lynkr 8.0.0 → 9.0.1

package/test/README.md

@@ -1,212 +0,0 @@
-# Test Suite Documentation
-
-All tests for the Lynkr project are consolidated in this `test/` directory.
-
-## Test Files
-
-### Unit Tests
-**File**: `routing.test.js`
-**Purpose**: Tests the hybrid routing logic in isolation
-**Run**: `DATABRICKS_API_KEY=test-key DATABRICKS_API_BASE=http://test.com node --test test/routing.test.js`
-**Coverage**: 10 tests
-- Routing with tier-based routing disabled (no TIER_* vars set)
-- Simple requests → Ollama
-- Complex requests → Cloud
-- Tool capability checks
-- Fallback configuration
-
----
-
-### Integration Tests
-**File**: `hybrid-routing-integration.test.js`
-**Purpose**: Tests configuration validation and metrics recording
-**Run**: `node --test test/hybrid-routing-integration.test.js`
-**Coverage**: 13 tests
-- Configuration validation (5 tests)
-- Metrics recording (6 tests)
-- Helper functions (2 tests)
-
----
-
-### Hybrid Routing Performance Tests
-**File**: `hybrid-routing-performance.test.js`
-**Purpose**: Measures performance overhead of hybrid routing
-**Run**: `node test/hybrid-routing-performance.test.js`
-**Key Metrics**:
-- Routing decision: <0.01ms (36.8M decisions/sec)
-- Metrics overhead: <0.01ms (43.6M ops/sec)
-- Combined overhead: <0.02ms (15.6M ops/sec)
-
----
-
-### System Performance Tests
-**File**: `performance-tests.js`
-**Purpose**: Tests system-wide performance optimizations
-**Run**: `DATABRICKS_API_KEY=test-key DATABRICKS_API_BASE=http://test.com node test/performance-tests.js`
-**Coverage**:
-- Database indexes (100% complete)
-- Persistent prompt cache
-- Regex pattern caching (4.5x faster)
-- Lazy loading
-- HTTP connection pooling
-- Response compression
-
----
-
-### Middleware Benchmarks
-**File**: `performance-benchmark.js`
-**Purpose**: Benchmarks middleware overhead
-**Run**: `DATABRICKS_API_KEY=test-key DATABRICKS_API_BASE=http://test.com node test/performance-benchmark.js`
-**Coverage**:
-- Metrics collection (3.4M ops/sec)
-- Circuit breakers (3.9M ops/sec)
-- Input validation (5.7M ops/sec)
-- Load shedding
-- Combined middleware stack
-
----
-
-## Running All Tests
-
-### Using npm scripts (Recommended)
-
-**Run all tests (unit + performance):**
-```bash
-npm test
-```
-
-**Run only unit/integration tests:**
-```bash
-npm run test:unit
-```
-
-**Run only performance tests:**
-```bash
-npm run test:performance
-```
-
-**Run only benchmarks:**
-```bash
-npm run test:benchmark
-```
-
-**Run quick smoke test (routing only):**
-```bash
-npm run test:quick
-```
-
-**Run everything including benchmarks:**
-```bash
-npm run test:all
-```
-
-### Manual execution (if needed)
-
-**Quick Test:**
-Run all unit and integration tests:
-```bash
-DATABRICKS_API_KEY=test-key DATABRICKS_API_BASE=http://test.com node --test test/
-```
-
-**Full Test Suite:**
-Run everything including performance tests:
-```bash
-# Unit + Integration tests
-DATABRICKS_API_KEY=test-key DATABRICKS_API_BASE=http://test.com node --test test/routing.test.js
-DATABRICKS_API_KEY=test-key DATABRICKS_API_BASE=http://test.com node --test test/hybrid-routing-integration.test.js
-
-# Performance tests
-node test/hybrid-routing-performance.test.js
-DATABRICKS_API_KEY=test-key DATABRICKS_API_BASE=http://test.com node test/performance-tests.js
-DATABRICKS_API_KEY=test-key DATABRICKS_API_BASE=http://test.com node test/performance-benchmark.js
-```
-
----
-
-## Test Organization
-
-```
-test/
-├── README.md                              ← This file
-├── routing.test.js                        ← Unit tests (10 tests)
-├── hybrid-routing-integration.test.js     ← Integration tests (13 tests)
-├── hybrid-routing-performance.test.js     ← Routing performance benchmarks
-├── performance-tests.js                   ← System performance tests
-└── performance-benchmark.js               ← Middleware benchmarks
-```
-
----
-
-## Important Notes
-
-### Not Test Files
-- `src/tests/` - This is **application code**, not tests! It provides test execution functionality as a feature.
-
-### Environment Variables
-Most tests require Databricks credentials (even though they're not used in actual API calls):
-```bash
-export DATABRICKS_API_KEY=test-key
-export DATABRICKS_API_BASE=http://test.com
-```
-
-### Test Results Summary
-
-| Test Type | Status | Count |
-|-----------|--------|-------|
-| Unit Tests | ✅ Passing | 10/10 |
-| Integration Tests | ✅ Passing | 13/13 |
-| Routing Performance | ✅ Complete | <0.02ms overhead |
-| System Performance | ✅ Complete | 100% optimizations |
-| Middleware Benchmarks | ✅ Complete | Acceptable overhead |
-
----
-
-## CI/CD Integration
-
-To run in CI/CD pipelines:
-
-```bash
-#!/bin/bash
-set -e
-
-# Run all tests using npm
-echo "Running test suite..."
-npm run test:all
-
-echo "All tests passed!"
-```
-
-Or for a faster CI pipeline (skip benchmarks):
-
-```bash
-#!/bin/bash
-set -e
-
-echo "Running tests..."
-npm test
-
-echo "Tests passed!"
-```
-
----
-
-## Adding New Tests
-
-When adding new tests, follow these conventions:
-
-1. **Unit tests**: Test individual functions in isolation
-   - Place in `test/`
-   - Use `node:test` framework
-   - Name: `feature-name.test.js`
-
-2. **Integration tests**: Test multiple components together
-   - Place in `test/`
-   - Use `node:test` framework
-   - Name: `feature-name-integration.test.js`
-
-3. **Performance tests**: Benchmark specific features
-   - Place in `test/`
-   - Use custom benchmark utilities
-   - Name: `feature-name-performance.test.js`
-
-4. **Always**: Document in this README!

package/test/azure-openai-config.test.js

@@ -1,213 +0,0 @@
-const assert = require("assert");
-const { describe, it, beforeEach, afterEach } = require("node:test");
-
-describe("Azure OpenAI Configuration Tests", () => {
-  let originalConfig;
-
-  beforeEach(() => {
-    // Clear module cache
-    delete require.cache[require.resolve("../src/config")];
-
-    // Store original config
-    originalConfig = { ...process.env };
-
-    // Set Azure OpenAI environment variables to empty strings to override .env values
-    // (deleting them would cause dotenv to reload from .env file)
-    process.env.AZURE_OPENAI_ENDPOINT = "";
-    process.env.AZURE_OPENAI_API_KEY = "";
-    process.env.AZURE_OPENAI_DEPLOYMENT = "";
-    process.env.AZURE_OPENAI_API_VERSION = "";
-
-    // Prevent .env TIER_* values from being picked up by dotenv
-    process.env.TIER_SIMPLE = "";
-    process.env.TIER_MEDIUM = "";
-    process.env.TIER_COMPLEX = "";
-    process.env.TIER_REASONING = "";
-  });
-
-  afterEach(() => {
-    // Restore original environment
-    process.env = originalConfig;
-  });
-
-  describe("Configuration Loading", () => {
-    it("should load Azure OpenAI configuration with all values set", () => {
-      process.env.AZURE_OPENAI_ENDPOINT = "https://test-resource.openai.azure.com";
-      process.env.AZURE_OPENAI_API_KEY = "test-api-key";
-      process.env.AZURE_OPENAI_DEPLOYMENT = "gpt-4o";
-      process.env.AZURE_OPENAI_API_VERSION = "2024-08-01-preview";
-      process.env.MODEL_PROVIDER = "databricks";
-      process.env.DATABRICKS_API_KEY = "test-key";
-      process.env.DATABRICKS_API_BASE = "http://test.com";
-
-      const config = require("../src/config");
-
-      assert.strictEqual(config.azureOpenAI.endpoint, "https://test-resource.openai.azure.com");
-      assert.strictEqual(config.azureOpenAI.apiKey, "test-api-key");
-      assert.strictEqual(config.azureOpenAI.deployment, "gpt-4o");
-      assert.strictEqual(config.azureOpenAI.apiVersion, "2024-08-01-preview");
-    });
-
-    it("should use default values when optional fields not set", () => {
-      process.env.AZURE_OPENAI_ENDPOINT = "https://test-resource.openai.azure.com";
-      process.env.AZURE_OPENAI_API_KEY = "test-api-key";
-      // Keep as empty strings (don't delete) to prevent dotenv from reloading
-      process.env.AZURE_OPENAI_DEPLOYMENT = "";
-      process.env.AZURE_OPENAI_API_VERSION = "";
-      process.env.MODEL_PROVIDER = "databricks";
-      process.env.DATABRICKS_API_KEY = "test-key";
-      process.env.DATABRICKS_API_BASE = "http://test.com";
-
-      const config = require("../src/config");
-
-      assert.strictEqual(config.azureOpenAI.deployment, "gpt-4o");
-      assert.strictEqual(config.azureOpenAI.apiVersion, "2024-08-01-preview");
-    });
-
-    it("should load null values when Azure OpenAI not configured", () => {
-      // Keep as empty strings (don't delete) to prevent dotenv from reloading
-      process.env.AZURE_OPENAI_ENDPOINT = "";
-      process.env.AZURE_OPENAI_API_KEY = "";
-      process.env.MODEL_PROVIDER = "databricks";
-      process.env.DATABRICKS_API_KEY = "test-key";
-      process.env.DATABRICKS_API_BASE = "http://test.com";
-
-      const config = require("../src/config");
-
-      assert.strictEqual(config.azureOpenAI.endpoint, null);
-      assert.strictEqual(config.azureOpenAI.apiKey, null);
-      assert.strictEqual(config.azureOpenAI.deployment, "gpt-4o"); // default
-      assert.strictEqual(config.azureOpenAI.apiVersion, "2024-08-01-preview"); // default
-    });
-  });
-
-  describe("Primary Provider Validation", () => {
-    it("should accept azure-openai as MODEL_PROVIDER", () => {
-      process.env.MODEL_PROVIDER = "azure-openai";
-      process.env.AZURE_OPENAI_ENDPOINT = "https://test-resource.openai.azure.com";
-      process.env.AZURE_OPENAI_API_KEY = "test-api-key";
-
-      const config = require("../src/config");
-
-      assert.strictEqual(config.modelProvider.type, "azure-openai");
-    });
-
-    it("should throw error when azure-openai is primary provider without endpoint", () => {
-      process.env.MODEL_PROVIDER = "azure-openai";
-      process.env.AZURE_OPENAI_ENDPOINT = ""; // Empty string, not deleted
-      process.env.AZURE_OPENAI_API_KEY = "test-api-key";
-
-      assert.throws(() => {
-        require("../src/config");
-      }, /AZURE_OPENAI_ENDPOINT and AZURE_OPENAI_API_KEY/);
-    });
-
-    it("should throw error when azure-openai is primary provider without API key", () => {
-      process.env.MODEL_PROVIDER = "azure-openai";
-      process.env.AZURE_OPENAI_ENDPOINT = "https://test-resource.openai.azure.com";
-      process.env.AZURE_OPENAI_API_KEY = ""; // Empty string, not deleted
-
-      assert.throws(() => {
-        require("../src/config");
-      }, /AZURE_OPENAI_ENDPOINT and AZURE_OPENAI_API_KEY/);
-    });
-
-    it("should throw error when azure-openai is primary provider without both", () => {
-      process.env.MODEL_PROVIDER = "azure-openai";
-      process.env.AZURE_OPENAI_ENDPOINT = ""; // Empty string, not deleted
-      process.env.AZURE_OPENAI_API_KEY = ""; // Empty string, not deleted
-
-      assert.throws(() => {
-        require("../src/config");
-      }, /AZURE_OPENAI_ENDPOINT and AZURE_OPENAI_API_KEY/);
-    });
-  });
-
-  describe("Fallback Provider Validation", () => {
-    it("should accept azure-openai as fallback provider with credentials", () => {
-      process.env.MODEL_PROVIDER = "ollama";
-      process.env.OLLAMA_ENDPOINT = "http://localhost:11434";
-      process.env.OLLAMA_MODEL = "qwen2.5-coder:latest";
-      process.env.FALLBACK_ENABLED = "true";
-      process.env.FALLBACK_PROVIDER = "azure-openai";
-      process.env.AZURE_OPENAI_ENDPOINT = "https://test-resource.openai.azure.com";
-      process.env.AZURE_OPENAI_API_KEY = "test-api-key";
-      process.env.DATABRICKS_API_KEY = "test-key";
-      process.env.DATABRICKS_API_BASE = "http://test.com";
-
-      const config = require("../src/config");
-
-      assert.strictEqual(config.modelProvider.fallbackProvider, "azure-openai");
-    });
-
-    it("should warn when azure-openai is fallback but credentials missing", () => {
-      process.env.MODEL_PROVIDER = "ollama";
-      process.env.OLLAMA_ENDPOINT = "http://localhost:11434";
-      process.env.OLLAMA_MODEL = "qwen2.5-coder:latest";
-      process.env.FALLBACK_ENABLED = "true";
-      process.env.FALLBACK_PROVIDER = "azure-openai";
-      // Set to empty strings instead of deleting (dotenv.config() in config module would reload from .env)
-      process.env.AZURE_OPENAI_ENDPOINT = "";
-      process.env.AZURE_OPENAI_API_KEY = "";
-      process.env.DATABRICKS_API_KEY = "test-key";
-      process.env.DATABRICKS_API_BASE = "http://test.com";
-      // Enable tier routing so fallback validation runs
-      process.env.TIER_SIMPLE = "ollama:llama3.2";
-      process.env.TIER_MEDIUM = "ollama:llama3.2";
-      process.env.TIER_COMPLEX = "ollama:llama3.2";
-      process.env.TIER_REASONING = "ollama:llama3.2";
-
-      // Should warn but not throw (fallback misconfigured warning)
-      const config = require("../src/config");
-      assert.strictEqual(config.modelProvider.fallbackProvider, "azure-openai");
-    });
-  });
-
-  describe("Deployment and API Version Defaults", () => {
-    it("should use gpt-4o as default deployment", () => {
-      // Keep as empty string (don't delete) to prevent dotenv from reloading from .env
-      process.env.AZURE_OPENAI_DEPLOYMENT = "";
-      process.env.MODEL_PROVIDER = "azure-openai";
-      process.env.AZURE_OPENAI_ENDPOINT = "https://test-resource.openai.azure.com";
-      process.env.AZURE_OPENAI_API_KEY = "test-api-key";
-
-      const config = require("../src/config");
-
-      assert.strictEqual(config.azureOpenAI.deployment, "gpt-4o");
-    });
-
-    it("should use custom deployment when specified", () => {
-      process.env.AZURE_OPENAI_DEPLOYMENT = "gpt-5";
-      process.env.MODEL_PROVIDER = "databricks";
-      process.env.DATABRICKS_API_KEY = "test-key";
-      process.env.DATABRICKS_API_BASE = "http://test.com";
-
-      const config = require("../src/config");
-
-      assert.strictEqual(config.azureOpenAI.deployment, "gpt-5");
-    });
-
-    it("should use 2024-08-01-preview as default API version", () => {
-      // Keep as empty string (don't delete) to prevent dotenv from reloading from .env
-      process.env.AZURE_OPENAI_API_VERSION = "";
-      process.env.MODEL_PROVIDER = "databricks";
-      process.env.DATABRICKS_API_KEY = "test-key";
-      process.env.DATABRICKS_API_BASE = "http://test.com";
-
-      const config = require("../src/config");
-
-      assert.strictEqual(config.azureOpenAI.apiVersion, "2024-08-01-preview");
-    });
-
-    it("should use custom API version when specified", () => {
-      process.env.AZURE_OPENAI_API_VERSION = "2025-01-01-preview";
-      process.env.MODEL_PROVIDER = "databricks";
-      process.env.DATABRICKS_API_KEY = "test-key";
-      process.env.DATABRICKS_API_BASE = "http://test.com";
-
-      const config = require("../src/config");
-
-      assert.strictEqual(config.azureOpenAI.apiVersion, "2025-01-01-preview");
-    });
-  });
-});

package/test/azure-openai-error-resilience.test.js

@@ -1,238 +0,0 @@
-const assert = require("assert");
-const { describe, it } = require("node:test");
-
-describe("Azure OpenAI Error Resilience Tests", () => {
-  describe("Error Response Structure", () => {
-    it("should recognize 401 authentication error", () => {
-      const errorResponse = {
-        status: 401,
-        json: {
-          error: {
-            message: "Incorrect API key provided",
-            type: "invalid_request_error",
-            code: "invalid_api_key"
-          }
-        }
-      };
-
-      assert.strictEqual(errorResponse.status, 401);
-      assert.strictEqual(errorResponse.json.error.code, "invalid_api_key");
-    });
-
-    it("should recognize 403 permission denied error", () => {
-      const errorResponse = {
-        status: 403,
-        json: {
-          error: {
-            message: "The API deployment for this resource does not exist",
-            type: "invalid_request_error",
-            code: "DeploymentNotFound"
-          }
-        }
-      };
-
-      assert.strictEqual(errorResponse.status, 403);
-    });
-
-    it("should recognize 404 deployment not found error", () => {
-      const errorResponse = {
-        status: 404,
-        json: {
-          error: {
-            message: "The API deployment for this resource does not exist",
-            type: "invalid_request_error",
-            code: "DeploymentNotFound"
-          }
-        }
-      };
-
-      assert.strictEqual(errorResponse.status, 404);
-      assert.strictEqual(errorResponse.json.error.code, "DeploymentNotFound");
-    });
-
-    it("should recognize 429 rate limit error with Retry-After header", () => {
-      const errorResponse = {
-        status: 429,
-        headers: {
-          "retry-after": "2",
-          "x-ratelimit-remaining-tokens": "0",
-          "x-ratelimit-remaining-requests": "0"
-        },
-        json: {
-          error: {
-            message: "Rate limit reached",
-            type: "rate_limit_error",
-            code: "rate_limit_exceeded"
-          }
-        }
-      };
-
-      assert.strictEqual(errorResponse.status, 429);
-      assert.strictEqual(errorResponse.headers["retry-after"], "2");
-      assert.strictEqual(errorResponse.json.error.code, "rate_limit_exceeded");
-    });
-
-    it("should recognize 400 content filter error", () => {
-      const errorResponse = {
-        status: 400,
-        json: {
-          error: {
-            message: "The response was filtered due to the prompt triggering Azure OpenAI's content management policy",
-            type: "invalid_request_error",
-            code: "content_filter"
-          }
-        }
-      };
-
-      assert.strictEqual(errorResponse.status, 400);
-      assert.strictEqual(errorResponse.json.error.code, "content_filter");
-    });
-
-    it("should recognize 500 internal server error", () => {
-      const errorResponse = {
-        status: 500,
-        json: {
-          error: {
-            message: "The server had an error while processing your request",
-            type: "server_error",
-            code: "internal_error"
-          }
-        }
-      };
-
-      assert.strictEqual(errorResponse.status, 500);
-      assert.strictEqual(errorResponse.json.error.type, "server_error");
-    });
-
-    it("should recognize 503 service unavailable error", () => {
-      const errorResponse = {
-        status: 503,
-        json: {
-          error: {
-            message: "The service is temporarily unavailable",
-            type: "server_error",
-            code: "service_unavailable"
-          }
-        }
-      };
-
-      assert.strictEqual(errorResponse.status, 503);
-    });
-  });
-
-  describe("Missing Choices Array Validation", () => {
-    it("should detect missing choices array", () => {
-      const invalidResponse = {
-        id: "chatcmpl-123",
-        object: "chat.completion",
-        model: "gpt-4o"
-        // choices array missing
-      };
-
-      assert.strictEqual(invalidResponse.choices, undefined);
-    });
-
-    it("should detect empty choices array", () => {
-      const invalidResponse = {
-        id: "chatcmpl-123",
-        object: "chat.completion",
-        model: "gpt-4o",
-        choices: []
-      };
-
-      assert.strictEqual(invalidResponse.choices.length, 0);
-    });
-
-    it("should validate valid choices array", () => {
-      const validResponse = {
-        choices: [
-          {
-            message: {
-              role: "assistant",
-              content: "Hello"
-            },
-            finish_reason: "stop"
-          }
-        ]
-      };
-
-      assert.ok(validResponse.choices?.length > 0);
-      assert.ok(validResponse.choices[0].message);
-    });
-  });
-
-  describe("Network Error Categorization", () => {
-    it("should categorize ETIMEDOUT as timeout error", () => {
-      const error = new Error("Request timeout");
-      error.code = "ETIMEDOUT";
-
-      assert.strictEqual(error.code, "ETIMEDOUT");
-    });
-
-    it("should categorize ECONNREFUSED as connection error", () => {
-      const error = new Error("Connection refused");
-      error.code = "ECONNREFUSED";
-
-      assert.strictEqual(error.code, "ECONNREFUSED");
-    });
-
-    it("should categorize ENOTFOUND as DNS error", () => {
-      const error = new Error("DNS lookup failed");
-      error.code = "ENOTFOUND";
-
-      assert.strictEqual(error.code, "ENOTFOUND");
-    });
-  });
-
-  describe("Retry Strategy", () => {
-    it("should identify retryable 5xx errors", () => {
-      const retryableStatuses = [500, 502, 503, 504];
-
-      for (const status of retryableStatuses) {
-        assert.ok(status >= 500 && status < 600);
-      }
-    });
-
-    it("should identify non-retryable 4xx errors", () => {
-      const nonRetryableStatuses = [400, 401, 403, 404];
-
-      for (const status of nonRetryableStatuses) {
-        assert.ok(status >= 400 && status < 500);
-        assert.notStrictEqual(status, 429); // 429 is retryable
-      }
-    });
-
-    it("should treat 429 as retryable with backoff", () => {
-      const status = 429;
-
-      assert.strictEqual(status, 429);
-      assert.ok(status === 429); // Special handling for rate limits
-    });
-  });
-
-  describe("Malformed JSON Response", () => {
-    it("should handle truncated JSON response", () => {
-      const truncatedJSON = '{"choices":[{"message":{"role":"assistant","content":"Hello';
-
-      assert.throws(() => {
-        JSON.parse(truncatedJSON);
-      }, SyntaxError);
-    });
-
-    it("should handle empty response body", () => {
-      const emptyBody = "";
-
-      assert.throws(() => {
-        JSON.parse(emptyBody);
-      }, SyntaxError);
-    });
-
-    it("should handle non-JSON response", () => {
-      const htmlError = "<html><body>Error 503</body></html>";
-
-      assert.throws(() => {
-        JSON.parse(htmlError);
-      }, SyntaxError);
-    });
-  });
-});