Support for randcase about pyvsc HOT 4 CLOSED

Hi @ishitapvips,
This is an interesting question, given that Python doesn't have a builtin case statement. I could see a couple of approaches. Perhaps you can provide feedback on which would fit best in your application.
Fundamentally, rand-case is the combination of weighted selection and a case statement. We could model this in Python with a weighted selection from a collection of values that represent the case options.
Let's say we're replacing a rand-case statement like the following:

randcase
    3 : x = 1;
    1 : x = 2;
    4 : x = 3;
endcase

This approach would look something like:

branch = vsc.select({ 1 : 3, 2 : 1, 3 : 4})
if branch == 1:
  x = 1
elif branch == 2:
  x = 2
elif branch == 3:
  x = 3

In other words, we specify a map with the key being the branch and the value being the branch weight. select returns the selected branch, and we use a regular if/else statement.

Another approach would be to use lambda expressions. Lambda expressions are pretty limited in Python, so you would likely need to put the actual implementation of the case branches in separate functions. That might look like this:

vsc.randcase([
  [3, lambda : do_branch1()],
  [1, lambda : do_branch2()],
  [4, lambda : do_branch3()]
  ])

So, a couple approaches to doing this in Python with PyVSC. Any thoughts on which approach would work best for your application?

Thanks and Best Regards,
Matthew

from pyvsc.

Hi Matthew,

As per your convenience, any of the approaches will work for us. But lambda function would be straight forward so can you implement that one?

Thanks & Regards,
Ishita

from pyvsc.

Hi Ishita,
I've added support for a 'randselect' method that performs a weighted selection between a list of lambda functions. Here's an example:

I'll leave this issue open until you're able to confirm that it works for you.

Best Regards,
Matthew

from pyvsc.

Hi Matthew,

Randselect is working fine for me right now.
If any issue is found later will comment here.

Thanks & Regards,
Ishita Shah

from pyvsc.