* feat(python): query retained pyc code-object facts * test(python): bind complete pyc selection benchmark tasks * test(python): include server instructions in token accounting * docs(perf): record pyc selected-facts task measurements * test(python): keep oracle dependencies separate from selected targets * docs(perf): record guarded oracle requalification * test(python): narrow complete Evidence through its owning contract
273 lines
9.2 KiB
JavaScript
273 lines
9.2 KiB
JavaScript
/** Same selected code-object question through production stdio MCP delivery. */
|
|
import assert from "node:assert/strict";
|
|
import { createHash } from "node:crypto";
|
|
import { createWriteStream } from "node:fs";
|
|
import { mkdir, mkdtemp, readFile, writeFile } from "node:fs/promises";
|
|
import { tmpdir } from "node:os";
|
|
import { join, resolve } from "node:path";
|
|
import { finished } from "node:stream/promises";
|
|
import { parseArgs } from "node:util";
|
|
import { Client } from "@modelcontextprotocol/client";
|
|
import { StdioClientTransport } from "@modelcontextprotocol/client/stdio";
|
|
import { createRetainedQueryCaller } from "./retained-query-calls.mjs";
|
|
import { finalizeRetainedQueryReceipt } from "./retained-query-finalization.mjs";
|
|
|
|
const { values } = parseArgs({
|
|
options: {
|
|
runtime: { type: "string" },
|
|
identity: { type: "string" },
|
|
oracle: { type: "string" },
|
|
output: { type: "string" },
|
|
lane: { type: "string", default: "complete" },
|
|
"page-size": { type: "string", default: "64" },
|
|
"two-step": { type: "boolean", default: false },
|
|
"probe-support": { type: "boolean", default: false },
|
|
},
|
|
});
|
|
for (const key of ["runtime", "oracle", "output"])
|
|
assert.ok(values[key] && values[key] === resolve(values[key]), key);
|
|
assert.ok(["complete", "selected"].includes(values.lane));
|
|
const pageSize = Number(values["page-size"]);
|
|
assert.ok(Number.isSafeInteger(pageSize) && pageSize > 0);
|
|
const expected = JSON.parse(await readFile(values.oracle, "utf8"));
|
|
const hash = (value) => createHash("sha256").update(value).digest("hex");
|
|
assert.equal(
|
|
hash(await readFile(expected.artifact.path)),
|
|
expected.artifact.sha256,
|
|
);
|
|
await mkdir(values.output, { mode: 0o700 });
|
|
const account = await mkdtemp(join(tmpdir(), "rea-pyc-views-"));
|
|
const client = new Client({ name: "pyc-views", version: "1" });
|
|
const transport = new StdioClientTransport({
|
|
command: process.execPath,
|
|
args: [join(values.runtime, "scripts/rea.mjs"), "mcp"],
|
|
maxBufferSize: 10485760,
|
|
env: {
|
|
HOME: account,
|
|
TMPDIR: account,
|
|
PATH: process.env.PATH,
|
|
REA_LOG_LEVEL: "silent",
|
|
NODE_OPTIONS: "--max-old-space-size=2048",
|
|
NODE_DISABLE_COMPILE_CACHE: "1",
|
|
UV_THREADPOOL_SIZE: "2",
|
|
},
|
|
stderr: "pipe",
|
|
});
|
|
const serverLog = createWriteStream(join(values.output, "server.log"));
|
|
const serverLogFinished = finished(serverLog);
|
|
serverLogFinished.catch(() => {});
|
|
transport.stderr?.pipe(serverLog);
|
|
const report = {
|
|
status: "running",
|
|
lane: values.lane,
|
|
artifact: expected.artifact,
|
|
selected_names: expected.names,
|
|
calls: [],
|
|
phases: {},
|
|
client: "SDK exact discovered toolDefinition, default validator",
|
|
task: "Return every matching code object for exact names with original indices and selected fields; export complete source Evidence.",
|
|
bytecode_execution: false,
|
|
page_size: pageSize,
|
|
direct_first_view: values.lane === "selected" && !values["two-step"],
|
|
};
|
|
const verifyRuntime = async (phase) => {
|
|
if (values.identity === undefined) return;
|
|
const started = performance.now();
|
|
const bytes = await readFile(values.identity);
|
|
const identity = JSON.parse(bytes);
|
|
for (const entry of [...identity.compiled_files, ...identity.entrypoints])
|
|
assert.equal(
|
|
hash(await readFile(join(values.runtime, entry.path))),
|
|
entry.sha256,
|
|
entry.path,
|
|
);
|
|
for (const entry of identity.driver_files)
|
|
assert.equal(
|
|
hash(await readFile(new URL(entry.path, import.meta.url))),
|
|
entry.sha256,
|
|
);
|
|
assert.equal(hash(await readFile(process.execPath)), identity.node_sha256);
|
|
report.runtime_identity_sha256 ??= hash(bytes);
|
|
assert.equal(hash(bytes), report.runtime_identity_sha256);
|
|
report.phases[`runtime_validation_${phase}_ms`] = performance.now() - started;
|
|
};
|
|
const transcript = [];
|
|
let definitions;
|
|
const caller = createRetainedQueryCaller(
|
|
{
|
|
callTool(params, options) {
|
|
return client.callTool(params, {
|
|
...options,
|
|
toolDefinition: definitions.get(params.name),
|
|
});
|
|
},
|
|
},
|
|
(receipt, event) => {
|
|
report.calls.push(receipt);
|
|
transcript.push(event);
|
|
},
|
|
);
|
|
const observe = async (name, input, phase) => {
|
|
const result = await caller(name, input, phase);
|
|
if (result.isError === true) assert.fail(JSON.stringify(result));
|
|
assert.ok(result.structuredContent);
|
|
return result.structuredContent;
|
|
};
|
|
const columns = new Set(["source_index", ...expected.fields]);
|
|
const select = (row) =>
|
|
Object.fromEntries(Object.entries(row).filter(([key]) => columns.has(key)));
|
|
try {
|
|
await verifyRuntime("before");
|
|
const started = performance.now();
|
|
await client.connect(transport);
|
|
report.server_instructions = client.getInstructions() ?? "";
|
|
const catalog = await client.listTools();
|
|
definitions = new Map(catalog.tools.map((tool) => [tool.name, tool]));
|
|
await writeFile(join(values.output, "catalog.json"), JSON.stringify(catalog));
|
|
report.phases.discovery_ms = performance.now() - started;
|
|
report.catalog_tools = catalog.tools.length;
|
|
const selected = values.lane === "selected";
|
|
const selectedView = (offset) => ({
|
|
kind: "python-pyc",
|
|
collection: "code_objects",
|
|
names: expected.names,
|
|
fields: expected.fields,
|
|
offset,
|
|
limit: pageSize,
|
|
});
|
|
const response = await observe(
|
|
"inspect_python_pyc",
|
|
{
|
|
path: expected.artifact.path,
|
|
...(selected
|
|
? {
|
|
detail: "summary",
|
|
...(values["two-step"] ? {} : { view: selectedView(0) }),
|
|
}
|
|
: {}),
|
|
},
|
|
"inspect",
|
|
);
|
|
const parentId = selected
|
|
? response.normalized_result.parent_evidence_id
|
|
: response.evidence_id;
|
|
report.parent_evidence_id = parentId;
|
|
let answer;
|
|
if (!selected) {
|
|
const source = response.normalized_result;
|
|
assert.deepEqual(source.artifact, expected.artifact);
|
|
assert.equal(source.coverage.status, "complete");
|
|
const rows = source.code_objects.map((row, source_index) =>
|
|
select({ ...row, source_index }),
|
|
);
|
|
assert.deepEqual(rows, expected.rows);
|
|
answer = rows.filter((row) => expected.names.includes(row.name));
|
|
} else {
|
|
answer = [];
|
|
let offset = 0;
|
|
let page = values["two-step"] ? undefined : response.normalized_result;
|
|
do {
|
|
page ??= (
|
|
await observe(
|
|
"inspect_analysis_view",
|
|
{
|
|
source: { kind: "retained-evidence", evidence_id: parentId },
|
|
view: selectedView(offset),
|
|
},
|
|
"selected-page",
|
|
)
|
|
).normalized_result;
|
|
assert.equal(page.kind, "python-pyc");
|
|
assert.deepEqual(page.pyc.artifact, expected.artifact);
|
|
assert.equal(page.pyc.coverage.status, "complete");
|
|
assert.equal(page.pyc.coverage.code_objects, expected.rows.length);
|
|
assert.equal(page.coverage.total, expected.answer.length);
|
|
answer.push(...page.items);
|
|
if (page.coverage.next_offset === null) break;
|
|
assert.ok(page.coverage.next_offset > offset);
|
|
assert.ok(page.coverage.next_offset <= expected.answer.length);
|
|
offset = page.coverage.next_offset;
|
|
page = undefined;
|
|
} while (true);
|
|
}
|
|
assert.deepEqual(answer, expected.answer);
|
|
report.answer_sha256 = hash(JSON.stringify(answer));
|
|
report.answer_count = answer.length;
|
|
await writeFile(join(values.output, "answer.json"), JSON.stringify(answer));
|
|
await observe(
|
|
"export_evidence_bundle",
|
|
{ path: join(values.output, "evidence.json") },
|
|
"export",
|
|
);
|
|
const bundle = JSON.parse(
|
|
await readFile(join(values.output, "evidence.json"), "utf8"),
|
|
);
|
|
const parent = bundle.records.find(
|
|
(record) => record.evidence_id === parentId,
|
|
);
|
|
assert.ok(parent);
|
|
assert.equal(parent.operation, "inspect_python_pyc");
|
|
assert.deepEqual(parent.normalized_result.artifact, expected.artifact);
|
|
assert.equal(parent.normalized_result.coverage.status, "complete");
|
|
assert.deepEqual(
|
|
parent.normalized_result.code_objects.map((row, source_index) =>
|
|
select({ ...row, source_index }),
|
|
),
|
|
expected.rows,
|
|
);
|
|
report.source_report_sha256 = hash(JSON.stringify(parent.normalized_result));
|
|
report.complete_source_export_verified = true;
|
|
if (values["probe-support"]) {
|
|
report.support = {};
|
|
for (const [label, name, input] of [
|
|
[
|
|
"summary_delivery",
|
|
"inspect_python_pyc",
|
|
{ path: expected.artifact.path, detail: "summary" },
|
|
],
|
|
[
|
|
"summary_view",
|
|
"inspect_analysis_view",
|
|
{
|
|
source: { kind: "retained-evidence", evidence_id: parentId },
|
|
view: { kind: "summary" },
|
|
},
|
|
],
|
|
]) {
|
|
try {
|
|
const reply = await caller(name, input, "capability-probe");
|
|
report.support[label] = { is_error: reply.isError === true, reply };
|
|
} catch (cause) {
|
|
report.support[label] = { is_error: true, message: String(cause) };
|
|
}
|
|
}
|
|
}
|
|
report.status = "passed";
|
|
} catch (cause) {
|
|
report.status = "failed";
|
|
report.error = String(cause);
|
|
} finally {
|
|
try {
|
|
await verifyRuntime("after");
|
|
} catch (cause) {
|
|
report.status = "failed";
|
|
report.runtime_identity_failure = String(cause);
|
|
}
|
|
await finalizeRetainedQueryReceipt({
|
|
client,
|
|
transport,
|
|
serverLog,
|
|
serverLogFinished,
|
|
account,
|
|
report,
|
|
output: values.output,
|
|
transcript,
|
|
});
|
|
if (
|
|
report.status !== "passed" ||
|
|
report.cleanup_failures.length ||
|
|
report.receipt_failures.length
|
|
)
|
|
process.exitCode = 1;
|
|
}
|