Fixed-point notation is not behaving acording to the documentation

I want to format some values with a fixed precision of 3 unless it's an integer. In that case I don't want any decimal point or trailing 0s.

Acording to the docs, the 'f' type in string formating should remove the decimal point if no digits follow it:

If no digits follow the decimal point, the decimal point is also removed unless the # option is used.

But testing it with python3.8 I get the following results:

>>> f'{123:.3f}'
'123.000'
>>> f'{123.0:.3f}'
'123.000'

Am I misunderstanding something? How could I achive the desired result without using if else checks?

Solution

In order to forcefully achieve both your desired outputs with the same f-string expression, you could apply some kung-fu like

i = 123
f"{i:.{3*isinstance(i, float)}f}"
# '123'

i = 123.0
f"{i:.{3*isinstance(i, float)}f}"
# '123.000'

But this won't improve your code in terms of readability. There's no harm in being more explicit.