Update lispytest.py

[gabrielsroka]

add tests for +-*/ ref https://web.mit.edu/scheme_v9.2/doc/mit-scheme-ref/Numerical-operations.html#Numerical-operations

[norvig]

Thanks for this work. You are absolutely correct that modern Schemes have these multi-argument capabilities, and your implementation and tests are just right.

However, I decided that I wanted to keep things as simple as possible, and I implemented just the "essential procedures" from the R^3RS specification; these extensions are considered inessential in that version of Scheme. So I think I will keep it as simple as possible.

[gabrielsroka]

EDIT: i see you mention this here

Lispy is not very complete compared to the Scheme standard

ok. according to r3rs https://people.csail.mit.edu/jaffer/r3rs_6.html#SEC28 set! is essential. you mention it in several places on http://norvig.com/lispy.html and include an implementation:

    elif op == 'set!':           # assignment
        (symbol, exp) = args
        env.find(symbol)[symbol] = eval(exp, env)

but it's missing from lis.py (so are several other pieces of "essential syntax", eg, cond, let, etc)

http://norvig.com/lispy.html has links to r5rs, but above, you mentioned r3rs. is it both a subset and a superset then? :)

thanks!

[norvig]

Excellent questions, @gabrielsroka !

Very well, then I contradict myself, I am large, I contain multitudes.

I think you're convincing me that, sometime in the new year, I need to sit down and bring some consistency to this (and update to Python 3, and add type hints). Thank you for bringing this up, and happy new year!

[gabrielsroka]

Happy new year to you, too!

I think I just figured out part of the problem: https://github.com/norvig/pytudes/blob/main/py/lis.py is different than: http://norvig.com/lis.py

eg, one has set! and the other one doesn't.

there are also differences in lispy.py and and lispytest.py

I smell a merge... :)