Visual metasearch engine built with React, Redux, Express, and TypeScript.
Spresso Search scrapes search results from Google using the node x-ray library, and uses the same library to scrape obtain meta-information on webpages (preview images, favicons). There is a screenshot feature, which takes screenshots of sites that don't have meta preview images in their HTML. There is also a text-outline feature, powered by node-unfluff, which scrapes text content from web pages(ideal for articles & other text-rich pages), allowing the user to read the contents of a web page in clean, formatted text and without leaving the Spresso Search site.
To run Spresso Search locally, first clone the repo with:
git clone https://github.com/JoshuaScript/spresso-search.git
cd into its directory:
Install the dependencies with
yarn start to run the client side code. The app should be visible on port 3000.
Open a new terminal tab/window in the same directory, and run
cd server to go into the server directory.
node ./build/server.js to start the server. The app is now ready for use.
If making any modifications to the server's TypeScript code, you should start the TypeScript compiler in watch mode with
tsc -w so your changes can be tracked in the JS build.