Language feature: list, set, dict comprehensions

[tusharsadhwani]

Current implementation supports lists, sets and dicts, but it doesn't support their comprehensions.

Code like:

my_list = [i*2 for i in range(10)]
my_set = {i*j for i in range(10) for j in range(10)}
my_dict = {i: i*2 for i in range(10) if i % 2 == 0}

[Jacco]

Did this one too :-) Will make a PR.

[Jacco]

I'm already making good progress with the dict and list comprehensions. I ran into the fact that the list comprehensions can be repeated:

            text = (("Hi", "Steve!"), ("What's", "up?"))
            print([word for sentence in text for word in sentence])

I checked the ast python generates:

Module(
   body=[
      Assign(
         targets=[
            Name(id='text', ctx=Store())],
         value=Tuple(
            elts=[
               Tuple(
                  elts=[
                     Constant(value='Hi'),
                     Constant(value='Steve!')],
                  ctx=Load()),
               Tuple(
                  elts=[
                     Constant(value="What's"),
                     Constant(value='up?')],
                  ctx=Load())],
            ctx=Load())),
      Expr(
         value=ListComp(
            elt=Name(id='word', ctx=Load()),
            generators=[
               comprehension(
                  target=Name(id='sentence', ctx=Store()),
                  iter=Name(id='text', ctx=Load()),
                  ifs=[],
                  is_async=0),
               comprehension(
                  target=Name(id='word', ctx=Store()),
                  iter=Name(id='sentence', ctx=Load()),
                  ifs=[],
                  is_async=0)]))],
   type_ignores=[])

My problem is that I am struggling to make the interpreter work with the generators list in ListComp. Probably I can do it in the non-lazy way but there must be a better way :-)

@tusharsadhwani

[tusharsadhwani]

What problem are you facing with the generators list? For list and dict comprehensions we shouldn't need generators just yet.

[Jacco]

For the case

[word for sentence in text for word in sentence]

The first generator is [for sentence in text]
The second generator is [for word in sentence]

If the interpreter handles this correctly it should:

1. get an item from the first generator
2. yield items from the item fetched until those are finished then repeat step 1

Now two 2 generators this might seem easy but what if there are 3 or 4 etc.

I am not sure how to implement that.

[tusharsadhwani]

Sorry for delaying the response, I've been away for a while. It is possible to do this part without generators as well! I'll try to produce some example code, but that will be after Monday. Thanks again!

…

On Mon, Dec 4, 2023, 6:57 PM Jacco Kulman ***@***.***> wrote: For the case [word for sentence in text for word in sentence] The first generator is [for sentence in text] The second generator is [for word in sentence] If the interpreter handles this correctly it should: 1. get an item from the first generator 2. yield items from the item fetched until those are finished then repeat step 1 Now two 2 generators this might seem easy but what if there are 3 or 4 etc. I am not sure how to implement that. — Reply to this email directly, view it on GitHub <#6 (comment)>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/AKLGU44QDJ54ZNBPM2JLWNDYHW3DBAVCNFSM6AAAAAA3S27DV6VHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHMYTQMZYGQ4DKOJRGA> . You are receiving this because you were mentioned.Message ID: ***@***.***>