Files
ThothII/backend/test/workspaces-schema.test.ts
T

131 lines
4.4 KiB
TypeScript

import { expect, test } from "vitest";
import { parseWorkspaceYaml, serializeWorkspaceYaml } from "../src/workspaces/schema.js";
export const validYaml = `workspace:
schema_version: 1
id: psd-clinical
name: Policlinico San Donato
description: Clinical data warehouse workspace
language: it
dwh:
engine: postgres
database: postgres
schema: datawarehouse
port: 5432
timeout_ms: 5000
supported_transports:
- postgres_direct
- rest_api
- ssh_tunnel
semantic_index:
vector_store:
engine: pgvector
database: postgres
schema: vectors
collection: clinical_documents
dimensions: 768
distance: cosine
port: 5432
timeout_ms: 5000
supported_transports:
- pgvector_direct
- rest_api
- ssh_tunnel
embedding:
provider: ollama_compatible
model: nomic-embed-text-v2-moe
dimensions: 768
timeout_ms: 5000
llm_policy:
default: zai/glm-5.2
allowed:
- zai/glm-5.2
- openai/gpt-5
`;
test("rejects a workspace whose embedding dimensions differ from its collection", () => {
expect(() => parseWorkspaceYaml(validYaml.replace("dimensions: 768", "dimensions: 1536")))
.toThrow(/dimensions/i);
});
test("rejects an LLM default outside its allowlist", () => {
expect(() => parseWorkspaceYaml(validYaml.replace("- zai/glm-5.2", "- openai/gpt-5")))
.toThrow(/allowlist/i);
});
test("rejects unknown keys and invalid immutable IDs", () => {
expect(() => parseWorkspaceYaml(validYaml.replace(" language: it", " language: it\n label: PSD")))
.toThrow(/unrecognized key/i);
expect(() => parseWorkspaceYaml(validYaml.replace("id: psd-clinical", "id: PSD")))
.toThrow(/id/i);
});
test("accepts optional connection ports and timeouts but rejects unsafe values", () => {
expect(parseWorkspaceYaml(validYaml).dwh.port).toBe(5432);
expect(() => parseWorkspaceYaml(validYaml.replace("port: 5432", "port: 0")))
.toThrow(/port/i);
expect(() => parseWorkspaceYaml(validYaml.replace("port: 5432", "port: 65536")))
.toThrow(/port/i);
expect(() => parseWorkspaceYaml(validYaml.replace("timeout_ms: 5000", "timeout_ms: 0")))
.toThrow(/timeout/i);
});
test("requires explicit vector database and schema identities with strict diagnostic declarations", () => {
const diagnosticWorkspace = validYaml.replace(
" engine: pgvector\n database: postgres",
" engine: pgvector\n database: vector_database",
).replace(
"llm_policy:\n",
"diagnostics:\n"
+ " dwh_rest:\n"
+ " method: POST\n"
+ " path: /rpc/ping\n"
+ " auth: bearer\n"
+ " response:\n"
+ " database: database\n"
+ " schema: schema\n"
+ " vector_rest:\n"
+ " metadata:\n"
+ " method: GET\n"
+ " path: /metadata\n"
+ " auth: bearer\n"
+ " response:\n"
+ " collection: collection\n"
+ " dimensions: dimensions\n"
+ " distance: distance\n"
+ " reversible_probe:\n"
+ " method: POST\n"
+ " path: /rpc/diagnostic_vector_probe\n"
+ " auth: bearer\n"
+ " embedding:\n"
+ " method: GET\n"
+ " path: /models\n"
+ " auth: none\n"
+ " response:\n"
+ " model: model\n"
+ " dimensions: dimensions\n"
+ "llm_policy:\n",
);
expect(parseWorkspaceYaml(diagnosticWorkspace).semantic_index.vector_store).toMatchObject({
database: "vector_database",
schema: "vectors",
});
expect(() => parseWorkspaceYaml(diagnosticWorkspace.replace("database: vector_database", 'database: " "')))
.toThrow(/database/i);
expect(() => parseWorkspaceYaml(diagnosticWorkspace.replace("schema: vectors", 'schema: " "')))
.toThrow(/schema/i);
expect(() => parseWorkspaceYaml(diagnosticWorkspace.replace("method: POST", "method: PATCH")))
.toThrow(/method/i);
expect(() => parseWorkspaceYaml(diagnosticWorkspace.replace(" distance: distance", " distance: distance\n extra: ignored")))
.toThrow(/unrecognized key/i);
});
test("serializes canonical YAML that parses back to the same workspace", () => {
const workspace = parseWorkspaceYaml(validYaml);
const serialized = serializeWorkspaceYaml(workspace);
expect(serializeWorkspaceYaml(parseWorkspaceYaml(serialized))).toBe(serialized);
expect(parseWorkspaceYaml(serialized)).toEqual(workspace);
});