonl', lines: [{ type: 'session', parentSession: '{{SID}}' }] }, + ], + }) + const result = await runScenario( + { steps: [...boot, { op: 'prompt', text: 'go' }] }, + { agent: AGENT, mode: 'replay', fixtureFile }, + ) + expect(result.sessionLogs.map(l => [l.id, l.createdAt])).toEqual([ + [result.sessionId, 900], + ['', 0], + ['aaaaaaaa-0000-4000-8000-000000000000', 500], + ['cccccccc-0000-4000-8000-000000000000', 500], + ]) + expect(result.sessionLogs[1]?.parentSession).toBe(result.sessionId) + }) + + it('treats an empty log file as a header-less primary with default fields', { timeout: 20_000 }, async () => { + const { fixtureFile } = await scenario({ logs: [{ file: 'b/empty.jsonl', lines: [] }] }) + const result = await runScenario( + { steps: boot }, + { agent: AGENT, mode: 'replay', fixtureFile }, + ) + expect(result.sessionLogs.map(l => [l.id, l.createdAt, l.parentSession])).toEqual([['', 0, undefined]]) + }) + + it('yields no logs when the sessions root vanished', { timeout: 20_000 }, async () => { + const { fixtureFile } = await scenario({ deleteSessionsRoot: true }) + const result = await runScenario( + { steps: boot }, + { agent: AGENT, mode: 'replay', fixtureFile }, + ) + expect(result.sessionLogs).toHaveLength(0) + }) +}) diff --git a/packages/support/acp-snapshot/tests/normalize.spec.ts b/packages/support/acp-snapshot/tests/normalize.spec.ts index 56c15306a3..8ebd1412b9 100644 --- a/packages/support/acp-snapshot/tests/normalize.spec.ts +++ b/packages/support/acp-snapshot/tests/normalize.spec.ts @@ -107,6 +107,17 @@ describe('normalizeSessionLog', () => { const out = normalizeSessionLog(`${header({})}\n${ev}\n`, ctx) expect(out).toContain('"durationMs":88') }) + + it('tolerates records missing the volatile fields it would zero', () => { + const bareHeader = JSON.stringify({ type: 'session', id: 's' }) + const timeless = JSON.stringify({ type: 'note', seq: 1 }) + const bareHook = JSON.stringify({ type: 'hook/result', seq: 2, time: 5, data: { decision: 'allow' } }) + const nullDataHook = JSON.stringify({ type: 'hook/result', seq: 3, time: 6, data: null }) + const out = normalizeSessionLog(`${bareHeader}\n${timeless}\n${bareHook}\n${nullDataHook}\n`, ctx) + expect(out).toContain('"type":"note","seq":1') + expect(out).toContain('"decision":"allow"') + expect(out).not.toContain('durationMs') + }) }) describe('scrubRequestHeaders', () => { @@ -135,6 +146,40 @@ describe('scrubRequestHeaders', () => { expect(out).not.toContain('{{tools}}') }) + it('scrubs a header carrying only one of system/tools, leaving the other absent', () => { + const systemOnly = scrubRequestHeaders(`${headerLine}\n${headerEvent({ system: 'secret prompt' })}\n`) + expect(systemOnly).toContain('"system":"{{system}}"') + expect(systemOnly).not.toContain('{{tools}}') + const toolsOnly = scrubRequestHeaders(`${headerLine}\n${headerEvent({ tools: [{ name: 't' }] })}\n`) + expect(toolsOnly).toContain('"tools":"{{tools}}"') + expect(toolsOnly).not.toContain('{{system}}') + }) + + it('leaves a delta with no scrubbable payload byte-identical (config-only, or non-array shapes)', () => { + const configOnly = JSON.stringify({ type: 'request/header-delta', seq: 8, time: 9, data: { config: { model: 'm2' } } }) + const oddShapes = JSON.stringify({ type: 'request/header-delta', seq: 9, time: 9, data: { system: { insert: 'not-an-array' }, tools: null } }) + const headerless = JSON.stringify({ type: 'request/header', seq: 10, time: 9, data: { reason: 'initial' } }) + const nullData = JSON.stringify({ type: 'request/header', seq: 11, time: 9, data: null }) + const raw = `${headerLine}\n${configOnly}\n${oddShapes}\n${headerless}\n${nullData}\n` + expect(scrubRequestHeaders(raw)).toBe(raw) + }) + + it('scrubs a one-sided tools delta and passes non-object schema entries through', () => { + const addedOnly = JSON.stringify({ + type: 'request/header-delta', seq: 8, time: 9, + data: { tools: { added: [null, 'weird', { name: 'x', description: 'D' }] } }, + }) + const out = scrubRequestHeaders(`${headerLine}\n${addedOnly}\n`) + // Non-object entries survive untouched; the object entry keeps only name. + expect(out).toContain('"added":[null,"weird",{"name":"x","description":"{{tools}}"}]') + const changedOnly = JSON.stringify({ + type: 'request/header-delta', seq: 8, time: 9, + data: { tools: { changed: [{ name: 'y', parameters: {} }] } }, + }) + expect(scrubRequestHeaders(`${headerLine}\n${changedOnly}\n`)) + .toContain('"changed":[{"name":"y","parameters":"{{tools}}"}]') + }) + it('scrubs a header-delta system payload but keeps its line positions and arity', () => { const delta = JSON.stringify({ type: 'request/header-delta', seq: 8, time: 9, diff --git a/packages/support/acp-snapshot/tests/suite.spec.ts b/packages/support/acp-snapshot/tests/suite.spec.ts new file mode 100644 index 0000000000..0bc15aeec2 --- /dev/null +++ b/packages/support/acp-snapshot/tests/suite.spec.ts @@ -0,0 +1,145 @@ +import { cpSync, mkdtempSync } from 'node:fs' +import { rm } from 'node:fs/promises' +import { tmpdir } from 'node:os' +import { join } from 'node:path' +import { fileURLToPath } from 'node:url' +import { afterAll, describe, expect, it } from 'vitest' +import { defineAcpSnapshotSuite, type Scenario } from '../src/index.ts' +import { childFixturePaths, fixtureContext, headerDeltaCount, normalizedHeaders } from '../src/suite.ts' + +/** + * Unit tests for the suite factory, by running it: two synthetic suites over + * the scripted fake ACP bin (./fixtures/fake-acp-agent.ts) register REAL + * describe/it trees at collection time, so every factory path — golden and log + * compares, the per-suite header pin and its uniformity guard, record-mode + * fixture write-back, skip semantics, and the fixture guard block — executes + * as an ordinary green test. The pure helpers get direct cases below. + * + * The replay suite runs against the committed fixtures in ./fixtures/suite. + * The record suite runs against a TEMP COPY of ./fixtures/record-suite + * (record mode writes session fixtures back into its snapshots dir; a run must + * never touch the committed tree). To re-bootstrap the record tree's goldens + * after changing the fake bin's output, run this spec once with + * `ACP_SNAPSHOT_SPEC_BOOTSTRAP=1` (points the record suite at the committed + * tree so vitest creates/updates the goldens and the write-back lands there), + * then commit the result. + */ + +const AGENT = { + binScript: fileURLToPath(new URL('./fixtures/fake-acp-agent.ts', import.meta.url)), + configPath: fileURLToPath(new URL('./fixtures/fake-acp-agent.ts', import.meta.url)), + tsconfigPath: fileURLToPath(new URL('../../../../tsconfig.json', import.meta.url)), +} + +const REPLAY_DIR = fileURLToPath(new URL('./fixtures/suite', import.meta.url)) +const RECORD_SRC = fileURLToPath(new URL('./fixtures/record-suite', import.meta.url)) + +const REPLAY_SCENARIOS: Scenario[] = [ + { name: 'pin-turn', hasModelTurn: true, recorded: true, pinsHeader: true }, + { name: 'plain-turn', hasModelTurn: true, recorded: true, childSessions: 1 }, + { name: 'no-model', hasModelTurn: false, recorded: false }, + { name: 'blocked-log', hasModelTurn: false, comparesLog: true, recorded: false }, + { name: 'authored-error', hasModelTurn: true, recorded: false }, +] + +const RECORD_SCENARIOS: Scenario[] = [ + { name: 'rec-pin', hasModelTurn: true, recorded: true, pinsHeader: true }, + { name: 'rec-child', hasModelTurn: true, recorded: true, childSessions: 1 }, + // recorded:false in record mode → registered but skipped (never re-recorded). + { name: 'rec-skip', hasModelTurn: true, recorded: false }, +] + +// Record mode mutates its snapshots dir, so run it on a throwaway copy — +// except under the documented bootstrap knob, which regenerates the committed +// fixtures/goldens in place. +const BOOTSTRAP = process.env.ACP_SNAPSHOT_SPEC_BOOTSTRAP === '1' +const recordDir = BOOTSTRAP ? RECORD_SRC : mkdtempSync(join(tmpdir(), 'acp-snap-record-suite-')) +if (!BOOTSTRAP) cpSync(RECORD_SRC, recordDir, { recursive: true }) +afterAll(async () => { + if (!BOOTSTRAP) await rm(recordDir, { recursive: true, force: true }) +}) + +describe('defineAcpSnapshotSuite: replay mode', () => { + defineAcpSnapshotSuite({ agent: AGENT, snapshotsDir: REPLAY_DIR, scenarios: REPLAY_SCENARIOS, mode: 'replay' }) +}) + +// The record suite's tests run in registration order: rec-pin re-records the +// pinned fixture FIRST, so rec-child's uniformity guard reads the fresh pin. +describe('defineAcpSnapshotSuite: record mode', () => { + defineAcpSnapshotSuite({ agent: AGENT, snapshotsDir: recordDir, scenarios: RECORD_SCENARIOS, mode: 'record' }) +}) + +describe('defineAcpSnapshotSuite: registration contract', () => { + it('throws when no scenario pins the request-header content', () => { + expect(() => { + defineAcpSnapshotSuite({ + agent: AGENT, + snapshotsDir: REPLAY_DIR, + scenarios: [{ name: 'pinless', hasModelTurn: true, recorded: true }], + mode: 'replay', + }) + }).toThrow(/no scenario pins/) + }) +}) + +describe('childFixturePaths', () => { + it('yields one sibling path per child, 1-based', () => { + expect(childFixturePaths('/snap/s', 2)).toEqual(['/snap/s/session.1.jsonl', '/snap/s/session.2.jsonl']) + }) + + it('yields nothing for a single-session scenario', () => { + expect(childFixturePaths('/snap/s', 0)).toEqual([]) + }) +}) + +describe('fixtureContext', () => { + it('reads the fixture header id and cwd', () => { + const ctx = fixtureContext('{"type":"session","id":"abc","cwd":"/rec"}\n{"type":"turn/start"}\n') + expect(ctx).toEqual({ sessionIds: ['abc'], cwd: '/rec' }) + }) + + it('yields no session ids for a header without a string id', () => { + expect(fixtureContext('{"type":"session","cwd":"/rec"}\n').sessionIds).toEqual([]) + }) + + it('falls back to an impossible sentinel cwd (never the empty string)', () => { + const ctx = fixtureContext('{"type":"session","id":"abc"}\n') + expect(ctx.cwd).toBe('\0no-cwd\0') + expect(ctx.cwd).not.toBe('') + }) + + it('treats an empty fixture as an empty header', () => { + expect(fixtureContext('')).toEqual({ sessionIds: [], cwd: '\0no-cwd\0' }) + }) +}) + +describe('normalizedHeaders', () => { + const header = (system: string): string => JSON.stringify({ + type: 'request/header', seq: 0, time: 9, data: { header: { config: { model: 'm' }, system }, reason: 'initial' }, + }) + + it('extracts every request/header payload in log order, normalized', () => { + const id = '11111111-2222-4333-8444-555555555555' + const log = `${JSON.stringify({ type: 'session', id, createdAt: 5, cwd: '/w' })}\n${header('one')}\n` + + `${JSON.stringify({ type: 'turn/start', seq: 1, time: 9, data: { turn: 1 } })}\n${header('two')}\n` + const headers = normalizedHeaders(log, { sessionIds: [id], cwd: '/w' }) + expect(headers).toEqual([ + { config: { model: 'm' }, system: 'one' }, + { config: { model: 'm' }, system: 'two' }, + ]) + }) + + it('yields nothing for a log without header events', () => { + const log = `${JSON.stringify({ type: 'session', id: 'a', createdAt: 5 })}\n` + expect(normalizedHeaders(log, { sessionIds: [], cwd: '/w' })).toEqual([]) + }) +}) + +describe('headerDeltaCount', () => { + it('counts request/header-delta events, ignoring blanks and other lines', () => { + const delta = JSON.stringify({ type: 'request/header-delta', seq: 2, time: 9, data: {} }) + const other = JSON.stringify({ type: 'request/header', seq: 0, time: 9, data: {} }) + expect(headerDeltaCount(`${other}\n\n${delta}\n${delta}\n`)).toBe(2) + expect(headerDeltaCount(`${other}\n`)).toBe(0) + }) +})