So how is JSON a graph? Or maybe the answer is just, No. Not generally. I'm confused by your question.

No. JSON is not a graph.

Any visualization of a graph from JSON would be a specific encoding of a graph (probably using IDs) which is not generally interpreted as a graph.

{ Left:{}, Right:{} }

Has a tree with 3 nodes

If you don't parse the JSON, you tend to traverse it as a graph

If cyclicallity is needed specifically, cyclic graphs do exist, but it was my impression that JSON doesn't have these, or I'm unable to imagine how that would work. While two fields may have the same 'value' in a JSON object, I don't know how I could express cyclicallity using normal JSON syntax to say that two identical values would be linked cyclically.

[0] https://www.rfc-editor.org/rfc/rfc6901

https://tree-merge.com/merging-categories-4-independant-diy-...

The conclusion I came to here is that nested/indented lists are actually quite nice to visualise trees. JSON text is already a good visual UI.

Now, this is light years from PlantUML, but Vega abd Vega-Lite can do just about anything with JSON data. It's the most powerful textual graph description vocabulary I've ever seen, by far, but it's too verbose for my average user.

Having said that I really like this new JSONCRACK a lot, now that I've played with it a little.

Not sure how to get rid of the enclosing PlantUML pragmas:

  @startjson
  {
     "fruit":"Apple",
     "size":"Large",
     "color": ["Red", "Green"]
  }
  @endjson

  //Asciidoc follows
  [plantuml]
  ----
  @startjson
  include::../path/to/fruit.json[]
  @endjson
  ----
  ```

Also, a little meta, but I like that the github page was linked to rather than directly linking to the product's page. It lets me know upfront that this is open source something I'm dealing with. +1 to the submitter

Particularly this digraph() example: https://observablehq.com/@d3/visiting-a-d3-hierarchy. With a bit of modification away from the GraphViz format, the result fits straight into one of the many graph component visualization libraries like react-flow.

You could visually diff a whole lot of reports, pretty much instantly. Our PM had a paid tool that could do this, but wasn't sure what it was.

[0] https://github.com/Wilfred/difftastic [1] https://difftastic.wilfred.me.uk/languages_supported.html

[0] https://github.com/seperman/deepdiff

Once I wrote a simple script to convert json files to `path : value\n` form; that is getting this

."a" : 1 ."b" : {} ."b"."a" : "" ."b"."c" : [] ."b"."c"[0] : true

from the json { "a": 1, "b": { "a": "", "c": [true] } }

It made better diffs but it was also harder to read.

[0] https://marketplace.visualstudio.com/items?itemName=richie5u...

https://github.com/TomConlin/json_to_paths

https://www.json-buddy.com/