GROQ: Flatten nested array and extract specific property (usedTime) from tasks

Based on your query structure, you're dealing with nested arrays and trying to flatten them while extracting a specific property. The key issue is that you need to properly flatten the nested tasks[].tasks[] structure and then project the usedTime property.

Here's how to achieve what you're after:

*[_type == "internalCustomers" && 'myMail@mail.com' in users[]->email]{
  "totalHours": tasks[].tasks[].usedTime
}

This syntax tasks[].tasks[].usedTime will:

1. Iterate through the outer tasks array
2. For each item, iterate through the inner tasks array
3. Extract the usedTime property from each task
4. Automatically flatten the result into a one-dimensional array

The double array projection ([][]) handles the flattening for you - GROQ automatically flattens nested arrays when you chain array projections like this, as explained in the pipeline components documentation.

Filtering out null values

If you want GROQ to omit null/undefined values (which it sounds like you do), you can add a filter using the defined() function:

*[_type == "internalCustomers" && 'myMail@mail.com' in users[]->email]{
  "totalHours": tasks[].tasks[defined(usedTime)].usedTime
}

This filters the inner tasks array to only include items where usedTime is defined before extracting the values.

Why your attempts returned null

Looking at your attempts:

• tasks[].tasks->usedTime - The -> operator is for following references, not accessing properties
• tasks[].tasks[]->usedTime - Same issue with ->
• tasks[].tasks[]->.usedTime - Same reference operator issue
• tasks[].tasks.usedTime - Missing the second [] to properly iterate through the nested array

The correct syntax for accessing properties in nested arrays is to use [] for each array level, then access the property directly with dot notation. The -> operator is specifically for dereferencing document references, not for property access.