faculty-api.ts

import { APIGatewayProxyHandler } from 'aws-lambda';
import 'source-map-support/register';
import * as cheerio from 'cheerio';
import Axios from 'axios';

const BASE_URL = 'https://www.cs.utah.edu';

const SELECTORS = {
  FACULTY_ENTRIES: 'table#people:nth-child(3) tr.professor',
  NAME: '#info > tbody > tr:nth-child(1) > td > h8',
  OFFICE: '#info > tbody > tr:nth-child(2) > td:nth-child(2)',
  PHONE: '#info > tbody > tr:nth-child(3) > td:nth-child(2)',
  EMAIL: '#info > tbody > tr:nth-child(4) > td:nth-child(2)',
  RESEARCH_INTERESTS: '#info2 > tbody > tr:nth-child(2)',
  THUMBNAIL: 'img'
}

function extractFaculty(html) {
  const $ = cheerio.load(html);
  const faculty = [];
  const entries = $(SELECTORS.FACULTY_ENTRIES);
  entries.each((_, elem) => {
    const facultyEntry = {
      name: $(SELECTORS.NAME, elem).text(),
      thumbnail: `${BASE_URL}${$(SELECTORS.THUMBNAIL, elem).attr('src')}`,
      contactInfo: {
        office: $(SELECTORS.OFFICE, elem).text(),
        phone: $(SELECTORS.PHONE, elem).text(),
        email: $(SELECTORS.EMAIL, elem).text(),
      },
      researchInterests: $(SELECTORS.RESEARCH_INTERESTS, elem).text()
    };
    faculty.push(facultyEntry);
  });

  return faculty;
}

export const harvest: APIGatewayProxyHandler = async (_, _context) => {
  const res = await Axios.get(`${BASE_URL}/people/faculty/`);
  const resHTML = res.data;
  const faculty = extractFaculty(resHTML);

  return {
    statusCode: 200,
    body: JSON.stringify({
      faculty
    })
  };
}