tad

Command line tool to test solutions to algorithms and data structures problems.

The goal of the project is to make it as simple as possible to develop and test solutions for algorithmic problems from websites such as HackerRank, LeetCode, GeeksForGeeks, and others.

tad takes a bunch of input and output files you give it and runs the source code of your solution against them
If you have Docker installed, tad can use it to compile and run most of the languages
tad can show you a line by line difference between your answer and the correct one
tad can launch an interactive debugger to debug your solution, if supported by the language

How it works

tad is a small binary written in Rust. It does not know how to run the solutions for any languages or how to compile them but it relies on pre-built container images or an explicitly provided commands for that.

When launched, tad finds and parses a configuration, looks up input and output files from test cases, runs the given solution against these test cases using a specified way, and finally, shows where the error is.

Requirements

If you want to use preconfigured environments to run the solutions, you need to have Docker installed. Other than that tad is a lightweight statically compiled binary that has no other dependencies.

Usage

For a directory containing a solution to the problem and a pair of input and output files

solution.py
input01.txt
output01.txt

tad can be launched like this:

$ tad solution.py
[1/3] Collecting cases [100%]
[2/3] Executing [100%]
[3/3] Verifying results [100%]
01 : Pass!

Specifying configuration options

tad knows that this solution file is written in Python3 because we've specified it inside the comment in the file:

Example code is taken from HackerRank

# tad: env=python3 inputs=input*.txt outputs=output*.txt
data = input()
result = "answer: " + data
f = open(os.environ['OUTPUT_PATH'], 'w')
f.write(str(result) + '\n')

When env parameter is specified, tad uses Docker to run this solution. But tad can also rely purely on command line options to specify the same parameters. Consider this:

$ tad solution.py --env python3 --inputs 'input*.txt' --outputs 'output*.txt'

This will achieve the same result without having to specify this parameters in the comment line of the solution's source code.

A third way to specify these parameters is to put them in a config file called tad.yaml:

env: python3
inputs:
  - input*.txt
outputs:
  - output*.txt

Again, having the tad.yaml file like the one above placed in the same directory will allow tad to parse these paramters from the config file, rater than looking for them in the source code or in the command line options.

Finally, tad can pick up the configuration options from the environment variables. Example below works in any common Linux/MacOS shell:

$ TAD_ENV=python3 TAD_INPUTS=input*.txt TAD_OUTPUT=output*.txt tad solution.py

Launch against specific test case

Just specify the name of the test case as a second positional argument:

$ tad solution.py 01

Do I have to use Docker?

When the env parameter is specified, tad will attempt to compile and launch the solution in the container. tad relies on Docker being instaleld on the system in order to do that.

Using containers to run and test solutions has multiple benefints

You get out of the box suppot for lnaugages
Ability to run solutions in interactive debug mode using the IDE of your choice
Automatic limitation of memory and cpu for each container
Predictable environment that matches the test environments of platforms like HackerRank 1:1

However, you might need to run the solution without Docker. If you want to do this, you can explicitly specify the command required to run the solution. When you do that, tad ignores any env, debug or even path to the solution. But tad will still run the supplied command against the found test cases and verify the output.

$ tad -i='input*.txt' -o 'output*.txt' --command "python3 solution.py"

Notice how we've shortened the syntax of options this time: Here -o is the same as --output and that rule applies to all other commands as well. See tad --help for details.

The shell command above will launch whatever is given to the command as it is.

This ability is also useful if you compile and run a solution in a language that does not have an out of the box container implementation in tad.

Note: If you'd like a language to be added and you can tell us how to compile and run it, we'd really appreciate you openning an issue with a request for new language and instructions on running it. ❤️

Installation

You have multiple options when installing tad.

If you have Rust installed

cargo install tad

Get the binary from the GitHub release

License

This project is licensed under the terms of the Apache-2.0 license.

vduseev/tad

tad