People:

I'll post a reviewed version of the PEP.

The only differences with the previous one will be the treatmen to float in both explicit and implicit construction:

In Explicit construction:

You can not call Decimal with a float. Instead you must use a method: Decimal.fromFloat(). The syntax:

Decimal.fromFloat(floatNumber, positions)

where floatNumber is the float number origin of the construction and positions is the positions after the decimal point where you apply a round-half-up rounding. In this way you can do, for example:

Decimal.fromFloat(1.1, 2): The same that doing Decimal('1.1').
Decimal.fromFloat(1.1, 16): The same that doing

Decimal('1.1000000000000001'). Decimal.fromFloat(1.1): The same that doing Decimal('110000000000000008881784197001252...e-51').

In Implicit construction:

Raise TypeError. You can not mix Decimal and float.

If you're ok, I'll post this.

Thank you!

. Facundo