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

package/enqueue_links/enqueue_links.js

@@ -1,271 +0,0 @@
-import ow from 'ow';
-import { getDomain } from 'tldts';
-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.
- *
- * 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 var EnqueueStrategy;
-(function (EnqueueStrategy) {
-    /**
-     * Matches any URLs found
-     */
-    EnqueueStrategy["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.
-     */
-    EnqueueStrategy["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.
-     */
-    EnqueueStrategy["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.
-     */
-    EnqueueStrategy["SameOrigin"] = "same-origin";
-})(EnqueueStrategy || (EnqueueStrategy = {}));
-/**
- * This function enqueues the urls provided to the {@link RequestQueue} provided. If you want to automatically find and enqueue links,
- * you should use the context-aware `enqueueLinks` function provided on the crawler contexts.
- *
- * Optionally, the function allows you to filter the target links' URLs using an array of globs or regular expressions
- * and override settings of the enqueued {@link Request} objects.
- *
- * **Example usage**
- *
- * ```javascript
- * await enqueueLinks({
- *   urls: aListOfFoundUrls,
- *   requestManager,
- *   selector: 'a.product-detail',
- *   globs: [
- *       'https://www.example.com/handbags/*',
- *       'https://www.example.com/purses/*'
- *   ],
- * });
- * ```
- *
- * @param options All `enqueueLinks()` parameters are passed via an options object.
- * @returns Promise that resolves to {@link BatchAddRequestsResult} object.
- */
-export async function enqueueLinks(options) {
-    if (!options || Object.keys(options).length === 0) {
-        throw new RangeError([
-            'enqueueLinks() was called without the required options. You can only do that when you use the `crawlingContext.enqueueLinks()` method in request handlers.',
-            'Check out our guide on how to use enqueueLinks() here: https://crawlee.dev/js/docs/examples/crawl-relative-links',
-        ].join('\n'));
-    }
-    ow(options, ow.object.exactShape({
-        urls: ow.array.ofType(ow.string),
-        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,
-        userData: ow.optional.object,
-        label: ow.optional.string,
-        pseudoUrls: ow.optional.array.ofType(ow.any(ow.string, ow.object.hasKeys('purl'))),
-        globs: ow.optional.array.ofType(ow.any(ow.string, ow.object.hasKeys('glob'))),
-        exclude: ow.optional.array.ofType(ow.any(ow.string, ow.regExp, ow.object.hasKeys('glob'), ow.object.hasKeys('regexp'))),
-        regexps: ow.optional.array.ofType(ow.any(ow.regExp, ow.object.hasKeys('regexp'))),
-        transformRequestFunction: ow.optional.function,
-        strategy: ow.optional.string.oneOf(Object.values(EnqueueStrategy)),
-        waitForAllRequestsToBeAdded: ow.optional.boolean,
-    }));
-    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) {
-        for (const excl of exclude) {
-            if (typeof excl === 'string' || 'glob' in excl) {
-                urlExcludePatternObjects.push(...constructGlobObjectsFromGlobs([excl]));
-            }
-            else if (excl instanceof RegExp || 'regexp' in excl) {
-                urlExcludePatternObjects.push(...constructRegExpObjectsFromRegExps([excl]));
-            }
-        }
-    }
-    if (pseudoUrls?.length) {
-        serviceLocator.getLogger().deprecated('`pseudoUrls` option is deprecated, use `globs` or `regexps` instead');
-        urlPatternObjects.push(...constructRegExpObjectsFromPseudoUrls(pseudoUrls));
-    }
-    if (globs?.length) {
-        urlPatternObjects.push(...constructGlobObjectsFromGlobs(globs));
-    }
-    if (regexps?.length) {
-        urlPatternObjects.push(...constructRegExpObjectsFromRegExps(regexps));
-    }
-    if (!urlPatternObjects.length) {
-        options.strategy ??= EnqueueStrategy.SameHostname;
-    }
-    const enqueueStrategyPatterns = [];
-    if (options.baseUrl) {
-        const url = new URL(options.baseUrl);
-        switch (options.strategy) {
-            case EnqueueStrategy.SameHostname:
-                // We need to get the origin of the passed in domain in the event someone sets baseUrl
-                // to an url like https://example.com/deep/default/path and one of the found urls is an
-                // absolute relative path (/path/to/page)
-                enqueueStrategyPatterns.push({ glob: ignoreHttpSchema(`${url.origin}/**`) });
-                break;
-            case EnqueueStrategy.SameDomain: {
-                // Get the actual hostname from the base url
-                const baseUrlHostname = getDomain(url.hostname, { mixedInputs: false });
-                if (baseUrlHostname) {
-                    // We have a hostname, so we can use it to match all links on the page that point to it and any subdomains of it
-                    url.hostname = baseUrlHostname;
-                    enqueueStrategyPatterns.push({ glob: ignoreHttpSchema(`${url.origin.replace(baseUrlHostname, `*.${baseUrlHostname}`)}/**`) }, { glob: ignoreHttpSchema(`${url.origin}/**`) });
-                }
-                else {
-                    // We don't have a hostname (can happen for ips for instance), so reproduce the same behavior
-                    // as SameDomainAndSubdomain
-                    enqueueStrategyPatterns.push({ glob: ignoreHttpSchema(`${url.origin}/**`) });
-                }
-                break;
-            }
-            case EnqueueStrategy.SameOrigin: {
-                // The same behavior as SameHostname, but respecting the protocol of the URL
-                enqueueStrategyPatterns.push({ glob: `${url.origin}/**` });
-                break;
-            }
-            case EnqueueStrategy.All:
-            default:
-                enqueueStrategyPatterns.push({ glob: `http{s,}://**` });
-                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 && options.respectRobotsTxtFile !== false) {
-        const robotsUserAgent = typeof options.respectRobotsTxtFile === 'object' ? (options.respectRobotsTxtFile.userAgent ?? '*') : '*';
-        const skippedRequests = [];
-        requestOptions = requestOptions.filter((request) => {
-            if (robotsTxtFile.isAllowed(request.url, robotsUserAgent)) {
-                return true;
-            }
-            skippedRequests.push(request);
-            return false;
-        });
-        await reportSkippedRequests(skippedRequests, 'robotsTxt');
-    }
-    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) {
-            filteredOptions = filterRequestOptionsByPatterns(requestOptions, enqueueStrategyPatterns.length > 0 ? enqueueStrategyPatterns : undefined, urlExcludePatternObjects, options.strategy, (url) => skippedRequests.push(url));
-        }
-        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));
-    }
-    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: [] };
-}
-/**
- * @internal
- * This method helps resolve the baseUrl that will be used for filtering in {@link enqueueLinks}.
- * - If a user provides a base url, we always return it
- * - If a user specifies {@link EnqueueStrategy.All} strategy, they do not care if the newly found urls are on the original
- *   request domain, or a redirected one
- * - In all other cases, we return the domain of the original request as that's the one we need to use for filtering
- */
-export function resolveBaseUrlForEnqueueLinksFiltering({ enqueueStrategy, finalRequestUrl, originalRequestUrl, userProvidedBaseUrl, }) {
-    // User provided base url takes priority
-    if (userProvidedBaseUrl) {
-        return userProvidedBaseUrl;
-    }
-    const originalUrlOrigin = new URL(originalRequestUrl).origin;
-    const finalUrlOrigin = new URL(finalRequestUrl ?? originalRequestUrl).origin;
-    // We can assume users want to go off the domain in this case
-    if (enqueueStrategy === EnqueueStrategy.All) {
-        return finalUrlOrigin;
-    }
-    // If the user wants to ensure the same domain is accessed, regardless of subdomains, we check to ensure the domains match
-    // Returning undefined here is intentional! If the domains don't match, having no baseUrl in enqueueLinks will cause it to not enqueue anything
-    // which is the intended behavior (since we went off domain)
-    if (enqueueStrategy === EnqueueStrategy.SameDomain) {
-        const originalHostname = getDomain(originalUrlOrigin, { mixedInputs: false });
-        const finalHostname = getDomain(finalUrlOrigin, { mixedInputs: false });
-        if (originalHostname === finalHostname) {
-            return finalUrlOrigin;
-        }
-        return undefined;
-    }
-    // Always enqueue urls that are from the same origin in all other cases, as the filtering happens on the original request url, even if there was a redirect
-    // before actually finding the urls
-    return originalUrlOrigin;
-}
-/**
- * Internal function that changes the enqueue globs to match both http and https
- */
-function ignoreHttpSchema(pattern) {
-    return pattern.replace(/^(https?):\/\//, 'http{s,}://');
-}

package/enqueue_links/index.d.ts

@@ -1,2 +0,0 @@
-export * from './enqueue_links.js';
-export * from './shared.js';

package/enqueue_links/index.js

@@ -1,2 +0,0 @@
-export * from './enqueue_links.js';
-export * from './shared.js';

package/enqueue_links/shared.d.ts

@@ -1,83 +0,0 @@
-import type { Awaitable } from '@crawlee/types';
-import type { RequestOptions } from '../request.js';
-import type { EnqueueLinksOptions } from './enqueue_links.js';
-export { tryAbsoluteURL } from '@crawlee/utils';
-export type UrlPatternObject = {
-    glob?: string;
-    regexp?: RegExp;
-} & Pick<RequestOptions, 'method' | 'payload' | 'label' | 'userData' | 'headers'>;
-export type PseudoUrlObject = {
-    purl: string;
-} & Pick<RequestOptions, 'method' | 'payload' | 'label' | 'userData' | 'headers'>;
-export type PseudoUrlInput = string | PseudoUrlObject;
-export type GlobObject = {
-    glob: string;
-} & Pick<RequestOptions, 'method' | 'payload' | 'label' | 'userData' | 'headers'>;
-export type GlobInput = string | GlobObject;
-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
- */
-export declare function updateEnqueueLinksPatternCache(item: GlobInput | RegExpInput | PseudoUrlInput, pattern: RegExpObject | GlobObject): void;
-/**
- * Helper factory used in the `enqueueLinks()` and enqueueLinksByClickingElements() function
- * to construct RegExps from PseudoUrl strings.
- * @ignore
- */
-export declare function constructRegExpObjectsFromPseudoUrls(pseudoUrls: readonly PseudoUrlInput[]): RegExpObject[];
-/**
- * Helper factory used in the `enqueueLinks()` and enqueueLinksByClickingElements() function
- * to construct Glob objects from Glob pattern strings.
- * @ignore
- */
-export declare function constructGlobObjectsFromGlobs(globs: readonly GlobInput[]): GlobObject[];
-/**
- * @internal
- */
-export declare function validateGlobPattern(glob: string): string;
-/**
- * Helper factory used in the `enqueueLinks()` and enqueueLinksByClickingElements() function
- * to check RegExps input and return valid RegExps.
- * @ignore
- */
-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 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' | 'sessionId' | 'strategy'>): RequestOptions[];
-/**
- * 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, `'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 | 'skip' | 'unchanged';
-}
-/**
- * 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

@@ -1,221 +0,0 @@
-import { URL } from 'node:url';
-import { Minimatch } from 'minimatch';
-import { purlToRegExp } from '@apify/pseudo_url';
-export { tryAbsoluteURL } from '@crawlee/utils';
-const MAX_ENQUEUE_LINKS_CACHE_SIZE = 1000;
-/**
- * To enable direct use of the Actor UI `globs`/`regexps`/`pseudoUrls` output while keeping high performance,
- * all the regexps from the output are only constructed once and kept in a cache
- * by the `enqueueLinks()` function.
- * @ignore
- */
-const enqueueLinksPatternCache = new Map();
-/**
- * @ignore
- */
-export function updateEnqueueLinksPatternCache(item, pattern) {
-    enqueueLinksPatternCache.set(item, pattern);
-    if (enqueueLinksPatternCache.size > MAX_ENQUEUE_LINKS_CACHE_SIZE) {
-        const key = enqueueLinksPatternCache.keys().next().value;
-        enqueueLinksPatternCache.delete(key);
-    }
-}
-/**
- * Helper factory used in the `enqueueLinks()` and enqueueLinksByClickingElements() function
- * to construct RegExps from PseudoUrl strings.
- * @ignore
- */
-export function constructRegExpObjectsFromPseudoUrls(pseudoUrls) {
-    return pseudoUrls.map((item) => {
-        // Get pseudoUrl object from cache.
-        let regexpObject = enqueueLinksPatternCache.get(item);
-        if (regexpObject)
-            return regexpObject;
-        if (typeof item === 'string') {
-            regexpObject = { regexp: purlToRegExp(item) };
-        }
-        else {
-            const { purl, ...requestOptions } = item;
-            regexpObject = { regexp: purlToRegExp(purl), ...requestOptions };
-        }
-        updateEnqueueLinksPatternCache(item, regexpObject);
-        return regexpObject;
-    });
-}
-/**
- * Helper factory used in the `enqueueLinks()` and enqueueLinksByClickingElements() function
- * to construct Glob objects from Glob pattern strings.
- * @ignore
- */
-export function constructGlobObjectsFromGlobs(globs) {
-    return globs
-        .filter((glob) => {
-        // Skip possibly nullish, empty strings
-        if (!glob) {
-            return false;
-        }
-        if (typeof glob === 'string') {
-            return glob.trim().length > 0;
-        }
-        if (glob.glob) {
-            return glob.glob.trim().length > 0;
-        }
-        return false;
-    })
-        .map((item) => {
-        // Get glob object from cache.
-        let globObject = enqueueLinksPatternCache.get(item);
-        if (globObject)
-            return globObject;
-        if (typeof item === 'string') {
-            globObject = { glob: validateGlobPattern(item) };
-        }
-        else {
-            const { glob, ...requestOptions } = item;
-            globObject = { glob: validateGlobPattern(glob), ...requestOptions };
-        }
-        updateEnqueueLinksPatternCache(item, globObject);
-        return globObject;
-    });
-}
-/**
- * @internal
- */
-export function validateGlobPattern(glob) {
-    const globTrimmed = glob.trim();
-    if (globTrimmed.length === 0)
-        throw new Error(`Cannot parse Glob pattern '${globTrimmed}': it must be an non-empty string`);
-    return globTrimmed;
-}
-/**
- * Helper factory used in the `enqueueLinks()` and enqueueLinksByClickingElements() function
- * to check RegExps input and return valid RegExps.
- * @ignore
- */
-export function constructRegExpObjectsFromRegExps(regexps) {
-    return regexps.map((item) => {
-        // Get regexp object from cache.
-        let regexpObject = enqueueLinksPatternCache.get(item);
-        if (regexpObject)
-            return regexpObject;
-        if (item instanceof RegExp) {
-            regexpObject = { regexp: item };
-        }
-        else {
-            regexpObject = item;
-        }
-        updateEnqueueLinksPatternCache(item, regexpObject);
-        return 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 function filterRequestOptionsByPatterns(requestOptions, includePatterns, excludePatterns = [], strategy, onSkippedUrl) {
-    const excludeMatchers = excludePatterns.map(createPatternObjectMatcher);
-    const includeMatchers = includePatterns?.length ? includePatterns.map(createPatternObjectMatcher) : undefined;
-    return requestOptions
-        .filter(({ url }) => {
-        const matchesExclude = excludeMatchers.some(({ match }) => match(url));
-        if (matchesExclude) {
-            onSkippedUrl?.(url);
-        }
-        return !matchesExclude;
-    })
-        .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((opts) => opts !== null);
-}
-/**
- * @ignore
- */
-export function createRequestOptions(sources, options = {}) {
-    return sources
-        .map((src) => typeof src === 'string'
-        ? { url: src, enqueueStrategy: options.strategy }
-        : { ...src, enqueueStrategy: options.strategy })
-        .filter(({ url }) => {
-        try {
-            return new URL(url, options.baseUrl).href;
-        }
-        catch (err) {
-            return false;
-        }
-    })
-        .map((requestOptions) => {
-        requestOptions.url = new URL(requestOptions.url, options.baseUrl).href;
-        requestOptions.userData ??= options.userData ?? {};
-        if (typeof options.label === 'string') {
-            requestOptions.userData = {
-                ...requestOptions.userData,
-                label: options.label,
-            };
-        }
-        if (options.skipNavigation) {
-            requestOptions.skipNavigation = true;
-        }
-        if (options.sessionId) {
-            requestOptions.sessionId = options.sessionId;
-        }
-        return requestOptions;
-    });
-}
-/**
- * @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);
-}