Scrape course description from UW CSE course page and store as JSON. Implemented using node.js.

app.js

var cheerio = require('cheerio');
var http = require('http');
var fs = require('fs');
var url = require('url');

var pageUrl = 'http://www.washington.edu/students/crscat/cse.html';

http.get(pageUrl, function(res) {
	res.setEncoding('utf8');
	var data = '';
	res.on('data', function(chunk) {
		data += chunk;
	});
	res.on('end', function() {
		parseCoursePage(cheerio.load(data), pageUrl);
	});
});

function parseCoursePage($, rootUrl) {
	var entries = [];

	$('a[name]').each(function(index, elm) {
		var courseId = $(elm).text();
		var courseName = $(elm).parent().text().substring(courseId.length);
		var courseDescription = $(elm).parent().parent().text().substring(
			courseId.length + courseName.length
			);

		var credits = 'unspecified';
		var areasOfKnowledge = [];

		var creditsStart = courseName.indexOf('(');
		var creditsEnd = courseName.indexOf(')');
		if (creditsStart >= 0 && creditsEnd >= 0 && creditsStart < creditsEnd) {
			// Get the credits
			credits = courseName.substring(creditsStart + 1, creditsEnd);

			// Get the areas of knowledge
			if (courseName.length >= creditsEnd + 1) {
				var aokText = courseName.substring(creditsEnd + 1);
				var aokTokens = aokText.split(',');
				for (var i = 0; i < aokTokens.length; i++) {
					var code = aokTokens[i].trim();
					if (code.length > 0) {
						areasOfKnowledge.push(code);
					}
				}
			}

			// Isolate course name
			courseName = courseName.substring(0, creditsStart - 1);
		}

		// Get instructor links
		var courseLinks = [];
		$(elm).parent().parent().children('a').each(function(linkIndex, linkObj) {
			var href = $(linkObj).attr('href');
			if (href != null && href.length > 0) {
				courseLinks.push(url.resolve(rootUrl, href.trim()));
			}
		});

		// Get prerequisites
		var prereqs = 'None';
		var parsedPr = /Prerequisite:(.*?)\.(?!\d)/.exec(courseDescription);
		if (parsedPr != null) {
			prereqs = parsedPr[1].trim();
		}
		console.log(prereqs);

		entries.push({
			id: courseId.trim(),
			name: courseName.trim(),
			credits: credits.trim(),
			aok: areasOfKnowledge,
			description: courseDescription.trim(),
			links: courseLinks,
			prereqs: prereqs
		});
	});

	saveResults(entries, './courses.json');
}

function saveResults(results, dest) {
	console.log('Saving results to ' + dest);
	var stream = fs.createWriteStream(dest, { encoding: 'utf8' });
	stream.write(JSON.stringify(results, null, 2));
}

package.json

{
	"name": "uw-cse-course-scraper",
	"description": "Scrape UW CSE courses and put them in a structured format",
	"author": "Hakon Verespej <hakon@madrona.com>",
	"dependencies": {
		"cheerio": "0.12.4"
	},
	"devDependencies": {
	}
}

UW CSE Course Scraper

This code contains node.js javascript to scrape and parse the UW CSE course content provided at http://www.washington.edu/students/crscat/cse.html. The result is a json document containing the parsed content in a structured format.

To use,

npm install
node app.js