> For the complete documentation index, see [llms.txt](https://docs.augelab.com/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://docs.augelab.com/function-blocks/blocks-reference/data-logic/data-operations/data-memory.md).

# Data Memory

This function block keeps a value in memory and optionally updates it when requested. It is useful for preserving the last known data across runs or for freezing a value until an explicit update is provided.

## 📥 Inputs <a href="#inputs" id="inputs"></a>

`Generic` General input data to store in memory. Can be any data type (numbers, text, image, lists, dictionaries, etc.).\
`Update Data` Boolean input that controls whether the incoming data should replace the stored value.

## 📤 Outputs <a href="#outputs" id="outputs"></a>

`Generic` The currently stored value. If never updated, this may be empty/None.

## 🕹️ Controls <a href="#controls" id="controls"></a>

This block has no additional UI widgets. Behavior is controlled by the two sockets listed above.

## 🎯 Features <a href="#features" id="features"></a>

* Persistent in-flow memory: holds a value across evaluations and only changes when `Update Data` is true.
* Accepts any data type on the `Generic` socket.
* Simple freeze/unfreeze mechanic: set `Update Data` to False to freeze the stored value, True to allow updates.

## 📝 Usage Instructions <a href="#usage" id="usage"></a>

1. Feed the value you want to keep into the `Generic` input.
2. Toggle `Update Data` to True when you want the memory to accept the new input.
3. Set `Update Data` to False to freeze the stored value; the block will keep outputting that value until an update is allowed.

## 📊 Evaluation <a href="#evaluation" id="evaluation"></a>

On each evaluation the block checks the `Update Data` input. If True, it stores the current `Generic` input and outputs it. If False, it outputs the previously stored value unchanged.

## 💡 Tips and Tricks <a href="#tips-and-tricks" id="tips-and-tricks"></a>

* Use with `Is None` to detect missing inputs before writing into memory, keeping the stored value safe from accidental None overwrites.
* Pair with `Replace None` to provide fallback data when the input is empty, then control writes with `Update Data`.
* Combine with `Data Read Local` / `Data Write Local` or `Data Read Global` / `Data Write Global` for more persistent sharing between different parts of your flow.
* Use alongside `Debug Input` to log stored values during development and verify when memory updates happen.
* When processing batches, coordinate with `Batch Processing`, `Get Batch Size`, or `Batch Concatenation` to control whether memory should store whole batches or individual elements.
* For logging or export scenarios, feed the block output into `CSV Export`, `Image Logger`, or other export blocks to record the preserved data only when a confirmed update occurs.

## 🛠️ Troubleshooting <a href="#troubleshooting" id="troubleshooting"></a>

* If output never changes, ensure the `Update Data` input is being set to True when you expect an update.
* If stored value becomes None unexpectedly, check upstream blocks for None outputs and consider using `Replace None` before writing to memory.
* If multiple areas of your flow require the same value, prefer the read/write local/global blocks to avoid duplication and ensure consistent updates.