On Dec 12, 2011, at 4:51 PM, Peter van Hardenberg wrote: > Because we haven't heard from him in a while we've been using PL/V8 to > validate a JSON datatype simulated by a DOMAIN with a simple > acceptance function. (See below.) This is not ideally performant but > thanks to V8's JIT the JSON parser is actually reasonably good. > > I think releasing something simple and non-performant with reasonable > semantics would be the best next step. If it were up to me, I'd > probably even try to just land PL/V8 as PL/JavaScript for 9.2 if the > crash bugs and deal breakers can be sifted out.
Note that Claes Jakobsson has been working on a JSON data type using the Jansson JSON library. http://pgxn.org/dist/pg-json/ I’ve submitted a pull request renaming it to jansson-json (though the data type is still "json"): https://github.com/theory/pg-json/tree/pgxn Anyway, it seems like a decent start to an extensible type implemented entirely as an extension. Claes tells me he plans to add index support soonish, so it could get to be pretty robust before long. Just another stab at the problem to alert folks to. Best, David -- Sent via pgsql-hackers mailing list (pgsql-hackers@postgresql.org) To make changes to your subscription: http://www.postgresql.org/mailpref/pgsql-hackers
