Json dict

[DaddyWesker]

So, I've added rust implementation of json dict discussed here and in person with @Necr0x0Der . It is draft since I haven't finished writing tests and also there will be additions I'm sure.

Also, I've got metta implementation of get-all-keys and get-value-by-key functions

(= (get-all-keys $jsondict) (match $jsondict ($key $value) $value))

(= (get-value-by-key $jsondict $key) (match $jsondict ($key $value) $value))

I don't know if we need them and where to put them. Anyway, here is the usage of current functions:

!(bind! &jsondict (json-dict ((key1 value1) (key2 value2))))

!(assertEqual 
    (get-all-keys &jsondict) 
    (superpose (key1 key2)))

!(assertEqual 
    (get-value-by-key &jsondict key1) 
    value1)

[Necr0x0Der]

A minor comment
(= (get-all-keys $jsondict) (match $jsondict ($key $value) $value))
should be
(= (get-all-keys $jsondict) (match $jsondict ($key $value) $key))

[Necr0x0Der]

I've got metta implementation of get-all-keys and get-value-by-key functions

It would be more interesting to have metta implementation of creating a space with the corresponding content (maybe, adding the content via foldl-atom) and compare performance with Rust implementation

[DaddyWesker]

A minor comment (= (get-all-keys $jsondict) (match $jsondict ($key $value) $value)) should be (= (get-all-keys $jsondict) (match $jsondict ($key $value) $key))

Of course. My bad.

[DaddyWesker]

I've got metta implementation of get-all-keys and get-value-by-key functions

It would be more interesting to have metta implementation of creating a space with the corresponding content (maybe, adding the content via foldl-atom) and compare performance with Rust implementation

Okay, I'll test. Though I thought about using add atom function.

[Necr0x0Der]

On a conceptual level, the current implementation has nothing specific about jsons. Creation of a space with certain content (underwent reduction) is generally applicable. Such the function can be created and added to stdlib (say, new-space-reduct or something like that). It can be done separately from json-related stuff.
The next question is if we want to use such non-specific representation for spaces as dicts (just adding pairs and treat them as key-value pairs). Or we want it to be more specific, e.g. ((Key key1) (Value value1)). The latter will be less efficient, but it will prevent misuse of json-related functions.

[Necr0x0Der]

Though I thought about using add atom function.

Yes, folding with add-atom over the given tuple.

[DaddyWesker]

It was decided to make new pr and leave only metta-implemented functions. Closing this one.