SwiftData vertegenwoordigt de toekomst van data-persistentie op Apple-platforms. Geïntroduceerd op WWDC 2023 biedt dit framework een native Swift-syntax en een naadloze integratie met SwiftUI. Voor bestaande Core Data-toepassingen vormt de migratie een strategische stap richting moderner en beter onderhoudbare code.

Wat dit artikel behandelt

Deze gids beschrijft het volledige migratieproces van Core Data naar SwiftData: compatibiliteitsbeoordeling, modelconversie, datamigratiestrategieën en coexistence-patronen voor een geleidelijke overgang.

Beoordeling van de Migratiehaalbaarheid

Voor het starten van de migratie maakt een grondige beoordeling het mogelijk potentiële obstakels te identificeren. Core Data en SwiftData delen dezelfde SQLite-persistentie-engine, waardoor de gegevens volledig compatibel zijn.

MigrationAssessment.swiftswift

// Migration assessment checklist

/*
 FEATURES SUPPORTED BY SWIFTDATA:
 ✅ Simple models with basic properties
 ✅ One-to-one and one-to-many relationships
 ✅ Optional properties and default values
 ✅ Transformable attributes (via Codable)
 ✅ CloudKit synchronization (basic)
 ✅ Automatic lightweight migrations
 ✅ Class inheritance (iOS 26+)

 FEATURES REQUIRING ATTENTION:
 ⚠️ NSFetchedResultsController → @Query + manual observation
 ⚠️ NSCompoundPredicate → #Predicate with combined logic
 ⚠️ Dynamic predicates → Workarounds required

 UNSUPPORTED FEATURES:
 ❌ Advanced CloudKit Sharing
 ❌ Derived attributes
 ❌ Fetched properties
*/

// Example of typical Core Data model to migrate
import CoreData

// Existing Core Data entity
class CDTask: NSManagedObject {
    @NSManaged var id: UUID
    @NSManaged var title: String
    @NSManaged var isCompleted: Bool
    @NSManaged var createdAt: Date
    @NSManaged var priority: Int16
    @NSManaged var category: CDCategory?
}

class CDCategory: NSManagedObject {
    @NSManaged var id: UUID
    @NSManaged var name: String
    @NSManaged var color: String
    @NSManaged var tasks: NSSet?
}

Compatibiliteit op datablock-niveau betekent dat gebruikers hun bestaande informatie behouden na de migratie. Er treedt geen dataverlies op wanneer het proces correct wordt uitgevoerd.

Conversie van Core Data-modellen naar SwiftData

De eerste concrete stap bestaat uit het converteren van Core Data-entiteiten naar SwiftData-klassen. Xcode biedt een automatisch hulpmiddel, maar het begrip van het handmatige proces blijft essentieel.

TaskModel.swiftswift

import SwiftData

// SwiftData equivalent of CDTask
@Model
final class Task {
    // Properties with default values
    var id: UUID = UUID()
    var title: String = ""
    var isCompleted: Bool = false
    var createdAt: Date = Date()
    var priority: Int = 0

    // Optional relationship to Category
    var category: Category?

    // Explicit initializer recommended
    init(
        id: UUID = UUID(),
        title: String,
        isCompleted: Bool = false,
        createdAt: Date = Date(),
        priority: Int = 0,
        category: Category? = nil
    ) {
        self.id = id
        self.title = title
        self.isCompleted = isCompleted
        self.createdAt = createdAt
        self.priority = priority
        self.category = category
    }
}

Belangrijke verschillen met Core Data zijn het gebruik van de @Model-macro in plaats van NSManagedObject, en native Swift-types in plaats van Objective-C-types.

CategoryModel.swiftswift

import SwiftData

@Model
final class Category {
    var id: UUID = UUID()
    var name: String = ""
    var color: String = "blue"

    // Inverse relationship with delete rule
    @Relationship(deleteRule: .cascade, inverse: \Task.category)
    var tasks: [Task] = []

    init(id: UUID = UUID(), name: String, color: String = "blue") {
        self.id = id
        self.name = name
        self.color = color
    }
}

Type-mapping

Core Data-types converteren rechtstreeks: Int16 wordt Int, NSSet wordt [Model] en Date blijft Date. Transformable-attributen vereisen het toepassen van Codable.

Configuratie van de ModelContainer

De ModelContainer van SwiftData vervangt de NSPersistentContainer van Core Data. De configuratie bepaalt waar en hoe de gegevens worden opgeslagen.

ModelContainerSetup.swiftswift

import SwiftData
import SwiftUI

@main
struct TaskManagerApp: App {
    // SwiftData container configuration
    var sharedModelContainer: ModelContainer = {
        // Schema including all models
        let schema = Schema([
            Task.self,
            Category.self
        ])

        // Configuration with storage options
        let modelConfiguration = ModelConfiguration(
            schema: schema,
            isStoredInMemoryOnly: false,
            // Use the same store as Core Data
            url: URL.applicationSupportDirectory
                .appending(path: "TaskManager.sqlite")
        )

        do {
            return try ModelContainer(
                for: schema,
                configurations: [modelConfiguration]
            )
        } catch {
            fatalError("Could not create ModelContainer: \(error)")
        }
    }()

    var body: some Scene {
        WindowGroup {
            ContentView()
        }
        .modelContainer(sharedModelContainer)
    }
}

Het cruciale punt zit in de URL van de store: door hetzelfde SQLite-bestand als Core Data te gebruiken, kan SwiftData de bestaande gegevens lezen.

Coexistence-strategie tussen Core Data en SwiftData

Voor complexe applicaties vormt een geleidelijke migratie via coexistence van beide frameworks de veiligste aanpak. Beide stacks kunnen toegang krijgen tot hetzelfde SQLite-bestand.

CoexistenceSetup.swiftswift

import CoreData
import SwiftData

// Configuration for coexistence
class PersistenceController {
    static let shared = PersistenceController()

    // Shared store between Core Data and SwiftData
    private let storeURL: URL = {
        let appSupport = FileManager.default
            .urls(for: .applicationSupportDirectory, in: .userDomainMask)
            .first!
        return appSupport.appending(path: "TaskManager.sqlite")
    }()

    // MARK: - Core Data Stack (existing)

    lazy var persistentContainer: NSPersistentContainer = {
        let container = NSPersistentContainer(name: "TaskManager")

        // Configure to use shared store
        let description = NSPersistentStoreDescription(url: storeURL)
        description.setOption(
            true as NSNumber,
            forKey: NSPersistentHistoryTrackingKey
        )
        container.persistentStoreDescriptions = [description]

        container.loadPersistentStores { _, error in
            if let error = error as NSError? {
                fatalError("Core Data error: \(error)")
            }
        }

        return container
    }()

    // MARK: - SwiftData Stack (new)

    lazy var swiftDataContainer: ModelContainer = {
        let schema = Schema([Task.self, Category.self])

        let config = ModelConfiguration(
            schema: schema,
            url: storeURL,
            // Disable automatic migrations in coexistence
            allowsSave: true
        )

        do {
            return try ModelContainer(for: schema, configurations: [config])
        } catch {
            fatalError("SwiftData error: \(error)")
        }
    }()
}

Synchronisatie van wijzigingen

In coexistence-modus zijn wijzigingen door het ene framework niet onmiddellijk zichtbaar voor het andere. Een expliciete reload of herstart van de app kan nodig zijn.

Migratie van Queries: Van NSFetchRequest naar @Query

Het belangrijkste verschil betreft de manier waarop gegevens worden opgehaald. SwiftUI gebruikt de property wrapper @Query om @FetchRequest te vervangen.

QueryMigration.swiftswift

import SwiftUI
import SwiftData

// ❌ Old pattern with Core Data
struct OldTaskListView: View {
    @FetchRequest(
        sortDescriptors: [
            NSSortDescriptor(keyPath: \CDTask.createdAt, ascending: false)
        ],
        predicate: NSPredicate(format: "isCompleted == NO")
    )
    private var tasks: FetchedResults<CDTask>

    var body: some View {
        List(tasks) { task in
            Text(task.title)
        }
    }
}

// ✅ New pattern with SwiftData
struct NewTaskListView: View {
    // @Query with built-in sorting and filtering
    @Query(
        filter: #Predicate<Task> { !$0.isCompleted },
        sort: \Task.createdAt,
        order: .reverse
    )
    private var tasks: [Task]

    var body: some View {
        List(tasks) { task in
            TaskRowView(task: task)
        }
    }
}

struct TaskRowView: View {
    let task: Task

    var body: some View {
        HStack {
            // Priority indicator
            Circle()
                .fill(priorityColor)
                .frame(width: 8, height: 8)

            VStack(alignment: .leading) {
                Text(task.title)
                    .font(.headline)

                if let category = task.category {
                    Text(category.name)
                        .font(.caption)
                        .foregroundStyle(.secondary)
                }
            }

            Spacer()

            // Date badge
            Text(task.createdAt, style: .date)
                .font(.caption2)
                .foregroundStyle(.tertiary)
        }
    }

    private var priorityColor: Color {
        switch task.priority {
        case 3: return .red
        case 2: return .orange
        case 1: return .yellow
        default: return .gray
        }
    }
}

Klaar om je iOS gesprekken te halen?

Oefen met onze interactieve simulatoren, flashcards en technische tests.

Ontdek iOS

Omgaan met Dynamische Predicates

Een belangrijke uitdaging bij SwiftData betreft dynamische predicates. In tegenstelling tot Core Data, waar predicates ter plekke kunnen worden aangepast, vereist @Query alternatieve benaderingen.

DynamicPredicates.swiftswift

import SwiftUI
import SwiftData

// Solution 1: Use @Query with custom init
struct FilteredTasksView: View {
    @Query private var tasks: [Task]

    // Create view with specific filter
    init(showCompleted: Bool, categoryId: UUID?) {
        // Build predicate based on parameters
        var predicates: [Predicate<Task>] = []

        if !showCompleted {
            predicates.append(#Predicate { !$0.isCompleted })
        }

        if let categoryId {
            predicates.append(#Predicate { task in
                task.category?.id == categoryId
            })
        }

        // Combine predicates
        let combinedPredicate: Predicate<Task>?
        if predicates.isEmpty {
            combinedPredicate = nil
        } else if predicates.count == 1 {
            combinedPredicate = predicates[0]
        } else {
            // Manually combine for AND logic
            combinedPredicate = #Predicate<Task> { task in
                !task.isCompleted && task.category?.id == categoryId
            }
        }

        _tasks = Query(
            filter: combinedPredicate,
            sort: \Task.createdAt,
            order: .reverse
        )
    }

    var body: some View {
        List(tasks) { task in
            TaskRowView(task: task)
        }
    }
}

// Solution 2: View-side filtering with all results
struct SmartTaskListView: View {
    // Fetch all tasks
    @Query(sort: \Task.createdAt, order: .reverse)
    private var allTasks: [Task]

    // Filter state
    @State private var searchText = ""
    @State private var showCompleted = false
    @State private var selectedCategory: Category?

    // Computed filtering
    private var filteredTasks: [Task] {
        allTasks.filter { task in
            // Text filter
            let matchesSearch = searchText.isEmpty ||
                task.title.localizedCaseInsensitiveContains(searchText)

            // Status filter
            let matchesStatus = showCompleted || !task.isCompleted

            // Category filter
            let matchesCategory = selectedCategory == nil ||
                task.category?.id == selectedCategory?.id

            return matchesSearch && matchesStatus && matchesCategory
        }
    }

    var body: some View {
        NavigationStack {
            List(filteredTasks) { task in
                TaskRowView(task: task)
            }
            .searchable(text: $searchText)
            .toolbar {
                FilterMenu(
                    showCompleted: $showCompleted,
                    selectedCategory: $selectedCategory
                )
            }
        }
    }
}

Versiegebaseerde Schema-migraties

Wanneer het datamodel evolueert, gebruikt SwiftData VersionedSchema om complexe migraties te beheren.

VersionedSchemas.swiftswift

import SwiftData

// Version 1: Initial schema
enum TaskSchemaV1: VersionedSchema {
    static var versionIdentifier = Schema.Version(1, 0, 0)

    static var models: [any PersistentModel.Type] {
        [Task.self, Category.self]
    }

    @Model
    final class Task {
        var id: UUID = UUID()
        var title: String = ""
        var isCompleted: Bool = false
        var createdAt: Date = Date()
        var category: Category?

        init(title: String) {
            self.title = title
        }
    }

    @Model
    final class Category {
        var id: UUID = UUID()
        var name: String = ""

        @Relationship(deleteRule: .cascade, inverse: \Task.category)
        var tasks: [Task] = []

        init(name: String) {
            self.name = name
        }
    }
}

// Version 2: Added priority and notes fields
enum TaskSchemaV2: VersionedSchema {
    static var versionIdentifier = Schema.Version(2, 0, 0)

    static var models: [any PersistentModel.Type] {
        [Task.self, Category.self]
    }

    @Model
    final class Task {
        var id: UUID = UUID()
        var title: String = ""
        var isCompleted: Bool = false
        var createdAt: Date = Date()
        // New properties with default values
        var priority: Int = 0
        var notes: String = ""
        var category: Category?

        init(title: String, priority: Int = 0) {
            self.title = title
            self.priority = priority
        }
    }

    @Model
    final class Category {
        var id: UUID = UUID()
        var name: String = ""
        // New property
        var color: String = "blue"

        @Relationship(deleteRule: .cascade, inverse: \Task.category)
        var tasks: [Task] = []

        init(name: String, color: String = "blue") {
            self.name = name
            self.color = color
        }
    }
}

Het migratieplan definieert de volgorde van de versies en eventueel benodigde aangepaste migraties.

MigrationPlan.swiftswift

import SwiftData

enum TaskMigrationPlan: SchemaMigrationPlan {
    // Chronological order of schemas
    static var schemas: [any VersionedSchema.Type] {
        [TaskSchemaV1.self, TaskSchemaV2.self]
    }

    // Migration stages
    static var stages: [MigrationStage] {
        [migrateV1toV2]
    }

    // V1 → V2 migration: lightweight (properties with defaults)
    static let migrateV1toV2 = MigrationStage.lightweight(
        fromVersion: TaskSchemaV1.self,
        toVersion: TaskSchemaV2.self
    )
}

// Container configuration with migration
@main
struct TaskManagerApp: App {
    var sharedModelContainer: ModelContainer = {
        do {
            return try ModelContainer(
                for: TaskSchemaV2.Task.self, TaskSchemaV2.Category.self,
                migrationPlan: TaskMigrationPlan.self
            )
        } catch {
            fatalError("Migration failed: \(error)")
        }
    }()

    var body: some Scene {
        WindowGroup {
            ContentView()
        }
        .modelContainer(sharedModelContainer)
    }
}

Lightweight versus aangepaste migraties

SwiftData verwerkt automatisch lightweight-migraties (toevoegen van eigenschappen met default-waarden, hernoemen, verwijderen). Complexe migraties die data-transformatie vereisen, gebruiken MigrationStage.custom.

Vervanging van NSFetchedResultsController

Voor lijsten met secties of gedetailleerde observatie van wijzigingen vervangt @Query in combinatie met data-extractie de NSFetchedResultsController.

SectionedResults.swiftswift

import SwiftUI
import SwiftData

struct SectionedTaskListView: View {
    @Query(sort: \Task.createdAt, order: .reverse)
    private var tasks: [Task]

    // Grouping by category
    private var tasksByCategory: [(Category?, [Task])] {
        Dictionary(grouping: tasks) { $0.category }
            .map { ($0.key, $0.value) }
            .sorted { first, second in
                // Tasks without category last
                guard let firstName = first.0?.name else { return false }
                guard let secondName = second.0?.name else { return true }
                return firstName < secondName
            }
    }

    var body: some View {
        List {
            ForEach(tasksByCategory, id: \.0?.id) { category, categoryTasks in
                Section(header: SectionHeader(category: category)) {
                    ForEach(categoryTasks) { task in
                        TaskRowView(task: task)
                    }
                }
            }
        }
    }
}

struct SectionHeader: View {
    let category: Category?

    var body: some View {
        HStack {
            if let category {
                Circle()
                    .fill(Color(category.color))
                    .frame(width: 12, height: 12)
                Text(category.name)
            } else {
                Text("Uncategorized")
                    .foregroundStyle(.secondary)
            }
        }
    }
}

// Alternative: Grouping by date
struct DateGroupedTasksView: View {
    @Query(sort: \Task.createdAt, order: .reverse)
    private var tasks: [Task]

    private var tasksByDate: [(Date, [Task])] {
        let calendar = Calendar.current

        let grouped = Dictionary(grouping: tasks) { task in
            calendar.startOfDay(for: task.createdAt)
        }

        return grouped
            .map { ($0.key, $0.value) }
            .sorted { $0.0 > $1.0 }
    }

    var body: some View {
        List {
            ForEach(tasksByDate, id: \.0) { date, dateTasks in
                Section(header: Text(date, style: .date)) {
                    ForEach(dateTasks) { task in
                        TaskRowView(task: task)
                    }
                }
            }
        }
    }
}

CRUD-operaties met ModelContext

De ModelContext vervangt de NSManagedObjectContext voor alle aanmaak-, lees-, update- en verwijderoperaties.

CRUDOperations.swiftswift

import SwiftUI
import SwiftData

struct TaskManagementView: View {
    @Environment(\.modelContext) private var modelContext
    @Query private var tasks: [Task]
    @Query private var categories: [Category]

    @State private var newTaskTitle = ""
    @State private var selectedCategory: Category?

    var body: some View {
        NavigationStack {
            VStack {
                // Add form
                AddTaskForm(
                    title: $newTaskTitle,
                    category: $selectedCategory,
                    categories: categories,
                    onAdd: addTask
                )

                // Task list
                List {
                    ForEach(tasks) { task in
                        TaskRowView(task: task)
                            .swipeActions(edge: .trailing) {
                                Button(role: .destructive) {
                                    deleteTask(task)
                                } label: {
                                    Label("Delete", systemImage: "trash")
                                }
                            }
                            .swipeActions(edge: .leading) {
                                Button {
                                    toggleCompletion(task)
                                } label: {
                                    Label(
                                        task.isCompleted ? "Todo" : "Done",
                                        systemImage: task.isCompleted ? "circle" : "checkmark"
                                    )
                                }
                                .tint(task.isCompleted ? .orange : .green)
                            }
                    }
                }
            }
            .navigationTitle("Tasks")
        }
    }

    // CREATE
    private func addTask() {
        guard !newTaskTitle.isEmpty else { return }

        let task = Task(
            title: newTaskTitle,
            category: selectedCategory
        )

        // Insert into context
        modelContext.insert(task)

        // Explicit save (optional - autosave enabled by default)
        do {
            try modelContext.save()
        } catch {
            print("Save error: \(error)")
        }

        // Reset form
        newTaskTitle = ""
        selectedCategory = nil
    }

    // UPDATE
    private func toggleCompletion(_ task: Task) {
        // Direct modification - SwiftData tracks automatically
        task.isCompleted.toggle()

        // Automatic save handles persistence
    }

    // DELETE
    private func deleteTask(_ task: Task) {
        modelContext.delete(task)
    }
}

struct AddTaskForm: View {
    @Binding var title: String
    @Binding var category: Category?
    let categories: [Category]
    let onAdd: () -> Void

    var body: some View {
        VStack(spacing: 12) {
            TextField("New task...", text: $title)
                .textFieldStyle(.roundedBorder)

            HStack {
                Picker("Category", selection: $category) {
                    Text("None").tag(nil as Category?)
                    ForEach(categories) { cat in
                        Text(cat.name).tag(cat as Category?)
                    }
                }
                .pickerStyle(.menu)

                Button("Add", action: onAdd)
                    .buttonStyle(.borderedProminent)
                    .disabled(title.isEmpty)
            }
        }
        .padding()
    }
}

Unit Tests met SwiftData

Een robuuste teststrategie vergemakkelijkt de validatie van de migratie. SwiftData maakt het mogelijk om in-memory containers te creëren voor tests.

SwiftDataTests.swiftswift

import XCTest
import SwiftData
@testable import TaskManager

final class TaskModelTests: XCTestCase {
    var container: ModelContainer!
    var context: ModelContext!

    override func setUpWithError() throws {
        // In-memory container for tests
        let config = ModelConfiguration(isStoredInMemoryOnly: true)
        container = try ModelContainer(
            for: Task.self, Category.self,
            configurations: config
        )
        context = ModelContext(container)
    }

    override func tearDownWithError() throws {
        container = nil
        context = nil
    }

    func testCreateTask() throws {
        // Given
        let task = Task(title: "Test Task")

        // When
        context.insert(task)
        try context.save()

        // Then
        let descriptor = FetchDescriptor<Task>()
        let tasks = try context.fetch(descriptor)

        XCTAssertEqual(tasks.count, 1)
        XCTAssertEqual(tasks.first?.title, "Test Task")
    }

    func testTaskCategoryRelationship() throws {
        // Given
        let category = Category(name: "Work", color: "blue")
        let task = Task(title: "Meeting", category: category)

        // When
        context.insert(category)
        context.insert(task)
        try context.save()

        // Then
        XCTAssertEqual(task.category?.name, "Work")
        XCTAssertTrue(category.tasks.contains(task))
    }

    func testDeleteCategoryCascade() throws {
        // Given
        let category = Category(name: "Personal")
        let task1 = Task(title: "Task 1", category: category)
        let task2 = Task(title: "Task 2", category: category)

        context.insert(category)
        context.insert(task1)
        context.insert(task2)
        try context.save()

        // When
        context.delete(category)
        try context.save()

        // Then - cascade delete should remove tasks
        let descriptor = FetchDescriptor<Task>()
        let remainingTasks = try context.fetch(descriptor)

        XCTAssertEqual(remainingTasks.count, 0)
    }

    func testFilteredFetch() throws {
        // Given
        let task1 = Task(title: "Completed", isCompleted: true)
        let task2 = Task(title: "Pending", isCompleted: false)
        let task3 = Task(title: "Also Pending", isCompleted: false)

        [task1, task2, task3].forEach { context.insert($0) }
        try context.save()

        // When
        var descriptor = FetchDescriptor<Task>(
            predicate: #Predicate { !$0.isCompleted }
        )
        let pendingTasks = try context.fetch(descriptor)

        // Then
        XCTAssertEqual(pendingTasks.count, 2)
    }
}

Volledige Migratie-checklist

Hieronder een samenvatting van de stappen voor een succesvolle migratie:

MigrationChecklist.swiftswift

/*
 PHASE 1: PREPARATION
 □ Audit Core Data features in use
 □ Identify features not supported by SwiftData
 □ Create dedicated migration branch
 □ Back up test data

 PHASE 2: MODEL CONVERSION
 □ Convert NSManagedObject entities to @Model
 □ Adapt relationships with @Relationship
 □ Configure appropriate delete rules
 □ Add required default values

 PHASE 3: CONFIGURATION
 □ Create ModelContainer with existing store URL
 □ Configure versioned schema if needed
 □ Define migration plan
 □ Test in coexistence mode if applicable

 PHASE 4: CODE MIGRATION
 □ Replace @FetchRequest with @Query
 □ Adapt predicates to #Predicate
 □ Migrate NSFetchedResultsController to manual grouping
 □ Convert CRUD operations to ModelContext

 PHASE 5: VALIDATION
 □ Run all unit tests
 □ Test migration with real data
 □ Verify performance with Instruments
 □ Validate CloudKit sync (if applicable)

 PHASE 6: DEPLOYMENT
 □ Document breaking changes
 □ Prepare rollback plan
 □ Deploy to TestFlight
 □ Monitor post-deployment crashes
*/

Conclusie

De migratie van Core Data naar SwiftData vertegenwoordigt een natuurlijke evolutie voor moderne iOS-applicaties. Compatibiliteit op SQLite-store-niveau garandeert het behoud van gebruikersgegevens, terwijl de native Swift-syntax de code aanzienlijk vereenvoudigt.

Belangrijkste Punten

✅ SwiftData en Core Data delen dezelfde SQLite-engine
✅ Coexistence maakt een geleidelijke migratie mogelijk
✅ @Query vervangt @FetchRequest met eenvoudigere syntax
✅ Dynamische predicates vereisen alternatieve patronen
✅ VersionedSchema beheert de schema-evolutie
✅ In-memory tests vergemakkelijken de validatie
✅ iOS 26 brengt ondersteuning voor klasse-overerving
✅ Het is verstandig nieuwe projecten met SwiftData te starten, behalve bij specifieke Core Data-behoeften

Begin met oefenen!

Test je kennis met onze gespreksimulatoren en technische tests.

Maak mijn gratis account aan

Migratie van Core Data naar SwiftData: Stap-voor-stap-gids 2026