Missing type to str.format() behavior

Question

According to the Python docs here , when leaving the type out, it defaults to a type of 'g' for floating point arguments.

However,

print("{0:.2}".format(14.9))

prints "1.5e+01", while

print("{0:.2g}".format(14.9))

prints "15"

Is this simply an issue of the documentation being incorrect or is there another reason for it?

Answer 1

According to the source code , this is a documentation bug. The correct description for the behaviour without a floating point specifier is "like 'g', but always with at least one digit after the decimal point".

Answer 2

You linked the documentation of Python 2.7, but you actually used Python 3.x. In the documentation of Python 3.x , the behaviour is correctly documented.

The Python 2.7 documentation is faulty anyway:

>>> "{0:.2}".format(14.9)
'15.0'

>>> "{0:.2g}".format(14.9)
'15'