job-hoarder/lib/jazzScrape/parser.js

Job board aggregator to pull in standardized job postings from company job pages

"use strict";
var __importDefault = (this && this.__importDefault) || function (mod) {
    return (mod && mod.__esModule) ? mod : { "default": mod };
};
Object.defineProperty(exports, "__esModule", { value: true });
var cheerio_1 = __importDefault(require("cheerio"));
var utils_1 = require("../utils");
var JazzScrapeParser = /** @class */ (function () {
    function JazzScrapeParser() {
        var _this = this;
        /**
         * Parse jobs from request result
         * @param {string} data String of jobs
         * @returns {array} List of parsed jobs
         */
        this.parseJobs = function (data) {
            if (!data)
                throw new Error('No jobs to parse');
            var jobs = [];
            var $ = cheerio_1.default.load(data);
            $('.jobs-list ul li.list-group-item').each(function (i, elem) {
                var job = $(elem);
                var url = job.find('a').attr('href');
                var id = utils_1.getMatch(url, /.+apply\/(.+)\/.+/);
                var title = job.find('a').text().trim();
                var jobLocation = job.find('.fa-map-marker').parent().text();
                var department = job.find('.fa-sitemap').parent().text();
                jobs.push({
                    id: id,
                    url: url,
                    title: title,
                    jobLocation: jobLocation,
                    department: department,
                });
            });
            return jobs;
        };
        /**
         * Parsed ID from the Jazz URL
         * @param url URL to parse ID from
         */
        this.parseIdFromUrl = function (url) { return utils_1.getMatch(url, /.+apply\/(.+)\/.+/); };
        /**
         * Parses and normalizes job based on Embedded JSON object
         * @param parsedData JSON object embedded in HTML
         * @param $ Cheerio object to extract extra features
         */
        this.parseJobJSON = function (parsedData, $) {
            var title = parsedData.title, _a = parsedData.url, url = _a === void 0 ? '' : _a, loc = parsedData.jobLocation.address, pDate = parsedData.datePosted, description = parsedData.description;
            var id = _this.parseIdFromUrl(url);
            var datePosted = new Date(pDate);
            var jobLocation = loc ? loc.addressLocality + ", " + loc.addressRegion : undefined;
            var department = $('li[title="Department"]').text().trim();
            return {
                id: id,
                url: url,
                title: title,
                datePosted: datePosted,
                jobLocation: jobLocation,
                department: department,
                description: description,
            };
        };
        /**
         * Scrapes key fields from HTML
         * @param $ Cheerio object to extract features from
         */
        this.parseJobHTML = function ($) {
            var title = $('h1:not(.brand-text)').text().trim();
            var url = $('meta[property="og:url"]').attr('content');
            var id = _this.parseIdFromUrl(url);
            var jobLocation = $('li[title="Location"]').text().trim();
            var department = $('li[title="Department"]').text().trim();
            var description = $('div.description').first().html();
            return {
                id: id,
                url: url,
                title: title,
                jobLocation: jobLocation,
                department: department,
                description: description,
            };
        };
        /**
         * Parses job from request result
         * @param {string} data String of job result
         * @returns {object} Object of parsed job
         */
        this.parseJob = function (data) {
            if (!data)
                throw new Error('No job to parse');
            var $ = cheerio_1.default.load(data);
            // Check to see if the page includes a JSON object
            var parsedData = JSON.parse($('script[type="application/ld+json"]').html());
            if (parsedData && parsedData.url)
                return _this.parseJobJSON(parsedData, $);
            // Parse HTML as fallback
            return _this.parseJobHTML($);
        };
    }
    return JazzScrapeParser;
}());
exports.default = JazzScrapeParser;