ThothII/backend/test/workspaces-bindings.test.ts

import { chmodSync, mkdirSync, mkdtempSync, realpathSync, rmSync, symlinkSync, writeFileSync } from "node:fs";
import { tmpdir } from "node:os";
import { join } from "node:path";
import { afterEach, expect, test } from "vitest";
import { resolveBinding, resolveEvidenceBinding, resolveRuntimeBindings, supportsSessionRuntime } from "../src/workspaces/bindings.js";
import { parseWorkspaceYaml } from "../src/workspaces/schema.js";

const workspace = parseWorkspaceYaml(`workspace:
  schema_version: 2
  id: psd-clinical
  name: Policlinico San Donato
  language: it
dwh:
  engine: postgres
  database: postgres
  schema: datawarehouse
  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
    supported_transports: [pgvector_direct, rest_api, ssh_tunnel]
  embedding:
    provider: ollama_compatible
    model: nomic-embed-text-v2-moe
    dimensions: 768
llm_policy:
  allowed: [zai/glm-5.2]
`);
const workspaceV3 = parseWorkspaceYaml(`workspace:
  schema_version: 3
  id: psd-clinical
  name: Policlinico San Donato
  language: it
dwh:
  engine: postgres
  database: postgres
  schema: datawarehouse
  supported_transports: [postgres_direct, rest_api]
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:
  allowed: [zai/glm-5.2]
`);
const temporaryRoots: string[] = [];

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

function secretPath(name: string): { root: string; path: string } {
  const root = mkdtempSync(join(tmpdir(), "thoth-binding-"));
  temporaryRoots.push(root);
  const secrets = join(root, "secrets");
  mkdirSync(secrets);
  const path = join(secrets, name);
  writeFileSync(path, "");
  return { root: secrets, path };
}

test("marks a portable workspace non-activatable when its local REST key file is absent", () => {
  const result = resolveBinding(workspace, "DWH", {
    THT_WS_PSD_CLINICAL_DWH_TRANSPORT: "rest_api",
    THT_WS_PSD_CLINICAL_DWH_BASE_URL: "https://dwh.example.test",
  }, ["/run/secrets"]);

  expect(result.missing).toContain("THT_WS_PSD_CLINICAL_DWH_API_KEY_FILE");
});

test("does not require a REST secret file when its declared diagnostic uses auth none", () => {
  const unauthenticatedWorkspace = parseWorkspaceYaml(`workspace:
  schema_version: 2
  id: psd-clinical
  name: Policlinico San Donato
  language: it
dwh:
  engine: postgres
  database: postgres
  schema: datawarehouse
  supported_transports: [rest_api]
semantic_index:
  vector_store:
    engine: pgvector
    database: postgres
    schema: vectors
    collection: clinical_documents
    dimensions: 768
    distance: cosine
    supported_transports: [rest_api]
  embedding:
    provider: ollama_compatible
    model: nomic-embed-text-v2-moe
    dimensions: 768
diagnostics:
  dwh_rest:
    method: POST
    path: /rpc/ping
    auth: none
    response: { database: database, schema: schema }
  vector_rest:
    metadata:
      method: GET
      path: /vector/metadata
      auth: none
      response: { collection: collection, dimensions: dimensions, distance: distance }
  embedding:
    method: GET
    path: /models
    auth: none
    response: { model: model, dimensions: dimensions }
llm_policy:
  allowed: [zai/glm-5.2]
`);

  const bindings = resolveRuntimeBindings(unauthenticatedWorkspace, {
    THT_WS_PSD_CLINICAL_DWH_TRANSPORT: "rest_api",
    THT_WS_PSD_CLINICAL_DWH_BASE_URL: "https://dwh.example.test",
    THT_WS_PSD_CLINICAL_VECTOR_TRANSPORT: "rest_api",
    THT_WS_PSD_CLINICAL_VECTOR_BASE_URL: "https://vector.example.test",
    THT_WS_PSD_CLINICAL_EMBEDDING_BASE_URL: "https://embedding.example.test",
  }, ["/run/secrets"]);

  expect(bindings.dwh.missing).toEqual([]);
  expect(bindings.vector.missing).toEqual([]);
  expect(bindings.embedding.missing).toEqual([]);
});

test("resolves direct bindings from the stable workspace namespace", () => {
  const password = secretPath("dwh-password");
  const result = resolveBinding(workspace, "DWH", {
    THT_WS_PSD_CLINICAL_DWH_TRANSPORT: "postgres_direct",
    THT_WS_PSD_CLINICAL_DWH_HOST: "dwh.internal",
    THT_WS_PSD_CLINICAL_DWH_PORT: "5432",
    THT_WS_PSD_CLINICAL_DWH_USER: "thoth_reader",
    THT_WS_PSD_CLINICAL_DWH_PASSWORD_FILE: password.path,
  }, [password.root]);

  expect(result).toMatchObject({
    transport: "postgres_direct",
    missing: [],
    values: {
      THT_WS_PSD_CLINICAL_DWH_HOST: "dwh.internal",
      THT_WS_PSD_CLINICAL_DWH_PORT: "5432",
      THT_WS_PSD_CLINICAL_DWH_PASSWORD_FILE: realpathSync(password.path),
    },
  });
});

test("reports only a FILE variable name when a secret path is outside the configured roots", () => {
  const outside = secretPath("outside-password");
  const allowed = secretPath("allowed-password");
  const result = resolveBinding(workspace, "DWH", {
    THT_WS_PSD_CLINICAL_DWH_TRANSPORT: "postgres_direct",
    THT_WS_PSD_CLINICAL_DWH_HOST: "dwh.internal",
    THT_WS_PSD_CLINICAL_DWH_PORT: "5432",
    THT_WS_PSD_CLINICAL_DWH_USER: "thoth_reader",
    THT_WS_PSD_CLINICAL_DWH_PASSWORD_FILE: outside.path,
  }, [allowed.root]);

  expect(result.missing).toEqual(["THT_WS_PSD_CLINICAL_DWH_PASSWORD_FILE"]);
  expect(result.missing.join("\n")).not.toContain(outside.path);
});

test("reports an invalid optional secret file instead of silently dropping it", () => {
  const password = secretPath("dwh-password");
  const result = resolveBinding(workspace, "DWH", {
    THT_WS_PSD_CLINICAL_DWH_TRANSPORT: "postgres_direct",
    THT_WS_PSD_CLINICAL_DWH_HOST: "dwh.internal",
    THT_WS_PSD_CLINICAL_DWH_PORT: "5432",
    THT_WS_PSD_CLINICAL_DWH_USER: "thoth_reader",
    THT_WS_PSD_CLINICAL_DWH_PASSWORD_FILE: password.path,
    THT_WS_PSD_CLINICAL_DWH_TLS_CA_FILE: "relative-ca.pem",
  }, [password.root]);

  expect(result.missing).toContain("THT_WS_PSD_CLINICAL_DWH_TLS_CA_FILE");
  expect(result.values).not.toHaveProperty("THT_WS_PSD_CLINICAL_DWH_TLS_CA_FILE");
});

test("rejects a selected transport that the canonical workspace does not support", () => {
  const directOnly = {
    ...workspace,
    dwh: { ...workspace.dwh, supported_transports: ["postgres_direct"] },
  };
  const result = resolveBinding(directOnly, "DWH", {
    THT_WS_PSD_CLINICAL_DWH_TRANSPORT: "rest_api",
  }, ["/run/secrets"]);

  expect(result).toMatchObject({
    transport: "rest_api",
    missing: ["THT_WS_PSD_CLINICAL_DWH_TRANSPORT"],
  });
});

test("never treats a vector reader credential as the optional writer binding", () => {
  const readerKey = secretPath("vector-reader-key");
  const writerWorkspace = {
    ...workspace,
    semantic_index: { ...workspace.semantic_index, vector_writer: {} },
  };
  const resolveWriter = () => resolveBinding(writerWorkspace, "VECTOR_WRITER" as never, {
    THT_WS_PSD_CLINICAL_VECTOR_API_KEY_FILE: readerKey.path,
  }, [readerKey.root]);

  expect(resolveWriter).not.toThrow();
  expect(resolveWriter()).toMatchObject({
    missing: ["THT_WS_PSD_CLINICAL_VECTOR_WRITER_API_KEY_FILE"],
    values: {},
  });
});

test("fails closed for v3 external semantic bindings", () => {
  expect(() => resolveBinding(workspaceV3, "VECTOR", {}, ["/run/secrets"]))
    .not.toThrow();
  expect(() => resolveBinding(workspaceV3, "EMBEDDING", {}, ["/run/secrets"]))
    .not.toThrow();
  expect(() => resolveRuntimeBindings(workspaceV3, {
    THT_WS_PSD_CLINICAL_VECTOR_TRANSPORT: "rest_api",
    THT_WS_PSD_CLINICAL_VECTOR_BASE_URL: "https://vector.example.test",
    THT_WS_PSD_CLINICAL_EMBEDDING_BASE_URL: "https://embedding.example.test",
    THT_WS_PSD_CLINICAL_VECTOR_API_KEY_FILE: "/run/secrets/vector-api-key",
  }, ["/run/secrets"])).not.toThrow();
});

test("schema v3 ignores external semantic binding variables and reports only DWH requirements", () => {
  const password = secretPath("dwh-password");
  const bindings = resolveRuntimeBindings(workspaceV3, {
    THT_WS_PSD_CLINICAL_DWH_TRANSPORT: "postgres_direct",
    THT_WS_PSD_CLINICAL_DWH_HOST: "dwh.internal",
    THT_WS_PSD_CLINICAL_DWH_PORT: "5432",
    THT_WS_PSD_CLINICAL_DWH_USER: "reader",
    THT_WS_PSD_CLINICAL_DWH_PASSWORD_FILE: password.path,
    THT_WS_PSD_CLINICAL_VECTOR_TRANSPORT: "rest_api",
    THT_WS_PSD_CLINICAL_VECTOR_BASE_URL: "https://vector.example.test",
    THT_WS_PSD_CLINICAL_VECTOR_API_KEY_FILE: "/run/secrets/vector-api-key",
    THT_WS_PSD_CLINICAL_EMBEDDING_BASE_URL: "https://embedding.example.test",
    THT_WS_PSD_CLINICAL_EMBEDDING_API_KEY_FILE: "/run/secrets/embedding-api-key",
  }, [password.root]);

  expect(bindings.dwh.missing).toEqual([]);
  expect(bindings.vector.missing).toEqual([]);
  expect(bindings.embedding.missing).toEqual([]);
  expect(bindings.vector.values).toEqual({});
  expect(bindings.embedding.values).toEqual({});
});


function withEvidence(source: Record<string, unknown>) {
  return parseWorkspaceYaml(`workspace:
  schema_version: 3
  id: psd-clinical
  name: Policlinico San Donato
  language: it
dwh:
  engine: postgres
  database: postgres
  schema: datawarehouse
  supported_transports: [postgres_direct]
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: { allowed: [zai/glm-5.2] }
evidence:
  source: ${JSON.stringify(source)}
`);
}

const evidenceVariable = (suffix: string) => `THT_WS_PSD_CLINICAL_EVIDENCE_${suffix}`;

test.each([
  { type: "filesystem", uri: "workspace-content/psd-clinical/evidence" },
  { type: "http", uris: ["https://evidence.example.test/guide.md"], authentication: "none" },
  { type: "s3", uri: "s3://clinical-evidence/published/", credentials: "ambient" },
])("does not resolve Evidence variables for $type modes without file credentials", (source) => {
  expect(resolveEvidenceBinding(withEvidence(source), {
    [evidenceVariable("SIGNED_URLS_FILE")]: "/CANARY/http",
    [evidenceVariable("ACCESS_KEY_FILE")]: "/CANARY/access",
  }, ["/run/secrets"])).toEqual({ values: {}, missing: [] });
});

test("requires only a safe HTTP signed-URL file and never reads its contents", () => {
  const signed = secretPath("evidence-signed-urls");
  writeFileSync(signed.path, "CANARY-SIGNED-URL-CONTENT");
  const source = withEvidence({
    type: "http",
    uris: ["https://evidence.example.test/guide.md"],
    authentication: "signed_urls_file",
  });
  const variable = evidenceVariable("SIGNED_URLS_FILE");

  expect(resolveEvidenceBinding(source, {}, [signed.root]).missing).toEqual([variable]);
  const resolved = resolveEvidenceBinding(source, {
    [variable]: signed.path,
    [evidenceVariable("ACCESS_KEY_FILE")]: signed.path,
  }, [signed.root]);
  expect(resolved).toEqual({ values: { [variable]: realpathSync(signed.path) }, missing: [] });
  expect(JSON.stringify(resolved)).not.toContain("CANARY-SIGNED-URL-CONTENT");
});

test("canonicalizes an in-root Evidence symlink before passing it to the harness", () => {
  const signed = secretPath("evidence-signed-target");
  const link = join(signed.root, "signed-urls-link");
  symlinkSync(signed.path, link);
  const source = withEvidence({
    type: "http",
    uris: ["https://evidence.example.test/guide.md"],
    authentication: "signed_urls_file",
  });
  const variable = evidenceVariable("SIGNED_URLS_FILE");

  expect(resolveEvidenceBinding(source, { [variable]: link }, [signed.root])).toEqual({
    values: { [variable]: realpathSync(signed.path) }, missing: [],
  });
});

test("requires S3 access and secret files together while accepting an optional safe session token", () => {
  const access = secretPath("evidence-access");
  const secret = secretPath("evidence-secret");
  const token = secretPath("evidence-token");
  const source = withEvidence({
    type: "s3", uri: "s3://clinical-evidence/published/", credentials: "static_files",
  });
  const env = {
    [evidenceVariable("ACCESS_KEY_FILE")]: access.path,
    [evidenceVariable("SECRET_KEY_FILE")]: secret.path,
    [evidenceVariable("SESSION_TOKEN_FILE")]: token.path,
    [evidenceVariable("SIGNED_URLS_FILE")]: access.path,
  };

  expect(resolveEvidenceBinding(source, {
    [evidenceVariable("ACCESS_KEY_FILE")]: access.path,
  }, [access.root]).missing).toEqual([evidenceVariable("SECRET_KEY_FILE")]);
  expect(resolveEvidenceBinding(source, env, [access.root, secret.root, token.root])).toEqual({
    values: {
      [evidenceVariable("ACCESS_KEY_FILE")]: realpathSync(access.path),
      [evidenceVariable("SECRET_KEY_FILE")]: realpathSync(secret.path),
      [evidenceVariable("SESSION_TOKEN_FILE")]: realpathSync(token.path),
    },
    missing: [],
  });
});

test("rejects relative, missing, directory, unreadable, and escaping symlink Evidence paths", () => {
  const allowed = secretPath("valid");
  const outside = secretPath("outside");
  const directory = join(allowed.root, "directory");
  mkdirSync(directory);
  const link = join(allowed.root, "escape");
  symlinkSync(outside.path, link);
  const unreadable = join(allowed.root, "unreadable");
  writeFileSync(unreadable, "secret");
  chmodSync(unreadable, 0o000);
  const source = withEvidence({
    type: "http", uris: ["https://evidence.example.test/guide.md"], authentication: "signed_urls_file",
  });
  const variable = evidenceVariable("SIGNED_URLS_FILE");

  for (const path of ["relative", join(allowed.root, "missing"), directory, unreadable, link]) {
    expect(resolveEvidenceBinding(source, { [variable]: path }, [allowed.root])).toEqual({
      values: {}, missing: [variable],
    });
  }
  chmodSync(unreadable, 0o600);
});

test("includes Evidence binding completeness in session runtime support without changing v3 compatibility", () => {
  const unsigned = resolveRuntimeBindings(workspaceV3, {}, ["/run/secrets"]);
  expect(unsigned.evidence).toEqual({ values: {}, missing: [] });
  expect(supportsSessionRuntime(unsigned)).toBe(true);

  const signed = resolveRuntimeBindings(withEvidence({
    type: "http", uris: ["https://evidence.example.test/guide.md"], authentication: "signed_urls_file",
  }), {}, ["/run/secrets"]);
  expect(signed.evidence.missing).toEqual([evidenceVariable("SIGNED_URLS_FILE")]);
  expect(supportsSessionRuntime(signed)).toBe(false);
});