Description

A tool to convert single-file python programs to a driver, stub, and solution code. Also automatically generates test cases based on the provided driver code in the python program.

This repository also provides a script to optionally upload the generated files to Kodethon's problem library.

Program Format

The python programs have to be a single file with the driver code defined after the function definitions.

For example:

# Function definition(s)
def add(a, b):
    return a + b

# Driver code
add(1, 2)

Usage

Step 1

python lib/prepare.py DIR_PATH/PYTHON_FILE_NAME.py

Creates a folder called 'output' with a PYTHON_FILE_NAME folder inside of it. The below files should be created in the folder:

• driver.py
• solution.py
• seed.json
• skeleton.py

Step 2

python lib/mutate.py output/PYTHON_FILE_NAME

Creates a 'cases' folder in the PYTHON_FILE_NAME folder filled with generated test cases.

(Optional) The above two commands can be run as follows:

cd bin/mutate; sh problem.sh DIR_PATH/PYTHON_FILE_NAME.py

(Optional) To run prepare and mutate all problems in DIR_PATH run:

cd bin/mutate; sh assignment.sh DIR_PATH

The folder should have multiple directories; each directory should have a python folder containing the python file. The python file should have a unique name; it is advisable to name the file after the problem name.

Step 3

Upload the generated problem(s) to Kodethon.

1. Rename and update config/credentials.yml.sample to config/credentials.yml
2. Update the file accordingly
3. cd bin/upload; sh assignment.sh DIR_PATH

DIR_PATH should contain the folders named after the problem.

Getting Programs (Optional)

The programs can either be manually written or scraped from an external source.

Scraping

scrape.py and parse.py must be in the same directory.

Step 1

In scrape.py, edit pages variable to point to a dictionary of links, where the key is the subtopic and the value is a list of URLs.

Step 2

To scrape, call scrape_links and pass in the PAGES variable and the TOPIC as a string.

• Generates directories to path 'raw/TOPIC/SUBTOPIC' (parallel to scrape.py and parse.py) and creates an HTML file for raw scraped HTML from each link.

Parsing

In parse.py, change TOPIC to the TOPIC used in scrape.py.