feat: safe Evidence tree enumeration and bounded blob streaming (P6)

2026-08-13 12:28:49 +02:00
parent e80a8b35ec
commit 0c1033889a
2 changed files with 193 additions and 5 deletions
@@ -0,0 +1,129 @@
+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: "Evidence Test",
+    gitAuthorEmail: "evidence@example.invalid",
+    installationId: "test",
+    secretRoots: [],
+    maxImportBytes: 1024,
+    maxImportEntries: 1,
+  };
+}
+
+type EvidenceLayout = "tree" | "empty" | "root-file" | "root-symlink" | "nested-symlink";
+
+async function fixture(layout: EvidenceLayout): Promise<{ root: string; remote: string; commit: string }> {
+  const root = mkdtempSync(join(tmpdir(), "thoth-evidence-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", "Evidence Test"]);
+  await git(source, ["config", "user.email", "evidence@example.invalid"]);
+  writeFileSync(join(source, "thoth-workspaces.yaml"), "schema_version: 1\nworkspaces: [{id: research, name: Research}]\n");
+  mkdirSync(join(source, "research"), { recursive: true });
+  writeFileSync(join(source, "research", "workspace.yaml"), "workspace:\n  schema_version: 3\n  id: research\n");
+  const evidence = join(source, "research", "evidence");
+  if (layout === "tree") {
+    mkdirSync(join(evidence, "nested"), { recursive: true });
+    writeFileSync(join(evidence, "guide.md"), "# guide\n");
+    writeFileSync(join(evidence, "nested", "deep.md"), "# deep\n");
+  } else if (layout === "empty") {
+    mkdirSync(evidence, { recursive: true });
+  } else if (layout === "root-file") {
+    writeFileSync(evidence, "not a tree\n");
+  } else if (layout === "root-symlink") {
+    writeFileSync(join(source, "research", "target.md"), "# target\n");
+    symlinkSync("target.md", evidence);
+  } else if (layout === "nested-symlink") {
+    mkdirSync(evidence, { recursive: true });
+    writeFileSync(join(source, "research", "outside.md"), "# outside\n");
+    symlinkSync("../outside.md", join(evidence, "link.md"));
+  }
+  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, commit };
+}
+
+async function bootstrapped(fixture: { root: string; remote: string }): Promise<GitWorkspaceRepository> {
+  const repository = new GitWorkspaceRepository(config(join(fixture.root, "registry"), fixture.remote));
+  await repository.bootstrap();
+  return repository;
+}
+
+test("enumerates a regular Evidence tree with ordered relative paths", async () => {
+  const fixtureValue = await fixture("tree");
+  const repository = await bootstrapped(fixtureValue);
+
+  const objects = await repository.evidenceTreeObjects(fixtureValue.commit, "research");
+
+  expect(objects.map((entry) => entry.posixPath)).toEqual(["guide.md", "nested/deep.md"]);
+  expect(objects.every((entry) => /^[0-9a-f]{40}$/.test(entry.oid))).toBe(true);
+  expect(objects.every((entry) => entry.mode === "100644" || entry.mode === "100755")).toBe(true);
+});
+
+test("accepts an empty Evidence tree", async () => {
+  const fixtureValue = await fixture("empty");
+  const repository = await bootstrapped(fixtureValue);
+
+  await expect(repository.evidenceTreeObjects(fixtureValue.commit, "research")).resolves.toEqual([]);
+});
+
+test("refuses a non-tree Evidence root and symlinks at any depth", async () => {
+  for (const layout of ["root-file", "root-symlink", "nested-symlink"] as const) {
+    const fixtureValue = await fixture(layout);
+    const repository = await bootstrapped(fixtureValue);
+    await expect(repository.evidenceTreeObjects(fixtureValue.commit, "research"))
+      .rejects.toMatchObject({ code: "workspace_invalid" });
+  }
+});
+
+test("refuses malformed revisions and workspace ids", async () => {
+  const fixtureValue = await fixture("tree");
+  const repository = await bootstrapped(fixtureValue);
+
+  await expect(repository.evidenceTreeObjects("HEAD", "research"))
+    .rejects.toMatchObject({ code: "workspace_invalid" });
+  await expect(repository.evidenceTreeObjects(fixtureValue.commit, "../research"))
+    .rejects.toMatchObject({ code: "workspace_invalid" });
+});
+
+test("streams bounded Evidence blobs and refuses oversized objects", async () => {
+  const fixtureValue = await fixture("tree");
+  const repository = await bootstrapped(fixtureValue);
+  const [guide] = await repository.evidenceTreeObjects(fixtureValue.commit, "research");
+
+  const bytes = await repository.evidenceBlobBytes(guide.oid, 1024);
+  expect(bytes.toString("utf8")).toBe("# guide\n");
+
+  await expect(repository.evidenceBlobBytes(guide.oid, 1))
+    .rejects.toMatchObject({ code: "workspace_invalid" });
+});