Pretty self-explanatory. JSON is a lot easier to parse than CSV. :)

I'm curious, where do you need to parse the dumps such that JSON is an easier option? Thus far I've only heard people wanting to import the dumps into a spreadsheet, where CSV works best.

If I'm going to add the JSON output format, what kind of structure should it have? Would it be something like this?

{
    "title": [ "col 1", "col 2", "col 3" ],
    "data":
    [
        [ "foo", "bar", "baz" ],
        [ "cat", "dog", "penguin" ]
    ]
}

I was thinking something like this:

{
  "Minecraft": {
     "minecraft:acacia_door": {
        "BlockID" : 196,
        "Subtypes": false,
        "ItemID": null,
         etc
     },
     "minecraft:cobblestone": {
         ...
     }
  }, 
  "Chisel": {
      ....
  },
   ...
}

Hopefully my indentation isn't too wonky. Essentially, a tree structure with mods at the top, then the blocks in each mod, and then each block with key/value pairs for the rest of the data.

The reason I'm looking for this is that I'm planning to write a web-based palette generation helper for Lost Cities, and parsing JSON in Javascript is a non-issue. People will need to be able to paste in their own block lists for it to be useful, though.

Here is an example of the current output format: https://pastebin.com/uXxrqSEE

Ah okay... This dump variant will then need entirely its own dump code to make such a data structure. Would it be enough to only implement it for the block registry then? For your use case I assume that would be enough? It would probably be a fair amount of work to add this completely separate dump style for everything...

Yes, that's exactly what I was looking for. Thanks!!