@crawlee/utils 4.0.0-beta.99 → 4.0.0-rc.1

package/internals/schemas.js

@@ -0,0 +1,115 @@
+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' });
+const storageRequestShape = 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(),
+});
+/**
+ * Shape of a request stored in a request queue. Compiled: storage clients validate every request with it.
+ * @internal
+ */
+export const storageRequest = z.compile(storageRequestShape);
+/**
+ * {@link storageRequest} before an id is assigned.
+ * @internal
+ */
+export const storageRequestWithoutId = z.compile(storageRequestShape.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 = z.compile(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

@@ -1,4 +1,6 @@
-import type { BaseHttpClient, CrawleeLogger } from '@crawlee/types';
+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;
@@ -54,6 +56,13 @@ export interface ParseSitemapOptions {
      * 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.
      */
@@ -73,6 +82,7 @@ export declare function parseSitemap<T extends ParseSitemapOptions>(initialSourc
  * ```
  */
 export declare class Sitemap {
+    #private;
     readonly urls: string[];
     constructor(urls: string[]);
     /**
@@ -94,7 +104,6 @@ export declare class Sitemap {
      * @param proxyUrl URL of a proxy to be used for fetching sitemap contents
      */
     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,

package/internals/sitemap.js

@@ -3,31 +3,30 @@ import { PassThrough, pipeline, Readable, Transform } from 'node:stream';
 import { StringDecoder } from 'node:string_decoder';
 import { createGunzip } from 'node:zlib';
 import { FetchHttpClient } from '@crawlee/http-client';
-import { fileTypeStream } from 'file-type';
-import sax from 'sax';
-import MIMEType from 'whatwg-mimetype';
+import { MIMEType } from 'whatwg-mimetype';
 import { mergeAsyncIterables } from './iterables.js';
 import { RobotsTxtFile } from './robots.js';
+import { filterUrl } from './url.js';
 class SitemapTxtParser extends Transform {
-    decoder = new StringDecoder('utf8');
-    buffer = '';
+    #decoder = new StringDecoder('utf8');
+    #buffer = '';
     constructor() {
         super({
             readableObjectMode: true,
             transform: (chunk, _encoding, callback) => {
-                this.processBuffer(this.decoder.write(chunk), false);
+                this.#processBuffer(this.#decoder.write(chunk), false);
                 callback();
             },
             flush: (callback) => {
-                this.processBuffer(this.decoder.end(), true);
+                this.#processBuffer(this.#decoder.end(), true);
                 callback();
             },
         });
     }
-    processBuffer(input, finalize) {
-        this.buffer += input;
-        if (finalize || this.buffer.includes('\n')) {
-            const parts = this.buffer
+    #processBuffer(input, finalize) {
+        this.#buffer += input;
+        if (finalize || this.#buffer.includes('\n')) {
+            const parts = this.#buffer
                 .split('\n')
                 .map((part) => part.trim())
                 .filter((part) => part.length > 0);
@@ -35,104 +34,109 @@ class SitemapTxtParser extends Transform {
                 for (const url of parts) {
                     this.push({ type: 'url', loc: url });
                 }
-                this.buffer = '';
+                this.#buffer = '';
             }
             else if (parts.length > 0) {
                 for (const url of parts.slice(0, -1)) {
                     this.push({ type: 'url', loc: url });
                 }
-                this.buffer = parts.at(-1);
+                this.#buffer = parts.at(-1);
             }
         }
     }
 }
 class SitemapXmlParser extends Transform {
-    decoder = new StringDecoder('utf8');
-    parser = new sax.SAXParser(true);
-    rootTagName;
-    currentTag = undefined;
-    url = {};
-    constructor() {
+    #decoder = new StringDecoder('utf8');
+    #parser;
+    #rootTagName;
+    #currentTag = undefined;
+    #url = {};
+    static async create() {
+        const { SAXParser } = (await import('sax')).default;
+        return new SitemapXmlParser(new SAXParser(true));
+    }
+    constructor(parser) {
         super({
             readableObjectMode: true,
             transform: (chunk, _encoding, callback) => {
-                this.parser.write(this.decoder.write(chunk));
+                this.#parser.write(this.#decoder.write(chunk));
                 callback();
             },
             flush: (callback) => {
-                const rest = this.decoder.end();
+                const rest = this.#decoder.end();
                 if (rest.length > 0) {
-                    this.parser.write(rest);
+                    this.#parser.write(rest);
                 }
-                this.parser.end();
+                this.#parser.end();
                 callback();
             },
         });
-        this.parser.onopentag = this.onOpenTag.bind(this);
-        this.parser.onclosetag = this.onCloseTag.bind(this);
-        this.parser.ontext = this.onText.bind(this);
-        this.parser.oncdata = this.onText.bind(this);
-        this.parser.onerror = this.destroy.bind(this);
+        this.#parser = parser;
+        this.#parser.onopentag = this.#onOpenTag.bind(this);
+        this.#parser.onclosetag = this.#onCloseTag.bind(this);
+        this.#parser.ontext = this.#onText.bind(this);
+        this.#parser.oncdata = this.#onText.bind(this);
+        this.#parser.onerror = this.destroy.bind(this);
     }
-    onOpenTag(node) {
-        if (this.rootTagName !== undefined) {
+    #onOpenTag(node) {
+        if (this.#rootTagName !== undefined) {
             if (node.name === 'loc' ||
                 node.name === 'lastmod' ||
                 node.name === 'priority' ||
                 node.name === 'changefreq') {
-                this.currentTag = node.name;
+                this.#currentTag = node.name;
             }
         }
         if (node.name === 'urlset') {
-            this.rootTagName = 'urlset';
+            this.#rootTagName = 'urlset';
         }
         if (node.name === 'sitemapindex') {
-            this.rootTagName = 'sitemapindex';
+            this.#rootTagName = 'sitemapindex';
         }
     }
-    onCloseTag(name) {
+    #onCloseTag(name) {
         if (name === 'loc' || name === 'lastmod' || name === 'priority' || name === 'changefreq') {
-            this.currentTag = undefined;
+            this.#currentTag = undefined;
         }
         if (name === 'url') {
-            if (this.url.loc !== undefined) {
-                this.push({ type: 'url', ...this.url, loc: this.url.loc });
+            if (this.#url.loc !== undefined) {
+                this.push({ type: 'url', ...this.#url, loc: this.#url.loc });
             }
-            this.url = {};
+            this.#url = {};
         }
     }
-    onText(text) {
-        if (this.currentTag === 'loc') {
-            if (this.rootTagName === 'sitemapindex') {
+    #onText(text) {
+        if (this.#currentTag === 'loc') {
+            if (this.#rootTagName === 'sitemapindex') {
                 this.push({ type: 'sitemapUrl', url: text.trim() });
             }
-            if (this.rootTagName === 'urlset') {
-                this.url ??= {};
-                this.url.loc = text.trim();
+            if (this.#rootTagName === 'urlset') {
+                this.#url ??= {};
+                this.#url.loc = text.trim();
             }
         }
         text = text.trim();
-        if (this.currentTag === 'lastmod') {
+        if (this.#currentTag === 'lastmod') {
             const lastmod = new Date(text);
             if (!Number.isNaN(lastmod.getTime())) {
-                this.url.lastmod = lastmod;
+                this.#url.lastmod = lastmod;
             }
         }
-        if (this.currentTag === 'priority') {
-            this.url.priority = Number(text);
+        if (this.#currentTag === 'priority') {
+            this.#url.priority = Number(text);
         }
-        if (this.currentTag === 'changefreq') {
+        if (this.#currentTag === 'changefreq') {
             if (['always', 'hourly', 'daily', 'weekly', 'monthly', 'yearly', 'never'].includes(text)) {
-                this.url.changefreq = text;
+                this.#url.changefreq = text;
             }
         }
     }
 }
 export async function* parseSitemap(initialSources, proxyUrl, options) {
-    const { httpClient = new FetchHttpClient(), emitNestedSitemaps = false, maxDepth = Infinity, sitemapRetries = 3, timeoutMillis: timeout = 30000, reportNetworkErrors = true, nestedSitemapFilter, logger, } = options ?? {};
+    const { httpClient = new FetchHttpClient(), emitNestedSitemaps = false, maxDepth = Infinity, sitemapRetries = 3, timeoutMillis: timeout = 30000, reportNetworkErrors = true, nestedSitemapFilter, enqueueStrategy = 'same-hostname', logger, } = options ?? {};
     const sources = [...initialSources];
     const visitedSitemapUrls = new Set();
-    const createParser = (contentType = '', url) => {
+    const createParser = async (contentType = '', url) => {
         let mimeType;
         try {
             mimeType = new MIMEType(contentType);
@@ -141,7 +145,7 @@ export async function* parseSitemap(initialSources, proxyUrl, options) {
             mimeType = null;
         }
         if (mimeType?.isXML() || url?.pathname.endsWith('.xml')) {
-            return new SitemapXmlParser();
+            return SitemapXmlParser.create();
         }
         if (mimeType?.essence === 'text/plain' || url?.pathname.endsWith('.txt')) {
             return new SitemapTxtParser();
@@ -154,8 +158,10 @@ export async function* parseSitemap(initialSources, proxyUrl, options) {
             continue;
         }
         let items = null;
+        // Parent URL, parsed once and reused as the origin for the strategy checks below.
+        let sitemapUrl;
         if (source.type === 'url') {
-            const sitemapUrl = new URL(source.url);
+            sitemapUrl = new URL(source.url);
             visitedSitemapUrls.add(sitemapUrl.toString());
             let retriesLeft = sitemapRetries + 1;
             while (retriesLeft-- > 0) {
@@ -181,6 +187,7 @@ export async function* parseSitemap(initialSources, proxyUrl, options) {
                         if (sitemapResponse.body === null) {
                             break;
                         }
+                        const { fileTypeStream } = await import('file-type');
                         const streamWithType = await fileTypeStream(Readable.fromWeb(sitemapResponse.body));
                         if (streamWithType.fileType !== undefined) {
                             contentType = streamWithType.fileType.mime;
@@ -194,7 +201,7 @@ export async function* parseSitemap(initialSources, proxyUrl, options) {
                                 sitemapUrl.pathname = sitemapUrl.pathname.substring(0, sitemapUrl.pathname.length - 3);
                             }
                         }
-                        items = pipeline(streamWithType, isGzipped ? createGunzip() : new PassThrough(), createParser(contentType ?? undefined, sitemapUrl), (e) => {
+                        items = pipeline(streamWithType, isGzipped ? createGunzip() : new PassThrough(), await createParser(contentType ?? undefined, sitemapUrl), (e) => {
                             if (e !== undefined && e !== null) {
                                 error = { type: 'parser', error: e };
                             }
@@ -222,7 +229,7 @@ export async function* parseSitemap(initialSources, proxyUrl, options) {
             }
         }
         else if (source.type === 'raw') {
-            items = pipeline(Readable.from([source.content]), createParser('text/xml'), (error) => {
+            items = pipeline(Readable.from([source.content]), await createParser('text/xml'), (error) => {
                 if (error !== undefined) {
                     logger?.warning(`Malformed sitemap content: ${error}`);
                 }
@@ -231,18 +238,39 @@ export async function* parseSitemap(initialSources, proxyUrl, options) {
         if (items === null) {
             continue;
         }
+        // URL entries dropped by the enqueue strategy filter, reported in one warning per sitemap after
+        // the loop (per-entry warnings could flood the log; individual drops are logged at debug level).
+        let droppedUrlEntries = 0;
         for await (const item of items) {
             if (item.type === 'sitemapUrl' && !visitedSitemapUrls.has(item.url)) {
                 if (nestedSitemapFilter && !nestedSitemapFilter(item.url)) {
                     logger?.debug(`Skipping sitemap ${item.url} due to nestedSitemapFilter.`);
                     continue;
                 }
+                // Keep only nested sitemaps matching the strategy (and using http(s)) relative to the
+                // parent. Raw string sources have no parent URL, so the check is skipped.
+                if (source.type === 'url') {
+                    const { allowed, reason } = filterUrl(item.url, sitemapUrl, enqueueStrategy);
+                    if (!allowed) {
+                        logger?.warning(`Skipping nested sitemap ${item.url} (parent ${source.url}): ${reason}.`);
+                        continue;
+                    }
+                }
                 sources.push({ type: 'url', url: item.url, depth: (source.depth ?? 0) + 1 });
                 if (emitNestedSitemaps) {
                     yield { loc: item.url, originSitemapUrl: null };
                 }
             }
             if (item.type === 'url') {
+                // Keep only URL entries that match the enqueue strategy relative to the parent (see above).
+                if (source.type === 'url') {
+                    const { allowed, reason } = filterUrl(item.loc, sitemapUrl, enqueueStrategy);
+                    if (!allowed) {
+                        droppedUrlEntries++;
+                        logger?.debug(`Skipping sitemap URL ${item.loc} (parent ${source.url}): ${reason}.`);
+                        continue;
+                    }
+                }
                 yield {
                     ...item,
                     originSitemapUrl: source.type === 'url'
@@ -251,6 +279,9 @@ export async function* parseSitemap(initialSources, proxyUrl, options) {
                 };
             }
         }
+        if (droppedUrlEntries > 0 && source.type === 'url') {
+            logger?.warning(`Skipped ${droppedUrlEntries} URL(s) from sitemap ${source.url} not matching enqueue strategy '${enqueueStrategy}' (or using a non-http(s) scheme). Enable debug logs to see each skipped URL.`);
+        }
     }
 }
 /**
@@ -292,7 +323,7 @@ export class Sitemap {
      * @param proxyUrl URL of a proxy to be used for fetching sitemap contents
      */
     static async load(urls, proxyUrl, parseSitemapOptions) {
-        return await this.parse((Array.isArray(urls) ? urls : [urls]).map((url) => ({ type: 'url', url })), proxyUrl, parseSitemapOptions);
+        return await this.#parse((Array.isArray(urls) ? urls : [urls]).map((url) => ({ type: 'url', url })), proxyUrl, parseSitemapOptions);
     }
     /**
      * Parse XML sitemap content from a string and return URLs of referenced pages. If the sitemap references other sitemaps, they will be loaded via HTTP.
@@ -300,9 +331,9 @@ export class Sitemap {
      * @param proxyUrl URL of a proxy to be used for fetching sitemap contents
      */
     static async fromXmlString(content, proxyUrl, parseSitemapOptions) {
-        return await this.parse([{ type: 'raw', content }], proxyUrl, parseSitemapOptions);
+        return await this.#parse([{ type: 'raw', content }], proxyUrl, parseSitemapOptions);
     }
-    static async parse(sources, proxyUrl, parseSitemapOptions) {
+    static async #parse(sources, proxyUrl, parseSitemapOptions) {
         const urls = [];
         try {
             for await (const item of parseSitemap(sources, proxyUrl, parseSitemapOptions)) {
@@ -374,7 +405,8 @@ export async function* discoverValidSitemaps(urls, options = {}) {
                 httpClient,
                 logger,
             });
-            for (const sitemapUrl of robotsFile.getSitemaps()) {
+            // Surface all referenced sitemaps, including cross-host; scoping happens at load time.
+            for (const sitemapUrl of robotsFile.getSitemaps({ enqueueStrategy: 'all' })) {
                 if (addSitemapUrl(sitemapUrl)) {
                     yield sitemapUrl;
                 }
@@ -383,7 +415,7 @@ export async function* discoverValidSitemaps(urls, options = {}) {
         catch (err) {
             logger?.warning(`Failed to fetch robots.txt file for ${hostname}`, { error: err });
         }
-        const sitemapUrl = domainUrls.find((url) => /sitemap\.(?:xml|txt)(?:\.gz)?$/i.test(url));
+        const sitemapUrl = domainUrls.find((url) => /sitemap(?:_index)?\.(?:xml|txt)(?:\.gz)?$/i.test(url));
         if (sitemapUrl !== undefined) {
             if (addSitemapUrl(sitemapUrl)) {
                 yield sitemapUrl;
@@ -391,6 +423,11 @@ export async function* discoverValidSitemaps(urls, options = {}) {
         }
         else {
             const firstUrl = new URL(domainUrls[0]);
+            // The guessed candidates are well-known paths on the same origin, so the query string
+            // and fragment of the input URL are not part of them - carrying them over probes (and
+            // yields) URLs like `/sitemap.xml?page=2`, which servers may well answer with a 404.
+            firstUrl.search = '';
+            firstUrl.hash = '';
             const possibleSitemapPathnames = ['/sitemap.xml', '/sitemap.txt', '/sitemap_index.xml'];
             const candidateSitemapUrls = possibleSitemapPathnames.map((pathname) => {
                 firstUrl.pathname = pathname;

package/internals/social.d.ts

@@ -225,7 +225,7 @@ export declare const TWITTER_REGEX: RegExp;
  * ```
  * import { social } from 'crawlee';
  *
- * const matches = text.match(social.TWITTER_REGEX_STRING);
+ * const matches = text.match(social.TWITTER_REGEX_GLOBAL);
  * if (matches) console.log(`${matches.length} Twitter profiles found!`);
  * ```
  */

package/internals/social.js

@@ -326,7 +326,7 @@ export const TWITTER_REGEX = new RegExp(`^${TWITTER_REGEX_STRING}$`, 'i');
  * ```
  * import { social } from 'crawlee';
  *
- * const matches = text.match(social.TWITTER_REGEX_STRING);
+ * const matches = text.match(social.TWITTER_REGEX_GLOBAL);
  * if (matches) console.log(`${matches.length} Twitter profiles found!`);
  * ```
  */
@@ -590,7 +590,7 @@ export const DISCORD_REGEX_GLOBAL = new RegExp(DISCORD_REGEX_STRING, 'ig');
  * @return An object with the social handles.
  */
 export async function parseHandlesFromHtml(html, data = null) {
-    const cheerio = await import('cheerio');
+    const cheerio = await import('cheerio/slim');
     const result = {
         emails: [],
         phones: [],

package/internals/url.d.ts

@@ -1,4 +1,73 @@
 import type { SearchParams } from '@crawlee/types';
+/**
+ * The different enqueueing strategies available.
+ *
+ * Depending on the strategy you select, we will only check certain parts of the URLs found. Here is a diagram of each URL part and their name:
+ *
+ * ```md
+ * Protocol          Domain
+ * ┌────┐          ┌─────────┐
+ * https://example.crawlee.dev/...
+ * │       └─────────────────┤
+ * │             Hostname    │
+ * │                         │
+ * └─────────────────────────┘
+ *          Origin
+ *```
+ *
+ * - The `Protocol` is usually `http` or `https`
+ * - The `Domain` represents the path without any possible subdomains to a website. For example, `crawlee.dev` is the domain of `https://example.crawlee.dev/`
+ * - The `Hostname` is the full path to a website, including any subdomains. For example, `example.crawlee.dev` is the hostname of `https://example.crawlee.dev/`
+ * - The `Origin` is the combination of the `Protocol` and `Hostname`. For example, `https://example.crawlee.dev` is the origin of `https://example.crawlee.dev/`
+ */
+export declare enum EnqueueStrategy {
+    /**
+     * Matches any URLs found
+     */
+    All = "all",
+    /**
+     * Matches any URLs that have the same hostname.
+     * For example, `https://wow.example.com/hello` will be matched for a base url of `https://wow.example.com/`, but
+     * `https://example.com/hello` will not be matched.
+     *
+     * > This strategy will match both `http` and `https` protocols regardless of the base URL protocol.
+     */
+    SameHostname = "same-hostname",
+    /**
+     * Matches any URLs that have the same domain as the base URL.
+     * For example, `https://wow.an.example.com` and `https://example.com` will both be matched for a base url of
+     * `https://example.com`.
+     *
+     * > This strategy will match both `http` and `https` protocols regardless of the base URL protocol.
+     */
+    SameDomain = "same-domain",
+    /**
+     * Matches any URLs that have the same hostname and protocol.
+     * For example, `https://wow.example.com/hello` will be matched for a base url of `https://wow.example.com/`, but
+     * `http://wow.example.com/hello` will not be matched.
+     *
+     * > This strategy will ensure the protocol of the base URL is the same as the protocol of the URL to be enqueued.
+     */
+    SameOrigin = "same-origin"
+}
+/** Reusable suffix for log messages explaining why a non-`http(s)` URL was rejected. */
+export declare const UNSUPPORTED_SCHEME_MESSAGE = "unsupported URL scheme (only http and https are allowed)";
+/**
+ * Check whether `target` matches `origin` under the given enqueue `strategy`. The URL scheme is not
+ * considered here (use {@link filterUrl} for the combined scheme + strategy check).
+ *
+ * The `enqueueLinks` implementation in `@crawlee/core` matches the same strategies via glob patterns
+ * (see `packages/core/src/enqueue_links/enqueue_links.ts`) — keep the two in sync when changing either.
+ */
+export declare function matchesEnqueueStrategy(strategy: EnqueueStrategy | `${EnqueueStrategy}`, target: URL, origin: URL): boolean;
+/**
+ * Check whether `target` may be enqueued under `strategy` relative to `origin`: it must use an `http(s)`
+ * scheme and match the strategy. On rejection, `reason` is a human-readable message for log output.
+ */
+export declare function filterUrl(target: string | URL, origin: string | URL, strategy: EnqueueStrategy | `${EnqueueStrategy}`): {
+    allowed: boolean;
+    reason?: string;
+};
 /**
  * Appends search (query string) parameters to a URL, replacing the original value (if any).
  *