Press n or j to go to the next uncovered block, b, p or k for the previous block.
| 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 285 286 287 288 289 290 291 292 293 294 295 296 297 298 299 300 301 302 303 304 305 306 307 308 309 310 311 312 313 314 315 316 317 318 319 320 321 322 323 324 325 326 327 328 329 330 331 332 333 334 335 336 337 338 339 340 341 342 343 344 345 346 347 348 349 350 351 352 353 354 355 356 357 358 359 360 361 362 363 364 365 366 367 368 369 370 371 372 373 374 375 376 377 378 379 380 381 382 383 384 385 386 387 388 389 390 391 392 393 394 395 396 397 398 399 400 401 402 403 404 405 406 407 408 409 410 411 412 413 414 415 416 417 418 419 420 421 422 423 424 425 426 427 428 429 430 431 432 433 434 435 436 437 438 439 440 441 442 443 444 445 446 447 448 449 450 451 452 453 454 455 456 457 458 459 460 461 462 463 464 465 466 467 468 469 470 471 472 473 474 475 476 477 478 479 480 481 482 483 484 485 486 487 488 489 490 491 492 493 494 495 496 497 498 499 500 501 502 503 504 505 506 507 508 509 510 511 512 513 514 515 516 517 518 519 520 521 522 523 524 525 526 527 528 529 530 531 532 533 534 535 536 537 538 539 540 541 542 543 544 545 546 547 548 549 550 551 552 553 554 555 556 557 558 559 560 561 562 563 564 565 566 567 568 569 570 571 572 573 574 575 576 577 | 6x 6x 4053x 350x 350x 1x 349x 1x 348x 348x 350x 19x 329x 6x 6x 6x 6x 6x 6x 6x 6x 6x 6x 6x 100x 100x 100x 100x 33x 33x 33x 33x 2394x 2361x 2361x 2361x 2361x 2361x 2361x 2361x 2361x 33x 33x 33x 33x 33x 33x 33x 33x 33x 254x 254x 238x 477x 475x 159x 316x 314x 2x 475x 238x 54x 54x 41x 41x 13x 215x 215x 213x 2x 2x 2x 2x 2x 4x 4x 3x 3x 3x 1x 1x 1x 1x 2x 2x 1x 1x 1x 1x 1x 1x 1x 270x 270x 267x 38x 218x 218x 3x 3x 2x 2x 1x 215x 254x 270x 54x 39x 1x 38x 6x 258x 258x 258x 258x 258x 258x 4x 4x 4x 254x 254x 254x 254x 254x 215x 215x 215x 39x 39x 258x 311x 19x 19x 19x | /**
* @fileoverview Base European Parliament API Client
*
* Shared HTTP infrastructure for all EP API sub-clients:
* LRU caching, token-bucket rate limiting, timeout/retry logic,
* and structured audit/performance instrumentation.
*
* **ISMS Policies:**
* - SC-002 (Secure Coding Standards)
* - PE-001 (Performance Standards)
* - AU-002 (Audit Logging and Monitoring)
* - DP-001 (Data Protection and GDPR Compliance)
*
* @module clients/ep/baseClient
* @see https://github.com/Hack23/ISMS-PUBLIC/blob/main/Secure_Development_Policy.md
*/
import { fetch } from 'undici';
import { LRUCache } from 'lru-cache';
import { RateLimiter } from '../../utils/rateLimiter.js';
import { withRetry, withTimeoutAndAbort, TimeoutError } from '../../utils/timeout.js';
import { performanceMonitor } from '../../utils/performance.js';
import { DEFAULT_RATE_LIMIT_PER_MINUTE, DEFAULT_API_URL, USER_AGENT } from '../../config.js';
// ─── URL Validation (SSRF prevention) ────────────────────────────────────────
/** Exact hostnames that must never be used as API endpoints. */
const BLOCKED_HOSTS_EXACT = new Set(['localhost', '0.0.0.0', '::1']);
/**
* Patterns matching private / link-local / loopback / IPv6-mapped-IPv4 address
* prefixes. All patterns are tested against the bracket-stripped hostname.
*
* IPv6-mapped IPv4 patterns use the normalised hex form that the WHATWG URL
* parser emits (e.g. `::ffff:127.0.0.1` → `::ffff:7f00:1`).
*/
const BLOCKED_HOST_PATTERNS = [
/^fe[89ab][0-9a-f]:/i, // IPv6 link-local fe80::/10
/^f[cd][0-9a-f]{2}:/i, // IPv6 unique-local fc00::/7
/^::ffff:7f/i, // IPv6-mapped 127.x.x.x loopback
/^::ffff:a[0-9a-f]{2}:/i, // IPv6-mapped 10.x.x.x RFC-1918
/^::ffff:ac1[0-9a-f]:/i, // IPv6-mapped 172.16-31.x.x RFC-1918
/^::ffff:c0a8:/i, // IPv6-mapped 192.168.x.x RFC-1918
/^::ffff:a9fe:/i, // IPv6-mapped 169.254.x.x link-local
/^127\./, // IPv4 loopback 127.0.0.0/8
/^169\.254\./, // IPv4 link-local 169.254.0.0/16
/^10\./, // RFC-1918 10.0.0.0/8
/^172\.(1[6-9]|2\d|3[01])\./, // RFC-1918 172.16.0.0/12
/^192\.168\./, // RFC-1918 192.168.0.0/16
];
/**
* Returns true when `host` is a loopback, link-local, or RFC-1918 address that
* MUST NOT be used as an API endpoint (SSRF prevention).
* @private
*/
function isBlockedHost(host: string): boolean {
return BLOCKED_HOSTS_EXACT.has(host) || BLOCKED_HOST_PATTERNS.some(p => p.test(host));
}
/**
* Validates an EP API base URL to prevent SSRF via environment variable or
* constructor argument poisoning.
*
* Enforces HTTPS-only and blocks requests to localhost, loopback addresses
* (127.0.0.0/8), and known internal IP ranges (link-local 169.254.x.x,
* RFC-1918 10.x.x.x / 172.16-31.x.x / 192.168.x.x, and their IPv6-mapped
* IPv4 equivalents), as well as IPv6 loopback/link-local/unique-local ranges.
*
* @param url - The URL string to validate
* @param label - Label used in error messages (defaults to `'EP_API_URL'`)
* @returns The validated URL string (unchanged)
* @throws {Error} If the URL uses a non-HTTPS scheme or targets an internal host
*
* @security SSRF Prevention – ISMS Policy: SC-002, NE-001
*/
export function validateApiUrl(url: string, label = 'EP_API_URL'): string {
let parsed: URL;
try {
parsed = new URL(url);
} catch {
throw new Error(`${label} is not a valid URL`);
}
if (parsed.protocol !== 'https:') {
throw new Error(`${label} must use HTTPS protocol`);
}
// In the Node.js WHATWG URL API, .hostname includes the surrounding brackets
// for IPv6 addresses (e.g. new URL('https://[::1]/').hostname === '[::1]').
// Strip them before pattern matching so both IPv4 and IPv6 patterns are
// applied to the bare address string without brackets.
// Also strip a single trailing dot so the FQDN form "localhost." is treated
// identically to "localhost" (both resolve to loopback).
const rawHost = parsed.hostname.replace(/\.$/, '');
const host = rawHost.startsWith('[') && rawHost.endsWith(']')
? rawHost.slice(1, -1)
: rawHost;
if (isBlockedHost(host)) {
throw new Error(`${label} must not point to internal or loopback addresses`);
}
return url;
}
// ─── Default configuration constants ─────────────────────────────────────────
/** Default base URL for European Parliament Open Data Portal API v2 — derived from centralized config */
export const DEFAULT_EP_API_BASE_URL = DEFAULT_API_URL;
/** Default HTTP request timeout in milliseconds (10 seconds) */
export const DEFAULT_REQUEST_TIMEOUT_MS = 10_000;
/** Whether automatic retry on transient failures is enabled by default */
export const DEFAULT_RETRY_ENABLED = true;
/** Default maximum number of retry attempts for failed requests */
export const DEFAULT_MAX_RETRIES = 2;
/** Default base retry delay in milliseconds (exponential backoff starting point) */
export const DEFAULT_RETRY_BASE_DELAY_MS = 1_000;
/** Default maximum retry delay in milliseconds (caps exponential backoff growth) */
export const DEFAULT_RETRY_MAX_DELAY_MS = 30_000;
/** Default cache time-to-live in milliseconds (15 minutes) */
export const DEFAULT_CACHE_TTL_MS = 900_000;
/** Default maximum number of entries in the LRU response cache */
export const DEFAULT_MAX_CACHE_SIZE = 500;
/** Default rate limit token bucket size (requests per interval) — derived from centralized config */
export const DEFAULT_RATE_LIMIT_TOKENS = DEFAULT_RATE_LIMIT_PER_MINUTE;
/** Default rate limit interval unit */
export const DEFAULT_RATE_LIMIT_INTERVAL = 'minute' as const;
/** Maximum allowed response body size in bytes (10 MiB, 10×1024×1024) to prevent memory exhaustion */
export const DEFAULT_MAX_RESPONSE_BYTES = 10_485_760;
// ─── Exported error class ─────────────────────────────────────────────────────
/**
* API Error thrown when European Parliament API requests fail.
*
* @example
* ```typescript
* throw new APIError('EP API request failed: Not Found', 404, { endpoint: '/meps/999999' });
* ```
* @public
*/
export class APIError extends Error {
constructor(
message: string,
public statusCode?: number,
public details?: unknown
) {
super(message);
this.name = 'APIError';
}
}
// ─── Interfaces ───────────────────────────────────────────────────────────────
/**
* JSON-LD response format from EP API.
* @internal
*/
export interface JSONLDResponse<T = Record<string, unknown>>
extends Record<string, unknown> {
data: T[];
'@context': unknown[];
}
/**
* Configuration options for EP API clients.
* @internal
*/
export interface EPClientConfig {
/** Base URL for European Parliament API. @default DEFAULT_EP_API_BASE_URL */
baseURL?: string;
/** Cache time-to-live in milliseconds. @default 900000 */
cacheTTL?: number;
/** Maximum number of entries in LRU cache. @default 500 */
maxCacheSize?: number;
/** Custom rate limiter instance. */
rateLimiter?: RateLimiter;
/** Request timeout in milliseconds. @default 10000 */
timeoutMs?: number;
/** Enable automatic retry on transient failures. @default true */
enableRetry?: boolean;
/** Maximum number of retry attempts. @default 2 */
maxRetries?: number;
/** Maximum allowed response body size in bytes. @default 10485760 (10 MiB) */
maxResponseBytes?: number;
}
/**
* Pre-built shared resources passed from a facade to its sub-clients
* so they all operate on the same cache and rate-limiter bucket.
* @internal
*/
export interface EPSharedResources {
cache: LRUCache<string, Record<string, unknown>>;
rateLimiter: RateLimiter;
baseURL: string;
timeoutMs: number;
enableRetry: boolean;
maxRetries: number;
maxResponseBytes: number;
/** Shared cache hit/miss counters so all sub-clients contribute to aggregate stats. */
cacheCounters: { hits: number; misses: number };
}
// ─── Base client ──────────────────────────────────────────────────────────────
/**
* Base class for European Parliament API sub-clients.
*
* Holds the shared HTTP machinery: LRU cache, token-bucket rate limiter,
* timeout/abort controller, and retry logic. Sub-clients extend this class
* and call the protected `get()` helper for all HTTP requests.
*
* @class BaseEPClient
*/
export class BaseEPClient {
/** LRU cache for API responses. */
protected readonly cache: LRUCache<string, Record<string, unknown>>;
/** European Parliament API base URL. */
protected readonly baseURL: string;
/** Token bucket rate limiter. */
protected readonly rateLimiter: RateLimiter;
/** Request timeout in milliseconds. */
protected readonly timeoutMs: number;
/** Enable automatic retry on transient failures. */
protected readonly enableRetry: boolean;
/** Maximum number of retry attempts. */
protected readonly maxRetries: number;
/** Maximum allowed response body size in bytes. */
protected readonly maxResponseBytes: number;
/** Shared cache hit/miss counters (shared via EPSharedResources when used as sub-client). */
private readonly cacheCounters: { hits: number; misses: number };
/**
* Resolves all EPClientConfig options to their final values with defaults applied.
* Extracted to keep constructor complexity within limits.
* @private
*/
private static resolveConfig(config: EPClientConfig): {
baseURL: string;
cacheTTL: number;
maxCacheSize: number;
rateLimiter: RateLimiter;
timeoutMs: number;
enableRetry: boolean;
maxRetries: number;
maxResponseBytes: number;
} {
const rawBaseURL = config.baseURL ?? DEFAULT_EP_API_BASE_URL;
// Validate baseURL to prevent SSRF when sub-clients are instantiated directly
validateApiUrl(rawBaseURL, 'config.baseURL');
// Ensure baseURL always ends with '/' so that relative endpoints resolve correctly
// e.g. new URL('meps', 'https://host/api/v2') would drop 'v2' without the trailing slash
const baseURL = rawBaseURL.endsWith('/') ? rawBaseURL : `${rawBaseURL}/`;
return {
baseURL,
cacheTTL: config.cacheTTL ?? DEFAULT_CACHE_TTL_MS,
maxCacheSize: config.maxCacheSize ?? DEFAULT_MAX_CACHE_SIZE,
rateLimiter:
config.rateLimiter ??
new RateLimiter({
tokensPerInterval: DEFAULT_RATE_LIMIT_TOKENS,
interval: DEFAULT_RATE_LIMIT_INTERVAL,
}),
timeoutMs: config.timeoutMs ?? DEFAULT_REQUEST_TIMEOUT_MS,
enableRetry: config.enableRetry ?? DEFAULT_RETRY_ENABLED,
maxRetries: config.maxRetries ?? DEFAULT_MAX_RETRIES,
maxResponseBytes: config.maxResponseBytes ?? DEFAULT_MAX_RESPONSE_BYTES,
};
}
/**
* Creates a BaseEPClient.
*
* When `shared` is provided the constructor reuses those pre-built resources
* instead of allocating new ones; this is the mechanism used by the facade to
* ensure all sub-clients share one cache and one rate-limiter.
*
* @param config - Client configuration (used when `shared` is absent)
* @param shared - Pre-built shared resources (passed by facade to sub-clients)
*/
constructor(config: EPClientConfig = {}, shared?: EPSharedResources) {
if (shared !== undefined) {
this.cache = shared.cache;
this.rateLimiter = shared.rateLimiter;
this.baseURL = shared.baseURL;
this.timeoutMs = shared.timeoutMs;
this.enableRetry = shared.enableRetry;
this.maxRetries = shared.maxRetries;
this.maxResponseBytes = shared.maxResponseBytes;
this.cacheCounters = shared.cacheCounters;
} else {
const cfg = BaseEPClient.resolveConfig(config);
this.baseURL = cfg.baseURL;
this.cache = new LRUCache<string, Record<string, unknown>>({
max: cfg.maxCacheSize,
ttl: cfg.cacheTTL,
allowStale: false,
updateAgeOnGet: true,
});
this.rateLimiter = cfg.rateLimiter;
this.timeoutMs = cfg.timeoutMs;
this.enableRetry = cfg.enableRetry;
this.maxRetries = cfg.maxRetries;
this.maxResponseBytes = cfg.maxResponseBytes;
this.cacheCounters = { hits: 0, misses: 0 };
}
}
// ─── HTTP infrastructure ────────────────────────────────────────────────────
/**
* Builds the full request URL from endpoint + optional params.
* @private
*/
private buildRequestUrl(
endpoint: string,
params?: Record<string, unknown>
): URL {
const url = new URL(endpoint, this.baseURL);
if (params === undefined) return url;
for (const [key, value] of Object.entries(params)) {
if (value === undefined || value === null) continue;
let str: string;
if (typeof value === 'string') {
str = value;
} else if (typeof value === 'number' || typeof value === 'boolean') {
str = String(value);
} else {
str = JSON.stringify(value);
}
url.searchParams.append(key, str);
}
return url;
}
/**
* Returns true when an error should trigger a retry.
*
* Retries on:
* - 429 Too Many Requests (rate-limited by EP API; exponential backoff applied)
* - 5xx Server Errors (transient server failures)
* - Network / unknown errors
*
* Does NOT retry on 4xx client errors (except 429).
* @private
*/
private shouldRetryRequest(error: unknown): boolean {
Iif (error instanceof TimeoutError) return false;
if (error instanceof APIError) {
const code = error.statusCode ?? 500;
return code === 429 || code >= 500;
}
return true;
}
/**
* Reads the response body as a stream, enforcing the response size cap.
* Used as a fallback when the `content-length` header is absent (e.g. chunked
* transfer encoding). Accumulates all chunks and parses the result as JSON.
* @private
*/
private async readStreamedBody<T>(response: Response): Promise<T> {
const reader = (response.body as ReadableStream<Uint8Array> | null)?.getReader();
if (!reader) {
return response.json() as Promise<T>;
}
const chunks: Uint8Array[] = [];
let totalBytes = 0;
let readerReleased = false;
try {
for (;;) {
const result = await reader.read();
if (result.done) break;
const { value } = result;
totalBytes += value.byteLength;
if (totalBytes > this.maxResponseBytes) {
reader.releaseLock();
readerReleased = true;
await (response.body as ReadableStream<Uint8Array> | null)?.cancel();
throw new APIError(
`EP API response too large: exceeds limit of ${String(this.maxResponseBytes)} bytes`,
413
);
}
chunks.push(value);
}
} finally {
if (!readerReleased) {
reader.releaseLock();
}
}
const combined = new Uint8Array(totalBytes);
let offset = 0;
for (const chunk of chunks) {
combined.set(chunk, offset);
offset += chunk.byteLength;
}
return JSON.parse(new TextDecoder().decode(combined)) as T;
}
/**
* Executes the HTTP fetch with timeout/abort support and response size guard.
* @private
*/
private async fetchWithTimeout<T>(url: URL, endpoint: string): Promise<T> {
return withTimeoutAndAbort(
async (signal) => {
const response = await fetch(url.toString(), {
headers: {
Accept: 'application/ld+json',
'User-Agent': USER_AGENT,
},
signal,
});
if (!response.ok) {
throw new APIError(
`EP API request failed: ${response.statusText}`,
response.status
);
}
// Guard against oversized responses to prevent memory exhaustion.
const contentLength = response.headers.get('content-length');
if (contentLength !== null) {
const bytes = Number.parseInt(contentLength, 10);
if (Number.isFinite(bytes) && bytes > this.maxResponseBytes) {
// Cancel/drain the body before throwing so the underlying TCP
// connection can be returned to the pool and reused.
await response.body?.cancel();
throw new APIError(
`EP API response too large: ${String(bytes)} bytes exceeds limit of ${String(this.maxResponseBytes)} bytes`,
413
);
}
return response.json() as Promise<T>;
}
// No content-length header (chunked encoding) — stream the body and
// enforce the cap incrementally so oversized responses are aborted
// before they are fully buffered in memory.
return this.readStreamedBody<T>(response);
},
this.timeoutMs,
`EP API request to ${endpoint} timed out after ${String(this.timeoutMs)}ms`
);
}
/**
* Wraps a fetch call with the configured retry policy.
* @private
*/
private async fetchWithRetry<T>(url: URL, endpoint: string): Promise<T> {
return withRetry(
() => this.fetchWithTimeout<T>(url, endpoint),
{
maxRetries: this.enableRetry ? this.maxRetries : 0,
retryDelayMs: DEFAULT_RETRY_BASE_DELAY_MS,
maxDelayMs: DEFAULT_RETRY_MAX_DELAY_MS,
shouldRetry: (error) => this.shouldRetryRequest(error),
}
);
}
/**
* Converts a caught error to a typed {@link APIError}.
* @private
*/
private toAPIError(error: unknown, endpoint: string): APIError {
if (error instanceof TimeoutError) {
return new APIError(
`EP API request to ${endpoint} timed out after ${String(this.timeoutMs)}ms`,
408
);
}
if (error instanceof APIError) return error;
return new APIError(
`EP API request error: ${error instanceof Error ? error.message : 'Unknown error'}`,
undefined,
error
);
}
/**
* Executes a cached, rate-limited GET request to the EP API.
*
* @template T - Expected response type (extends `Record<string, unknown>`)
* @param endpoint - API endpoint path (relative to `baseURL`)
* @param params - Optional query parameters
* @returns Promise resolving to the typed API response
* @throws {APIError} On HTTP errors, network failures, or parse failures
* @protected
*/
protected async get<T extends Record<string, unknown>>(
endpoint: string,
params?: Record<string, unknown>
): Promise<T> {
// Consume one rate-limit token; waits asynchronously up to 5 s before giving up
const rlResult = await this.rateLimiter.removeTokens(1);
Iif (!rlResult.allowed) {
throw new APIError(
`Rate limit exceeded. Retry after ${String(rlResult.retryAfterMs)}ms`,
429,
{ retryAfterMs: rlResult.retryAfterMs, remainingTokens: rlResult.remainingTokens }
);
}
const cacheKey = this.getCacheKey(endpoint, params);
// Cache hit path
const cacheStart = performance.now();
const cached = this.cache.get(cacheKey);
if (cached !== undefined) {
this.cacheCounters.hits++;
performanceMonitor.recordDuration('ep_api_cache_hit', performance.now() - cacheStart);
return cached as T;
}
this.cacheCounters.misses++;
const url = this.buildRequestUrl(endpoint, params);
const requestStart = performance.now();
try {
const data = await this.fetchWithRetry<T>(url, endpoint);
performanceMonitor.recordDuration('ep_api_request', performance.now() - requestStart);
this.cache.set(cacheKey, data);
return data;
} catch (error) {
performanceMonitor.recordDuration('ep_api_request_failed', performance.now() - requestStart);
throw this.toAPIError(error, endpoint);
}
}
/**
* Generates a deterministic cache key.
* @param endpoint - API endpoint path
* @param params - Optional query parameters
* @returns JSON string used as cache key
* @private
*/
private getCacheKey(
endpoint: string,
params?: Record<string, unknown>
): string {
return JSON.stringify({ endpoint, params });
}
// ─── Public cache helpers ───────────────────────────────────────────────────
/**
* Clears all entries from the LRU cache.
* @public
*/
clearCache(): void {
this.cache.clear();
}
/**
* Returns cache statistics for monitoring and debugging.
* @returns `{ size, maxSize, hitRate, hits, misses }`
* @public
*/
getCacheStats(): { size: number; maxSize: number; hitRate: number; hits: number; misses: number } {
const { hits, misses } = this.cacheCounters;
const total = hits + misses;
return {
size: this.cache.size,
maxSize: this.cache.max,
hitRate: total > 0 ? hits / total : 0,
hits,
misses,
};
}
}
|