Getting started from zero

Prerequisites

To work with the SketchResponse codebase, there are a few technology prerequisites that you will need installed on your machine.

In order to clone the repository you will need to install Git on your computer. You may also want to create a Github account, though you should not need one to simply clone the SketchResponse repository from the command line. You will need to be logged into a Github account if you want to clone the repository through your browser. You will also need a Github account if you want to contribute changes back to the project.
The Sketch Tool front-end is built using Node.js and npm (which is now packaged with Node). You will need to have both of these installed on your computer.
The Grader Library back-end is built using Python and relies on a couple python packages that you will need to install. We recommend you use the pip python package installer. If you are using an up-to-date python installation you probably already have it installed. If you already have another python package manager, we only use commonly available packages so it should not be a problem.

Note for Windows Users: When installing Python make sure to enable the option to add Python to the path. This is disabled by default for some reason.

You will also need to be comfortable using the command-line to execute commands.

Note for Windows Users: You should use the Git Bash shell to execute all commands in this document. You will need to run it as Administrator to execute the npm and pip installation commands below. You may also need to run the Git Bash shell as Administrator to run the local flask testing server. To do this, right-click Git Bash from the Start Menu and select Run as Administrator.

If you have statisfied all of the above prerequisites, then we can start working on the codebase.

Getting the codebase

Clone this repository to the directory of your choice:

By using HTTPS:

$ git clone https://github.com/SketchResponse/sketchresponse.git

By using SSH:

$ git clone git@github.com:SketchResponse/sketchresponse.git

Running either of the above commands will create a new directory called SketchResponse. This SketchResponse directory is your copy of the repository. The rest of this document will refer to it as the repo directory.

Run this command to change directories to the repo of the SketchResponse repository.
```
$ cd sketchresponse
```

Building the Sketch Tool front-end

Change to the sketch_tool directory:
```
$ cd sketch_tool
```
Install dependencies listed in package.json using npm. If you are in Windows make sure you are running the Git Bash shell as Administrator:
```
$ npm install
```

Build a distribution of the Sketch Tool. The distribution directory will be found in the /static/sketch_tool_dist/ directory.
```
$ npm run build
```
You can change the target browsers of the build by modifying the BrowserList entry in package.json. Currently the target is set to the last 2 versions of every major browser excluding IE.

Now you have a built distribution of the Sketch Tool front-end, we will use it later to let us test grading scripts. The front-end distribution is also what you will need to have hosted on a public server when deploying SketchResponse for your application.

Grader Library back-end dependencies

The SketchResponse grader library back-end has only two third party package requirements, which are listed in the requirements.txt file.

flask
numpy

You will need to be in the repo directory of the codebase. If you are still in the sketch_tool directory, run the following command to change back to the repo directory.

$ cd ..

Now that you are in the repo directory, you can run the following command to install the required python packages. If you are in Windows make sure you are running the Git Bash shell as Administrator

$ pip install -r requirements.txt

Numpy is used to support the mathematical computations of the grader library.

Flask is required to run a local server for convenience when implementing and testing new grading scripts. This lets you test your scripts without having to deploy them to another server, giving you more immediate feedback and simplifying the debugging process. See the Tutorial Test a Grading Script on a Local Server for details.

What's next

Now that you have the codebase installed and ready to use, lets start looking at how to configure the Sketch Tool and implement a couple grading scripts. Click here.

Getting Started