ThothII/backend/test/workspaces-git-annotations.test.ts

import { execFile } from "node:child_process";
import { mkdtempSync, mkdirSync, rmSync, symlinkSync, writeFileSync } from "node:fs";
import { tmpdir } from "node:os";
import { join } from "node:path";
import { promisify } from "node:util";
import { afterEach, expect, test } from "vitest";
import { GitWorkspaceRepository } from "../src/workspaces/git-repository.js";
import type { WorkspaceRegistryConfig } from "../src/workspaces/types.js";

const runFile = promisify(execFile);
const temporaryRoots: string[] = [];

afterEach(() => {
  temporaryRoots.splice(0).forEach((root) => rmSync(root, { recursive: true, force: true }));
});

async function git(cwd: string, args: string[]): Promise<string> {
  const { stdout } = await runFile("git", args, { cwd });
  return stdout.trim();
}

function config(root: string, remoteUrl: string): WorkspaceRegistryConfig {
  return {
    root,
    remoteUrl,
    branch: "main",
    gitAuthorName: "Workspace Registry Test",
    gitAuthorEmail: "workspace-registry@example.invalid",
    installationId: "test",
    secretRoots: [],
    maxImportBytes: 1024,
    maxImportEntries: 1,
  };
}

interface RepoFixture {
  root: string;
  remote: string;
  source: string;
  commit: string;
}

async function makeRepo(id: string, annotations: string | Buffer | "dir" | "symlink"): Promise<RepoFixture> {
  const root = mkdtempSync(join(tmpdir(), "thoth-annotations-git-"));
  temporaryRoots.push(root);
  const remote = join(root, "remote.git");
  const source = join(root, "source");
  await git(root, ["init", "--bare", "--initial-branch=main", remote]);
  mkdirSync(source);
  await git(source, ["init", "--initial-branch=main"]);
  await git(source, ["config", "user.name", "Annotations Test"]);
  await git(source, ["config", "user.email", "annotations@example.invalid"]);
  writeFileSync(join(source, "thoth-workspaces.yaml"),
    `schema_version: 1\nworkspaces: [{id: ${id}, name: Workspace}]\n`);
  mkdirSync(join(source, id, "schema"), { recursive: true });
  writeFileSync(join(source, id, "workspace.yaml"), `workspace:\n  schema_version: 4\n  id: ${id}\n`);
  const annotationsPath = join(source, id, "schema", "annotations.yaml");
  if (annotations === "dir") {
    mkdirSync(annotationsPath, { recursive: true });
    writeFileSync(join(annotationsPath, "child.txt"), "not a blob\n");
  } else if (annotations === "symlink") {
    writeFileSync(join(source, id, "target.yaml"), "tables: {}\n");
    symlinkSync("target.yaml", annotationsPath);
  } else {
    writeFileSync(annotationsPath, annotations);
  }
  await git(source, ["add", "-A"]);
  await git(source, ["commit", "-m", "initial"]);
  await git(source, ["remote", "add", "origin", remote]);
  await git(source, ["push", "origin", "main"]);
  const commit = await git(source, ["rev-parse", "HEAD"]);
  return { root, remote, source, commit };
}

async function bootstrapped(fixture: RepoFixture): Promise<GitWorkspaceRepository> {
  const repository = new GitWorkspaceRepository(config(join(fixture.root, "registry"), fixture.remote));
  await repository.bootstrap();
  return repository;
}

test("reads a regular annotation blob at the exact commit", async () => {
  const fixture = await makeRepo("research", "tables: {}\n");
  const repository = await bootstrapped(fixture);

  const object = await repository.annotationsObject(fixture.commit, "research");

  expect(object).toBeDefined();
  expect(object!.blobId).toMatch(/^[0-9a-f]{40}$/);
  expect(object!.contents.toString("utf8")).toBe("tables: {}\n");
});

test("returns undefined when the annotation object is absent", async () => {
  const fixture = await makeRepo("research", "tables: {}\n");
  const repository = await bootstrapped(fixture);

  await expect(repository.annotationsObject(fixture.commit, "absent")).resolves.toBeUndefined();
});

test("refuses a tree at the annotation path", async () => {
  const fixture = await makeRepo("clinical", "dir");
  const repository = await bootstrapped(fixture);

  await expect(repository.annotationsObject(fixture.commit, "clinical"))
    .rejects.toMatchObject({ code: "workspace_invalid" });
});

test("refuses a symlink at the annotation path", async () => {
  const fixture = await makeRepo("research", "symlink");
  const repository = await bootstrapped(fixture);

  await expect(repository.annotationsObject(fixture.commit, "research"))
    .rejects.toMatchObject({ code: "workspace_invalid" });
});

test("refuses oversized and non-UTF-8 annotation blobs", async () => {
  const oversized = await makeRepo("research", Buffer.concat([
    Buffer.from("tables: {}\n"),
    Buffer.alloc(16 * 1024 * 1024, 0x78),
  ]));
  const repository = await bootstrapped(oversized);
  await expect(repository.annotationsObject(oversized.commit, "research"))
    .rejects.toMatchObject({ code: "workspace_invalid" });

  const nonUtf8 = await makeRepo("research", Buffer.from([0x74, 0x61, 0x62, 0xff, 0xfe, 0x00]));
  const repository2 = await bootstrapped(nonUtf8);
  await expect(repository2.annotationsObject(nonUtf8.commit, "research"))
    .rejects.toMatchObject({ code: "workspace_invalid" });
});

test("refuses malformed ids and revisions before Git", async () => {
  const fixture = await makeRepo("research", "tables: {}\n");
  const repository = await bootstrapped(fixture);

  await expect(repository.annotationsObject(fixture.commit, "workspace-docs"))
    .rejects.toMatchObject({ code: "workspace_invalid" });
  await expect(repository.annotationsObject(fixture.commit, "../research"))
    .rejects.toMatchObject({ code: "workspace_invalid" });
  await expect(repository.annotationsObject("HEAD", "research"))
    .rejects.toMatchObject({ code: "workspace_invalid" });
});

test("redacts Git failures while reading the annotation blob", async () => {
  const fixture = await makeRepo("research", "tables: {}\n");
  const repository = await bootstrapped(fixture);
  rmSync(repository.repoPath, { recursive: true, force: true });

  const error = await repository.annotationsObject(fixture.commit, "research").catch((failure: unknown) => failure);
  expect(error).toMatchObject({ code: "git_unavailable", message: "Workspace Git operation failed" });
  expect((error as Error).message).not.toContain(fixture.root);
});