+We first wrote a web crawler to clone GitHub repositories of projects which depend on the most popular node modules on npmjs.org. Upon a user search, the GitHub repos are parsed for specific methods through a code snippet algorithm we developed using Esprima and Async libraries. After the documents have been parsed, the individual code snippets, along with the source code, are displayed on the front-end using factories, directives, filters, and modals (as well as highlight.js) to allow for the users to easily browse the results. Individual snippets also have an associated score that users can vote on after they log into GitHub.
0 commit comments