# Accessing Metadata Programmatically

* [Programs](#programs)
* [Nodes](#nodes)

Metadata can be manipulated and updated within both nodes and programs in Algoreus.

This feature can be employed for metadata-centric processing. For instance, if you have a dataset with a field that contains sensitive information and your organization enforces a policy to mask all such sensitive information, you can tag the sensitive field accordingly. Following this, a node can be designed that reads the metadata of all fields in the dataset and masks the content of a field if it is tagged as sensitive.

### Programs 

Metadata can be accessed by leveraging methods from the MetadataReader object. These are accessible through the appropriate program context object in your program, obtained by invoking the `getContext()` method within the initialize method of your program.

Here's an example of how you can fetch the metadata of an entity:

```
@Override
public void initialize() throws Exception {
  MapReduceContext context = getContext();

  // construct the metadata entity
  MetadataEntity entity = MetadataEntity.builder().append(MetadataEntity.NAMESPACE, "myNamespace")
    .appendAsType(MetadataEntity.DATASET, "myDataset").build();

  // get the metadata
  Map<MetadataScope, Metadata> metadata = context.getMetadata(entity);
}
```

Similarly, you can tag metadata to an entity through methods provided by the `MetadataWriter` object. This object is accessible through the same program context object.

Here's an example of how you can tag metadata to an entity:

```
@Override
public void initialize() throws Exception {
  MapReduceContext context = getContext();

  // construct the metadata entity
  MetadataEntity entity = MetadataEntity.builder().append(MetadataEntity.NAMESPACE, "myNamespace")
    .appendAsType(MetadataEntity.DATASET, "myDataset").build();

  // add a tag
  context.addTags(entity, "someTag");
}
```

### Nodes&#x20;

Within a node, metadata can be accessed through methods from the `MetadataReader` object, available via the relevant context provided in the `prepareRun` stage.

Here's an example of how you can retrieve the metadata of an entity within a node:

```
@Override
public void prepareRun(BatchSourceContext context) throws Exception {
  context.setInput(Input.ofDataset(config.tableName));

  // construct the metadata entity
  MetadataEntity entity = MetadataEntity.builder().append(MetadataEntity.NAMESPACE, "myNamespace")
    .appendAsType(MetadataEntity.DATASET, "myDataset").build();

  // get the metadata
  Map<MetadataScope, Metadata> metadata = context.getMetadata(entity);
}
```

***