Polish `AbstractJackson2Decoder#decodeToMono`

[ShenFeng312]

	@Override
	public Mono<Object> decodeToMono(Publisher<DataBuffer> input, ResolvableType elementType,
			@Nullable MimeType mimeType, @Nullable Map<String, Object> hints) {

		return Mono.deferContextual(contextView -> {

			Map<String, Object> hintsToUse = contextView.isEmpty() ? hints :
					Hints.merge(hints, ContextView.class.getName(), contextView);

			return DataBufferUtils.join(input, this.maxInMemorySize).flatMap(dataBuffer ->
					Mono.justOrEmpty(decode(dataBuffer, elementType, mimeType, hintsToUse)));
		});
	}

In decodeToMono, it seems that all DataBuffers need to be awaited before deserialization can proceed. This implies higher memory overhead. Why not use NonBlockingJsonParser instead like decode? It seems that this could save memory and avoid blocking.

[ShenFeng312]

        Jackson2JsonDecoder jackson2JsonDecoder = new Jackson2JsonDecoder();
        byte[][] jsonChunks = {
                "{\"name\":\"Alice\",\"age\":".getBytes(),
                "25,\"city\":\"New York\"}".getBytes()
        };

        DataBuffer[] buffers = new DataBuffer[jsonChunks.length];
        for (int i = 0; i < jsonChunks.length; i++) {
            buffers[i] = new DefaultDataBufferFactory().wrap(jsonChunks[i]);
        }
        Flux<Object> decode = jackson2JsonDecoder.decode(Flux.fromArray(buffers), ResolvableType.forType(Map.class), null, null);

        var o = decode.collectList().block();
        System.out.println(o.getFirst());

I tried converting using the above code, and it succeeded. I think we can optimize it well as long as we impose some limitations.