The following applications need to be installed:
- Git
- Node.js - this project tests against v4.2.2 LTS
- MongoDB
- A text editor like Notepad++ or Sublime Text 3
NOTE: Make sure your node version is compliant!
Make sure that the directories containing node, npm, mongod and mongo are added to the path making them accessible from the terminal.
After cloning the git repository run npm install from a terminal window. If this process fails complaining about bower, you will need to run npm install bower -g, then run npm install again. If you receive another error involving Git, then you will need to ensure that Git is installed and the *\Git\bin directory is added to your PATH.
You will likely also have to run npm install gulp -g in order to run gulp. This command installs gulp globally so it can be used in the command line.
Note: On UNIX based platforms, this will need to be run with sudo and on Windows, the command prompt will require administrative privileges.
gulp clean:modules will make sure that all the modules are the latest versions and ascertain that there are no unneeded modules downloaded. This clears out both npm and bower modules.
This will reduce the amount of conflicts rising from versioning and previous experiments.
To run the website server, you must open two terminal windows. In the first window, execute gulp mongo:start and leave this window open. This starts the mongo server. In the other window, execute gulp start:dev or gulp start:prod. Then navigate to localhost:8080.
Running the server in development mode disables LDAP login and HTTPS and falls back to hard-coded credentials. These hard-coded credentials are as follows:
- Guybrush Threepwood (username: test1) (password: test)
- Rick Sanchez (username: test2) (password: test)
- Dick Dickerson (username: test3) (password: test)
In order to run the server in production mode, you need to have access to the company network for LDAP purposes. You also need to create server\secrets\ldapAuth.js which will hold the credentials needed to bind to the LDAP server. Lastly, you will need to have certificates to run HTTPS. Please look up self-signed certificates in order to properly set up the production environment, or obtain certificates from a trusted Certificate Authority.
Any changes to the back-end server code while the server is running will result in a server restart. However, this DOES NOT automatically re-run client tests or JavaScript analysis commands, so be sure to run them yourself before submitting a PR.
Any changes to the front-end client code will not result in a server restart. You will, however, need to perform a hard-refresh in your browser window to pick up the changes.
gulp start:test will run the server in the test environment. This is identical to dev, but it uses a hard-coded port number of 7357 and does not generate data.
To run the client side tests, execute gulp test:client from a terminal window. This automatically runs when a gulp start:dev or gulp start:prod command is executed.
To run the server load tests, execute gulp start:test from a terminal window then open an additional terminal window and enter gulp test:load. This will submit a number of ideas to the server and provide timing statistics in milliseconds to the terminal window.
Running gulp jshint will lint any JavaScript file under the src folder (excluding src/lib), the server folder and gulpfile.js. Running gulp jscs will check the code style for the aforementioned files.
Running gulp code-check will run both of these commands.
Unlike gulp test:client, these commands do not automatically run when executing a gulp start:dev or gulp start:prod
The express server app uses mongodb to store users and ideas. There are two databases, flintandsteel for production and flintandsteel-dev for development. If the development database gets messy, gulp clean:db-dev can be used to drop the database.
At this point in time gulp clean:db-dev runs automatically when calling gulp generate:data, which runs automatically when executing gulp start:dev.
To start with some predefined data, gulp generate:data clears the development database (using gulp clean:db-dev) and populates the database with pre-populated ideas and users. This happens automatically if you run gulp start:dev.
Refer to CONTRIBUTING.md for directions on how to set up and contribute to this project.
Contributions using other channels will be disregarded and asked to be routed through the proper procedure.
Note: Do not push to the stable branch, committing to the stable branch will result in code being lost. The stable branch is created for milestone releases.