TL;DR: Use
Filter(COLLECTION, Defaults(Source))to load a data source’s empty schema into a collection so later patch operations align perfectly.
💡 Challenge
You need to load a data source structure into a collection, especially if you plan to patch the collection back to the data source later on.
✅ Solution
Use the Filter function together with Defaults() in Power Apps to load the empty structure of the data source into your collection efficiently.
🔧 How It’s Done
1. Load the empty schema
🔸 Use Filter(COLLECTION, Defaults(Source)) to create a blank record schema in your collection.
🔸 This ensures your collection matches the data source fields for smooth patch operations.
2. Populate the collection with the proper structure
🔸 Use ClearCollect(MyCollection, Filter(COLLECTION, Defaults(MyDataSource))).
🔸 Replace MyDataSource with your actual data source name.
🎉 Result
You’ve set up a structured collection that matches the data source schema, ensuring smoother data handling and an enhanced app experience.
🌟 Key Advantages
🔸 Validation & Data Integrity: Minimizes data mismatches and validation errors by matching the schema
🔸 Performance Boost: Works with in-memory collections, reducing direct calls to the data source
🔸 Data Preparation: Simplifies data transformation before sending it back
🛠️ FAQ
Q: Why use Defaults(MyDataSource) in the Filter function?
Defaults(MyDataSource) returns a blank record with the correct schema. Filtering with it creates a collection that matches the data source structure without retrieving actual data.
Q: Will this method retrieve any existing records from the data source?
No. Filter(COLLECTION, Defaults(MyDataSource)) only uses the Defaults record to shape your collection’s schema. To load actual data, use Filter(MyDataSource, ...) or Collect(MyCollection, MyDataSource).
Q: How do I patch data back to the data source using this structured collection?
Use Patch(MyDataSource, Defaults(MyDataSource), MyCollection) or loop through the collection to update or create records. Matching fields ensure smooth data operations.
Marcel Lehmann
Microsoft MVP