HP PPL and Statistical distributions

[Albert Chan]

Some issues with PYTHON program (assumed it is true Python):

1. Turning repeat until to while loop must ensure it run at least once.
Loop counts must also match. It might be safer translate repeat-until as:

while 1:
      ...
      if until_condition: break

2. True Python does not turn 1./0. to +inf, but raise ZeroDivisionError instead.
Why Python's ZeroDivisionError for floating-point type is a bad and unnecessary feature

In fact, Python seldom return float('inf'), for floating math operations.
(1e300*1e300 return +inf, but 1e300 ** 2 raise OverflowError ... weird)

>>> from math import *
>>> log(0)
Traceback (most recent call last):
File "<stdin>", line 1, in <module>
ValueError: math domain error
>>> exp(1000)
Traceback (most recent call last):
File "<stdin>", line 1, in <module>
OverflowError: math range error

We have to catch exceptions instead. see emece67 double_exponential.py

This is why quad.lua translated to PPL. It behave closer to Lua:

CAS> 1./0.         → +Inf
CAS> log(0.)       → -Inf
CAS> exp(1000.) → +Inf