Further compaction

[Dobes Vandermeer]

Have you considered making the encoding even more efficient by omitted key names for required fields, altogether?

e.g.

{
    "firstName": "Corrine",
    "lastName": "Ziemann",
    "title": "Ms.",
    "gender": "f",
    "zipCode": 75963,
    "countryCode": "en",
    "birthYear": 1960,
    "active": false,
    "shoppingCartItems": [
        {
            "productNumber": 29857,
            "amount": 1
        },
        {
            "productNumber": 53409,
            "amount": 6
        }
    ]
}

could become

["Corrine", "Ziemann", "Mrs.", false, [
        [29857, 1],
        [53409, 6]
    ],{
    "gender": "f",
    "zipCode": 75963,
    "countryCode": "en",
    "birthYear": 1960
}
]

Assuming that gender, zipCode, etc. are optional and the other fields are marked as required.

It seems like this could reduce the file size quite significantly.

Also you could consider applying the compression of graphql-crunch, which re-uses prior values if they are repeated.

[Daniel Meyer]

Sorry for the late reply.
Yes your proposal would increase the effiency. But also it would change the whole structure and make if impossible to keep compressQuery() working.
This module is mainly made to be used with NoSQL-Databases.