ThothII/backend/test/effective-config.test.ts

import { createHash } from "node:crypto";
import { expect, test } from "vitest";
import {
  buildCanonicalEffectiveConfig,
  canonicalEffectiveConfigJson,
  configFingerprint,
  effectiveConfigIdentity,
  inputFingerprint,
} from "../src/workspaces/effective-config.js";

const semanticRuntime = {
  internalQdrantUrl: "http://qdrant:6333",
  internalEmbeddingUrl: "http://embedding:11434",
  internalEmbeddingId: "ollama/qwen3-embedding:0.6b",
  internalEmbeddingModel: "qwen3-embedding:0.6b",
  internalEmbeddingDimensions: 1024,
};

function directRendered(): Record<string, unknown> {
  return {
    runtime_identity: {
      workspace_id: "psd-clinical",
      workspace_revision: "a".repeat(40),
      source_identity: "workspace://psd-clinical",
    },
    session_storage: { mode: "local" },
    profile: "server",
    language: "it",
    database: {
      host: "dwh.internal",
      port: 5432,
      database: "postgres",
      schema: "datawarehouse",
      user: "thoth_reader",
      password_file: "/run/secrets/dwh-password",
      ssl_ca_file: "/run/secrets/dwh-ca.pem",
      transport: "direct",
    },
    dwh: { type: "postgres_direct" },
    resources: {
      vector: {
        engine: "qdrant",
        base_url: "http://qdrant:6333",
        collection: "psd-clinical",
        dimensions: 1024,
        distance: "cosine",
        collection_lifecycle: "require_existing",
      },
      embeddings: {
        provider: "ollama_internal",
        base_url: "http://embedding:11434",
        model: "qwen3-embedding:0.6b",
        dimensions: 1024,
      },
    },
    roots: {
      sessions: "/data/sessions/psd-clinical/sessions",
      artifacts: "/data/sessions/psd-clinical/artifacts",
      indexes: "/data/sessions/psd-clinical/indexes",
    },
    paths: {
      sessions: "/data/sessions/psd-clinical/sessions",
      artifacts: "/data/sessions/psd-clinical/artifacts",
      indexes: "/data/sessions/psd-clinical/indexes",
      memory: "/data/sessions/psd-clinical/memory",
    },
    evidence: {
      sources: [{
        type: "filesystem",
        root: "/srv/registry/snapshots/rev/psd-clinical/evidence",
        patterns: ["**/*.md"],
        max_bytes: 10_485_760,
      }],
    },
  };
}

function restRendered(): Record<string, unknown> {
  return {
    ...directRendered(),
    database: {
      host: "localhost",
      port: 5432,
      database: "postgres",
      schema: "datawarehouse",
      user: "rest",
      password: "",
      transport: "rest",
    },
    dwh: {
      type: "thoth_rest",
      database: { database: "postgres", schema: "datawarehouse" },
      endpoint: {
        base_url: "https://dwh.example.test",
        api_key_file: "/run/secrets/dwh-api-key",
      },
    },
    rest: {
      base_url: "https://dwh.example.test",
      api_key_file: "/run/secrets/dwh-api-key",
    },
  };
}

const directCanonical =
  `{"schemaVersion":2,"dwh":{` +
  `"engine":"postgres","database":"postgres","schema":"datawarehouse",` +
  `"transport":"postgres_direct","host":"dwh.internal","port":5432,"user":"thoth_reader"},` +
  `"vector":{"collection":"psd-clinical","dimensions":1024,"distance":"cosine"},` +
  `"embedding":{"id":"ollama/qwen3-embedding:0.6b","model":"qwen3-embedding:0.6b","dimensions":1024},` +
  `"roots":{"artifacts":"/data/sessions/psd-clinical/artifacts",` +
  `"indexes":"/data/sessions/psd-clinical/indexes"}}`;

test("canonical effective config is deterministic and contains the expected key order", () => {
  const rendered = directRendered();
  const canonical = buildCanonicalEffectiveConfig(rendered);
  expect(canonicalEffectiveConfigJson(canonical)).toBe(directCanonical);
  expect(buildCanonicalEffectiveConfig(rendered)).toEqual(canonical);
});

test("canonical effective config excludes secrets, evidence, session storage, and runtime identity", () => {
  const json = canonicalEffectiveConfigJson(buildCanonicalEffectiveConfig(directRendered()));
  expect(json).not.toContain("password_file");
  expect(json).not.toContain("ssl_ca_file");
  expect(json).not.toContain("session_storage");
  expect(json).not.toContain("runtime_identity");
  expect(json).not.toContain("evidence");
  expect(json).not.toContain("sources");
  expect(json).not.toContain("collection_lifecycle");
  expect(json).not.toContain("base_url"); // vector/embedding service URLs are not identity
  expect(json).not.toContain("memory");
  expect(json).not.toContain('"sessions"');
});

test("REST transport canonicalizes to transport rest_api with baseUrl and no host/port", () => {
  const canonical = buildCanonicalEffectiveConfig(restRendered());
  const json = canonicalEffectiveConfigJson(canonical);
  expect(json).toContain(`"transport":"rest_api"`);
  expect(json).toContain(`"baseUrl":"https://dwh.example.test"`);
  expect(json).not.toContain(`"host":`);
  expect(json).not.toContain(`"port":`);
  expect(json).not.toContain("api_key_file");
});

test("identity and fingerprint helpers produce stable prefixed hex values", () => {
  const rendered = directRendered();
  const identity = effectiveConfigIdentity("psd-clinical", rendered);
  const cfg = configFingerprint(rendered);
  const input = inputFingerprint("psd-clinical", rendered);

  expect(identity).toMatch(/^workspace:\/\/psd-clinical@v1:[0-9a-f]{64}$/);
  expect(cfg).toBe("sha256:" + createHash("sha256").update(directCanonical).digest("hex"));
  expect(input).toBe("sha256:" + createHash("sha256").update(identity).digest("hex"));
  expect(input).not.toBe(cfg);
});

test("content-only or session_storage changes keep the same effective config identity", () => {
  const base = directRendered();
  const identityBefore = effectiveConfigIdentity("psd-clinical", base);
  const fingerprintBefore = configFingerprint(base);

  const contentOnly = {
    ...base,
    runtime_identity: {
      ...base.runtime_identity,
      workspace_revision: "b".repeat(40),
    },
    session_storage: { mode: "remote", url: "http://example.test" },
    evidence: {
      sources: [{
        type: "filesystem",
        root: "/srv/registry/snapshots/other/psd-clinical/evidence",
        patterns: ["**/*.txt"],
        max_bytes: 999,
      }],
    },
  };

  expect(effectiveConfigIdentity("psd-clinical", contentOnly)).toBe(identityBefore);
  expect(configFingerprint(contentOnly)).toBe(fingerprintBefore);
});

test("DWH-affecting changes alter the effective config identity", () => {
  const base = directRendered();
  const identityBefore = effectiveConfigIdentity("psd-clinical", base);

  const changedHost = { ...base, database: { ...(base.database as object), host: "dwh-two.internal" } };
  expect(effectiveConfigIdentity("psd-clinical", changedHost)).not.toBe(identityBefore);

  const changedDatabase = { ...base, database: { ...(base.database as object), database: "analytics" } };
  expect(effectiveConfigIdentity("psd-clinical", changedDatabase)).not.toBe(identityBefore);

  const changedCollection = { ...base, resources: { ...base.resources, vector: { ...(base.resources as Record<string, any>).vector, collection: "other" } } };
  expect(effectiveConfigIdentity("psd-clinical", changedCollection)).not.toBe(identityBefore);

  const changedEmbeddingIdentity = { ...base, resources: { ...base.resources, embeddings: { ...(base.resources as Record<string, any>).embeddings, model: "other-embedding" } } };
  expect(effectiveConfigIdentity("psd-clinical", changedEmbeddingIdentity)).not.toBe(identityBefore);

  const changedTransport = restRendered();
  expect(effectiveConfigIdentity("psd-clinical", changedTransport)).not.toBe(identityBefore);
});