@crawlee/core 4.0.0-beta.13 → 4.0.0-beta.131

package/storages/request_loader.d.ts

@@ -0,0 +1,101 @@
+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.
+     *
+     * This is a statement about what the *next fetch* would return, not about how much work is left, so it
+     * may report `true` while {@link IRequestLoader.getPendingCount} is non-zero - a loader that withholds
+     * requests for a while (as {@link ThrottlingRequestManager} does for a rate-limited domain) is empty
+     * for as long as it will not hand anything over. Use `isFinished()` to ask whether the work is done.
+     */
+    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>;
+}

package/storages/request_loader.js

@@ -0,0 +1 @@
+export {};

package/storages/request_manager.d.ts

@@ -0,0 +1,33 @@
+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>;
+}

package/storages/request_manager.js

@@ -0,0 +1 @@
+export {};

package/storages/request_manager_tandem.d.ts

@@ -0,0 +1,97 @@
+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;
+    /**
+     * @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>;
+}

package/storages/request_manager_tandem.js

@@ -0,0 +1,197 @@
+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);
+    }
+}