Package for easily adding a course on UWaterloo's QUEST.

## Project description

This python package allows one to easily interact with the University of Waterloo’s QUEST system to easily add a course. It does so by continuously querying the QUEST servers to add you into the given course until it succeeds.

Both Python 2 and Python 3 are supported, although Python 3 support requires 2to3 but that will be handled automatically by setup.py on installation.

## Install

There are currently two resources from which you can install UWaterloo AddCourse: the first is directly from the repository and the second is from PyPi.

### The Repository

You can install this package from source using:

git clone https://github.com/kcolford/uwaterloo-addcourse.git
python setup.py install

Note that when you install in this way, you will be getting the development version, not the official stable release (although it should still work as documented).

### PyPi

Using easy_install, you just have to invoke:

easy_install uwaterloo-addcourse

You can use pip to install this package with just:

pip install uwaterloo-addcourse

You can also navigate to our PyPi page and choose one of the download links at the bottom.

## Usage

There are two ways to use this API: the command line and the python interpreter. Note that when using the interpreter, you have more control over the queries you would have when using the command line. Although the command line is far more user friendly.

### Command Line

To add a class from the command line, simply invoke the helper script like so:

user@computer:~/\$ addcourse
Desired Course: cs246
QUEST ID: jsmith
...

Optional command line arguments to addcourse are as follows

-h, --help

show this help message and exit

--version

show program’s version number and exit

-c COURSE, --course COURSE

the course to try getting in to

-u USERID, --userid USERID

### Python Interpreter

Simply write a script like so:

#! /usr/bin/env python

addcourse('jsmith', 'password123', numbers('cs246'))

Note that the call to numbers fetches a list of class numbers that correspond to the given course code (in this case 'cs246'). You can then use a splicing or other list manipulations to delete classes you don’t want or add alternative classes that you do want. See pydoc addcourse for more information.

## Development

Development on UWaterloo AddCourse is relatively simple. We use GitHub and so all additions or improvements should be done through a pull request. Improvements of any kind will be gladly accepted.

Our current direction is looking into the following:

• Extending the utility offered by the QuestBrowser class to then use for other purposes including:

• A better user interface for using any feature of QUEST.

• Incorporating a javascript interpreter to allow us to easily function even after changes to QUEST are made.

• Making asynchronous queries to QUEST to do multiple things at once instead of just waiting for each query to slowly complete. Current issues surrounding this include:

• urllib is currently only synchronous, no asynchronous API is currently available at the time of this writing, possible solutions include:

1. Locating an asynchronous fork of urllib (possibly on PyPi).

2. Rolling our own asynchronous urllib fork (we could host it separately on PyPi).

3. Taking the source of urllib and refactoring it to only use the asynchronous IO interfaces instead of the synchronous ones (even just using select based IO can be a great improvement).

4. Using threading because although the GIL (Global Interpreter Lock) prevents speed ups of python code, it does allow switching from one execution context to another when a thread blocks on IO. This would effectively simulate option 3 but with the select loop running in kernelspace rather than in application code.

Of the aforementioned options, number 4 seems to be the most promising although introducing full threading is always a mixed bag of good and bad.

• Refactoring will/may have to be done in order to improve thread safety.

• Adding a command line option for a password. This is inherently unsafe as there are many distributions and installations of ps that allow viewing of the command line arguments given to a process by other users. That being said, it does make it very inconvenient for a user who wants to script the command line invocations of addcourse as they have to sit there and type in their password each time the command runs.

• We need to set up automated testing and incorporate continuous integration systems like travisCI. Problems that interfere with this currently are:

• We need a dummy QUEST login to test with because no one is going to leave their real QUEST user id and password in the repository for any one to steal and mess with. Also, the tests might mess up the account and that would be really bad if someone didn’t get their degree because of a typo made in a pull request.

• Finer control of progress messages and reports. Currently we just use python’s print statement to output messages to the user, but we may want to move towards the logging module in the builtin library. Note that originally, that was what we used, but the logging module proved to difficult and unwieldy to continue with.

Moving in this direction will allow us to control the verbosity of the API through setting the loglevel.

This file is part of UWaterloo-AddCourse.

UWaterloo-AddCourse is free software: you can redistribute it and/or modify it under the terms of the GNU General Public License as published by the Free Software Foundation, either version 3 of the License, or (at your option) any later version.

UWaterloo-AddCourse is distributed in the hope that it will be useful, but WITHOUT ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License for more details.

## Credits

Mark Petrick, for the support that inspired this project.

## Project details

### Source Distribution

Uploaded source
Uploaded 3 4
Uploaded 2 7