Transform array of objects to named object keys in GROQ query

Unfortunately, GROQ doesn't have built-in functionality to dynamically create objects with keys derived from field values. You cannot directly transform your array structure into an object keyed by the day field within a GROQ query.

The limitation is that GROQ doesn't support:

• Dynamic object key creation from field values
• Functions like object::merge() or similar reduce operations to build keyed objects
• Using computed values as object keys in projections

Your best options:

1. Transform client-side (recommended)

This is the most practical approach. Query the array as-is and transform it with a simple reduce:

const result = await client.fetch(`*[_type == "yourType"][0]{ hours }`);

const hoursKeyed = result.hours.reduce((acc, item) => {
  acc[item.day] = { from: item.from, to: item.to };
  return acc;
}, {});

2. Restructure your schema

If you frequently need this keyed format, consider storing it as an object from the start:

hours: {
  monday: { from: "9:00AM", to: "5:00PM" },
  tuesday: { from: "9:00AM", to: "5:00PM" },
  // ...
}

This eliminates transformation needs, though it's less flexible if days vary.

3. Hardcode projections (not recommended)

If you have a fixed set of days, you could manually project each:

*[_type == "location"]{
  "hours": {
    "monday": hours[day == "monday"][0],
    "tuesday": hours[day == "tuesday"][0],
    "wednesday": hours[day == "wednesday"][0]
    // ... etc
  }
}

This works but is tedious and unmaintainable.

The bottom line: GROQ excels at querying and projecting data, not complex restructuring. For transformations like array-to-keyed-object conversions, client-side processing is the right tool. It's cleaner, more maintainable, and only adds a few lines of code.