RSSHub/lib/routes/linkedin/utils.ts

import { load } from 'cheerio';
import dayjs from 'dayjs';

import { Job } from './models';

/**
 * Constants
 */
const BASE_URL = 'https://www.linkedin.com';

const KEYWORDS_QUERY_KEY = 'keywords';

const JOB_TYPES_QUERY_KEY = 'f_JT';

const JOB_TYPES = {
    F: 'full-time',
    P: 'part-time',
    C: 'contract',
};

const EXP_LEVELS_QUERY_KEY = 'f_E';

const EXP_LEVELS = {
    1: 'internship',
    2: 'entry',
    3: 'associate',
    4: 'mid-senior',
    5: 'director',
};

/**
 * Params parsing
 */

/**
 * Parse params in route into valid query params
 * e.g. /jobs/:keywords/:job_types/:exp_levels
 *   if job_types is 'C-T'
 *   the output search param is 'C,T', which can be
 *   used as search param in query
 * @param {String} params params in route (separated by '-')
 * @param {Object} map valid value map for this param
 * @returns search params separated by ',' that can be used
 *   as search param in url
 */
function parseParamsToSearchParams(params, map) {
    if (!params) {
        return '';
    } // Handle undefined params

    const validParamValues = params.split('-').filter((v) => v in map);
    return validParamValues.join(',');
}

/**
 * Parse params in route into human readable strings
 * e.g. /jobs/:keywords/:job_types/:exp_levels
 *   if job_types is 'C-T'
 *   the output is 'contract,full-time'
 * @param {String} params params in route (separated by '-')
 * @param {Object} map valid value map for this param
 * @returns param value strings separated by ','
 */
function parseParamsToString(params, map) {
    if (!params) {
        return '';
    } // Handle undefined params

    const validParamValues = params
        .split('-')
        .filter((v) => v in map)
        .map((v) => map[v]);
    return validParamValues.join(',');
}

/**
 * HTML page parsing
 */

/**
 * Parse job search page
 * Example page: https://www.linkedin.com/jobs-guest/jobs/api/seeMoreJobPostings/search?keywords=Software%20Engineer&location=United%20States&locationId=&geoId=103644278&sortBy=R&f_TPR=&position=1&pageNum=0
 *
 * @param {String} data HTML string of job search page
 * @returns {Job[]} Array of jobs with data filled
 */
function parseJobSearch(data) {
    const $ = load(data);

    // Parse data
    const jobs = $('li')
        .toArray()
        .map((elem) => {
            const elemHtml = $(elem);
            const link = elemHtml.find('a.base-card__full-link, a.base-card--link')?.attr('href')?.split('?')[0];
            const title = elemHtml.find('h3.base-search-card__title')?.text()?.trim();
            const company = elemHtml.find('h4.base-search-card__subtitle')?.text()?.trim();
            const location = elemHtml.find('span.job-search-card__location')?.text()?.trim();
            const pubDate = elemHtml.find('time')?.attr('datetime');

            return new Job(title, link, company, location, pubDate);
        });
    return jobs;
}

/**
 * Parse job detail page
 * Example page: https://www.linkedin.com/jobs/view/software-engineer-backend-junior-at-genies-3429649821?trk=public_jobs_topcard-title
 *
 * @param {String} data HTML string of job detail page
 * @returns {Job} Job details
 */
function parseJobDetail(data) {
    const job = new Job();
    const $ = load(data);

    job.recruiter = $('a.message-the-recruiter__cta').attr('href');
    job.description = $('div.description__text description__text--rich').text();

    return job;
}

const parseRouteParam = (searchParam: string | null): string => {
    if (!searchParam || typeof searchParam !== 'string') {
        return 'all';
    }
    return encodeURIComponent(searchParam.split(',').join('-'));
};

/**
 * Parse company profile page for posts
 * Example page: https://www.linkedin.com/company/google/
 *
 * @param {Cheerio} $ HTML string of company profile page
 * @returns {Array<JSON>} Array of company posts
 */
function parseCompanyPosts($) {
    const posts = $('ul.updates__list > li')
        .toArray() // Convert the Cheerio object to a plain array
        .map((elem) => {
            const elemHtml = $(elem);
            let link = elemHtml.find('a.main-feed-card__overlay-link').attr('href');

            // Reposts don't have an overlay link — construct URL from the activity URN
            if (!link) {
                const activityUrn = elemHtml.find('article[data-activity-urn]').attr('data-activity-urn');
                if (activityUrn) {
                    link = `https://www.linkedin.com/feed/update/${activityUrn}`;
                }
            }

            const text = elemHtml.find('p.attributed-text-segment-list__content').text().trim();
            const date = parseRelativeShorthandDate(elemHtml.find('time').text().trim());

            return { link, text, date };
        });

    return posts;
}

/**
 * Parse company profile page for its name
 * Example page: https://www.linkedin.com/company/google/
 *
 * @param {Cheerio} $ HTML string of company profile page
 * @returns {String} Company name
 */
function parseCompanyName($) {
    return $('h1.top-card-layout__title').text().trim();
}

/**
 * Parse relative date shorthand string into a Date object
 *
 * @param {String} shorthand The shorthand string representing the date
 * @returns {Date|null} The parsed date or null if the format is invalid
 */
function parseRelativeShorthandDate(shorthand) {
    const match = shorthand.match(/^(\d+)([wdmyh])$/);
    if (!match) {
        return null;
    }

    const [, amount, unit] = match;
    const unitMap = {
        w: 'week',
        d: 'day',
        m: 'month',
        y: 'year',
        h: 'hour',
    };

    return dayjs().subtract(Number.parseInt(amount), unitMap[unit]);
}

export {
    BASE_URL,
    EXP_LEVELS,
    EXP_LEVELS_QUERY_KEY,
    JOB_TYPES,
    JOB_TYPES_QUERY_KEY,
    KEYWORDS_QUERY_KEY,
    parseCompanyName,
    parseCompanyPosts,
    parseJobDetail,
    parseJobSearch,
    parseParamsToSearchParams,
    parseParamsToString,
    parseRouteParam,
};