In Cloud Data Fusion Studio, the Wrangler workspace lets you encode and
decode values across your dataset rows. Converting between formats helps you
safely handle binary payloads, format URL parameters, and prepare records for
downstream processing.

## Encode a row

You can use base encoding of data to store or transfer data in environments
that, for legacy reasons, are restricted to US-ASCII data. You might use it in
new applications without those legacy restrictions because it allows the
manipulation of objects with text editors.

You can apply the following encoding schemes, which are based
on [RFC-4648](https://tools.ietf.org/html/rfc4648),
to all values in a column:

- Base32
- Base64
- Hex
- URL

When you encode, Wrangler generates a new column with a name in the following
format: `<column>_encode_<type> except for url-encode`.

Cloud Data Fusion uses the following rules for the column values:

- If the column is `null`, the resulting column is also `null`.
- If the chosen column isn't found in the row, the row is skipped.
- If the column value doesn't have a string or byte data type, the transformation fails, and an error displays.

### Supported encoding options

Wrangler supports the following encoding options:

Encode base64
:   The **Base64** option adds the `encode64` directive as a transformation step
    to the recipe and creates a new column with encoded values.

Encode base32
:   The **Base32** option adds the `encode32` directive as a transformation step
    to the recipe and creates a new column with encoded values.

Encode hex
:   The **Hex** option adds the `encode_hex` directive as a transformation step
    to the recipe and creates a new column with encoded values.

Encode URL
:   The **URL** option adds the `url-encode` directive as a transformation step
    to the recipe and encodes the current column.

## Decode a row

You can use base decoding of data to store or transfer data in environments
that, for legacy reasons, are restricted to US-ASCII data. You might use it in
new applications without those legacy restrictions because it allows the
manipulation of objects with text editors.

You can apply the following decoding schemes, which are based
on [RFC-4648](https://tools.ietf.org/html/rfc4648),
to each value in a column:

- Base32
- Base64
- Hex
- URL

When you decode, Wrangler generates a new column with a name in the following
format: `<column>_encode_<type>`, except for `url-decode`.

Cloud Data Fusion uses the following rules for the column values:

- If the column is `null`, the resulting column is also `null`.
- If the chosen column isn't found in the row, the row is skipped.
- If the column value doesn't contain the string or byte array data types, the operation fails.

### Supported decoding options

Wrangler supports the following encoding options:

Decode base64
:   The **base64** option adds the `decode64` directive as a transformation step
    to the recipe and creates a new column with the decoded values.

Decode base32
:   The **base32** option adds the `decode32` directive as a transformation step
    to the recipe and creates a new column with the decoded values.

Decode hex
:   The **Hex** option adds the `decode hex` directive as a transformation step
    to the recipe and creates a new column with the decoded values.

Decode URL
:   The **URL** option adds the `url-decode` directive as a transformation step
    to the recipe and decodes the current column.

## What's next

- Learn more about [Wrangler directives](https://docs.cloud.google.com/data-fusion/docs/concepts/wrangler-overview#apply_directives).