copy-pasting the example result from XTDB OpenAPI Specification
into JSON-LD Playground
results in emptiness-or-validation errors (only visualization works) which is expected… no context.
btw… the only 2 libs handling this json-ld stuff in python, are last touched about
7-10y ago. Maybe the stuff did not change… like transit? that is also from 2014
btw in my current setup, seems xtdb is the bottleneck, and actually talking to it (i.e. EDN-ing or Transit-ing ), not the actual work on either side. Looking at above RDF stuff, i don’t think json-LD would be any faster.
Hey @svil it’s probably better to think of it as a regular JSON API that happens to borrow JSON-LD’s mechanism for encoding non-JSON types, e.g. encoding a duration as an object:
…where the use of @type, @value and xt: namespacing looks approximately like the JSON-LD conventions.
I don’t think there is anything else in the API currently which resembles or behaves like JSON-LD in particular - so the confusion here is almost certainly my fault due to over-emphasising the “LD” suffix too much - apologies!
Does that help?
how about gRPC ?
It’s definitely interesting - and somewhat similar to https://substrait.io/ - for the time being though we hope the JSON API has lowered the barrier for client experimentation and development. In contrast, and as far as I’m aware, the gRPC toolchains across various languages are still nowhere near as easy to get started with compared to JSON.