@crawlee/core 4.0.0-beta.7 → 4.0.0-beta.71

package/storages/request_loader.d.ts

@@ -0,0 +1,97 @@
+import type { Dictionary } from '@crawlee/types';
+import type { Request } from '../request.js';
+import type { IRequestManager } from './request_manager.js';
+import type { RequestQueueOperationInfo } from './request_queue.js';
+/**
+ * An abstract interface defining a read-only stream of requests to crawl.
+ *
+ * Request loaders are used to manage and provide access to a storage of crawling requests.
+ *
+ * Key responsibilities:
+ * - Fetching the next request to be processed.
+ * - Marking requests as handled once they are no longer in progress.
+ * - Managing state information such as the total and handled request counts.
+ *
+ * ## Request lifecycle contract
+ *
+ * Every request returned by {@link IRequestLoader.fetchNextRequest} is considered **in progress**
+ * until it is passed to {@link IRequestLoader.markRequestAsHandled}. Once you fetch a request, you are
+ * obligated to eventually mark it as handled — there is no way to hand a request back to a loader
+ * (only an {@link IRequestManager} can reclaim requests for a retry). "Handled" therefore means
+ * "finished with this request", whether processing succeeded or was abandoned after exhausting retries.
+ *
+ * Honoring this contract matters for three reasons:
+ * - **Restarts and migrations:** loaders that persist their state (see {@link IRequestLoader.persistState})
+ *   treat in-progress requests as interrupted and re-serve them after a restart. A request that is fetched
+ *   but never marked handled will be crawled again.
+ * - **Termination detection:** {@link IRequestLoader.isFinished} only resolves to `true` once nothing is
+ *   in progress. Leaving a request unmarked keeps the crawler running indefinitely.
+ * - **Bookkeeping:** the handled and pending counts are derived from the set of in-progress requests, so
+ *   skipping {@link IRequestLoader.markRequestAsHandled} corrupts {@link IRequestLoader.getHandledCount}
+ *   and {@link IRequestLoader.getPendingCount}.
+ *
+ * Concrete implementations such as {@link RequestList} or {@link SitemapRequestLoader} build on this interface.
+ * The {@link IRequestManager} interface extends it with the capability to enqueue and reclaim requests.
+ */
+export interface IRequestLoader {
+    /**
+     * Returns an approximation of the total number of requests in the loader (i.e. pending + handled).
+     */
+    getTotalCount(): Promise<number>;
+    /**
+     * Returns an approximation of the number of pending requests in the loader.
+     */
+    getPendingCount(): Promise<number>;
+    /**
+     * Returns the number of requests in the loader that have been handled.
+     */
+    getHandledCount(): Promise<number>;
+    /**
+     * Returns `true` if all requests were already handled and there are no more left.
+     */
+    isFinished(): Promise<boolean>;
+    /**
+     * Resolves to `true` if the next call to {@link IRequestLoader.fetchNextRequest} function
+     * would return `null`, otherwise it resolves to `false`.
+     * Note that even if the loader is empty, there might be some pending requests currently being processed.
+     */
+    isEmpty(): Promise<boolean>;
+    /**
+     * Gets the next {@link Request} to process, or `null` if there are no more pending requests.
+     *
+     * The returned request is marked as **in progress** and remains so until it is passed to
+     * {@link IRequestLoader.markRequestAsHandled}. The caller is responsible for eventually marking
+     * every fetched request as handled; otherwise the loader never considers itself finished and the
+     * request may be re-served after a restart. See the request lifecycle contract on {@link IRequestLoader}.
+     */
+    fetchNextRequest<T extends Dictionary = Dictionary>(): Promise<Request<T> | null>;
+    /**
+     * Can be used to iterate over the loader instance in a `for await .. of` loop.
+     * Provides an alternative for the repeated use of `fetchNextRequest`.
+     */
+    [Symbol.asyncIterator](): AsyncGenerator<Request>;
+    /**
+     * Marks a request previously returned by {@link IRequestLoader.fetchNextRequest} as handled,
+     * removing it from the set of in-progress requests.
+     *
+     * Call this once you are done with the request — whether processing succeeded or was abandoned after
+     * exhausting retries. Because a loader cannot take a request back, marking it handled is the only way to
+     * signal completion; failing to do so prevents {@link IRequestLoader.isFinished} from ever resolving to
+     * `true` and skews the handled and pending counts. See the request lifecycle contract on {@link IRequestLoader}.
+     */
+    markRequestAsHandled(request: Request): Promise<RequestQueueOperationInfo | void | null>;
+    /**
+     * Persists the current state of the loader into the default {@link KeyValueStore}.
+     *
+     * Not all loaders support persistence; implementations that do not should leave this `undefined`.
+     */
+    persistState?(): Promise<void>;
+    /**
+     * Combines the loader with a request manager to support adding and reclaiming requests.
+     *
+     * @param requestManager Request manager to combine the loader with. If not provided, the default
+     *  {@link RequestQueue} is used.
+     */
+    toTandem?(requestManager?: IRequestManager): Promise<IRequestManager>;
+}
+//# sourceMappingURL=request_loader.d.ts.map

package/storages/request_loader.d.ts.map

@@ -0,0 +1 @@
+{"version":3,"file":"request_loader.d.ts","sourceRoot":"","sources":["../../src/storages/request_loader.ts"],"names":[],"mappings":"AAAA,OAAO,KAAK,EAAE,UAAU,EAAE,MAAM,gBAAgB,CAAC;AAEjD,OAAO,KAAK,EAAE,OAAO,EAAE,MAAM,eAAe,CAAC;AAC7C,OAAO,KAAK,EAAE,eAAe,EAAE,MAAM,sBAAsB,CAAC;AAC5D,OAAO,KAAK,EAAE,yBAAyB,EAAE,MAAM,oBAAoB,CAAC;AAEpE;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;GA8BG;AACH,MAAM,WAAW,cAAc;IAC3B;;OAEG;IACH,aAAa,IAAI,OAAO,CAAC,MAAM,CAAC,CAAC;IAEjC;;OAEG;IACH,eAAe,IAAI,OAAO,CAAC,MAAM,CAAC,CAAC;IAEnC;;OAEG;IACH,eAAe,IAAI,OAAO,CAAC,MAAM,CAAC,CAAC;IAEnC;;OAEG;IACH,UAAU,IAAI,OAAO,CAAC,OAAO,CAAC,CAAC;IAE/B;;;;OAIG;IACH,OAAO,IAAI,OAAO,CAAC,OAAO,CAAC,CAAC;IAE5B;;;;;;;OAOG;IACH,gBAAgB,CAAC,CAAC,SAAS,UAAU,GAAG,UAAU,KAAK,OAAO,CAAC,OAAO,CAAC,CAAC,CAAC,GAAG,IAAI,CAAC,CAAC;IAElF;;;OAGG;IACH,CAAC,MAAM,CAAC,aAAa,CAAC,IAAI,cAAc,CAAC,OAAO,CAAC,CAAC;IAElD;;;;;;;;OAQG;IACH,oBAAoB,CAAC,OAAO,EAAE,OAAO,GAAG,OAAO,CAAC,yBAAyB,GAAG,IAAI,GAAG,IAAI,CAAC,CAAC;IAEzF;;;;OAIG;IACH,YAAY,CAAC,IAAI,OAAO,CAAC,IAAI,CAAC,CAAC;IAE/B;;;;;OAKG;IACH,QAAQ,CAAC,CAAC,cAAc,CAAC,EAAE,eAAe,GAAG,OAAO,CAAC,eAAe,CAAC,CAAC;CACzE"}

package/storages/request_loader.js

@@ -0,0 +1,2 @@
+export {};
+//# sourceMappingURL=request_loader.js.map

package/storages/request_loader.js.map

@@ -0,0 +1 @@
+{"version":3,"file":"request_loader.js","sourceRoot":"","sources":["../../src/storages/request_loader.ts"],"names":[],"mappings":""}

package/storages/request_manager.d.ts

@@ -0,0 +1,34 @@
+import type { Request, Source } from '../request.js';
+import type { IRequestLoader } from './request_loader.js';
+import type { AddRequestsBatchedOptions, AddRequestsBatchedResult, RequestQueueOperationInfo, RequestQueueOperationOptions } from './request_queue.js';
+export type RequestsLike = AsyncIterable<Source | string> | Iterable<Source | string> | (Source | string)[];
+/**
+ * Extends the read-only {@link IRequestLoader} interface with the capability to enqueue new requests
+ * and reclaim failed ones.
+ */
+export interface IRequestManager extends IRequestLoader {
+    /**
+     * Reclaims request to the provider if its processing failed.
+     * The request will be returned by some subsequent `fetchNextRequest()` call.
+     */
+    reclaimRequest(request: Request, options?: RequestQueueOperationOptions): Promise<RequestQueueOperationInfo | null>;
+    addRequest(requestLike: Source, options?: RequestQueueOperationOptions): Promise<RequestQueueOperationInfo>;
+    addRequestsBatched(requests: RequestsLike, options?: AddRequestsBatchedOptions): Promise<AddRequestsBatchedResult>;
+    /**
+     * Remove all requests from the queue but keep the queue itself, resetting it
+     * so it can be reused (e.g. across multiple `crawler.run()` calls).
+     *
+     * Implementations that do not support purging may leave this `undefined`.
+     */
+    purge?(): Promise<void>;
+    /**
+     * Tells the manager how long a consumer expects to hold a request fetched via `fetchNextRequest()`
+     * before marking it handled or reclaiming it (typically the request-handler timeout plus padding).
+     *
+     * Managers backed by a storage backend that reserves requests via locking use this to avoid handing
+     * the same request out again while it is still being processed. Implementations that do not need
+     * this hint may leave it `undefined`.
+     */
+    setExpectedRequestProcessingTimeSecs?(secs: number): Promise<void>;
+}
+//# sourceMappingURL=request_manager.d.ts.map

package/storages/request_manager.d.ts.map

@@ -0,0 +1 @@
+{"version":3,"file":"request_manager.d.ts","sourceRoot":"","sources":["../../src/storages/request_manager.ts"],"names":[],"mappings":"AAAA,OAAO,KAAK,EAAE,OAAO,EAAE,MAAM,EAAE,MAAM,eAAe,CAAC;AACrD,OAAO,KAAK,EAAE,cAAc,EAAE,MAAM,qBAAqB,CAAC;AAC1D,OAAO,KAAK,EACR,yBAAyB,EACzB,wBAAwB,EACxB,yBAAyB,EACzB,4BAA4B,EAC/B,MAAM,oBAAoB,CAAC;AAE5B,MAAM,MAAM,YAAY,GAAG,aAAa,CAAC,MAAM,GAAG,MAAM,CAAC,GAAG,QAAQ,CAAC,MAAM,GAAG,MAAM,CAAC,GAAG,CAAC,MAAM,GAAG,MAAM,CAAC,EAAE,CAAC;AAE5G;;;GAGG;AACH,MAAM,WAAW,eAAgB,SAAQ,cAAc;IACnD;;;OAGG;IACH,cAAc,CAAC,OAAO,EAAE,OAAO,EAAE,OAAO,CAAC,EAAE,4BAA4B,GAAG,OAAO,CAAC,yBAAyB,GAAG,IAAI,CAAC,CAAC;IAEpH,UAAU,CAAC,WAAW,EAAE,MAAM,EAAE,OAAO,CAAC,EAAE,4BAA4B,GAAG,OAAO,CAAC,yBAAyB,CAAC,CAAC;IAE5G,kBAAkB,CAAC,QAAQ,EAAE,YAAY,EAAE,OAAO,CAAC,EAAE,yBAAyB,GAAG,OAAO,CAAC,wBAAwB,CAAC,CAAC;IAEnH;;;;;OAKG;IACH,KAAK,CAAC,IAAI,OAAO,CAAC,IAAI,CAAC,CAAC;IAExB;;;;;;;OAOG;IACH,oCAAoC,CAAC,CAAC,IAAI,EAAE,MAAM,GAAG,OAAO,CAAC,IAAI,CAAC,CAAC;CACtE"}

package/storages/request_manager.js

@@ -0,0 +1,2 @@
+export {};
+//# sourceMappingURL=request_manager.js.map

package/storages/request_manager.js.map

@@ -0,0 +1 @@
+{"version":3,"file":"request_manager.js","sourceRoot":"","sources":["../../src/storages/request_manager.ts"],"names":[],"mappings":""}

package/storages/request_manager_tandem.d.ts

@@ -0,0 +1,107 @@
+import type { Dictionary } from '@crawlee/types';
+import type { Request, Source } from '../request.js';
+import type { IRequestLoader } from './request_loader.js';
+import type { IRequestManager, RequestsLike } from './request_manager.js';
+import type { AddRequestsBatchedOptions, AddRequestsBatchedResult, RequestQueueOperationInfo, RequestQueueOperationOptions } from './request_queue.js';
+/**
+ * A request manager that combines a {@link IRequestLoader} (such as a `RequestList`) with a writable
+ * {@link IRequestManager} (such as a `RequestQueue`).
+ * It first reads requests from the loader and then, when needed, transfers them in batches to the manager.
+ */
+export declare class RequestManagerTandem implements IRequestManager {
+    private log;
+    private requestLoader;
+    private requestManagerPromise?;
+    private resolvedRequestManager?;
+    private requestManagerFactory;
+    /**
+     * The latest expected request-processing time hinted via {@link setExpectedRequestProcessingTimeSecs}.
+     * Remembered so it can be applied to the writable manager once it is lazily resolved.
+     */
+    private expectedRequestProcessingSecs?;
+    /**
+     * @param requestLoader The read-only loader to read requests from first.
+     * @param requestManager The writable manager to transfer requests into and enqueue new ones. May be passed as a
+     *  factory function so that the tandem can be constructed synchronously and the manager opened lazily on first use
+     *  (e.g. a lazily-opened default {@link RequestQueue}).
+     */
+    constructor(requestLoader: IRequestLoader, requestManager: IRequestManager | (() => IRequestManager | Promise<IRequestManager>));
+    /**
+     * Resolves the writable request manager, opening it lazily (via the factory) on first use and memoizing the result.
+     * @private
+     */
+    private getRequestManager;
+    /**
+     * Transfers a single request from the read-only loader to the writable manager.
+     * If the transfer fails, the request is dropped (and logged) rather than reclaimed.
+     *
+     * @returns `true` if a request was successfully transferred (or there was nothing to transfer), and `false` if a
+     *  transfer was attempted but failed - in which case the caller should not fetch from the manager this round.
+     * @private
+     */
+    private transferNextRequestToQueue;
+    /**
+     * Fetches the next request from the request manager. If the manager is empty and the loader
+     * is not finished, it will transfer a request from the loader to the manager first.
+     * @inheritdoc
+     */
+    fetchNextRequest<T extends Dictionary = Dictionary>(): Promise<Request<T> | null>;
+    /**
+     * @inheritdoc
+     */
+    isFinished(): Promise<boolean>;
+    /**
+     * @inheritdoc
+     */
+    isEmpty(): Promise<boolean>;
+    /**
+     * @inheritdoc
+     */
+    getHandledCount(): Promise<number>;
+    /**
+     * @inheritdoc
+     */
+    getTotalCount(): Promise<number>;
+    /**
+     * @inheritdoc
+     */
+    getPendingCount(): Promise<number>;
+    /**
+     * @inheritdoc
+     */
+    [Symbol.asyncIterator](): AsyncGenerator<Request<Dictionary>, void, unknown>;
+    /**
+     * @inheritdoc
+     */
+    markRequestAsHandled(request: Request): Promise<RequestQueueOperationInfo | void | null>;
+    /**
+     * @inheritdoc
+     */
+    reclaimRequest(request: Request, options?: RequestQueueOperationOptions): Promise<RequestQueueOperationInfo | null>;
+    /**
+     * @inheritdoc
+     */
+    addRequest(requestLike: Source, options?: RequestQueueOperationOptions): Promise<RequestQueueOperationInfo>;
+    /**
+     * @inheritdoc
+     */
+    addRequestsBatched(requests: RequestsLike, options?: AddRequestsBatchedOptions): Promise<AddRequestsBatchedResult>;
+    /**
+     * Persists the state of the underlying read-only loader, if it supports persistence.
+     * @inheritdoc
+     */
+    persistState(): Promise<void>;
+    /**
+     * Purges the writable request manager so the tandem can be reused (e.g. across repeated `crawler.run()` calls).
+     * The read-only loader is immutable and cannot be purged, so only the manager side is reset.
+     * @inheritdoc
+     */
+    purge(): Promise<void>;
+    /**
+     * Forwards the hint to the writable request manager — that is where requests are fetched from and
+     * reserved. The manager is opened lazily, so the value is remembered and applied once it resolves.
+     * @inheritdoc
+     */
+    setExpectedRequestProcessingTimeSecs(secs: number): Promise<void>;
+}
+//# sourceMappingURL=request_manager_tandem.d.ts.map

package/storages/request_manager_tandem.d.ts.map

@@ -0,0 +1 @@
+{"version":3,"file":"request_manager_tandem.d.ts","sourceRoot":"","sources":["../../src/storages/request_manager_tandem.ts"],"names":[],"mappings":"AAAA,OAAO,KAAK,EAAE,UAAU,EAAE,MAAM,gBAAgB,CAAC;AAGjD,OAAO,KAAK,EAAE,OAAO,EAAE,MAAM,EAAE,MAAM,eAAe,CAAC;AAErD,OAAO,KAAK,EAAE,cAAc,EAAE,MAAM,qBAAqB,CAAC;AAC1D,OAAO,KAAK,EAAE,eAAe,EAAE,YAAY,EAAE,MAAM,sBAAsB,CAAC;AAC1E,OAAO,KAAK,EACR,yBAAyB,EACzB,wBAAwB,EACxB,yBAAyB,EACzB,4BAA4B,EAC/B,MAAM,oBAAoB,CAAC;AAE5B;;;;GAIG;AACH,qBAAa,oBAAqB,YAAW,eAAe;IACxD,OAAO,CAAC,GAAG,CAAgB;IAC3B,OAAO,CAAC,aAAa,CAAiB;IACtC,OAAO,CAAC,qBAAqB,CAAC,CAA2B;IACzD,OAAO,CAAC,sBAAsB,CAAC,CAAkB;IAEjD,OAAO,CAAC,qBAAqB,CAAmD;IAEhF;;;OAGG;IACH,OAAO,CAAC,6BAA6B,CAAC,CAAS;IAE/C;;;;;OAKG;gBAEC,aAAa,EAAE,cAAc,EAC7B,cAAc,EAAE,eAAe,GAAG,CAAC,MAAM,eAAe,GAAG,OAAO,CAAC,eAAe,CAAC,CAAC;IAOxF;;;OAGG;YACW,iBAAiB;IAe/B;;;;;;;OAOG;YACW,0BAA0B;IAyBxC;;;;OAIG;IACG,gBAAgB,CAAC,CAAC,SAAS,UAAU,GAAG,UAAU,KAAK,OAAO,CAAC,OAAO,CAAC,CAAC,CAAC,GAAG,IAAI,CAAC;IAmBvF;;OAEG;IACG,UAAU,IAAI,OAAO,CAAC,OAAO,CAAC;IAMpC;;OAEG;IACG,OAAO,IAAI,OAAO,CAAC,OAAO,CAAC;IAMjC;;OAEG;IACG,eAAe,IAAI,OAAO,CAAC,MAAM,CAAC;IAKxC;;OAEG;IACG,aAAa,IAAI,OAAO,CAAC,MAAM,CAAC;IAUtC;;OAEG;IACG,eAAe,IAAI,OAAO,CAAC,MAAM,CAAC;IASxC;;OAEG;IACI,CAAC,MAAM,CAAC,aAAa,CAAC;IAQ7B;;OAEG;IACG,oBAAoB,CAAC,OAAO,EAAE,OAAO,GAAG,OAAO,CAAC,yBAAyB,GAAG,IAAI,GAAG,IAAI,CAAC;IAI9F;;OAEG;IACG,cAAc,CAChB,OAAO,EAAE,OAAO,EAChB,OAAO,CAAC,EAAE,4BAA4B,GACvC,OAAO,CAAC,yBAAyB,GAAG,IAAI,CAAC;IAI5C;;OAEG;IACG,UAAU,CAAC,WAAW,EAAE,MAAM,EAAE,OAAO,CAAC,EAAE,4BAA4B,GAAG,OAAO,CAAC,yBAAyB,CAAC;IAIjH;;OAEG;IACG,kBAAkB,CACpB,QAAQ,EAAE,YAAY,EACtB,OAAO,CAAC,EAAE,yBAAyB,GACpC,OAAO,CAAC,wBAAwB,CAAC;IAIpC;;;OAGG;IACG,YAAY,IAAI,OAAO,CAAC,IAAI,CAAC;IAInC;;;;OAIG;IACG,KAAK,IAAI,OAAO,CAAC,IAAI,CAAC;IAI5B;;;;OAIG;IACG,oCAAoC,CAAC,IAAI,EAAE,MAAM,GAAG,OAAO,CAAC,IAAI,CAAC;CAI1E"}

package/storages/request_manager_tandem.js

@@ -0,0 +1,198 @@
+import { serviceLocator } from '../service_locator.js';
+/**
+ * A request manager that combines a {@link IRequestLoader} (such as a `RequestList`) with a writable
+ * {@link IRequestManager} (such as a `RequestQueue`).
+ * It first reads requests from the loader and then, when needed, transfers them in batches to the manager.
+ */
+export class RequestManagerTandem {
+    log;
+    requestLoader;
+    requestManagerPromise;
+    resolvedRequestManager;
+    requestManagerFactory;
+    /**
+     * The latest expected request-processing time hinted via {@link setExpectedRequestProcessingTimeSecs}.
+     * Remembered so it can be applied to the writable manager once it is lazily resolved.
+     */
+    expectedRequestProcessingSecs;
+    /**
+     * @param requestLoader The read-only loader to read requests from first.
+     * @param requestManager The writable manager to transfer requests into and enqueue new ones. May be passed as a
+     *  factory function so that the tandem can be constructed synchronously and the manager opened lazily on first use
+     *  (e.g. a lazily-opened default {@link RequestQueue}).
+     */
+    constructor(requestLoader, requestManager) {
+        this.log = serviceLocator.getLogger().child({ prefix: 'RequestManagerTandem' });
+        this.requestLoader = requestLoader;
+        this.requestManagerFactory = typeof requestManager === 'function' ? requestManager : () => requestManager;
+    }
+    /**
+     * Resolves the writable request manager, opening it lazily (via the factory) on first use and memoizing the result.
+     * @private
+     */
+    async getRequestManager() {
+        if (this.resolvedRequestManager === undefined) {
+            this.requestManagerPromise ??= Promise.resolve(this.requestManagerFactory());
+            this.resolvedRequestManager = await this.requestManagerPromise;
+            // Apply any hint received before the manager was resolved.
+            if (this.expectedRequestProcessingSecs !== undefined) {
+                await this.resolvedRequestManager.setExpectedRequestProcessingTimeSecs?.(this.expectedRequestProcessingSecs);
+            }
+        }
+        return this.resolvedRequestManager;
+    }
+    /**
+     * Transfers a single request from the read-only loader to the writable manager.
+     * If the transfer fails, the request is dropped (and logged) rather than reclaimed.
+     *
+     * @returns `true` if a request was successfully transferred (or there was nothing to transfer), and `false` if a
+     *  transfer was attempted but failed - in which case the caller should not fetch from the manager this round.
+     * @private
+     */
+    async transferNextRequestToQueue() {
+        const request = await this.requestLoader.fetchNextRequest();
+        if (request === null) {
+            return true;
+        }
+        const requestManager = await this.getRequestManager();
+        try {
+            await requestManager.addRequest(request, { forefront: true });
+            return true;
+        }
+        catch (error) {
+            this.log.exception(error, 'Adding request from the RequestLoader to the RequestManager failed, the request has been dropped.', { url: request.url, uniqueKey: request.uniqueKey });
+            return false;
+        }
+        finally {
+            // Mark it as handled so that the request doesn't get stuck in the `inProgress` state in the loader.
+            await this.requestLoader.markRequestAsHandled(request);
+        }
+    }
+    /**
+     * Fetches the next request from the request manager. If the manager is empty and the loader
+     * is not finished, it will transfer a request from the loader to the manager first.
+     * @inheritdoc
+     */
+    async fetchNextRequest() {
+        // First, try to transfer a request from the requestList
+        const [listEmpty, listFinished] = await Promise.all([
+            this.requestLoader.isEmpty(),
+            this.requestLoader.isFinished(),
+        ]);
+        if (!listEmpty && !listFinished) {
+            // If the transfer failed, the request was dropped; don't fetch from the manager this round (matching
+            // crawlee-python behaviour). The next `fetchNextRequest()` call will pick up where we left off.
+            if (!(await this.transferNextRequestToQueue())) {
+                return null;
+            }
+        }
+        // Try to fetch from manager after the transfer
+        return (await this.getRequestManager()).fetchNextRequest();
+    }
+    /**
+     * @inheritdoc
+     */
+    async isFinished() {
+        const requestManager = await this.getRequestManager();
+        const storagesFinished = await Promise.all([this.requestLoader.isFinished(), requestManager.isFinished()]);
+        return storagesFinished.every(Boolean);
+    }
+    /**
+     * @inheritdoc
+     */
+    async isEmpty() {
+        const requestManager = await this.getRequestManager();
+        const storagesEmpty = await Promise.all([this.requestLoader.isEmpty(), requestManager.isEmpty()]);
+        return storagesEmpty.every(Boolean);
+    }
+    /**
+     * @inheritdoc
+     */
+    async getHandledCount() {
+        // Since one of the stores needs to have priority when both are present, we query the request manager - the request loader will first be dumped into the manager and then left empty.
+        return (await this.getRequestManager()).getHandledCount();
+    }
+    /**
+     * @inheritdoc
+     */
+    async getTotalCount() {
+        const requestManager = await this.getRequestManager();
+        const [managerTotal, loaderTotal] = await Promise.all([
+            requestManager.getTotalCount(),
+            // count only pending to avoid double counting, requests marked as "handled" have been moved to requestManager
+            this.requestLoader.getPendingCount(),
+        ]);
+        return managerTotal + loaderTotal;
+    }
+    /**
+     * @inheritdoc
+     */
+    async getPendingCount() {
+        const requestManager = await this.getRequestManager();
+        const [managerPending, loaderPending] = await Promise.all([
+            requestManager.getPendingCount(),
+            this.requestLoader.getPendingCount(),
+        ]);
+        return managerPending + loaderPending;
+    }
+    /**
+     * @inheritdoc
+     */
+    async *[Symbol.asyncIterator]() {
+        while (true) {
+            const req = await this.fetchNextRequest();
+            if (!req)
+                break;
+            yield req;
+        }
+    }
+    /**
+     * @inheritdoc
+     */
+    async markRequestAsHandled(request) {
+        return (await this.getRequestManager()).markRequestAsHandled(request);
+    }
+    /**
+     * @inheritdoc
+     */
+    async reclaimRequest(request, options) {
+        return (await this.getRequestManager()).reclaimRequest(request, options);
+    }
+    /**
+     * @inheritdoc
+     */
+    async addRequest(requestLike, options) {
+        return (await this.getRequestManager()).addRequest(requestLike, options);
+    }
+    /**
+     * @inheritdoc
+     */
+    async addRequestsBatched(requests, options) {
+        return (await this.getRequestManager()).addRequestsBatched(requests, options);
+    }
+    /**
+     * Persists the state of the underlying read-only loader, if it supports persistence.
+     * @inheritdoc
+     */
+    async persistState() {
+        await this.requestLoader.persistState?.();
+    }
+    /**
+     * Purges the writable request manager so the tandem can be reused (e.g. across repeated `crawler.run()` calls).
+     * The read-only loader is immutable and cannot be purged, so only the manager side is reset.
+     * @inheritdoc
+     */
+    async purge() {
+        await (await this.getRequestManager()).purge?.();
+    }
+    /**
+     * Forwards the hint to the writable request manager — that is where requests are fetched from and
+     * reserved. The manager is opened lazily, so the value is remembered and applied once it resolves.
+     * @inheritdoc
+     */
+    async setExpectedRequestProcessingTimeSecs(secs) {
+        this.expectedRequestProcessingSecs = secs;
+        await this.resolvedRequestManager?.setExpectedRequestProcessingTimeSecs?.(secs);
+    }
+}
+//# sourceMappingURL=request_manager_tandem.js.map

package/storages/request_manager_tandem.js.map

@@ -0,0 +1 @@
+{"version":3,"file":"request_manager_tandem.js","sourceRoot":"","sources":["../../src/storages/request_manager_tandem.ts"],"names":[],"mappings":"AAIA,OAAO,EAAE,cAAc,EAAE,MAAM,uBAAuB,CAAC;AAUvD;;;;GAIG;AACH,MAAM,OAAO,oBAAoB;IACrB,GAAG,CAAgB;IACnB,aAAa,CAAiB;IAC9B,qBAAqB,CAA4B;IACjD,sBAAsB,CAAmB;IAEzC,qBAAqB,CAAmD;IAEhF;;;OAGG;IACK,6BAA6B,CAAU;IAE/C;;;;;OAKG;IACH,YACI,aAA6B,EAC7B,cAAoF;QAEpF,IAAI,CAAC,GAAG,GAAG,cAAc,CAAC,SAAS,EAAE,CAAC,KAAK,CAAC,EAAE,MAAM,EAAE,sBAAsB,EAAE,CAAC,CAAC;QAChF,IAAI,CAAC,aAAa,GAAG,aAAa,CAAC;QACnC,IAAI,CAAC,qBAAqB,GAAG,OAAO,cAAc,KAAK,UAAU,CAAC,CAAC,CAAC,cAAc,CAAC,CAAC,CAAC,GAAG,EAAE,CAAC,cAAc,CAAC;IAC9G,CAAC;IAED;;;OAGG;IACK,KAAK,CAAC,iBAAiB;QAC3B,IAAI,IAAI,CAAC,sBAAsB,KAAK,SAAS,EAAE,CAAC;YAC5C,IAAI,CAAC,qBAAqB,KAAK,OAAO,CAAC,OAAO,CAAC,IAAI,CAAC,qBAAqB,EAAE,CAAC,CAAC;YAC7E,IAAI,CAAC,sBAAsB,GAAG,MAAM,IAAI,CAAC,qBAAqB,CAAC;YAE/D,2DAA2D;YAC3D,IAAI,IAAI,CAAC,6BAA6B,KAAK,SAAS,EAAE,CAAC;gBACnD,MAAM,IAAI,CAAC,sBAAsB,CAAC,oCAAoC,EAAE,CACpE,IAAI,CAAC,6BAA6B,CACrC,CAAC;YACN,CAAC;QACL,CAAC;QACD,OAAO,IAAI,CAAC,sBAAsB,CAAC;IACvC,CAAC;IAED;;;;;;;OAOG;IACK,KAAK,CAAC,0BAA0B;QACpC,MAAM,OAAO,GAAG,MAAM,IAAI,CAAC,aAAa,CAAC,gBAAgB,EAAE,CAAC;QAE5D,IAAI,OAAO,KAAK,IAAI,EAAE,CAAC;YACnB,OAAO,IAAI,CAAC;QAChB,CAAC;QAED,MAAM,cAAc,GAAG,MAAM,IAAI,CAAC,iBAAiB,EAAE,CAAC;QAEtD,IAAI,CAAC;YACD,MAAM,cAAc,CAAC,UAAU,CAAC,OAAO,EAAE,EAAE,SAAS,EAAE,IAAI,EAAE,CAAC,CAAC;YAC9D,OAAO,IAAI,CAAC;QAChB,CAAC;QAAC,OAAO,KAAK,EAAE,CAAC;YACb,IAAI,CAAC,GAAG,CAAC,SAAS,CACd,KAAc,EACd,mGAAmG,EACnG,EAAE,GAAG,EAAE,OAAO,CAAC,GAAG,EAAE,SAAS,EAAE,OAAO,CAAC,SAAS,EAAE,CACrD,CAAC;YACF,OAAO,KAAK,CAAC;QACjB,CAAC;gBAAS,CAAC;YACP,oGAAoG;YACpG,MAAM,IAAI,CAAC,aAAa,CAAC,oBAAoB,CAAC,OAAO,CAAC,CAAC;QAC3D,CAAC;IACL,CAAC;IAED;;;;OAIG;IACH,KAAK,CAAC,gBAAgB;QAClB,wDAAwD;QACxD,MAAM,CAAC,SAAS,EAAE,YAAY,CAAC,GAAG,MAAM,OAAO,CAAC,GAAG,CAAC;YAChD,IAAI,CAAC,aAAa,CAAC,OAAO,EAAE;YAC5B,IAAI,CAAC,aAAa,CAAC,UAAU,EAAE;SAClC,CAAC,CAAC;QAEH,IAAI,CAAC,SAAS,IAAI,CAAC,YAAY,EAAE,CAAC;YAC9B,qGAAqG;YACrG,gGAAgG;YAChG,IAAI,CAAC,CAAC,MAAM,IAAI,CAAC,0BAA0B,EAAE,CAAC,EAAE,CAAC;gBAC7C,OAAO,IAAI,CAAC;YAChB,CAAC;QACL,CAAC;QAED,+CAA+C;QAC/C,OAAO,CAAC,MAAM,IAAI,CAAC,iBAAiB,EAAE,CAAC,CAAC,gBAAgB,EAAK,CAAC;IAClE,CAAC;IAED;;OAEG;IACH,KAAK,CAAC,UAAU;QACZ,MAAM,cAAc,GAAG,MAAM,IAAI,CAAC,iBAAiB,EAAE,CAAC;QACtD,MAAM,gBAAgB,GAAG,MAAM,OAAO,CAAC,GAAG,CAAC,CAAC,IAAI,CAAC,aAAa,CAAC,UAAU,EAAE,EAAE,cAAc,CAAC,UAAU,EAAE,CAAC,CAAC,CAAC;QAC3G,OAAO,gBAAgB,CAAC,KAAK,CAAC,OAAO,CAAC,CAAC;IAC3C,CAAC;IAED;;OAEG;IACH,KAAK,CAAC,OAAO;QACT,MAAM,cAAc,GAAG,MAAM,IAAI,CAAC,iBAAiB,EAAE,CAAC;QACtD,MAAM,aAAa,GAAG,MAAM,OAAO,CAAC,GAAG,CAAC,CAAC,IAAI,CAAC,aAAa,CAAC,OAAO,EAAE,EAAE,cAAc,CAAC,OAAO,EAAE,CAAC,CAAC,CAAC;QAClG,OAAO,aAAa,CAAC,KAAK,CAAC,OAAO,CAAC,CAAC;IACxC,CAAC;IAED;;OAEG;IACH,KAAK,CAAC,eAAe;QACjB,qLAAqL;QACrL,OAAO,CAAC,MAAM,IAAI,CAAC,iBAAiB,EAAE,CAAC,CAAC,eAAe,EAAE,CAAC;IAC9D,CAAC;IAED;;OAEG;IACH,KAAK,CAAC,aAAa;QACf,MAAM,cAAc,GAAG,MAAM,IAAI,CAAC,iBAAiB,EAAE,CAAC;QACtD,MAAM,CAAC,YAAY,EAAE,WAAW,CAAC,GAAG,MAAM,OAAO,CAAC,GAAG,CAAC;YAClD,cAAc,CAAC,aAAa,EAAE;YAC9B,8GAA8G;YAC9G,IAAI,CAAC,aAAa,CAAC,eAAe,EAAE;SACvC,CAAC,CAAC;QACH,OAAO,YAAY,GAAG,WAAW,CAAC;IACtC,CAAC;IAED;;OAEG;IACH,KAAK,CAAC,eAAe;QACjB,MAAM,cAAc,GAAG,MAAM,IAAI,CAAC,iBAAiB,EAAE,CAAC;QACtD,MAAM,CAAC,cAAc,EAAE,aAAa,CAAC,GAAG,MAAM,OAAO,CAAC,GAAG,CAAC;YACtD,cAAc,CAAC,eAAe,EAAE;YAChC,IAAI,CAAC,aAAa,CAAC,eAAe,EAAE;SACvC,CAAC,CAAC;QACH,OAAO,cAAc,GAAG,aAAa,CAAC;IAC1C,CAAC;IAED;;OAEG;IACH,KAAK,CAAC,CAAC,CAAC,MAAM,CAAC,aAAa,CAAC;QACzB,OAAO,IAAI,EAAE,CAAC;YACV,MAAM,GAAG,GAAG,MAAM,IAAI,CAAC,gBAAgB,EAAE,CAAC;YAC1C,IAAI,CAAC,GAAG;gBAAE,MAAM;YAChB,MAAM,GAAG,CAAC;QACd,CAAC;IACL,CAAC;IAED;;OAEG;IACH,KAAK,CAAC,oBAAoB,CAAC,OAAgB;QACvC,OAAO,CAAC,MAAM,IAAI,CAAC,iBAAiB,EAAE,CAAC,CAAC,oBAAoB,CAAC,OAAO,CAAC,CAAC;IAC1E,CAAC;IAED;;OAEG;IACH,KAAK,CAAC,cAAc,CAChB,OAAgB,EAChB,OAAsC;QAEtC,OAAO,CAAC,MAAM,IAAI,CAAC,iBAAiB,EAAE,CAAC,CAAC,cAAc,CAAC,OAAO,EAAE,OAAO,CAAC,CAAC;IAC7E,CAAC;IAED;;OAEG;IACH,KAAK,CAAC,UAAU,CAAC,WAAmB,EAAE,OAAsC;QACxE,OAAO,CAAC,MAAM,IAAI,CAAC,iBAAiB,EAAE,CAAC,CAAC,UAAU,CAAC,WAAW,EAAE,OAAO,CAAC,CAAC;IAC7E,CAAC;IAED;;OAEG;IACH,KAAK,CAAC,kBAAkB,CACpB,QAAsB,EACtB,OAAmC;QAEnC,OAAO,CAAC,MAAM,IAAI,CAAC,iBAAiB,EAAE,CAAC,CAAC,kBAAkB,CAAC,QAAQ,EAAE,OAAO,CAAC,CAAC;IAClF,CAAC;IAED;;;OAGG;IACH,KAAK,CAAC,YAAY;QACd,MAAM,IAAI,CAAC,aAAa,CAAC,YAAY,EAAE,EAAE,CAAC;IAC9C,CAAC;IAED;;;;OAIG;IACH,KAAK,CAAC,KAAK;QACP,MAAM,CAAC,MAAM,IAAI,CAAC,iBAAiB,EAAE,CAAC,CAAC,KAAK,EAAE,EAAE,CAAC;IACrD,CAAC;IAED;;;;OAIG;IACH,KAAK,CAAC,oCAAoC,CAAC,IAAY;QACnD,IAAI,CAAC,6BAA6B,GAAG,IAAI,CAAC;QAC1C,MAAM,IAAI,CAAC,sBAAsB,EAAE,oCAAoC,EAAE,CAAC,IAAI,CAAC,CAAC;IACpF,CAAC;CACJ"}