JS16_ProjectA

In this project we will lay the foundations for our system by integrating data from multiple sources into a central database. The database will serve the apps and the visualization tool that will be developed in other projects.

Links

• Main page: https://www.got.show
• API accessable at: https://api.got.show/api/
• Documentation about API endpoints: https://api.got.show/doc/
• Main wiki entry: https://rostlab.org/owiki/index.php/Javascript_technology_2016#Project_A

Developer information

Documentation

We are using apidoc to generate documentation for the RESTful API service. To get started follow these instructions:

• Open a terminal and cd into the checked out git repository folder
• Install the tool globally: sudo npm install apidoc -g
• Generate the documentation: apidoc -i app/ -o apidoc/
• Open the HTML file inside the apidoc folder or go to http://127.0.0.1:8080/doc/ if you already have set up the project

Setup NodeJS & MongoDB

• Install nodejs and mongodb on your local machine (https://docs.mongodb.org/manual/tutorial/install-mongodb-on-ubuntu/ and https://nodejs.org/en/download/package-manager/#debian-and-ubuntu-based-linux-distributions)
• Clone this project to a folder on your hard drive, open a console and change into the folder you just checked out
• Run sudo npm install to install any sub-modules required
• Copy the config file in cfg to config.json and edit it
  • You can leave username and password empty on default configurations
  • Use 127.0.0.1 and port 27017 for default configurations
  • In order to stream real-time Twitter data, please register your Twitter account at http://apps.twitter.com and insert your API keys into the config.json. Never upload your API keys to GitHub. By default, config.json is on .gitignore.
• Start local MongoDB server with mongod
  • You can specifc the port and folder you want to use: mongod --dbpath /your/db/path/here --port 27017
• Run nodejs app.js to start the server
• Node should show in console Mongoose connected - Node server is listening on port 8080
• If needed, you can start MongoDB shell via mongo. Then type show dbs to see all databases. Type use db_name_here to switch to preferred database. With show collections you can see all tables (in NoSQL tables are called collections). With db.collection_name.find() you can output the collection content.

Scraping and filling the database

x is in the following a placeholder and has to be replaced by the intended collection. (e.g. characters)

• To delete the collection and fill it again (new _ids are set!) with newly scraped data use: npm run refill --collection=x
• To update the collection with newly scraped data (manual edits are overwritten!) use: npm run update --collection=x
• To only add new properties/entries to the collection from a newly scrap use: npm run safeUpdate --collection=x

Available Collections:

• 'ages',
• 'characters',
• 'episodes',
• 'cities', (uses 'data/cities.json')
• 'continents', (uses 'data/continents.json')
• 'cultures',
• 'events',
• 'houses',
• 'regions',
• 'characterLocations', (requires cities collection to be filled)
• 'characterPaths', (requires characters collection to be filled)
• 'characterImages' (requires characters collection to be filled)
• 'characterPlods' (requires characters collection to be filled)

Updating the pageRank of characters

x is a placeholder for the file containing the pageRanks. (e.g. data/pageRanks.json)

• Requirements: Characters and characterImages collections are up-to-date.
• Run: npm run updatePageRanks --update=characters --file=x