Last active
March 12, 2020 15:50
-
-
Save NikolaRusakov/02e192c82a490964e52ffc1b2df6c5a9 to your computer and use it in GitHub Desktop.
Scrape of LinkedIn profile, the manual way
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
| const cheerio = require('cheerio') | |
| // TODO: feed me with selected dom element with class of .profile-detail | |
| const detailsCs = { data: ''} | |
| const ce = cheerio.load(detailsCs.data, { | |
| withDomLvl1: true, | |
| normalizeWhitespace: true, | |
| decodeEntities: true | |
| }); | |
| let jsonframe = require('jsonframe-cheerio'); | |
| jsonframe(ce); // initializes the plugin | |
| const skills = { | |
| topSkills: { | |
| heading: '.pv-skill-categories-section .pv-profile-section__card-heading', | |
| skills: { | |
| _s: ".pv-skill-category-entity__top-skill", | |
| _d: [{ | |
| "name": '.pv-skill-category-entity__name-text', | |
| "count": '.pv-skill-category-entity__endorsement-count', | |
| "highlight": '.custom-highlight' | |
| }] | |
| } | |
| }, | |
| otherSkills: { | |
| sections: [".pv-skill-category-list h3"], | |
| skills: { | |
| _s: '#skill-categories-expanded .pv-skill-category-list', | |
| _d: [{ | |
| heading: '.pv-skill-categories-section__secondary-skill-heading', | |
| skill: { | |
| _s: '.pv-skill-category-entity--secondary', | |
| _d: [{ | |
| "name": '.pv-skill-category-entity__name-text', | |
| "count": '.pv-skill-category-entity__endorsement-count', | |
| }] | |
| } | |
| }] | |
| } | |
| } | |
| }; | |
| const positions = { | |
| positions: { | |
| heading: '#experience-section .pv-profile-section__card-heading', | |
| experience: { | |
| _s: '.pv-profile-section__card-item-v2', | |
| _d: [{ | |
| employedDuration: '.pv-entity__date-range', | |
| duration: '.pv-entity__bullet-item-v2', | |
| company: '.pv-entity__secondary-title', | |
| companyLogo: 'img @ src', | |
| position: 'h3', | |
| location: '.pv-entity__location', | |
| details: '.pv-entity__extra-details' | |
| }] | |
| } | |
| }, | |
| education: { | |
| heading: '#education-section .pv-profile-section__card-header', | |
| data: { | |
| _s: '#education-section .pv-entity__summary-info', | |
| _d: [{ | |
| schoolName: '.pv-entity__school-name', | |
| degreeName: '.pv-entity__degree-name .pv-entity__comma-item', | |
| fieldOfStudy: '.pv-entity__fos .pv-entity__comma-item', | |
| degree: '.pv-entity__grade .pv-entity__comma-item' | |
| }] | |
| } | |
| }, | |
| skills | |
| }; | |
| const expList = ce('.profile-detail').scrape(positions, {string: true}) | |
| console.log(expList); |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment