On Tue, 06 Dec 2011 19:15:26 -0500, Jonathan M Davis <[email protected]> 
wrote:
On Tuesday, December 06, 2011 18:26:39 Adam D. Ruppe wrote:
Kai Meyer Wrote:
> I like it. Any reason something like this doesn't already exist in
> std.json?
I think it's just that nobody has done updates to std.json for a while, but
I don't really know.

I believe that Robert Jacques has a revised version of std.json that he
intends to put up for review, but it relies on changes that he's made to
std.variant which also need be reviewed.

- Jonathan M Davis

Correct. Andrei suggested that since JSONValue is a tagged union, it should 
really use std.Variant.Algebraic as its representation, which necessitated 
fixing both Variant and Algebraic so that was possible. The code and 
documentation links are below if you'd like to try it out/comment on it. The 
only thing unintuitive in my revision from your use case is creating an empty 
JSON object or array:

JSON.Value myObject = JSON.Value[string].init;
JSON.Value myArray  = JSON.Value[].init;

Since this is undocumented :( I think I'm going to add some aliases 
(JSON.Object_init, JSON.Array_init) to simplify things for people.

https://jshare.johnshopkins.edu/rjacque2/public_html/json2.mht
https://jshare.johnshopkins.edu/rjacque2/public_html/variant.mht

https://jshare.johnshopkins.edu/rjacque2/public_html/json2.d
https://jshare.johnshopkins.edu/rjacque2/public_html/variant.d

Reply via email to