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

package/enqueue_links/enqueue_links.js

@@ -1,7 +1,8 @@
 import ow from 'ow';
 import { getDomain } from 'tldts';
-import log from '@apify/log';
-import { constructGlobObjectsFromGlobs, constructRegExpObjectsFromPseudoUrls, constructRegExpObjectsFromRegExps, createRequestOptions, createRequests, filterRequestsByPatterns, } from './shared.js';
+import { Request } from '../request.js';
+import { serviceLocator } from '../service_locator.js';
+import { applyRequestTransform, constructGlobObjectsFromGlobs, constructRegExpObjectsFromPseudoUrls, constructRegExpObjectsFromRegExps, createRequestOptions, filterRequestOptionsByPatterns, } from './shared.js';
 /**
  * The different enqueueing strategies available.
  *
@@ -66,7 +67,7 @@ export var EnqueueStrategy;
  * ```javascript
  * await enqueueLinks({
  *   urls: aListOfFoundUrls,
- *   requestQueue,
+ *   requestManager,
  *   selector: 'a.product-detail',
  *   globs: [
  *       'https://www.example.com/handbags/*',
@@ -87,11 +88,13 @@ export async function enqueueLinks(options) {
     }
     ow(options, ow.object.exactShape({
         urls: ow.array.ofType(ow.string),
-        requestQueue: ow.object.hasKeys('fetchNextRequest', 'addRequest'),
+        requestManager: ow.object.hasKeys('addRequestsBatched'),
         robotsTxtFile: ow.optional.object.hasKeys('isAllowed'),
+        respectRobotsTxtFile: ow.optional.any(ow.boolean, ow.object.exactShape({ userAgent: ow.optional.string })),
         onSkippedRequest: ow.optional.function,
         forefront: ow.optional.boolean,
         skipNavigation: ow.optional.boolean,
+        sessionId: ow.optional.string,
         limit: ow.optional.number,
         selector: ow.optional.string,
         baseUrl: ow.optional.string,
@@ -105,7 +108,9 @@ export async function enqueueLinks(options) {
         strategy: ow.optional.string.oneOf(Object.values(EnqueueStrategy)),
         waitForAllRequestsToBeAdded: ow.optional.boolean,
     }));
-    const { requestQueue, limit, urls, pseudoUrls, exclude, globs, regexps, transformRequestFunction, forefront, waitForAllRequestsToBeAdded, robotsTxtFile, onSkippedRequest, } = options;
+    const { requestManager, limit, urls, 
+    // oxlint-disable-next-line typescript/no-deprecated -- still accepted for backwards compat
+    pseudoUrls, exclude, globs, regexps, transformRequestFunction, forefront, waitForAllRequestsToBeAdded, robotsTxtFile, onSkippedRequest, } = options;
     const urlExcludePatternObjects = [];
     const urlPatternObjects = [];
     if (exclude?.length) {
@@ -119,7 +124,7 @@ export async function enqueueLinks(options) {
         }
     }
     if (pseudoUrls?.length) {
-        log.deprecated('`pseudoUrls` option is deprecated, use `globs` or `regexps` instead');
+        serviceLocator.getLogger().deprecated('`pseudoUrls` option is deprecated, use `globs` or `regexps` instead');
         urlPatternObjects.push(...constructRegExpObjectsFromPseudoUrls(pseudoUrls));
     }
     if (globs?.length) {
@@ -167,44 +172,61 @@ export async function enqueueLinks(options) {
                 break;
         }
     }
+    async function reportSkippedRequests(skippedRequests, reason) {
+        if (onSkippedRequest && skippedRequests.length > 0) {
+            await Promise.all(skippedRequests.map((request) => {
+                return onSkippedRequest({
+                    url: request.url,
+                    reason: request.skippedReason ?? reason,
+                });
+            }));
+        }
+    }
     let requestOptions = createRequestOptions(urls, options);
-    if (robotsTxtFile) {
+    if (robotsTxtFile && options.respectRobotsTxtFile !== false) {
+        const robotsUserAgent = typeof options.respectRobotsTxtFile === 'object' ? (options.respectRobotsTxtFile.userAgent ?? '*') : '*';
         const skippedRequests = [];
         requestOptions = requestOptions.filter((request) => {
-            if (robotsTxtFile.isAllowed(request.url)) {
+            if (robotsTxtFile.isAllowed(request.url, robotsUserAgent)) {
                 return true;
             }
             skippedRequests.push(request);
             return false;
         });
-        if (onSkippedRequest && skippedRequests.length > 0) {
-            await Promise.all(skippedRequests.map((request) => {
-                return onSkippedRequest({ url: request.url, reason: 'robotsTxt' });
-            }));
-        }
+        await reportSkippedRequests(skippedRequests, 'robotsTxt');
     }
-    if (transformRequestFunction) {
-        requestOptions = requestOptions
-            .map((request) => transformRequestFunction(request))
-            .filter((r) => !!r);
-    }
-    function createFilteredRequests() {
-        // No user provided patterns means we can skip an extra filtering step
+    async function createFilteredRequests() {
+        const skippedRequests = [];
+        // Step 1: Filter request options by exclude patterns, user patterns (globs/regexps), and strategy patterns.
+        // Pattern-level options (label, userData, method, etc.) are merged during this step.
+        let filteredOptions;
         if (urlPatternObjects.length === 0) {
-            return createRequests(requestOptions, enqueueStrategyPatterns, urlExcludePatternObjects, options.strategy);
+            filteredOptions = filterRequestOptionsByPatterns(requestOptions, enqueueStrategyPatterns.length > 0 ? enqueueStrategyPatterns : undefined, urlExcludePatternObjects, options.strategy, (url) => skippedRequests.push(url));
         }
-        // Generate requests based on the user patterns first
-        const generatedRequestsFromUserFilters = createRequests(requestOptions, urlPatternObjects, urlExcludePatternObjects, options.strategy);
-        // ...then filter them by the enqueue links strategy (making this an AND check)
-        return filterRequestsByPatterns(generatedRequestsFromUserFilters, enqueueStrategyPatterns);
+        else {
+            // Filter by user patterns first (with exclude)
+            const afterUserPatterns = filterRequestOptionsByPatterns(requestOptions, urlPatternObjects, urlExcludePatternObjects, options.strategy, (url) => skippedRequests.push(url));
+            // ...then filter by the enqueue links strategy (making this an AND check)
+            filteredOptions = filterRequestOptionsByPatterns(afterUserPatterns, enqueueStrategyPatterns.length > 0 ? enqueueStrategyPatterns : undefined, [], options.strategy, (url) => skippedRequests.push(url));
+        }
+        await reportSkippedRequests(skippedRequests.map((url) => ({ url })), 'filters');
+        // Step 2: Apply transformRequestFunction on request options - it has the highest priority
+        if (transformRequestFunction) {
+            const skippedByTransform = [];
+            filteredOptions = applyRequestTransform(filteredOptions, transformRequestFunction, (r) => skippedByTransform.push(r));
+            await reportSkippedRequests(skippedByTransform, 'transform');
+        }
+        // Step 3: Create Request instances from the final request options
+        return filteredOptions.map((opts) => new Request(opts));
     }
-    let requests = createFilteredRequests();
-    if (limit)
-        requests = requests.slice(0, limit);
-    const { addedRequests } = await requestQueue.addRequestsBatched(requests, {
+    const { addedRequests, requestsOverLimit } = await requestManager.addRequestsBatched(await createFilteredRequests(), {
         forefront,
         waitForAllRequestsToBeAdded,
+        maxNewRequests: limit,
     });
+    if (requestsOverLimit?.length !== undefined && requestsOverLimit.length > 0) {
+        await reportSkippedRequests(requestsOverLimit.map((r) => ({ url: typeof r === 'string' ? r : r.url })), 'enqueueLimit');
+    }
     return { processedRequests: addedRequests, unprocessedRequests: [] };
 }
 /**
@@ -247,4 +269,3 @@ export function resolveBaseUrlForEnqueueLinksFiltering({ enqueueStrategy, finalR
 function ignoreHttpSchema(pattern) {
     return pattern.replace(/^(https?):\/\//, 'http{s,}://');
 }
-//# sourceMappingURL=enqueue_links.js.map

package/enqueue_links/index.d.ts

@@ -1,3 +1,2 @@
 export * from './enqueue_links.js';
 export * from './shared.js';
-//# sourceMappingURL=index.d.ts.map

package/enqueue_links/index.js

@@ -1,3 +1,2 @@
 export * from './enqueue_links.js';
 export * from './shared.js';
-//# sourceMappingURL=index.js.map

package/enqueue_links/shared.d.ts

@@ -1,5 +1,5 @@
+import type { Awaitable } from '@crawlee/types';
 import type { RequestOptions } from '../request.js';
-import { Request } from '../request.js';
 import type { EnqueueLinksOptions } from './enqueue_links.js';
 export { tryAbsoluteURL } from '@crawlee/utils';
 export type UrlPatternObject = {
@@ -18,6 +18,11 @@ export type RegExpObject = {
     regexp: RegExp;
 } & Pick<RequestOptions, 'method' | 'payload' | 'label' | 'userData' | 'headers'>;
 export type RegExpInput = RegExp | RegExpObject;
+export type SkippedRequestReason = 'robotsTxt' | 'limit' | 'enqueueLimit' | 'filters' | 'transform' | 'redirect' | 'depth';
+export type SkippedRequestCallback = (args: {
+    url: string;
+    reason: SkippedRequestReason;
+}) => Awaitable<void>;
 /**
  * @ignore
  */
@@ -45,23 +50,34 @@ export declare function validateGlobPattern(glob: string): string;
  */
 export declare function constructRegExpObjectsFromRegExps(regexps: readonly RegExpInput[]): RegExpObject[];
 /**
+ * Filters request options by URL patterns and merges pattern-level options (label, userData, method, payload, headers)
+ * from the first matching pattern into each RequestOptions entry.
+ *
+ * When `includePatterns` is empty/undefined, all options pass through (only exclude filtering applies).
  * @ignore
  */
-export declare function createRequests(requestOptions: (string | RequestOptions)[], urlPatternObjects?: UrlPatternObject[], excludePatternObjects?: UrlPatternObject[], strategy?: EnqueueLinksOptions['strategy']): Request[];
-export declare function filterRequestsByPatterns(requests: Request[], patterns?: UrlPatternObject[]): Request[];
+export declare function filterRequestOptionsByPatterns(requestOptions: RequestOptions[], includePatterns: UrlPatternObject[] | undefined, excludePatterns?: UrlPatternObject[], strategy?: EnqueueLinksOptions['strategy'], onSkippedUrl?: (url: string) => void): RequestOptions[];
 /**
  * @ignore
  */
-export declare function createRequestOptions(sources: readonly (string | Record<string, unknown>)[], options?: Pick<EnqueueLinksOptions, 'label' | 'userData' | 'baseUrl' | 'skipNavigation' | 'strategy'>): RequestOptions[];
+export declare function createRequestOptions(sources: readonly (string | Record<string, unknown>)[], options?: Pick<EnqueueLinksOptions, 'label' | 'userData' | 'baseUrl' | 'skipNavigation' | 'sessionId' | 'strategy'>): RequestOptions[];
 /**
- * Takes an Apify {@link RequestOptions} object and changes its attributes in a desired way. This user-function is used
- * {@link enqueueLinks} to modify requests before enqueuing them.
+ * Takes a {@link RequestOptions} object and changes its attributes in a desired way. This user-function is used
+ * by {@link enqueueLinks} to modify request options before they are converted to {@link Request} instances.
  */
 export interface RequestTransform {
     /**
      * @param original Request options to be modified.
-     * @returns The modified request options to enqueue.
+     * @returns The modified request options to enqueue, `'unchanged'` to keep the original options as-is,
+     *   or a falsy value / `'skip'` to exclude the request from the queue.
      */
-    (original: RequestOptions): RequestOptions | false | undefined | null;
+    (original: RequestOptions): RequestOptions | false | undefined | null | 'skip' | 'unchanged';
 }
-//# sourceMappingURL=shared.d.ts.map
+/**
+ * Applies a {@link RequestTransform} function to a list of request options.
+ * Options for which the transform returns a falsy value are removed from the list.
+ * @param onSkipped Called with the original request options when the transform returns a falsy value (i.e. the request is skipped).
+ * @ignore
+ * @internal
+ */
+export declare function applyRequestTransform(requestOptions: RequestOptions[], transformFn: RequestTransform, onSkipped?: (requestOptions: RequestOptions) => void): RequestOptions[];

package/enqueue_links/shared.js

@@ -1,7 +1,6 @@
 import { URL } from 'node:url';
-import { minimatch } from 'minimatch';
+import { Minimatch } from 'minimatch';
 import { purlToRegExp } from '@apify/pseudo_url';
-import { Request } from '../request.js';
 export { tryAbsoluteURL } from '@crawlee/utils';
 const MAX_ENQUEUE_LINKS_CACHE_SIZE = 1000;
 /**
@@ -110,51 +109,37 @@ export function constructRegExpObjectsFromRegExps(regexps) {
     });
 }
 /**
+ * Filters request options by URL patterns and merges pattern-level options (label, userData, method, payload, headers)
+ * from the first matching pattern into each RequestOptions entry.
+ *
+ * When `includePatterns` is empty/undefined, all options pass through (only exclude filtering applies).
  * @ignore
  */
-export function createRequests(requestOptions, urlPatternObjects, excludePatternObjects = [], strategy) {
+export function filterRequestOptionsByPatterns(requestOptions, includePatterns, excludePatterns = [], strategy, onSkippedUrl) {
+    const excludeMatchers = excludePatterns.map(createPatternObjectMatcher);
+    const includeMatchers = includePatterns?.length ? includePatterns.map(createPatternObjectMatcher) : undefined;
     return requestOptions
-        .map((opts) => ({ url: typeof opts === 'string' ? opts : opts.url, opts }))
         .filter(({ url }) => {
-        return !excludePatternObjects.some((excludePatternObject) => {
-            const { regexp, glob } = excludePatternObject;
-            return (regexp && url.match(regexp)) || (glob && minimatch(url, glob, { nocase: true }));
-        });
+        const matchesExclude = excludeMatchers.some(({ match }) => match(url));
+        if (matchesExclude) {
+            onSkippedUrl?.(url);
+        }
+        return !matchesExclude;
     })
-        .map(({ url, opts }) => {
-        if (!urlPatternObjects || !urlPatternObjects.length) {
-            return new Request(typeof opts === 'string' ? { url: opts, enqueueStrategy: strategy } : { ...opts });
-        }
-        for (const urlPatternObject of urlPatternObjects) {
-            const { regexp, glob, ...requestRegExpOptions } = urlPatternObject;
-            if ((regexp && url.match(regexp)) || (glob && minimatch(url, glob, { nocase: true }))) {
-                const request = typeof opts === 'string'
-                    ? { url: opts, ...requestRegExpOptions, enqueueStrategy: strategy }
-                    : { ...opts, ...requestRegExpOptions, enqueueStrategy: strategy };
-                return new Request(request);
+        .map((opts) => {
+        if (!includeMatchers) {
+            return { ...opts, enqueueStrategy: strategy };
+        }
+        for (const { match, glob, regexp, ...patternOptions } of includeMatchers) {
+            if (match(opts.url)) {
+                return { ...opts, ...patternOptions, enqueueStrategy: strategy };
             }
         }
         // didn't match any positive pattern
+        onSkippedUrl?.(opts.url);
         return null;
     })
-        .filter((request) => request);
-}
-export function filterRequestsByPatterns(requests, patterns) {
-    if (!patterns?.length) {
-        return requests;
-    }
-    const filtered = [];
-    for (const request of requests) {
-        for (const urlPatternObject of patterns) {
-            const { regexp, glob } = urlPatternObject;
-            if ((regexp && request.url.match(regexp)) || (glob && minimatch(request.url, glob, { nocase: true }))) {
-                filtered.push(request);
-                // Break the pattern loop, as we already matched this request once
-                break;
-            }
-        }
-    }
-    return filtered;
+        .filter((opts) => opts !== null);
 }
 /**
  * @ignore
@@ -184,7 +169,53 @@ export function createRequestOptions(sources, options = {}) {
         if (options.skipNavigation) {
             requestOptions.skipNavigation = true;
         }
+        if (options.sessionId) {
+            requestOptions.sessionId = options.sessionId;
+        }
         return requestOptions;
     });
 }
-//# sourceMappingURL=shared.js.map
+/**
+ * @ignore
+ */
+function createPatternObjectMatcher(urlPatternObject) {
+    const { regexp, glob } = urlPatternObject;
+    let match;
+    if (regexp) {
+        match = (url) => regexp.test(url);
+    }
+    else if (glob) {
+        const m = new Minimatch(glob, { nocase: true });
+        match = (url) => m.match(url);
+    }
+    else {
+        match = () => false;
+    }
+    return { ...urlPatternObject, match };
+}
+/**
+ * Applies a {@link RequestTransform} function to a list of request options.
+ * Options for which the transform returns a falsy value are removed from the list.
+ * @param onSkipped Called with the original request options when the transform returns a falsy value (i.e. the request is skipped).
+ * @ignore
+ * @internal
+ */
+export function applyRequestTransform(requestOptions, transformFn, onSkipped) {
+    return requestOptions
+        .map((opts) => {
+        const transformed = transformFn(opts);
+        if (transformed === 'skip') {
+            onSkipped?.(opts);
+            return null;
+        }
+        if (transformed === 'unchanged') {
+            return opts;
+        }
+        if (!transformed) {
+            onSkipped?.(opts);
+            return null;
+        }
+        return transformed;
+    })
+        .filter((r) => r !== null);
+}

package/errors.d.ts

@@ -14,6 +14,26 @@ export declare class CriticalError extends NonRetryableError {
  */
 export declare class MissingRouteError extends CriticalError {
 }
+/**
+ * Thrown when a request's `userData` does not match the {@link RouteSchemas|Standard Schema} registered for its label.
+ *
+ * As the `userData` does not change between attempts, this error is non-retryable.
+ */
+export declare class RequestValidationError extends NonRetryableError {
+    readonly label: string | symbol;
+    readonly issues: readonly {
+        readonly message: string;
+        readonly path?: readonly (PropertyKey | {
+            key: PropertyKey;
+        })[];
+    }[];
+    constructor(label: string | symbol, issues: readonly {
+        readonly message: string;
+        readonly path?: readonly (PropertyKey | {
+            key: PropertyKey;
+        })[];
+    }[]);
+}
 /**
  * Errors of `RetryRequestError` type will always be retried by the crawler.
  *
@@ -23,11 +43,40 @@ export declare class RetryRequestError extends Error {
     constructor(message?: string);
 }
 /**
- * Errors of `SessionError` type will trigger a session rotation.
+ * Errors of `SessionError` type retire the session associated with the request and trigger a regular retry.
  *
- * This error doesn't respect the `maxRequestRetries` option and has a separate limit of `maxSessionRotations`.
+ * The retry counts towards the `maxRequestRetries` limit, just like any other error.
  */
-export declare class SessionError extends RetryRequestError {
+export declare class SessionError extends Error {
+    constructor(message?: string);
+}
+/**
+ * Thrown when a requested session is not found in the referenced SessionPool.
+ */
+export declare class MissingSessionError extends Error {
+    constructor(sessionId?: string);
+}
+export declare class ContextPipelineInterruptedError extends Error {
     constructor(message?: string);
 }
-//# sourceMappingURL=errors.d.ts.map
+export declare class ContextPipelineInitializationError extends Error {
+    constructor(error: unknown, options?: ErrorOptions);
+}
+export declare class ContextPipelineCleanupError extends CriticalError {
+    constructor(error: unknown, options?: ErrorOptions);
+}
+export declare class RequestHandlerError extends Error {
+    constructor(error: unknown, options?: ErrorOptions);
+}
+/**
+ * Thrown when attempting to set a different service instance after one has already been retrieved.
+ */
+export declare class ServiceConflictError extends Error {
+    constructor(serviceName: string, newValue: unknown, existingValue: unknown);
+}
+/**
+ * Thrown by crawlers when `skipNavigation` is used on a request.
+ * Subclasses can catch this error to skip their own navigation-dependent logic.
+ */
+export declare class NavigationSkippedError extends NonRetryableError {
+}

package/errors.js

@@ -1,3 +1,4 @@
+import { inspectValue } from './debug.js';
 /**
  * Errors of `NonRetryableError` type will never be retried by the crawler.
  */
@@ -14,6 +15,28 @@ export class CriticalError extends NonRetryableError {
  */
 export class MissingRouteError extends CriticalError {
 }
+/**
+ * Thrown when a request's `userData` does not match the {@link RouteSchemas|Standard Schema} registered for its label.
+ *
+ * As the `userData` does not change between attempts, this error is non-retryable.
+ */
+export class RequestValidationError extends NonRetryableError {
+    label;
+    issues;
+    constructor(label, issues) {
+        const details = issues
+            .map((issue) => {
+            const path = (issue.path ?? [])
+                .map((segment) => (typeof segment === 'object' ? segment.key : segment))
+                .join('.');
+            return `- ${path ? `${path}: ` : ''}${issue.message}`;
+        })
+            .join('\n');
+        super(`Request userData for label '${String(label)}' failed schema validation:\n${details}`);
+        this.label = label;
+        this.issues = issues;
+    }
+}
 /**
  * Errors of `RetryRequestError` type will always be retried by the crawler.
  *
@@ -25,13 +48,55 @@ export class RetryRequestError extends Error {
     }
 }
 /**
- * Errors of `SessionError` type will trigger a session rotation.
+ * Errors of `SessionError` type retire the session associated with the request and trigger a regular retry.
  *
- * This error doesn't respect the `maxRequestRetries` option and has a separate limit of `maxSessionRotations`.
+ * The retry counts towards the `maxRequestRetries` limit, just like any other error.
  */
-export class SessionError extends RetryRequestError {
+export class SessionError extends Error {
     constructor(message) {
-        super(`Detected a session error, rotating session... ${message ? `\n${message}` : ''}`);
+        super(`Detected a session error, retiring session... ${message ? `\n${message}` : ''}`);
     }
 }
-//# sourceMappingURL=errors.js.map
+/**
+ * Thrown when a requested session is not found in the referenced SessionPool.
+ */
+export class MissingSessionError extends Error {
+    constructor(sessionId) {
+        super(`The current SessionPool instance couldn't find a valid session${sessionId ? ` for the following id: ${sessionId}.` : '.'}`);
+    }
+}
+export class ContextPipelineInterruptedError extends Error {
+    constructor(message) {
+        super(`Request handling was interrupted during context initialization ${message ? ` - ${message}` : ''}`);
+    }
+}
+export class ContextPipelineInitializationError extends Error {
+    constructor(error, options) {
+        super(undefined, { cause: error, ...options });
+    }
+}
+export class ContextPipelineCleanupError extends CriticalError {
+    constructor(error, options) {
+        super(undefined, { cause: error, ...options });
+    }
+}
+export class RequestHandlerError extends Error {
+    constructor(error, options) {
+        super(undefined, { cause: error, ...options });
+    }
+}
+/**
+ * Thrown when attempting to set a different service instance after one has already been retrieved.
+ */
+export class ServiceConflictError extends Error {
+    constructor(serviceName, newValue, existingValue) {
+        super(`Service ${serviceName} is already in use. ` +
+            `Existing value: ${inspectValue(existingValue)}, attempted new value: ${inspectValue(newValue)}.`);
+    }
+}
+/**
+ * Thrown by crawlers when `skipNavigation` is used on a request.
+ * Subclasses can catch this error to skip their own navigation-dependent logic.
+ */
+export class NavigationSkippedError extends NonRetryableError {
+}

package/events/event_manager.d.ts

@@ -1,28 +1,55 @@
 import { AsyncEventEmitter } from '@vladfrangu/async_event_emitter';
 import type { BetterIntervalID } from '@apify/utilities';
-import { Configuration } from '../configuration.js';
+export interface EventManagerOptions {
+    /** Interval between emitted `persistState` events in milliseconds. */
+    persistStateIntervalMillis: number;
+}
 export declare const enum EventType {
     PERSIST_STATE = "persistState",
     SYSTEM_INFO = "systemInfo",
     MIGRATING = "migrating",
     ABORTING = "aborting",
-    EXIT = "exit"
+    EXIT = "exit",
+    STATUS_MESSAGE = "statusMessage"
+}
+export type EventTypeName = EventType | 'systemInfo' | 'persistState' | 'migrating' | 'aborting' | 'exit' | 'statusMessage';
+/**
+ * Payload emitted with the {@link EventType.STATUS_MESSAGE|`statusMessage`} event.
+ *
+ * The crawler broadcasts these whenever it wants to report its progress (e.g. periodically, or on
+ * start/finish). Consumers such as the Apify SDK can listen for the event and propagate the message
+ * to the platform. This keeps the crawler decoupled from any specific status-reporting backend.
+ */
+export interface EventStatusMessageData {
+    /**
+     * Identifies the crawler that emitted the message.
+     *
+     * Either the user-provided `id` from the crawler options, or a randomly generated one.
+     * Since a single event manager may be shared by multiple crawlers, consumers can use this
+     * to attribute the message to a specific crawler instance.
+     */
+    crawlerId: string;
+    /** The human-readable status message. */
+    message: string;
+    /** Whether this is the final status message of the run. */
+    isStatusMessageTerminal?: boolean;
+    /** The log level the message was logged with. */
+    level?: 'DEBUG' | 'INFO' | 'WARNING' | 'ERROR';
 }
-export type EventTypeName = EventType | 'systemInfo' | 'persistState' | 'migrating' | 'aborting' | 'exit';
 interface Intervals {
     persistState?: BetterIntervalID;
     systemInfo?: BetterIntervalID;
 }
 export declare abstract class EventManager {
-    readonly config: Configuration;
     protected events: AsyncEventEmitter<{}>;
     protected initialized: boolean;
     protected intervals: Intervals;
 // @ts-ignore optional peer dependency or compatibility with es2022
-    protected log: import("@apify/log").Log;
-    constructor(config?: Configuration);
+    protected log: import("@crawlee/types").CrawleeLogger;
+    private persistStateIntervalMillis;
+    constructor(options: EventManagerOptions);
     /**
-     * Initializes the event manager by creating the `persistState` event interval.
+     * Initializes the event manager by starting the `persistState` event interval.
      * This is automatically called at the beginning of `crawler.run()`.
      */
     init(): Promise<void>;
@@ -49,4 +76,3 @@ export declare abstract class EventManager {
     waitForAllListenersToComplete(): Promise<boolean>;
 }
 export {};
-//# sourceMappingURL=event_manager.d.ts.map

package/events/event_manager.js

@@ -1,7 +1,6 @@
 import { AsyncEventEmitter } from '@vladfrangu/async_event_emitter';
-import log from '@apify/log';
 import { betterClearInterval, betterSetInterval } from '@apify/utilities';
-import { Configuration } from '../configuration.js';
+import { serviceLocator } from '../service_locator.js';
 export var EventType;
 (function (EventType) {
     EventType["PERSIST_STATE"] = "persistState";
@@ -9,30 +8,30 @@ export var EventType;
     EventType["MIGRATING"] = "migrating";
     EventType["ABORTING"] = "aborting";
     EventType["EXIT"] = "exit";
+    EventType["STATUS_MESSAGE"] = "statusMessage";
 })(EventType || (EventType = {}));
 export class EventManager {
-    config;
     events = new AsyncEventEmitter();
     initialized = false;
     intervals = {};
-    log = log.child({ prefix: 'Events' });
-    constructor(config = Configuration.getGlobalConfig()) {
-        this.config = config;
+    log = serviceLocator.getLogger().child({ prefix: 'Events' });
+    persistStateIntervalMillis;
+    constructor(options) {
+        this.persistStateIntervalMillis = options.persistStateIntervalMillis;
         this.events.setMaxListeners(50);
     }
     /**
-     * Initializes the event manager by creating the `persistState` event interval.
+     * Initializes the event manager by starting the `persistState` event interval.
      * This is automatically called at the beginning of `crawler.run()`.
      */
     async init() {
         if (this.initialized) {
             return;
         }
-        const persistStateIntervalMillis = this.config.get('persistStateIntervalMillis');
         this.intervals.persistState = betterSetInterval((intervalCallback) => {
             this.emit("persistState" /* EventType.PERSIST_STATE */, { isMigrating: false });
             intervalCallback();
-        }, persistStateIntervalMillis);
+        }, this.persistStateIntervalMillis);
         this.initialized = true;
     }
     /**
@@ -86,4 +85,3 @@ export class EventManager {
         return this.events.waitForAllListenersToComplete();
     }
 }
-//# sourceMappingURL=event_manager.js.map

package/events/index.d.ts

@@ -1,3 +1,2 @@
 export * from './event_manager.js';
 export * from './local_event_manager.js';
-//# sourceMappingURL=index.d.ts.map

package/events/index.js

@@ -1,3 +1,2 @@
 export * from './event_manager.js';
 export * from './local_event_manager.js';
-//# sourceMappingURL=index.js.map