# Passthrough null values

**URL:** <https://forum.confluent.io/t/passthrough-null-values/37820>\
**Category:** Kafka Streams\
**Created:** [23 May 2025 14:45 UTC](https://forum.confluent.io/t/passthrough-null-values/37820 "2025-05-23T14:45:37Z")\
**Posts on this page:** 3\
**Page:** 1

<div class="post-metadata">

**Author:** ![foal](https://sea1.discourse-cdn.com/flex019/user_avatar/forum.confluent.io/foal/32/3692_2.png) [@foal](https://forum.confluent.io/u/foal)\
**Post date:** [23 May 2025 14:45 UTC](https://forum.confluent.io/t/passthrough-null-values/37820/1 "2025-05-23T14:45:37Z")

</div>

Hi,

Is it possible to passthrough null values (KTable) even if no internal state for such key?

Right now, the Kafka streams application optimizes it.

Best,  
Stanislav

---

<div class="post-metadata">

**Author:** ![mjsax](https://sea1.discourse-cdn.com/flex019/user_avatar/forum.confluent.io/mjsax/32/3113_2.png) [@mjsax](https://forum.confluent.io/u/mjsax)\
**Post date:** [25 May 2025 16:04 UTC](https://forum.confluent.io/t/passthrough-null-values/37820/2 "2025-05-25T16:04:51Z")

</div>

Not 100% sure, but I believe if you would disable caching it might pass through all records?

Not sure what your overall application looks like, but using a manual `process()` (or `transform()` in older versions) would also allow you to do this – but not sure if this would fix into your overall application code.

You could also read the data as `KStream`, and do a “broadcast” and later merge…? Something like this:

```auto
KStream input = ...

KStreams nullStream = input.filter(/*only keep nulls*/);
KStream tableStream = input.toTable(...).toStream();

KStream merged = tableStream.merge(nullStream);

```

---

<div class="post-metadata">

**Author:** ![foal](https://sea1.discourse-cdn.com/flex019/user_avatar/forum.confluent.io/foal/32/3692_2.png) [@foal](https://forum.confluent.io/u/foal)\
**Post date:** [25 May 2025 16:18 UTC](https://forum.confluent.io/t/passthrough-null-values/37820/3 "2025-05-25T16:18:43Z")

</div>

> “broadcast” and later merge

Nice idea. I will try.

Thank you!