Understand any codebase instantly. System intelligence for codebases, built for humans and AI.
TypeScript aiai-memoryclaude-codeclicode-analysiscode-mappingcode-memorycodebase-understandingcodexdeveloper-toolsllmopenclawpersistent-memoryprogram-analysissoftware-architecturesystem-intelligencesystem-mapping
// Copyright 2026 Ix Infrastructure Inc.
import { describe, expect, it, vi } from "vitest";
import { IxClient } from "../../client/api.js";
/**
* `ix context` resolves its target and then has to ask the backend for the
* graph around it. It used to ask by NAME, which made the backend re-run the
* search the resolver had just done — the single most expensive call the CLI
* makes, measured at 10.6 s against 0.06 s for the id form on the same graph.
*
* Worse than slow: the re-derived seeds can land on a different node of the
* same name. `README.md` resolved to one id and came back seeded on
* `mcp/node_modules/serve-static/README.md`, with the resolved id in neither
* the seed list nor the returned nodes.
*/
describe("contextForNode", () => {
function capture() {
const calls: Array<{ url: string; body: any }> = [];
vi.stubGlobal("fetch", async (url: any, init: any) => {
calls.push({ url: String(url), body: JSON.parse(init.body) });
return new Response(JSON.stringify({ nodes: [], edges: [], claims: [] }), {
status: 200,
headers: { "Content-Type": "application/json" },
});
});
return calls;
}
it("sends the node id as a slice, never a name", async () => {
const calls = capture();
await new IxClient("http://x").contextForNode("11111111-2222-3333-4444-555555555555");
const body = calls[0].body;
expect(body.slices).toEqual([
{
type: "nodes",
nodeIds: ["11111111-2222-3333-4444-555555555555"],
expand: true,
hops: 1,
},
]);
// The whole point: no free-text query for the backend to re-search.
expect(body.query).toBeUndefined();
vi.unstubAllGlobals();
});
it("passes depth through rather than pinning it to full", async () => {
const calls = capture();
await new IxClient("http://x").contextForNode("11111111-2222-3333-4444-555555555555", {
depth: "compact",
asOfRev: 2000,
});
// `--depth` has to keep meaning what it says: compact and standard return
// the summarized graph, full returns whole nodes plus claims. Hardcoding
// full would silently ignore the flag and change every bundle's size.
expect(calls[0].body.depth).toBe("compact");
expect(calls[0].body.asOfRev).toBe(2000);
vi.unstubAllGlobals();
});
});
describe("context.ts call sites", () => {
it("has no by-name context call left in the command", async () => {
const { readFileSync } = await import("node:fs");
const { fileURLToPath } = await import("node:url");
const { dirname, join } = await import("node:path");
const src = readFileSync(
join(dirname(fileURLToPath(import.meta.url)), "..", "commands", "context.ts"),
"utf8",
);
// Three call sites, and all must be by id. Leaving one behind makes `--diff`
// compare a by-id bundle against a by-name one and report the entire graph
// as changed on the first run after upgrade. The third is `--from-issue`,
// which seeds from the node its starting point resolved to.
expect(src).not.toContain("client.query(resolved.name");
expect(src.match(/client\.contextForNode\(resolved\.id/g) ?? []).toHaveLength(3);
});
});