astpp.parseprint gives a wrong value sometimes, in case of Starred nodes

[Bhumil Haria]

astpp.prettyprint() seems to fail for specific (not all) cases of Starred nodes. In my tests, it works for Store contexts, but fails for Load ones.

Repro Steps:

>>> from astpp import parseprint
>>> parseprint('a,b,c = [ *[1,2], 3]')
Module(body=[
    Assign(targets=[
        Tuple(elts=[
            Name(id='a', ctx=Store()),
            Name(id='b', ctx=Store()),
            Name(id='c', ctx=Store()),
          ], ctx=Store()),
      ], value=List(elts=[
        Starred(value=List(elts=[
            Num(n=1),
            Num(n=2),
          ], ctx=Load()), ctx=Load()),
        Num(n=3),
      ], ctx=Load())),
  ])

For the same input text, however, the AST module output differs.

>>> import ast
>>> t = ast.parse('a,b,c = [ *[1,2], 3]')
>>> t.body[0].value
<_ast.Tuple at 0x7f9ec2bc6278>

So the t.body[0].value shows List node in one case and Tuple node in another.

Do let me know if any more details are needed.

PS: I've been using this excellent AST pretty printer (and even better documentation) for the past few days. It's great!

[Thomas Kluyver]

That's weird. parseprint is just calling astparse and then printing the results. I see a List for both ways:

 3│ ast.parse('a,b,c = [ *[1,2], 3]').body[0].value
 3> <_ast.List at 0x7f921251ac18>

 4│ astpp.parseprint('a,b,c = [ *[1,2], 3]')
Module(body=[
    Assign(targets=[
        Tuple(elts=[
            Name(id='a', ctx=Store()),
            Name(id='b', ctx=Store()),
            Name(id='c', ctx=Store()),
          ], ctx=Store()),
      ], value=List(elts=[
        Starred(value=List(elts=[
            Num(n=1),
            Num(n=2),
          ], ctx=Load()), ctx=Load()),
        Num(n=3),
      ], ctx=Load())),
  ])

Is it possible you ran them in two different versions of Python?

[Bhumil Haria]

No, I ran them in the same shell, should've taken a screenshot.

However, when I try the same thing in a fresh shell, I do see that they're the same. I'm not sure what happened, so I'm closing this for now. Sorry for wasting your time.

If I see this again I'll investigate more to find more concrete repro steps, and share results.

[Bhumil Haria]

Could not reliably reproduce this, resolving as invalid