[BUG] Muilt-input map task accept list input if it is the default argument

[dansola]

Describe the bug

Currently Flyte does not support partial task with lists as inputs.

e.g.

@task
def mult_sum(my_val: int, my_list: list[int]) -> int:
    return sum(my_val * v for v in my_list)


@workflow
def wf() -> list[int]:
    my_list = [1, 2, 3]
    my_vals = [1, 2]
    partial_task = functools.partial(mult_sum, my_list=my_list)
    return map_task(partial_task)(my_val=my_vals)

ValueError: Cannot use a partial task with lists as inputs

However if my_list is a default workflow argument like so:

@workflow
def wf(my_list: list[int] = [1, 2, 3]) -> list[int]:
    my_vals = [1, 2]
    partial_task = functools.partial(mult_sum, my_list=my_list)
    return map_task(partial_task)(my_val=my_vals)

No error is shown - in fact it works correctly when you run locally.

When you run remotely you get a somewhat ambiguous error without a traceback"

input arrays have different lengths: expecting '3' found '2'

Expected behavior

Ideally it would be nice to support partial task with lists as inputs, but if not, it would be great to have a similar compile time/local runtime error ValueError saying that Cannot use a partial task with lists as inputs.

Additional context to reproduce

No response

Screenshots

No response

Are you sure this issue hasn't been raised already?

• Yes

Have you read the Code of Conduct?

• Yes