Hint or advice how to improve docs about `reprlib.Repr.indent` feature

Documentation
1

This is a pull request I’m currently working on:

I appreciate the review and feedback that I received. Unfortunately, I’m not sure how to improve the docs paragraph. I already tried to be as precise as possible. So I was hoping someone with more experience regarding writing docs can give me a hint or advice which sentence needs change and which part can stay the way it is?

Here are the reprlib.Repr.indent features that I tried to describe in the docs paragraph:

>>> from reprlib import aRepr
>>> example = {"one": ["two", "three"], 4: "five"}
>>> aRepr.indent = None
>>> print(aRepr.repr(example))
{'one': ['two', 'three'], 4: 'five'}
>>> aRepr.indent = 2
>>> print(aRepr.repr(example))
{
  'one': [
    'two',
    'three',
  ],
  4: 'five',
}
>>> aRepr.indent = 4
>>> print(aRepr.repr(example))
{
    'one': [
        'two',
        'three',
    ],
    4: 'five',
}
>>> aRepr.indent = "...."
>>> print(aRepr.repr(example))
{
....'one': [
........'two',
........'three',
....],
....4: 'five',
}
>>>
2

I’ll post my suggestions for docs refactoring later today. +1 on the feature itself

3

Overall, the feature is certainly a very helpful one. Thanks for contributing it!

I left a review with a suggested revision of the added docs, to clarify the text and add examples to illustrate.

Also, a quick manual test of a handful of simple cases for indent (0, empty string, True, False, negative integers, and more clearly incompatible types like lists, object() and floats) revealed either a bug, potentially unexpected behavior or a less than user-friendly error in each case, which adequate tests coverage of various inputs would help expose. My review provides more detail and recommendations for each.