refactor: remove workspace migration utilities

2026-08-10 21:32:01 +02:00
parent c2f9b03973
commit fa24f43fd4
5 changed files with 43 additions and 487 deletions
@@ -0,0 +1,43 @@
+import { execFileSync } from "node:child_process";
+import { existsSync, readFileSync } from "node:fs";
+import { expect, test } from "vitest";
+
+test("declares a durable isolated registry volume and only read-only Git credential mounts", () => {
+  const compose = readFileSync(new URL("../../compose.yaml", import.meta.url), "utf8");
+  const development = readFileSync(new URL("../../docker-compose.dev.yml", import.meta.url), "utf8");
+  const gitHttps = readFileSync(new URL("../../deploy/compose.git-https.yaml", import.meta.url), "utf8");
+  const gitSsh = readFileSync(new URL("../../deploy/compose.git-ssh.yaml", import.meta.url), "utf8");
+  const dockerfile = readFileSync(new URL("../../docker/core.Dockerfile", import.meta.url), "utf8");
+  const smoke = readFileSync(new URL("../../scripts/workspace-registry-smoke.sh", import.meta.url), "utf8");
+
+  for (const source of [compose, development]) {
+    expect(source).toContain("THT_WORKSPACE_REGISTRY_ROOT: /data/workspace-registry");
+    expect(source).toContain("THT_WORKSPACE_GIT_REMOTE: ${THT_WORKSPACE_GIT_REMOTE:?set THT_WORKSPACE_GIT_REMOTE}");
+    expect(source).toContain("workspace-registry:/data/workspace-registry");
+  }
+  expect(compose).not.toMatch(/workspace-registry-git-(?:credentials|ca|ssh-key|known-hosts):ro/);
+  expect(gitHttps).toMatch(/workspace-registry-git-credentials:ro/);
+  expect(gitHttps).toMatch(/workspace-registry-git-ca:ro/);
+  expect(gitSsh).toMatch(/workspace-registry-git-ssh-key:ro/);
+  expect(gitSsh).toMatch(/workspace-registry-git-known-hosts:ro/);
+  expect(dockerfile).toMatch(/mkdir -p[^\n]*\/data\/workspace-registry/);
+  expect(dockerfile).toMatch(/chown -R thoth:thoth \/home\/thoth\/\.pi \/data/);
+  expect(smoke).toContain('core_remote="/fixtures/offline.git"');
+  expect(smoke).toContain('"degraded":true');
+  expect(smoke).toContain('core_remote="/fixtures/remote.git"');
+});
+
+test("workspace registry smoke image cleanup is scoped to the per-run image identity", () => {
+  const output = execFileSync("bash", ["scripts/workspace-registry-smoke.sh"], {
+    cwd: new URL("../..", import.meta.url),
+    env: { ...process.env, WORKSPACE_REGISTRY_SMOKE_SELF_TEST: "image-cleanup-identity" },
+    encoding: "utf8",
+  });
+
+  expect(output).toContain("workspace registry smoke image cleanup identity self-test passed");
+});
+
+test("workspace migration source modules are absent from the live backend boundary", () => {
+  expect(existsSync(new URL("../src/workspaces/migrate-legacy.ts", import.meta.url))).toBe(false);
+  expect(existsSync(new URL("../src/workspaces/migrate-v2-qdrant.ts", import.meta.url))).toBe(false);
+});
@@ -1,114 +0,0 @@
-import { execFileSync } from "node:child_process";
-import { existsSync, readFileSync, rmSync, writeFileSync } from "node:fs";
-import { mkdtemp } from "node:fs/promises";
-import { tmpdir } from "node:os";
-import { join } from "node:path";
-import { afterEach, expect, test } from "vitest";
-import {
-  main,
-  migrateLegacyWorkspace,
-  writeMigratedWorkspace,
-} from "../src/workspaces/migrate-legacy.js";
-import * as workspaceSchema from "../src/workspaces/schema.js";
-import { parseWorkspaceYaml } from "../src/workspaces/schema.js";
-
-const temporaryRoots: string[] = [];
-
-afterEach(() => {
-  temporaryRoots.splice(0).forEach((root) => rmSync(root, { recursive: true, force: true }));
-});
-
-function readFixture(name: string): string {
-  return readFileSync(new URL(`../../harness/workspaces/${name}`, import.meta.url), "utf8");
-}
-
-test("migrates the current local PSD descriptor without copying secret values", () => {
-  const result = migrateLegacyWorkspace(readFixture("local.yaml"), { id: "local", collection: "local" });
-
-  expect(result.workspace.workspace).toMatchObject({ id: "local", schema_version: 3, language: "it" });
-  expect(result.workspace).not.toHaveProperty("evidence");
-  expect(JSON.stringify(result)).not.toMatch(/password:|api_key:|\$\{THT_/i);
-});
-
-test("migrates a legacy descriptor only with an explicit target collection into schema v3", () => {
-  const result = migrateLegacyWorkspace(readFixture("tht.example.yaml"), { id: "example", collection: "shared" });
-  const isOperationalWorkspace = (workspaceSchema as { isOperationalWorkspace?: unknown }).isOperationalWorkspace;
-
-  expect(result.workspace.workspace.schema_version).toBe(3);
-  expect(parseWorkspaceYaml(result.source).workspace.schema_version).toBe(3);
-  expect(isOperationalWorkspace).toBeTypeOf("function");
-  expect((isOperationalWorkspace as (workspace: ReturnType<typeof parseWorkspaceYaml>) => boolean)(
-    parseWorkspaceYaml(result.source),
-  )).toBe(true);
-  expect(parseWorkspaceYaml(result.source)).toMatchObject({
-    semantic_index: { vector_store: { engine: "qdrant", collection: "shared" } },
-  });
-});
-
-test("requires an explicit target collection for legacy migration", () => {
-  expect(() => migrateLegacyWorkspace(readFixture("local.yaml"), { id: "local" } as never)).toThrow(
-    /collection/i,
-  );
-});
-
-test("writes versioned repository artifacts atomically without replacing a prior migration", async () => {
-  const root = await mkdtemp(join(tmpdir(), "thoth-workspace-migrate-"));
-  temporaryRoots.push(root);
-  const migration = migrateLegacyWorkspace(readFixture("local.yaml"), { id: "local", collection: "local" });
-
-  const destination = await writeMigratedWorkspace(migration, root);
-
-  expect(destination).toBe(join(root, "workspaces", "local.yaml"));
-  expect(parseWorkspaceYaml(readFileSync(destination, "utf8"))).toMatchObject({ workspace: { id: "local" } });
-  await expect(writeMigratedWorkspace(migration, root)).rejects.toThrow(/already exists/i);
-  expect(existsSync(destination)).toBe(true);
-});
-
-test("CLI accepts an explicit valid ID when a legacy filename contains dots", async () => {
-  const root = await mkdtemp(join(tmpdir(), "thoth-workspace-migrate-"));
-  temporaryRoots.push(root);
-  const input = join(root, "psd.clinical.yaml");
-  writeFileSync(input, readFixture("local.yaml"));
-
-  await main(["--input", input, "--output", root, "--id", "psd-clinical", "--collection", "psd-clinical"]);
-
-  const destination = join(root, "workspaces", "psd-clinical.yaml");
-  expect(parseWorkspaceYaml(readFileSync(destination, "utf8"))).toMatchObject({
-    workspace: { id: "psd-clinical", schema_version: 3 },
-  });
-});
-
-test("declares a durable isolated registry volume and only read-only Git credential mounts", () => {
-  const compose = readFileSync(new URL("../../compose.yaml", import.meta.url), "utf8");
-  const development = readFileSync(new URL("../../docker-compose.dev.yml", import.meta.url), "utf8");
-  const gitHttps = readFileSync(new URL("../../deploy/compose.git-https.yaml", import.meta.url), "utf8");
-  const gitSsh = readFileSync(new URL("../../deploy/compose.git-ssh.yaml", import.meta.url), "utf8");
-  const dockerfile = readFileSync(new URL("../../docker/core.Dockerfile", import.meta.url), "utf8");
-  const smoke = readFileSync(new URL("../../scripts/workspace-registry-smoke.sh", import.meta.url), "utf8");
-
-  for (const source of [compose, development]) {
-    expect(source).toContain("THT_WORKSPACE_REGISTRY_ROOT: /data/workspace-registry");
-    expect(source).toContain("THT_WORKSPACE_GIT_REMOTE: ${THT_WORKSPACE_GIT_REMOTE:?set THT_WORKSPACE_GIT_REMOTE}");
-    expect(source).toContain("workspace-registry:/data/workspace-registry");
-  }
-  expect(compose).not.toMatch(/workspace-registry-git-(?:credentials|ca|ssh-key|known-hosts):ro/);
-  expect(gitHttps).toMatch(/workspace-registry-git-credentials:ro/);
-  expect(gitHttps).toMatch(/workspace-registry-git-ca:ro/);
-  expect(gitSsh).toMatch(/workspace-registry-git-ssh-key:ro/);
-  expect(gitSsh).toMatch(/workspace-registry-git-known-hosts:ro/);
-  expect(dockerfile).toMatch(/mkdir -p[^\n]*\/data\/workspace-registry/);
-  expect(dockerfile).toMatch(/chown -R thoth:thoth \/home\/thoth\/\.pi \/data/);
-  expect(smoke).toContain('core_remote="/fixtures/offline.git"');
-  expect(smoke).toContain('"degraded":true');
-  expect(smoke).toContain('core_remote="/fixtures/remote.git"');
-});
-
-test("workspace registry smoke image cleanup is scoped to the per-run image identity", () => {
-  const output = execFileSync("bash", ["scripts/workspace-registry-smoke.sh"], {
-    cwd: new URL("../..", import.meta.url),
-    env: { ...process.env, WORKSPACE_REGISTRY_SMOKE_SELF_TEST: "image-cleanup-identity" },
-    encoding: "utf8",
-  });
-
-  expect(output).toContain("workspace registry smoke image cleanup identity self-test passed");
-});
@@ -1,102 +0,0 @@
-import { expect, test } from "vitest";
-import { migrateWorkspaceV2ToV3 } from "../src/workspaces/migrate-v2-qdrant.js";
-import { parseWorkspaceYaml } from "../src/workspaces/schema.js";
-
-const workspaceV2Yaml = `workspace:
-  schema_version: 2
-  id: psd-clinical
-  name: Policlinico San Donato
-  description: Clinical data warehouse workspace
-  language: it
-dwh:
-  engine: postgres
-  database: postgres
-  schema: datawarehouse
-  supported_transports:
-    - postgres_direct
-    - rest_api
-semantic_index:
-  vector_store:
-    engine: pgvector
-    database: postgres
-    schema: vectors
-    collection: clinical_documents
-    dimensions: 768
-    distance: inner_product
-    supported_transports:
-      - pgvector_direct
-      - rest_api
-  embedding:
-    provider: openai_compatible
-    model: text-embedding-3-large
-    dimensions: 768
-llm_policy:
-  default: zai/glm-5.2
-  allowed:
-    - zai/glm-5.2
-diagnostics:
-  dwh_rest:
-    method: POST
-    path: /rpc/dwh
-    auth: bearer
-    response:
-      database: database
-      schema: schema
-  vector_rest:
-    metadata:
-      method: GET
-      path: /vector
-      auth: bearer
-      response:
-        collection: collection
-        dimensions: dimensions
-        distance: distance
-  embedding:
-    method: GET
-    path: /models
-    auth: none
-    response:
-      model: model
-      dimensions: dimensions
-`;
-
-test("migrates a schema v2 workspace to the internal qdrant schema v3 shape", () => {
-  const legacy = parseWorkspaceYaml(workspaceV2Yaml);
-
-  const migrated = migrateWorkspaceV2ToV3(legacy, "psd-clinical");
-
-  expect(migrated).not.toHaveProperty("evidence");
-  expect(migrated).toMatchObject({
-    workspace: { schema_version: 3, id: "psd-clinical", name: "Policlinico San Donato" },
-    dwh: legacy.dwh,
-    semantic_index: {
-      vector_store: {
-        engine: "qdrant",
-        collection: "psd-clinical",
-        dimensions: 1024,
-        distance: "cosine",
-      },
-      embedding: {
-        provider: "ollama_internal",
-        model: "qwen3-embedding:0.6b",
-        dimensions: 1024,
-      },
-    },
-    llm_policy: legacy.llm_policy,
-    diagnostics: {
-      dwh_rest: legacy.diagnostics?.dwh_rest,
-    },
-  });
-});
-
-test("drops vector and embedding diagnostics and transports during v2 to v3 migration", () => {
-  const legacy = parseWorkspaceYaml(workspaceV2Yaml);
-
-  const migrated = migrateWorkspaceV2ToV3(legacy, "psd-clinical");
-
-  expect(migrated.diagnostics).toEqual({
-    dwh_rest: legacy.diagnostics?.dwh_rest,
-  });
-  expect(migrated.semantic_index.vector_store).not.toHaveProperty("supported_transports");
-  expect(migrated.semantic_index.embedding).not.toHaveProperty("timeout_ms");
-});