Iterable data types within quantum kernels

[marwafar]

Required prerequisites

• [X] Search the issue tracker to check if your feature has already been mentioned or rejected in other issues.

Describe the feature

If I run the example below:

import cudaq

qubits_num=4

@cudaq.kernel
def circuit():

    qubits=cudaq.qvector(qubits_num)

    s='0001'

    for i in s:
        if i=='1':
            x(qubits[i])

    mz(qubits)

result=cudaq.sample(circuit,shots_count=5000)
print(result)

I get the following error:

cudaq.kernel.ast_bridge.CompilerError: test.py:11: error: {} iterable type not supported.
     (offending source -> for i in s:
    if i == '1':
        x(qubits[i]))

There are many cases where we need to use if statement. Currently, we have to repeat the code multiple time for each if statement.

Edit from bettinaheim: We have a similar case of "not supported" when iterating over a register of qubits. Let's define more broadly what iterations to support within kernels. Re string, see also https://github.com/NVIDIA/cuda-quantum/issues/1452.

[amccaskey]

Just a note here - this is an error with iterating through a string, which we have not implement yet. This is not an error with if statements. If you update to s = [0, 0, 0, 1] this code should work fine.

[bettinaheim]

I updated this to more broadly capture iterable data types within kernels, to also include e.g. iterating over a register of qubits.