/help50

This is help50, a command-line tool that helps students understand error messages.

Primary LanguagePython

Contributing

To contribute a sample error message that help50 does yet support, open an issue and provide, ideally, both the error message and the command (and code, if any) that triggered it, both formatted in Markdown like code.

Better yet, to contribute to a helper for foo, implement helpers/foo.py per the below, where lines will be an array of strings (i.e., lines of stderr and/or stdout potentially from foo) for which user needs help, before must be a subset of lines that this helper has matched on, and after must be an array of strings that help user understand before. Helper must return None if it does not recognize lines[0]. Only if helper recognizes lines[0] may it look at lines[1:].

import re
def help(lines):
    ...
    # if helper recognizes lines[0]
        ...
        return (before, after)

Here are open issues if you'd like to solve one or more!

Development

  1. Install Docker Engine
    • If running Ubuntu, also install the latest version of Docker Compose
  2. Execute docker-compose build in a terminal
  3. Execute docker-compose up in a terminal

The provided docker-compose.yml will "mount" the repository within the container (at /srv/www) so that you can make changes to files locally that will be reflected inside the container.

For a shell within the container, execute

docker exec -it help50_web bash -l

after the container has been started (with docker-compose up).

Usage

Web

Visit http://localhost:8080/ in a browser

Command Line

Assuming a container is running and listening at http://localhost:8080/ and you're inside of the container (via docker exec -it help50_web bash -l):

  • help50 ./foo
  • CC=clang help50 make foo
  • help50 clang -o foo foo.c
  • ./foo |& help50
  • CC=clang make foo |& help50
  • clang -o foo foo.c |& help50

TODO

  • Create client and server subdirectories? But Elastic Beanstalk might only look in root of repo.
  • Add support for usernames and hearts via a MySQL database.
  • Create tests (that get executed before deployment).