> ## Documentation Index
> Fetch the complete documentation index at: https://docs.prophecy.ai/llms.txt
> Use this file to discover all available pages before exploring further.

# ColumnParser

> Parse XML or JSON inside a table

<Panel>
  <Info>
    Dependencies:

    * ProphecySparkBasicsPython 0.2.27+
    * ProphecyLibsPython 1.9.16+
    * ProphecyLibsScala 8.2.1+
  </Info>

  <Info>
    Cluster requirements:

    * UC dedicated clusters 14.3+ supported
    * UC standard clusters 14.3+ supported
    * Livy clusters not supported
  </Info>
</Panel>

The ColumnParser lets you parse XML or JSON that is included in a column of your table.

## Parameters

| Parameter          | Description                                                              |
| ------------------ | ------------------------------------------------------------------------ |
| Source Column Name | Name of the column that contains the XML or JSON records.                |
| Parser Type        | Format of the column you want to parse (XML or JSON).                    |
| Parsing Method     | How Prophecy will derive the schema that will be used to parse the data. |

When you select a parsing method, you have three options:

* **Parse automatically.** Prophecy infers the schema by reading the first 40 records.
* **Parse from sample record.** Prophecy uses the schema that you provide in the sample record.
* **Parse from schema.** Prophecy uses the schema that you provide in the form of a schema struct.

## Output

The schema of the ColumnParser gem output includes the parsed content as a **struct** data type, in addition to all of the input columns.

<img src="https://mintcdn.com/prophecy-62973bd0/5tvw_2e98LqB5rz7/data-engineering/gems/transform/img/new-output-struct.png?fit=max&auto=format&n=5tvw_2e98LqB5rz7&q=85&s=935c3535f8b51e1789019f6e1e2a8837" alt="New output struct" width="2620" height="1508" data-path="data-engineering/gems/transform/img/new-output-struct.png" />

## Example code

<Tip>
  To see the compiled code of your project, [switch to the Code view](/data-engineering/development/pipelines/pipelines#project-editor) in the project header.
</Tip>

This example shows the code to parse XML.

<CodeGroup>
  ```python example.py theme={null}
  def xml_column_parser(spark: SparkSession, in0: DataFrame) -> DataFrame:
   from prophecy.libs.utils import xml_parse

   return xml_parse(in0, "XML", "parseAuto", None, None)
  ```
</CodeGroup>