@crawlee/utils 4.0.0-beta.13 → 4.0.0-beta.130

package/internals/open_graph_parser.js

@@ -1,4 +1,3 @@
-import { load } from 'cheerio';
 /**
  * To be used with the spread operator. Ensures that the item is defined, and is not empty.
  *
@@ -255,32 +254,32 @@ const OPEN_GRAPH_PROPERTIES = [
         outputName: 'articleInfo',
         children: [
             {
-                name: 'music:published_time',
+                name: 'article:published_time',
                 outputName: 'publishedTime',
                 children: [],
             },
             {
-                name: 'music:modified_time',
+                name: 'article:modified_time',
                 outputName: 'modifiedTime',
                 children: [],
             },
             {
-                name: 'music:expiration_time',
+                name: 'article:expiration_time',
                 outputName: 'expirationTime',
                 children: [],
             },
             {
-                name: 'music:author',
+                name: 'article:author',
                 outputName: 'author',
                 children: [],
             },
             {
-                name: 'music:section',
+                name: 'article:section',
                 outputName: 'section',
                 children: [],
             },
             {
-                name: 'music:tag',
+                name: 'article:tag',
                 outputName: 'tag',
                 children: [],
             },
@@ -366,7 +365,8 @@ const parseOpenGraphProperty = (property, $) => {
         }, {}),
     };
 };
-export function parseOpenGraph(item, additionalProperties) {
+export async function parseOpenGraph(item, additionalProperties) {
+    const { load } = await import('cheerio');
     const $ = typeof item === 'string' ? load(item) : item;
     return [...(additionalProperties || []), ...OPEN_GRAPH_PROPERTIES].reduce((acc, curr) => {
         return {
@@ -375,4 +375,3 @@ export function parseOpenGraph(item, additionalProperties) {
         };
     }, {});
 }
-//# sourceMappingURL=open_graph_parser.js.map

package/internals/robots.d.ts

@@ -1,4 +1,15 @@
+import type { BaseHttpClient } from '@crawlee/http-client';
+import type { CrawleeLogger } from '@crawlee/types';
 import { Sitemap } from './sitemap.js';
+import { type EnqueueStrategy } from './url.js';
+export interface RobotsTxtFileSitemapsOptions {
+    /**
+     * Keep only sitemap URLs matching this strategy relative to the robots.txt host; non-`http(s)` schemes
+     * are always dropped. Pass `'all'` to disable host filtering.
+     * @default 'same-hostname'
+     */
+    enqueueStrategy?: EnqueueStrategy | `${EnqueueStrategy}`;
+}
 /**
  * Loads and queries information from a [robots.txt file](https://en.wikipedia.org/wiki/Robots.txt).
  *
@@ -18,15 +29,23 @@ import { Sitemap } from './sitemap.js';
  * ```
  */
 export declare class RobotsTxtFile {
-    private robots;
-    private proxyUrl?;
+    #private;
     private constructor();
     /**
      * Determine the location of a robots.txt file for a URL and fetch it.
      * @param url the URL to fetch robots.txt for
-     * @param [proxyUrl] a proxy to be used for fetching the robots.txt file
+     * @param [options] additional options
+     * @param [options.signal] an AbortSignal to cancel the request
+     * @param [options.timeoutMillis] timeout in milliseconds for the request
+     * @param [options.proxyUrl] a proxy to be used for fetching the robots.txt file
      */
-    static find(url: string, proxyUrl?: string): Promise<RobotsTxtFile>;
+    static find(url: string, options?: {
+        signal?: AbortSignal;
+        timeoutMillis?: number;
+        proxyUrl?: string;
+        httpClient?: BaseHttpClient;
+        logger?: CrawleeLogger;
+    }): Promise<RobotsTxtFile>;
     /**
      * Allows providing the URL and robots.txt content explicitly instead of loading it from the target site.
      * @param url the URL for robots.txt file
@@ -34,7 +53,12 @@ export declare class RobotsTxtFile {
      * @param [proxyUrl] a proxy to be used for fetching the robots.txt file
      */
     static from(url: string, content: string, proxyUrl?: string): RobotsTxtFile;
-    protected static load(url: string, proxyUrl?: string): Promise<RobotsTxtFile>;
+    private static load;
+    /**
+     * Get crawl delay for a given user agent.
+     * @param [userAgent] relevant user agent, default to `*`
+     */
+    getCrawlDelay(userAgent?: string): number | undefined;
     /**
      * Check if a URL should be crawled by robots.
      * @param url the URL to check against the rules in robots.txt
@@ -42,17 +66,19 @@ export declare class RobotsTxtFile {
      */
     isAllowed(url: string, userAgent?: string): boolean;
     /**
-     * Get URLs of sitemaps referenced in the robots file.
+     * Get URLs of sitemaps referenced in the robots file, filtered by `options.enqueueStrategy` relative to
+     * the robots.txt host (default `'same-hostname'`; pass `'all'` to disable). Non-`http(s)` schemes are
+     * always dropped.
      */
-    getSitemaps(): string[];
+    getSitemaps(options?: RobotsTxtFileSitemapsOptions): string[];
     /**
-     * Parse all the sitemaps referenced in the robots file.
+     * Parse all the sitemaps referenced in the robots file. `options` are forwarded to `getSitemaps`
+     * and the sitemap parser.
      */
-    parseSitemaps(): Promise<Sitemap>;
+    parseSitemaps(options?: RobotsTxtFileSitemapsOptions): Promise<Sitemap>;
     /**
      * Get all URLs from all the sitemaps referenced in the robots file. A shorthand for `(await robots.parseSitemaps()).urls`.
+     * `options` are forwarded to `parseSitemaps`.
      */
-    parseUrlsFromSitemaps(): Promise<string[]>;
+    parseUrlsFromSitemaps(options?: RobotsTxtFileSitemapsOptions): Promise<string[]>;
 }
-export { RobotsTxtFile as RobotsFile };
-//# sourceMappingURL=robots.d.ts.map

package/internals/robots.js

@@ -1,7 +1,7 @@
-import { gotScraping } from 'got-scraping';
+import { FetchHttpClient } from '@crawlee/http-client';
 import robotsParser from 'robots-parser';
 import { Sitemap } from './sitemap.js';
-let HTTPError;
+import { filterUrl } from './url.js';
 /**
  * Loads and queries information from a [robots.txt file](https://en.wikipedia.org/wiki/Robots.txt).
  *
@@ -21,22 +21,29 @@ let HTTPError;
  * ```
  */
 export class RobotsTxtFile {
-    robots;
-    proxyUrl;
-    constructor(robots, proxyUrl) {
-        this.robots = robots;
-        this.proxyUrl = proxyUrl;
+    #url;
+    #robots;
+    #proxyUrl;
+    #logger;
+    constructor(url, robots, proxyUrl, logger) {
+        this.#url = url;
+        this.#robots = robots;
+        this.#proxyUrl = proxyUrl;
+        this.#logger = logger;
     }
     /**
      * Determine the location of a robots.txt file for a URL and fetch it.
      * @param url the URL to fetch robots.txt for
-     * @param [proxyUrl] a proxy to be used for fetching the robots.txt file
+     * @param [options] additional options
+     * @param [options.signal] an AbortSignal to cancel the request
+     * @param [options.timeoutMillis] timeout in milliseconds for the request
+     * @param [options.proxyUrl] a proxy to be used for fetching the robots.txt file
      */
-    static async find(url, proxyUrl) {
+    static async find(url, options) {
         const robotsTxtFileUrl = new URL(url);
         robotsTxtFileUrl.pathname = '/robots.txt';
         robotsTxtFileUrl.search = '';
-        return RobotsTxtFile.load(robotsTxtFileUrl.toString(), proxyUrl);
+        return RobotsTxtFile.load(robotsTxtFileUrl.toString(), options);
     }
     /**
      * Allows providing the URL and robots.txt content explicitly instead of loading it from the target site.
@@ -46,35 +53,40 @@ export class RobotsTxtFile {
      */
     static from(url, content, proxyUrl) {
         // @ts-ignore
-        return new RobotsTxtFile(robotsParser(url, content), proxyUrl);
+        return new RobotsTxtFile(url, robotsParser(url, content), proxyUrl);
     }
-    static async load(url, proxyUrl) {
-        if (!HTTPError) {
-            HTTPError = (await import('got-scraping')).HTTPError;
-        }
-        try {
-            const response = await gotScraping({
-                url,
-                proxyUrl,
-                method: 'GET',
-                responseType: 'text',
-            });
-            // @ts-ignore
-            return new RobotsTxtFile(robotsParser(url.toString(), response.body), proxyUrl);
+    static async load(url, options) {
+        const { proxyUrl, logger, httpClient = new FetchHttpClient() } = options || {};
+        const response = await httpClient.sendRequest(new Request(url, { method: 'GET' }), {
+            proxyUrl,
+            timeoutMillis: options?.timeoutMillis,
+            signal: options?.signal,
+        });
+        if (response.status < 200 || response.status >= 300) {
+            throw new Error(`Failed to load robots.txt from ${url}: HTTP ${response.status}`);
         }
-        catch (e) {
-            if (e instanceof HTTPError && e.response.statusCode === 404) {
-                return new RobotsTxtFile({
-                    isAllowed() {
-                        return true;
-                    },
-                    getSitemaps() {
-                        return [];
-                    },
-                }, proxyUrl);
-            }
-            throw e;
+        if (response.status === 404) {
+            return new RobotsTxtFile(url, {
+                isAllowed() {
+                    return true;
+                },
+                getSitemaps() {
+                    return [];
+                },
+                getCrawlDelay() {
+                    return undefined;
+                },
+            }, proxyUrl, logger);
         }
+        // @ts-ignore
+        return new RobotsTxtFile(url, robotsParser(url.toString(), await response.text()), proxyUrl, logger);
+    }
+    /**
+     * Get crawl delay for a given user agent.
+     * @param [userAgent] relevant user agent, default to `*`
+     */
+    getCrawlDelay(userAgent = '*') {
+        return this.#robots.getCrawlDelay(userAgent);
     }
     /**
      * Check if a URL should be crawled by robots.
@@ -82,27 +94,39 @@ export class RobotsTxtFile {
      * @param [userAgent] relevant user agent, default to `*`
      */
     isAllowed(url, userAgent = '*') {
-        return this.robots.isAllowed(url, userAgent) ?? true; // `undefined` means that there is no explicit rule for the requested URL - assume it's allowed
+        return this.#robots.isAllowed(url, userAgent) ?? true; // `undefined` means that there is no explicit rule for the requested URL - assume it's allowed
     }
     /**
-     * Get URLs of sitemaps referenced in the robots file.
+     * Get URLs of sitemaps referenced in the robots file, filtered by `options.enqueueStrategy` relative to
+     * the robots.txt host (default `'same-hostname'`; pass `'all'` to disable). Non-`http(s)` schemes are
+     * always dropped.
      */
-    getSitemaps() {
-        return this.robots.getSitemaps();
+    getSitemaps(options = {}) {
+        const { enqueueStrategy = 'same-hostname' } = options;
+        const sitemaps = [];
+        for (const sitemapUrl of this.#robots.getSitemaps()) {
+            // `filterUrl` tolerates an unparseable origin (returns not-allowed) rather than throwing.
+            const { allowed, reason } = filterUrl(sitemapUrl, this.#url, enqueueStrategy);
+            if (!allowed) {
+                this.#logger?.warning(`Skipping sitemap ${sitemapUrl} listed in robots.txt at ${this.#url}: ${reason}.`);
+                continue;
+            }
+            sitemaps.push(sitemapUrl);
+        }
+        return sitemaps;
     }
     /**
-     * Parse all the sitemaps referenced in the robots file.
+     * Parse all the sitemaps referenced in the robots file. `options` are forwarded to `getSitemaps`
+     * and the sitemap parser.
      */
-    async parseSitemaps() {
-        return Sitemap.load(this.robots.getSitemaps(), this.proxyUrl);
+    async parseSitemaps(options = {}) {
+        return Sitemap.load(this.getSitemaps(options), this.#proxyUrl, { ...options, logger: this.#logger });
     }
     /**
      * Get all URLs from all the sitemaps referenced in the robots file. A shorthand for `(await robots.parseSitemaps()).urls`.
+     * `options` are forwarded to `parseSitemaps`.
      */
-    async parseUrlsFromSitemaps() {
-        return (await this.parseSitemaps()).urls;
+    async parseUrlsFromSitemaps(options = {}) {
+        return (await this.parseSitemaps(options)).urls;
     }
 }
-// to stay backwards compatible
-export { RobotsTxtFile as RobotsFile };
-//# sourceMappingURL=robots.js.map

package/internals/schemas.d.ts

@@ -0,0 +1,114 @@
+import { BaseHttpClient } from '@crawlee/http-client';
+import type { Dictionary } from '@crawlee/types';
+import { z } from 'zod';
+/**
+ * Accepts any object (including arrays and functions).
+ * @internal
+ */
+export declare const anyObject: z.ZodCustom<Dictionary, Dictionary>;
+/**
+ * Accepts any array without validating its items (cheap for huge arrays).
+ * @internal
+ */
+export declare const anyArray: z.ZodCustom<unknown[], unknown[]>;
+/**
+ * Accepts any function.
+ * @internal
+ */
+export declare const anyFunction: z.ZodCustom<(...args: any[]) => unknown, (...args: any[]) => unknown>;
+/**
+ * Mirrors `ow.number`: `Infinity` is a valid number, `NaN` is not.
+ * @internal
+ */
+export declare const anyNumber: z.ZodCustom<number, number>;
+/**
+ * Accepts any object (including functions) that has all the given keys, own or inherited.
+ * @internal
+ */
+export declare function objectWithKeys(keys: string[], message?: string): z.ZodType<Dictionary>;
+/**
+ * Accepts only instances of {@link BaseHttpClient} (all Crawlee HTTP clients extend it).
+ * @internal
+ */
+export declare const httpClient: z.ZodCustom<BaseHttpClient, BaseHttpClient>;
+/**
+ * Accepts any object implementing the CrawleeLogger interface.
+ * @internal
+ */
+export declare const logger: z.ZodType<Dictionary, unknown, z.core.$ZodTypeInternals<Dictionary, unknown>>;
+/**
+ * Accepts any typed array (`Uint8Array`, `Float64Array`, ...), but not a `DataView`.
+ * @internal
+ */
+export declare const typedArray: z.ZodCustom<NodeJS.TypedArray<ArrayBufferLike>, NodeJS.TypedArray<ArrayBufferLike>>;
+/**
+ * Accepts any non-null, non-array object.
+ * @internal
+ */
+export declare const plainObject: z.ZodCustom<Record<string, unknown>, Record<string, unknown>>;
+/**
+ * Shape of a request stored in a request queue.
+ * @internal
+ */
+export declare const storageRequest: z.ZodObject<{
+    id: z.ZodString;
+    url: z.ZodURL;
+    uniqueKey: z.ZodString;
+    method: z.ZodOptional<z.ZodString>;
+    retryCount: z.ZodOptional<z.ZodNumber>;
+    handledAt: z.ZodOptional<z.ZodUnion<readonly [z.ZodString, z.ZodDate]>>;
+}, z.core.$loose>;
+/**
+ * {@link storageRequest} before an id is assigned.
+ * @internal
+ */
+export declare const storageRequestWithoutId: z.ZodObject<{
+    url: z.ZodURL;
+    uniqueKey: z.ZodString;
+    method: z.ZodOptional<z.ZodString>;
+    retryCount: z.ZodOptional<z.ZodNumber>;
+    handledAt: z.ZodOptional<z.ZodUnion<readonly [z.ZodString, z.ZodDate]>>;
+}, z.core.$loose>;
+/**
+ * `z.array(item)` whose top-level type error names the element type — ``expected an array of numbers`` —
+ * instead of zod's bare `expected array`. Element failures keep zod's per-index messages, and `elements`
+ * is a human-readable plural (`'numbers'`, `'URL patterns'`), since element types cannot be introspected.
+ * @internal
+ */
+export declare function arrayOf<TItem extends z.ZodType>(item: TItem, elements: string): z.ZodArray<TItem>;
+/**
+ * Batch of {@link storageRequestWithoutId}.
+ * @internal
+ */
+export declare const storageRequestBatch: z.ZodArray<z.ZodObject<{
+    url: z.ZodURL;
+    uniqueKey: z.ZodString;
+    method: z.ZodOptional<z.ZodString>;
+    retryCount: z.ZodOptional<z.ZodNumber>;
+    handledAt: z.ZodOptional<z.ZodUnion<readonly [z.ZodString, z.ZodDate]>>;
+}, z.core.$loose>>;
+/**
+ * Options of request queue add/update operations.
+ * @internal
+ */
+export declare const requestQueueOperationOptions: z.ZodObject<{
+    forefront: z.ZodOptional<z.ZodBoolean>;
+}, z.core.$strip>;
+/**
+ * Options of key-value store `listKeys`.
+ * @internal
+ */
+export declare const keyValueStoreListKeysOptions: z.ZodObject<{
+    prefix: z.ZodOptional<z.ZodString>;
+    exclusiveStartKey: z.ZodOptional<z.ZodString>;
+    limit: z.ZodOptional<z.ZodNumber>;
+}, z.core.$strip>;
+/**
+ * Options of dataset item listing.
+ * @internal
+ */
+export declare const datasetListItemsOptions: z.ZodObject<{
+    desc: z.ZodOptional<z.ZodBoolean>;
+    limit: z.ZodOptional<z.ZodNumber>;
+    offset: z.ZodOptional<z.ZodNumber>;
+}, z.core.$strip>;

package/internals/schemas.js

@@ -0,0 +1,114 @@
+import { BaseHttpClient } from '@crawlee/http-client';
+import { z } from 'zod';
+/**
+ * Accepts any object (including arrays and functions).
+ * @internal
+ */
+export const anyObject = z.custom((value) => (typeof value === 'object' && value !== null) || typeof value === 'function', { message: 'Invalid input: expected object' });
+/**
+ * Accepts any array without validating its items (cheap for huge arrays).
+ * @internal
+ */
+export const anyArray = z.custom(Array.isArray, { message: 'Invalid input: expected array' });
+/**
+ * Accepts any function.
+ * @internal
+ */
+export const anyFunction = z.custom((value) => typeof value === 'function', {
+    message: 'Invalid input: expected function',
+});
+/**
+ * Mirrors `ow.number`: `Infinity` is a valid number, `NaN` is not.
+ * @internal
+ */
+export const anyNumber = z.custom((value) => typeof value === 'number' && !Number.isNaN(value), {
+    message: 'Invalid input: expected number',
+});
+/**
+ * Accepts any object (including functions) that has all the given keys, own or inherited.
+ * @internal
+ */
+export function objectWithKeys(keys, message) {
+    return z.custom((value) => ((typeof value === 'object' && value !== null) || typeof value === 'function') &&
+        keys.every((key) => key in value), {
+        message: message ?? `Invalid input: expected an object with keys ${keys.map((key) => `'${key}'`).join(', ')}`,
+    });
+}
+/**
+ * Accepts only instances of {@link BaseHttpClient} (all Crawlee HTTP clients extend it).
+ * @internal
+ */
+export const httpClient = z.instanceof(BaseHttpClient);
+/**
+ * Accepts any object implementing the CrawleeLogger interface.
+ * @internal
+ */
+export const logger = objectWithKeys(['child', 'info', 'error', 'warning'], "Expected an object implementing the CrawleeLogger interface (missing one of 'child', 'info', 'error', 'warning'), got something else.");
+/**
+ * Accepts any typed array (`Uint8Array`, `Float64Array`, ...), but not a `DataView`.
+ * @internal
+ */
+export const typedArray = z.custom((value) => ArrayBuffer.isView(value) && !(value instanceof DataView), { message: 'Invalid input: expected a typed array' });
+/**
+ * Accepts any non-null, non-array object.
+ * @internal
+ */
+export const plainObject = z.custom((value) => typeof value === 'object' && value !== null && !Array.isArray(value), { message: 'Invalid input: expected an object' });
+/**
+ * Shape of a request stored in a request queue.
+ * @internal
+ */
+export const storageRequest = z.looseObject({
+    id: z.string(),
+    url: z.url({ protocol: /^https?$/ }),
+    uniqueKey: z.string(),
+    method: z.string().optional(),
+    retryCount: z.number().int().optional(),
+    handledAt: z.union([z.string(), z.date()]).optional(),
+});
+/**
+ * {@link storageRequest} before an id is assigned.
+ * @internal
+ */
+export const storageRequestWithoutId = storageRequest.omit({ id: true });
+/**
+ * `z.array(item)` whose top-level type error names the element type — ``expected an array of numbers`` —
+ * instead of zod's bare `expected array`. Element failures keep zod's per-index messages, and `elements`
+ * is a human-readable plural (`'numbers'`, `'URL patterns'`), since element types cannot be introspected.
+ * @internal
+ */
+export function arrayOf(item, elements) {
+    return z.array(item, {
+        error: (issue) => issue.code === 'invalid_type' ? `Invalid input: expected an array of ${elements}` : undefined,
+    });
+}
+/**
+ * Batch of {@link storageRequestWithoutId}.
+ * @internal
+ */
+export const storageRequestBatch = arrayOf(storageRequestWithoutId, 'requests');
+/**
+ * Options of request queue add/update operations.
+ * @internal
+ */
+export const requestQueueOperationOptions = z.object({
+    forefront: z.boolean().optional(),
+});
+/**
+ * Options of key-value store `listKeys`.
+ * @internal
+ */
+export const keyValueStoreListKeysOptions = z.object({
+    prefix: z.string().optional(),
+    exclusiveStartKey: z.string().optional(),
+    limit: z.number().int().gt(0).optional(),
+});
+/**
+ * Options of dataset item listing.
+ * @internal
+ */
+export const datasetListItemsOptions = z.object({
+    desc: z.boolean().optional(),
+    limit: z.number().int().optional(),
+    offset: z.number().int().optional(),
+});

package/internals/sitemap.d.ts

@@ -1,5 +1,6 @@
-// @ts-ignore optional peer dependency or compatibility with es2022
-import type { Delays } from 'got-scraping';
+import type { BaseHttpClient } from '@crawlee/http-client';
+import type { CrawleeLogger } from '@crawlee/types';
+import { type EnqueueStrategy } from './url.js';
 interface SitemapUrlData {
     loc: string;
     lastmod?: Date;
@@ -36,9 +37,36 @@ export interface ParseSitemapOptions {
      */
     sitemapRetries?: number;
     /**
-     * Network timeouts for sitemap fetching. See [Got documentation](https://github.com/sindresorhus/got/blob/main/documentation/6-timeout.md) for more details.
+     * Timeout settings for network requests when fetching sitemaps. By default this is `30000` milliseconds (30 seconds).
      */
-    networkTimeouts?: Delays;
+    timeoutMillis?: number;
+    /**
+     * If true, the parser will log a warning if it fails to fetch a sitemap due to a network error
+     * @default true
+     */
+    reportNetworkErrors?: boolean;
+    /**
+     * Custom HTTP client to be used for fetching sitemaps.
+     */
+    httpClient?: BaseHttpClient;
+    /**
+     * Optional filter for nested sitemap URLs discovered in sitemap index files.
+     * Called with the URL of each child sitemap before it is fetched.
+     * Return `true` to include the sitemap, `false` to skip it.
+     * If not provided, all nested sitemaps are followed.
+     */
+    nestedSitemapFilter?: (sitemapUrl: string) => boolean;
+    /**
+     * Keep only sitemap-derived URLs (nested `<sitemap>` and `<url>` entries) matching this strategy
+     * relative to the parent sitemap URL; non-`http(s)` schemes are always dropped. Skipped for raw string
+     * sources (no parent URL). Pass `'all'` to disable host filtering.
+     * @default 'same-hostname'
+     */
+    enqueueStrategy?: EnqueueStrategy | `${EnqueueStrategy}`;
+    /**
+     * Optional logger for reporting warnings during sitemap parsing.
+     */
+    logger?: CrawleeLogger;
 }
 export declare function parseSitemap<T extends ParseSitemapOptions>(initialSources: SitemapSource[], proxyUrl?: string, options?: T): AsyncIterable<T['emitNestedSitemaps'] extends true ? SitemapUrl | NestedSitemap : SitemapUrl>;
 /**
@@ -62,7 +90,7 @@ export declare class Sitemap {
      * @param url The domain URL to fetch the sitemap for.
      * @param proxyUrl A proxy to be used for fetching the sitemap file.
      */
-    static tryCommonNames(url: string, proxyUrl?: string): Promise<Sitemap>;
+    static tryCommonNames(url: string, proxyUrl?: string, parseSitemapOptions?: ParseSitemapOptions): Promise<Sitemap>;
     /**
      * Fetch sitemap content from given URL or URLs and return URLs of referenced pages.
      * @param urls sitemap URL(s)
@@ -74,8 +102,46 @@ export declare class Sitemap {
      * @param content XML sitemap content
      * @param proxyUrl URL of a proxy to be used for fetching sitemap contents
      */
-    static fromXmlString(content: string, proxyUrl?: string): Promise<Sitemap>;
-    protected static parse(sources: SitemapSource[], proxyUrl?: string, parseSitemapOptions?: ParseSitemapOptions): Promise<Sitemap>;
+    static fromXmlString(content: string, proxyUrl?: string, parseSitemapOptions?: ParseSitemapOptions): Promise<Sitemap>;
+    private static parse;
 }
+/**
+ * Given a list of URLs, discover related sitemap files for these domains by checking the `robots.txt` file,
+ * the default `sitemap.xml` & `sitemap.txt` files and the URLs themselves.
+ * @param `urls` The list of URLs to discover sitemaps for.
+ * @param `options` Options for sitemap discovery
+ * @returns An async iterable with the discovered sitemap URLs.
+ */
+export declare function discoverValidSitemaps(urls: string[], options?: {
+    /**
+     * Proxy URL to be used for network requests.
+     */
+    proxyUrl?: string;
+    /**
+     * Timeout in milliseconds for the entire `discoverValidSitemaps` call.
+     * An `AbortController` is created internally and its signal is passed to every HTTP request,
+     * so the whole discovery operation is cancelled once the timeout elapses.
+     * Defaults to `60_000` ms (60 seconds) to prevent indefinite hangs.
+     */
+    timeoutMillis?: number;
+    /**
+     * An external `AbortSignal` to cancel the entire discovery operation.
+     * If both `signal` and `timeout` are provided, the operation is cancelled
+     * when either the signal is aborted or the timeout elapses (whichever comes first).
+     */
+    signal?: AbortSignal;
+    /**
+     * Timeout in milliseconds for each individual HTTP request during discovery.
+     * Defaults to `20000` ms (20 seconds).
+     */
+    requestTimeoutMillis?: number;
+    /**
+     * HTTP client to be used for network requests.
+     */
+    httpClient?: BaseHttpClient;
+    /**
+     * Optional logger for reporting warnings during sitemap discovery.
+     */
+    logger?: CrawleeLogger;
+}): AsyncIterable<string>;
 export {};
-//# sourceMappingURL=sitemap.d.ts.map