Examples of correct and incorrect code

[inad9300]

I find myself very often searching for examples online, since TSLint's docs are lacking them. For TSLint to be useful, one must know what its rules are about, and this is currently not provided by the documentation alone.

The ESLint team did a great job in this regard, and I think you should take inspiration from them, e.g. https://eslint.org/docs/rules/no-empty.

If this is not under consideration already, please consider it 🙂

[ajafff]

Closing as a duplicate of #1329

For the time being the easiest way to understand a rule is to look at the tests.

[inad9300]

Thanks for the housekeeping efforts, and sorry for the duplicates.

[inad9300]

Actually, can you guide me a bit on how to understand the tests? They are somewhat cryptic.

[ajafff]

The test consist of different line types:

Code lines everything that is not listed below.

Message substitution (or template)
[name]: message are typically located at the bottom of the file. They define names that can be reused in error markup

Error Markup
~~~~~ [error message] the squiggles are located below a code line at the position the error is shown. The error message is either the real message or the name of a message substitution.
~nil means the error has a width of 0.
~~~ or ~nil without an error message indicate an error that continues on the next line.
[foo % ('a', 'b')] interpolates the strings a and b into the message substitution identified by foo. The substitution is done by node's util.format.

Example:

foo(b);
    ~ [err % ('a', 'b')]
eval('some code');
~~~~~~~~~~~~~~~~~ [Usage of 'eval' is forbidden.]

[err]: Use '%s' instead of '%s'.

The above means there is an error Use 'a' instead of 'b'. on the first line at the identifier b.
There's also an error on the eval() call. That message is not substituted.

[adidahiya]

the testing markup is documented here: https://palantir.github.io/tslint/develop/testing-rules/

[inad9300]

Thanks to both 😄