Filter array of subarrays: drop table columns

[PsychoTeras]

Hi all,

There is a problem I can't solve. I have the following JSON:

{
    "rows": [
        [
            "A1",
            "B1",
            "C1"
        ],
        [
            "A2",
            "B2",
            "C2"
        ],
        [
            "A3",
            "B3",
            "C3"
        ]
    ],
    "columns": [
        "A",
        "B",
        "C"
    ]
}

You can see that it is essentially a table with columns and rows.
So, what I need is to drop some columns and its data. For example, I want to keep column "B" only:

{
    "rows": [
        [
            "B1"
        ],
        [
            "B2"
        ],
        [
            "B3"
        ]
    ],
    "columns": [
        "B"
    ]
}

Ideally, I would like to specify the names of the columns that I want to keep. But it is also ok to hardcode their indexes.

Thanks in advance,
Riz

[Courela]

This almost works:

{
    "rows": {
        "#loop($.rows)": [ "#currentvalueatpath($[?/(@ =~ //^B.*$///)])" ]
    },
    "columns": "#valueof($.columns[?/(@ =~ //^B.*$///)])"
}

And I say almost, because the result would be this:

{
    "rows": [["B1"], ["B2"], ["B3"]],
    "columns": "B"
}

"columns" value is not an array, only if you have more that one matching element, like this input:

{
    "rows": [["A1", "B1", "C1"], ["A2", "B2", "C2"], ["A3", "B3", "C3"]],
    "columns": ["A", "B", "C", "B1"]
}

Output:

{
    "rows": [["B1"], ["B2"], ["B3"]],
    "columns": ["B","B1"]
}