@crawlee/core 4.0.0-beta.9 → 4.0.0-beta.91

package/events/local_event_manager.d.ts

@@ -1,7 +1,19 @@
-import { EventManager } from './event_manager.js';
+import type { Configuration } from '../configuration.js';
+import { EventManager, type EventManagerOptions } from './event_manager.js';
+export interface LocalEventManagerOptions extends EventManagerOptions {
+    /** Interval between emitted `systemInfo` events in milliseconds. */
+    systemInfoIntervalMillis: number;
+}
 export declare class LocalEventManager extends EventManager {
+    private systemInfoIntervalMillis;
+    constructor(options: LocalEventManagerOptions);
+    /**
+     * Creates a new `LocalEventManager` based on the provided `Configuration`.
+     * Uses the global configuration from the service locator if none is provided.
+     */
+    static fromConfiguration(configuration?: Configuration): LocalEventManager;
     /**
-     * Initializes the EventManager and sets up periodic `systemInfo` and `persistState` events.
+     * Initializes the EventManager and sets up periodic `systemInfo` events.
      * This is automatically called at the beginning of `crawler.run()`.
      */
     init(): Promise<void>;
@@ -23,5 +35,5 @@ export declare class LocalEventManager extends EventManager {
     private createSystemInfo;
     private createCpuInfo;
     private createMemoryInfo;
+    private getMemoryInfo;
 }
-//# sourceMappingURL=local_event_manager.d.ts.map

package/events/local_event_manager.js

@@ -1,10 +1,25 @@
-import { getCurrentCpuTicksV2, getMemoryInfo, isContainerized } from '@crawlee/utils';
-import log from '@apify/log';
 import { betterClearInterval, betterSetInterval } from '@apify/utilities';
+import { serviceLocator } from '../service_locator.js';
 import { EventManager } from './event_manager.js';
 export class LocalEventManager extends EventManager {
+    systemInfoIntervalMillis;
+    constructor(options) {
+        super(options);
+        this.systemInfoIntervalMillis = options.systemInfoIntervalMillis;
+    }
+    /**
+     * Creates a new `LocalEventManager` based on the provided `Configuration`.
+     * Uses the global configuration from the service locator if none is provided.
+     */
+    static fromConfiguration(configuration) {
+        const resolvedConfiguration = configuration ?? serviceLocator.getConfiguration();
+        return new LocalEventManager({
+            persistStateIntervalMillis: resolvedConfiguration.persistStateIntervalMillis,
+            systemInfoIntervalMillis: resolvedConfiguration.systemInfoIntervalMillis,
+        });
+    }
     /**
-     * Initializes the EventManager and sets up periodic `systemInfo` and `persistState` events.
+     * Initializes the EventManager and sets up periodic `systemInfo` events.
      * This is automatically called at the beginning of `crawler.run()`.
      */
     async init() {
@@ -12,9 +27,8 @@ export class LocalEventManager extends EventManager {
             return;
         }
         await super.init();
-        const systemInfoIntervalMillis = this.config.get('systemInfoIntervalMillis');
         this.emitSystemInfoEvent = this.emitSystemInfoEvent.bind(this);
-        this.intervals.systemInfo = betterSetInterval(this.emitSystemInfoEvent.bind(this), systemInfoIntervalMillis);
+        this.intervals.systemInfo = betterSetInterval(this.emitSystemInfoEvent.bind(this), this.systemInfoIntervalMillis);
     }
     /**
      * @inheritDoc
@@ -31,7 +45,7 @@ export class LocalEventManager extends EventManager {
      */
     async emitSystemInfoEvent(intervalCallback) {
         const info = await this.createSystemInfo({
-            maxUsedCpuRatio: this.config.get('maxUsedCpuRatio'),
+            maxUsedCpuRatio: serviceLocator.getConfiguration().maxUsedCpuRatio,
         });
         this.events.emit("systemInfo" /* EventType.SYSTEM_INFO */, info);
         intervalCallback();
@@ -40,7 +54,8 @@ export class LocalEventManager extends EventManager {
      * @internal
      */
     async isContainerizedWrapper() {
-        return this.config.get('containerized', await isContainerized());
+        const { isContainerized } = await import('../system-info/runtime.js');
+        return serviceLocator.getConfiguration().containerized ?? (await isContainerized());
     }
     /**
      * Creates a SystemInfo object based on local metrics.
@@ -53,7 +68,11 @@ export class LocalEventManager extends EventManager {
         };
     }
     async createCpuInfo(options) {
-        const usedCpuRatio = await getCurrentCpuTicksV2(await this.isContainerizedWrapper());
+        const { getCurrentCpuTicksV2 } = await import('../system-info/cpu-info.js');
+        const usedCpuRatio = await getCurrentCpuTicksV2({
+            containerized: await this.isContainerizedWrapper(),
+            logger: serviceLocator.getLogger(),
+        });
         return {
             cpuCurrentUsage: usedCpuRatio * 100,
             isCpuOverloaded: usedCpuRatio > options.maxUsedCpuRatio,
@@ -61,15 +80,22 @@ export class LocalEventManager extends EventManager {
     }
     async createMemoryInfo() {
         try {
-            const memInfo = await getMemoryInfo(await this.isContainerizedWrapper());
+            const memInfo = await this.getMemoryInfo();
             return {
+                memTotalBytes: memInfo.totalBytes,
                 memCurrentBytes: memInfo.mainProcessBytes + memInfo.childProcessesBytes,
             };
         }
         catch (err) {
-            log.exception(err, 'Memory snapshot failed.');
+            this.log.exception(err, 'Memory snapshot failed.');
             return {};
         }
     }
+    async getMemoryInfo() {
+        const { getMemoryInfo } = await import('../system-info/memory-info.js');
+        return getMemoryInfo({
+            containerized: await this.isContainerizedWrapper(),
+            logger: serviceLocator.getLogger(),
+        });
+    }
 }
-//# sourceMappingURL=local_event_manager.js.map

package/index.d.ts

@@ -1,19 +1,22 @@
+export * from './debug.js';
 export * from './errors.js';
 export * from './autoscaling/index.js';
 export * from './configuration.js';
+export * from './service_locator.js';
 export * from './crawlers/index.js';
 export * from './enqueue_links/index.js';
 export * from './events/index.js';
-export * from './http_clients/index.js';
 export * from './log.js';
+export * from './owned_or_injected.js';
 export * from './proxy_configuration.js';
 export * from './request.js';
 export * from './router.js';
 export * from './serialization.js';
 export * from './session_pool/index.js';
 export * from './storages/index.js';
+export * from './memory-storage/index.js';
 export * from './validators.js';
 export * from './cookie_utils.js';
+export * from './recoverable_state.js';
 export { PseudoUrl } from '@apify/pseudo_url';
-export { Dictionary, Awaitable, Constructor, StorageClient, Cookie, QueueOperationInfo } from '@crawlee/types';
-//# sourceMappingURL=index.d.ts.map
+export type { StorageBackend } from '@crawlee/types';

package/index.js

@@ -1,18 +1,21 @@
+export * from './debug.js';
 export * from './errors.js';
 export * from './autoscaling/index.js';
 export * from './configuration.js';
+export * from './service_locator.js';
 export * from './crawlers/index.js';
 export * from './enqueue_links/index.js';
 export * from './events/index.js';
-export * from './http_clients/index.js';
 export * from './log.js';
+export * from './owned_or_injected.js';
 export * from './proxy_configuration.js';
 export * from './request.js';
 export * from './router.js';
 export * from './serialization.js';
 export * from './session_pool/index.js';
 export * from './storages/index.js';
+export * from './memory-storage/index.js';
 export * from './validators.js';
 export * from './cookie_utils.js';
+export * from './recoverable_state.js';
 export { PseudoUrl } from '@apify/pseudo_url';
-//# sourceMappingURL=index.js.map

package/log.d.ts

@@ -1,3 +1,82 @@
-import log, { Log, Logger, LoggerJson, LoggerOptions, LoggerText, LogLevel } from '@apify/log';
-export { log, Log, LoggerOptions, LogLevel, Logger, LoggerJson, LoggerText };
-//# sourceMappingURL=log.d.ts.map
+import type { CrawleeLogger, CrawleeLoggerOptions } from '@crawlee/types';
+import type { LoggerOptions } from '@apify/log';
+import log, { Log, Logger, LoggerJson, LoggerText, LogLevel } from '@apify/log';
+export type { CrawleeLogger, CrawleeLoggerOptions };
+/**
+ * Abstract base class for custom Crawlee logger implementations.
+ *
+ * Subclasses must implement two methods:
+ * - {@link BaseCrawleeLogger.logWithLevel} — the core logging dispatch
+ * - {@link BaseCrawleeLogger.createChild} — how to create a child logger instance
+ *
+ * All other `CrawleeLogger` methods (`error`, `warning`, `info`, `debug`, etc.)
+ * are derived automatically. Level filtering is entirely the responsibility of the
+ * underlying library — `logWithLevel()` is called for every message.
+ *
+ * **Example — Winston adapter:**
+ * ```typescript
+ * const CRAWLEE_TO_WINSTON = { 1: 'error', 2: 'warn', 3: 'warn', 4: 'info', 5: 'debug', 6: 'debug' };
+ *
+ * class WinstonAdapter extends BaseCrawleeLogger {
+ *     constructor(private logger: winston.Logger, options?: Partial<CrawleeLoggerOptions>) {
+ *         super(options);
+ *     }
+ *
+ *     logWithLevel(level: number, message: string, data?: Record<string, unknown>): void {
+ *         this.logger.log(CRAWLEE_TO_WINSTON[level] ?? 'info', message, data);
+ *     }
+ *
+ *     protected createChild(options: Partial<CrawleeLoggerOptions>): CrawleeLogger {
+ *         return new WinstonAdapter(this.logger.child({ prefix: options.prefix }), { ...this.getOptions(), ...options });
+ *     }
+ * }
+ * ```
+ */
+export declare abstract class BaseCrawleeLogger implements CrawleeLogger {
+    private options;
+    private readonly warningsLogged;
+    constructor(options?: Partial<CrawleeLoggerOptions>);
+    /**
+     * Core logging method. Subclasses must implement this to dispatch log messages
+     * to the underlying logger (Winston, Pino, console, etc.).
+     *
+     * Level filtering is the responsibility of the underlying library — this method
+     * is called for every message regardless of the current level.
+     *
+     * @param level Crawlee log level (use {@link LogLevel} constants)
+     * @param message The log message
+     * @param data Optional structured data to attach to the log entry
+     */
+    abstract logWithLevel(level: number, message: string, data?: Record<string, unknown>): void;
+    /**
+     * Creates a child logger instance. Subclasses must implement this to define
+     * how child loggers are created for the underlying logger.
+     */
+    protected abstract createChild(options: Partial<CrawleeLoggerOptions>): CrawleeLogger;
+    getOptions(): CrawleeLoggerOptions;
+    setOptions(options: Partial<CrawleeLoggerOptions>): void;
+    child(options: Partial<CrawleeLoggerOptions>): CrawleeLogger;
+    error(message: string, data?: Record<string, unknown>): void;
+    exception(exception: Error, message: string, data?: Record<string, unknown>): void;
+    softFail(message: string, data?: Record<string, unknown>): void;
+    warning(message: string, data?: Record<string, unknown>): void;
+    warningOnce(message: string): void;
+    info(message: string, data?: Record<string, unknown>): void;
+    debug(message: string, data?: Record<string, unknown>): void;
+    perf(message: string, data?: Record<string, unknown>): void;
+    deprecated(message: string): void;
+}
+/**
+ * Adapter that wraps `@apify/log`'s {@link Log} instance to implement the {@link CrawleeLogger} interface.
+ *
+ * This is the default logger used by Crawlee when no custom logger is configured.
+ * Users who want to use a different logging library should implement {@link BaseCrawleeLogger} directly.
+ */
+export declare class ApifyLogAdapter extends BaseCrawleeLogger {
+    private readonly apifyLog;
+    constructor(apifyLog: Log, options?: Partial<CrawleeLoggerOptions>);
+    logWithLevel(level: number, message: string, data?: Record<string, unknown>): void;
+    protected createChild(options: Partial<CrawleeLoggerOptions>): CrawleeLogger;
+}
+export { log, Log, LogLevel, Logger, LoggerJson, LoggerText };
+export type { LoggerOptions };

package/log.js

@@ -1,3 +1,104 @@
 import log, { Log, Logger, LoggerJson, LoggerText, LogLevel } from '@apify/log';
+/**
+ * Abstract base class for custom Crawlee logger implementations.
+ *
+ * Subclasses must implement two methods:
+ * - {@link BaseCrawleeLogger.logWithLevel} — the core logging dispatch
+ * - {@link BaseCrawleeLogger.createChild} — how to create a child logger instance
+ *
+ * All other `CrawleeLogger` methods (`error`, `warning`, `info`, `debug`, etc.)
+ * are derived automatically. Level filtering is entirely the responsibility of the
+ * underlying library — `logWithLevel()` is called for every message.
+ *
+ * **Example — Winston adapter:**
+ * ```typescript
+ * const CRAWLEE_TO_WINSTON = { 1: 'error', 2: 'warn', 3: 'warn', 4: 'info', 5: 'debug', 6: 'debug' };
+ *
+ * class WinstonAdapter extends BaseCrawleeLogger {
+ *     constructor(private logger: winston.Logger, options?: Partial<CrawleeLoggerOptions>) {
+ *         super(options);
+ *     }
+ *
+ *     logWithLevel(level: number, message: string, data?: Record<string, unknown>): void {
+ *         this.logger.log(CRAWLEE_TO_WINSTON[level] ?? 'info', message, data);
+ *     }
+ *
+ *     protected createChild(options: Partial<CrawleeLoggerOptions>): CrawleeLogger {
+ *         return new WinstonAdapter(this.logger.child({ prefix: options.prefix }), { ...this.getOptions(), ...options });
+ *     }
+ * }
+ * ```
+ */
+export class BaseCrawleeLogger {
+    options;
+    warningsLogged = new Set();
+    constructor(options = {}) {
+        this.options = options;
+    }
+    getOptions() {
+        return this.options;
+    }
+    setOptions(options) {
+        this.options = { ...this.options, ...options };
+    }
+    child(options) {
+        return this.createChild(options);
+    }
+    error(message, data) {
+        this.logWithLevel(LogLevel.ERROR, message, data);
+    }
+    exception(exception, message, data) {
+        this.logWithLevel(LogLevel.ERROR, `${message}: ${exception.message}`, {
+            ...data,
+            stack: exception.stack,
+            exception,
+        });
+    }
+    softFail(message, data) {
+        this.logWithLevel(LogLevel.SOFT_FAIL, message, data);
+    }
+    warning(message, data) {
+        this.logWithLevel(LogLevel.WARNING, message, data);
+    }
+    warningOnce(message) {
+        if (!this.warningsLogged.has(message)) {
+            this.warningsLogged.add(message);
+            this.warning(message);
+        }
+    }
+    info(message, data) {
+        this.logWithLevel(LogLevel.INFO, message, data);
+    }
+    debug(message, data) {
+        this.logWithLevel(LogLevel.DEBUG, message, data);
+    }
+    perf(message, data) {
+        this.logWithLevel(LogLevel.PERF, `[PERF] ${message}`, data);
+    }
+    deprecated(message) {
+        this.warningOnce(`[DEPRECATED] ${message}`);
+    }
+}
+/**
+ * Adapter that wraps `@apify/log`'s {@link Log} instance to implement the {@link CrawleeLogger} interface.
+ *
+ * This is the default logger used by Crawlee when no custom logger is configured.
+ * Users who want to use a different logging library should implement {@link BaseCrawleeLogger} directly.
+ */
+export class ApifyLogAdapter extends BaseCrawleeLogger {
+    apifyLog;
+    constructor(apifyLog, options) {
+        super(options ?? {});
+        this.apifyLog = apifyLog;
+    }
+    logWithLevel(level, message, data) {
+        this.apifyLog.internal(level, message, data);
+    }
+    createChild(options) {
+        return new ApifyLogAdapter(this.apifyLog.child({ prefix: options.prefix ?? null }), {
+            ...this.getOptions(),
+            ...options,
+        });
+    }
+}
 export { log, Log, LogLevel, Logger, LoggerJson, LoggerText };
-//# sourceMappingURL=log.js.map

package/memory-storage/consts.d.ts

@@ -0,0 +1,4 @@
+/**
+ * Length of id property of a Request instance in characters.
+ */
+export declare const REQUEST_ID_LENGTH = 15;

package/memory-storage/consts.js

@@ -0,0 +1,4 @@
+/**
+ * Length of id property of a Request instance in characters.
+ */
+export const REQUEST_ID_LENGTH = 15;

package/memory-storage/index.d.ts

@@ -0,0 +1 @@
+export * from './memory-storage.js';

package/memory-storage/index.js

@@ -0,0 +1 @@
+export * from './memory-storage.js';

package/memory-storage/memory-storage.d.ts

@@ -0,0 +1,46 @@
+import type * as storage from '@crawlee/types';
+import type { CrawleeLogger } from '@crawlee/types';
+import { DatasetBackend } from './resource-clients/dataset.js';
+import { KeyValueStoreBackend } from './resource-clients/key-value-store.js';
+import { RequestQueueBackend } from './resource-clients/request-queue.js';
+export interface MemoryStorageOptions {
+    /**
+     * Optional logger for MemoryStorageBackend warnings.
+     */
+    logger?: CrawleeLogger;
+}
+export declare class MemoryStorageBackend implements storage.StorageBackend {
+    readonly logger?: CrawleeLogger;
+    /**
+     * Unique per-instance cache partition key. Mirrors the way `FileSystemStorageBackend` partitions its
+     * cache by storage directory: two distinct `MemoryStorageBackend` instances must not share cached backends.
+     */
+    private readonly instanceCacheKey;
+    readonly keyValueStoreBackendCache: KeyValueStoreBackend[];
+    readonly datasetBackendCache: DatasetBackend[];
+    readonly requestQueueBackendCache: RequestQueueBackend[];
+    constructor(options?: MemoryStorageOptions);
+    /**
+     * Return a per-instance unique cache key so that distinct `MemoryStorageBackend` instances get separate
+     * cache partitions in the storage backend cache.
+     */
+    getStorageBackendCacheKey(): string;
+    private static resolveStorageKey;
+    createDatasetBackend(options?: storage.StorageIdentifier): Promise<storage.DatasetBackend>;
+    createKeyValueStoreBackend(options?: storage.StorageIdentifier): Promise<storage.KeyValueStoreBackend>;
+    createRequestQueueBackend(options?: storage.StorageIdentifier): Promise<RequestQueueBackend>;
+    storageExists(id: string, type: 'Dataset' | 'KeyValueStore' | 'RequestQueue'): Promise<boolean>;
+    /**
+     * Cleans up the default storages before the run starts. For the in-memory storage this simply
+     * resets the in-memory state of the cached default dataset, key-value store and request queue.
+     *
+     * As with `FileSystemStorageBackend`, the run's input (the `INPUT` key in the default key-value
+     * store) is preserved — only the rest of the default storages is cleared.
+     */
+    purge(): Promise<void>;
+    /**
+     * This method should be called at the end of the process. The in-memory storage holds no resources
+     * that outlive the process (no file handles, no cross-process locks), so there is nothing to do.
+     */
+    teardown(): Promise<void>;
+}

package/memory-storage/memory-storage.js

@@ -0,0 +1,136 @@
+import { randomUUID } from 'node:crypto';
+import { DatasetBackend } from './resource-clients/dataset.js';
+import { KeyValueStoreBackend } from './resource-clients/key-value-store.js';
+import { RequestQueueBackend } from './resource-clients/request-queue.js';
+export class MemoryStorageBackend {
+    logger;
+    /**
+     * Unique per-instance cache partition key. Mirrors the way `FileSystemStorageBackend` partitions its
+     * cache by storage directory: two distinct `MemoryStorageBackend` instances must not share cached backends.
+     */
+    instanceCacheKey = `MemoryStorageBackend:${randomUUID()}`;
+    keyValueStoreBackendCache = [];
+    datasetBackendCache = [];
+    requestQueueBackendCache = [];
+    constructor(options = {}) {
+        this.logger = options.logger;
+    }
+    /**
+     * Return a per-instance unique cache key so that distinct `MemoryStorageBackend` instances get separate
+     * cache partitions in the storage backend cache.
+     */
+    getStorageBackendCacheKey() {
+        return this.instanceCacheKey;
+    }
+    static resolveStorageKey(options) {
+        const isAlias = 'alias' in options && !!options.alias;
+        const rawKey = isAlias ? options.alias : (options.name ?? options.id);
+        // Normalize the internal __default__ alias to the user-facing 'default' name.
+        const cacheKey = rawKey === '__default__' ? 'default' : rawKey;
+        return { isAlias, cacheKey };
+    }
+    async createDatasetBackend(options = {}) {
+        const { isAlias, cacheKey } = MemoryStorageBackend.resolveStorageKey(options);
+        if (cacheKey) {
+            const found = this.datasetBackendCache.find((store) => store.id === cacheKey ||
+                store.name?.toLowerCase() === cacheKey.toLowerCase() ||
+                store.cacheKey.toLowerCase() === cacheKey.toLowerCase());
+            if (found) {
+                return found;
+            }
+        }
+        const newStore = new DatasetBackend({
+            name: isAlias ? undefined : cacheKey,
+            cacheKey,
+            storageBackend: this,
+        });
+        this.datasetBackendCache.push(newStore);
+        return newStore;
+    }
+    async createKeyValueStoreBackend(options = {}) {
+        const { isAlias, cacheKey } = MemoryStorageBackend.resolveStorageKey(options);
+        if (cacheKey) {
+            const found = this.keyValueStoreBackendCache.find((store) => store.id === cacheKey ||
+                store.name?.toLowerCase() === cacheKey.toLowerCase() ||
+                store.cacheKey.toLowerCase() === cacheKey.toLowerCase());
+            if (found) {
+                return found;
+            }
+        }
+        const newStore = new KeyValueStoreBackend({
+            name: isAlias ? undefined : cacheKey,
+            cacheKey,
+            storageBackend: this,
+        });
+        this.keyValueStoreBackendCache.push(newStore);
+        return newStore;
+    }
+    async createRequestQueueBackend(options = {}) {
+        const { isAlias, cacheKey } = MemoryStorageBackend.resolveStorageKey(options);
+        if (cacheKey) {
+            const found = this.requestQueueBackendCache.find((queue) => queue.id === cacheKey ||
+                queue.name?.toLowerCase() === cacheKey.toLowerCase() ||
+                queue.cacheKey.toLowerCase() === cacheKey.toLowerCase());
+            if (found) {
+                return found;
+            }
+        }
+        const newStore = new RequestQueueBackend({
+            name: isAlias ? undefined : cacheKey,
+            cacheKey,
+            storageBackend: this,
+        });
+        this.requestQueueBackendCache.push(newStore);
+        return newStore;
+    }
+    async storageExists(id, type) {
+        let backends;
+        switch (type) {
+            case 'Dataset':
+                backends = this.datasetBackendCache;
+                break;
+            case 'KeyValueStore':
+                backends = this.keyValueStoreBackendCache;
+                break;
+            case 'RequestQueue':
+                backends = this.requestQueueBackendCache;
+                break;
+            default:
+                return false;
+        }
+        // In-memory storage only knows about backends in its cache.
+        return backends.some((store) => store.id === id);
+    }
+    /**
+     * Cleans up the default storages before the run starts. For the in-memory storage this simply
+     * resets the in-memory state of the cached default dataset, key-value store and request queue.
+     *
+     * As with `FileSystemStorageBackend`, the run's input (the `INPUT` key in the default key-value
+     * store) is preserved — only the rest of the default storages is cleared.
+     */
+    async purge() {
+        // The run default is opened via `{ alias: '__default__' }`, which `resolveStorageKey`
+        // normalizes to `cacheKey === 'default'` (with `name === undefined`) — that is the clause
+        // that actually matches it. The `name === 'default'` clause additionally covers a store a user
+        // explicitly opened via `{ name: 'default' }`. (`'__default__'` never reaches `cacheKey`,
+        // as it is always normalized to `'default'` first, so it does not need to be checked here.)
+        const isDefault = (store) => store.name === 'default' || store.cacheKey === 'default';
+        const purgeDefaults = async (cache, purgeStore) => {
+            await Promise.all(cache.filter(isDefault).map(async (store) => purgeStore(store)));
+        };
+        await Promise.all([
+            // Preserve the run input (INPUT) when purging the default key-value store, matching
+            // `FileSystemStorageBackend`.
+            purgeDefaults(this.keyValueStoreBackendCache, async (store) => store.purgeExceptInput()),
+            purgeDefaults(this.datasetBackendCache, async (store) => store.purge()),
+            purgeDefaults(this.requestQueueBackendCache, async (store) => store.purge()),
+        ]);
+    }
+    /**
+     * This method should be called at the end of the process. The in-memory storage holds no resources
+     * that outlive the process (no file handles, no cross-process locks), so there is nothing to do.
+     */
+    async teardown() {
+        // Nothing to tear down for in-memory storage.
+    }
+}

package/memory-storage/resource-clients/common/base-client.d.ts

@@ -0,0 +1,4 @@
+export declare class BaseClient {
+    id: string;
+    constructor(id: string);
+}

package/memory-storage/resource-clients/common/base-client.js

@@ -0,0 +1,6 @@
+export class BaseClient {
+    id;
+    constructor(id) {
+        this.id = id;
+    }
+}

package/memory-storage/resource-clients/dataset.d.ts

@@ -0,0 +1,40 @@
+import type * as storage from '@crawlee/types';
+import type { Dictionary } from '@crawlee/types';
+import type { MemoryStorageBackend } from '../memory-storage.js';
+import { BaseClient } from './common/base-client.js';
+export interface DatasetBackendOptions {
+    id?: string;
+    name?: string;
+    /**
+     * The key used for cache lookup. When provided, takes precedence over `name` and `id`.
+     * This allows alias-opened storages to have a cache key that differs from their
+     * metadata `name` (which is `undefined` for unnamed storages).
+     */
+    cacheKey?: string;
+    storageBackend: MemoryStorageBackend;
+}
+export declare class DatasetBackend<Data extends Dictionary = Dictionary> extends BaseClient implements storage.DatasetBackend<Data> {
+    name?: string;
+    /**
+     * The key used for cache lookup. For named storages, this equals the name. For alias (unnamed)
+     * storages, this is the alias string. Falls back to id.
+     */
+    cacheKey: string;
+    createdAt: Date;
+    accessedAt: Date;
+    modifiedAt: Date;
+    itemCount: number;
+    private readonly datasetEntries;
+    private readonly storageBackend;
+    constructor(options: DatasetBackendOptions);
+    getMetadata(): Promise<storage.DatasetInfo>;
+    drop(): Promise<void>;
+    purge(): Promise<void>;
+    getData(options?: storage.DatasetBackendListOptions): Promise<storage.PaginatedList<Data>>;
+    private getDataPage;
+    pushData(items: Data[]): Promise<void>;
+    toDatasetInfo(): storage.DatasetInfo;
+    private generateLocalEntryName;
+    private getStartAndEndIndexes;
+    private updateTimestamps;
+}