Latest version of our Java SDK (2.3.3 at the time of writing) introduces a new class: RawQueryExecutor
. It gives you the direct result of a N1QL or FTS query as a JSON String or JsonObject, just like if you were using the REST API directly. The answer is not wrapped in rows. As such you can pass the result of the method directly to your client. No need for additional mapping or anything.
This is particularly useful for N1QL queries as it allows you to query exactly what you need through the SELECT clause. It is very efficient combined with the right covering indexes.
Here's the asynchronous code sample:
1 2 3 |
AsyncRawQueryExecutor asyncRawQueryExecutor= new AsyncRawQueryExecutor("default", "", bucket.core()); Observable<String> result = asyncRawQueryExecutor.n1qlToRawJson(N1qlQuery.simple("Select name, abv, ibu, style FROM `beer-sample` WHERE ibu IS NOT MISSING ORDER BY ibu LIMIT 10 OFFSET 10")); |
and the synchronous one:
1 2 3 |
RawQueryExecutor rawQueryExecutor = new RawQueryExecutor("default", "", bucket.core(), bucket.environment()); String jsonString = rawQueryExecutor.n1qlToRawJson(N1qlQuery.simple("Select name, abv, ibu, style FROM `beer-sample` WHERE ibu IS NOT MISSING ORDER BY ibu LIMIT 10 OFFSET 10")); |
I would love to know if you find it useful and if you want this also for simpler operations. Tell us what you think!