# minishell test [![Build Status](https://api.travis-ci.com/cacharle/minishell_test.svg?branch=master)](https://travis-ci.com/cacharle/minishell_test) [![PyPI - Python Version](https://img.shields.io/pypi/pyversions/minishell-test)](https://pypi.org/project/minishell-test/) [![Documentation](https://readthedocs.org/projects/minishell-test/badge/?version=latest)](https://minishell-test.readthedocs.io)

Test for the minishell project of school 42.

![preview](https://i.imgur.com/98xh2xY.gif)

## Installation

### pip

```
$ pip3 install minishell-test
```

### Manual

```
$ git clone https://github.com/cacharle/minishell_test
$ cd minishell_test
$ pip3 install -e .
```

## Usage

```
$ minishell_test             # In your project directory
$ python3 -m minishell_test  # If you don't have ~/.brew/bin or ~/.local/bin in your PATH

$ minishell_test --help
usage: minishell_test [-h] [-p PATH] [-l] [-t COMMAND] [-k] [-r BEGIN END]
                      [--show-range] [-x] [-v] [-b] [-n] [-m] [-g]
                      [suite [suite ...]]

Test for the minishell project of school 42.

positional arguments:
  suite                 Test suites/group to run.
                        It tries to be smart and autocomplete the suite name
                        (e.g ./run int -> ./run preprocess/interpolation)

optional arguments:
  -h, --help            show this help message and exit
  -p PATH, --path PATH  Path to minishell directory
  -l, --list            Print available test suites
  -t COMMAND, --try-cmd COMMAND
                        Run a custom command like this test would
                        (the only environment variable passed to your executable are TERM and PATH)
  -k, --check-leaks     Run valgrind on tests (disable usual comparison with bash)
  -r BEGIN END, --range BEGIN END
                        Range of test index to run (imply --show-index)
  --show-range          Show test index (useful with --range)
  -x, --exit-first      Exit on first fail
  -v, --verbose         Increase verbosity level (e.g -vv == 2)
  -b, --bonus           Enable bonus tests
  -n, --no-bonus        Disable bonus tests
  -m, --make            Make minishell and exit
  -g, --pager           After running the test, display the result in a pager of your choice

Made by cacharle - https://cacharle.xyz
```

## Test compatibility

Your executable **must** support the `-c` option which allow to pass command as string.

```command
$ bash -c 'echo bonjour je suis'
bonjour je suis
$ ./minishell -c 'echo bonjour je suis'
bonjour je suis

$ bash -c 'ls'
README.md test.sh
$ ./minishell -c 'ls'
README.md test.sh
```

With this setup `argv[2]` is what you would usually get in `line` from `get_next_line`.
This allows you to set the prompt to whatever you want.

### Environement variables

My test only gives the `PATH` and `TERM` environment variables to your minishell.
**Please check that your project still work with those settings before messaging me on Slack or creating an issue**.
You can test this quickly with the `-t` option (e.g `minishell_test -t 'echo bonjour`).

## Bonus

* Force the bonus tests with `$ minishell_test -b`
* Change the `BONUS` variable in [config.py](minishell_test/config.py) to True
* Set the environment variable `MINISHELL_TEST_BONUS` to `yes`
  (e.g `echo 'export MINISHELL_TEST_BONUS=yes' >> ~/.zshrc`)

## Memory leaks

`$ minishell_test -k`, checkout the `--show-range`, `--range` and `-x` options to help
to select the tests to run since valgrind is really slow.

## Don't check error messages

If you don't want to copy bash syntax error message,
you can set the environment variable `MINISHELL_TEST_DONT_CHECK_ERROR_MESSAGE` to `yes`.
It will still test your exit status code but will discard any output on error tests.

## Linux

The tester will try to convert to output/status code of bash on Linux to the one on Mac.

---

## Add new tests

You can find the suites in the [minishell\_test/suites](minishell_test/suites) directory.

### Add individual test

In your suite function you can use the `test` function. With the following arguments:

1. Command to be tested (output and status will be compared to bash)
2. A command to setup the sandbox directory where the tested command will be run
3. List of files to watch (the content of each file will be compared)

```python
test("echo bonjour je suis")                                  # simple command
test("cat < somefile", setup="echo file content > somefile")  # setup
test("ls > somefile", setup="", files=["somefile"])           # watch a file
test("echo $A", exports={"A": "a"})                           # export variables
                                                              # in the environment
test("echo bonjour", hook=lambda s: s.replace("o", "a"))      # pass the shell output
                                                              # through a hook function

test("cat < somefile > otherfile",
     setup="echo file content > somefile",
     files=["otherfile"])
```

### Add Suite

A test suite is a group of related tests.

```python
@suite()  # @suite(bonus=True) if it's a bonus suite
def suite_yoursuitename(test):
    """ a description of the suite """
    test(...)
    test(...)
    test(...)
```

---

## Wildcard (or glob)

There is a commented glob suite in [minishell\_test/suites/preprocess.py](minishell_test/suites/preprocess.py).
Good luck handling `*'.*'`.