[Python] Why does decimal.Context.create_decimal_from_float() not round near-zero values?

Contrary to what the docs might suggest, create_decimal_from_float() does not always round to its specified precision. As the last line of the following example shows, when a near-zero value is passed as an argument, it does not get clamped. I would expect it to get rounded to zero.

>>> import decimal
>>> decimal.getcontext().prec = 6
>>> decimal.getcontext().create_decimal_from_float(0.123456789123456789)
Decimal('0.123457')
>>> decimal.getcontext().create_decimal_from_float(0.000000000000000001)
Decimal('1.00000E-18')


I have discovered this behavior while trying to convert the result of sinus of 180 degrees to a decimal and had a similar result.

From my perspective, it looks like a bug. But is there something that I am missing? Is there any workaround to make sure the argument gets rounded properly?

Used Python version is 3.10.

Continue reading...