Job Parameters in AWS Glue Notebooks

Question

How can a define a custom job parameters when working with an aws glue notebook?
I tried to use the %%configure magic but it does not seems to work. I want to define some parameters, resolve them with getResolvedOptions and making them overridable when running the notebook.

Andrea Campolonghi · Answer

A question. What does -f mean after the %%configure magic?

Gabriel Olaleye · Answer

Hello Andrea,

In AWS Glue Notebooks, you can define custom job parameters and make them overridable when running the notebook by using the `glueContext` object. You can use the `getResolvedOptions` method to access the parameters and set their default values. Here's a step-by-step guide:

1. **Import AWS Glue Libraries**:

At the beginning of your AWS Glue Notebook, import the necessary libraries and create a `glueContext` object. This context will be used to access and resolve custom job parameters.

```python
   import sys
   from awsglue.context import GlueContext
   from pyspark.context import SparkContext
   
   sc = SparkContext()
   glueContext = GlueContext(sc)
   ```

2. **Define Custom Job Parameters**:

Below the context setup, define your custom job parameters using `getResolvedOptions`. You can provide default values for these parameters. Here's an example:

```python
   from awsglue.utils import getResolvedOptions
   
   # Define your custom parameters with default values
   args = getResolvedOptions(sys.argv, ['myParam1', 'myParam2'])
   myParam1 = args['myParam1'] if 'myParam1' in args else 'default_value1'
   myParam2 = args['myParam2'] if 'myParam2' in args else 'default_value2'
   ```

3. **Access and Use the Parameters**:

Now that you have defined your custom parameters, you can access and use them in your notebook as needed. For example:

```python
   print(f"Custom Parameter 1: {myParam1}")
   print(f"Custom Parameter 2: {myParam2}")
   ```

4. **Running the Notebook**:

When you run the notebook, you can override these custom parameters by providing them as command-line arguments. For example:

```
   --myParam1 my_value1 --myParam2 my_value2
   ```

These values will override the default values you defined in the notebook.

5. **Using `%%configure` Magic (Optional)**:

While `%%configure` is typically used for configuring the Spark environment in AWS Glue Notebooks, it can also be used to set custom parameters if needed. Here's an example:

```python
   %%configure -f
   {
       "myParam1": "custom_value1",
       "myParam2": "custom_value2"
   }
   ```

This will set the parameters to the specified values for the duration of the notebook's session.

By following these steps, you can define custom job parameters in AWS Glue Notebooks, set default values, and make them overridable when running the notebook by passing command-line arguments or using the `%%configure` magic command.

Please give me a thumbs up if my suggestion help

Job Parameters in AWS Glue Notebooks

Relevant content