Coding/Learn/dart/dart records
Dart & Flutter Engineering Complete Course
01.Introduction & Execution Architecture
02.Syntax, Statements & Code Organization
03.Variables, Type Inference, final & const
04.Core Built-In Types: int, double, num, bool, String
05.String Interpolation, Multiline & Raw Strings
06.Operators: Arithmetic, Type Test, Ternary & Cascade
07.Conditional Branching: if, else if, else & Guards
08.Loops: standard for, for-in, while & do-while
09.Switch Statements, Switch Expressions & Exhaustiveness
10.Loop Control: break, continue & Labeled Jumps
11.Pattern Matching & Destructuring in Dart 3
12.Error Handling: try, catch, on, finally & assert
13.Functions, Return Types & Arrow Syntax (=>)
14.Named Parameters & the required Keyword
15.Optional Positional Parameters & Default Values
16.Anonymous Functions, Lambdas & typedefs
17.Lexical Scope & Stateful Closures
18.Synchronous Generators: sync* and yield
19.Lists: Fixed-Length, Growable & Index Operations
20.Sets: Unique Values & Set Theory Algebra
21.Maps: Key-Value Dictionaries & Lookups
22.Collection-if, Collection-for & Spread Operators
23.Iterables & Higher-Order Methods: map, where, fold
24.Generics: Type Parameters, Bounds & Methods
25.Classes & Objects: Members, Instance Methods & this
26.Constructors: Initializing Formals & Initializer Lists
27.Named, Redirecting & Factory Constructors
28.Const Constructors & Canonical Instance Caching
29.Getters, Setters & Private Fields (_)
30.Inheritance, Method Overriding (@override) & super
31.Abstract Classes & Method Contracts
32.Implicit Interfaces & the implements Keyword
33.Mixins: Multi-Source Composition with with & on
34.Extension Methods: Extending External APIs
35.Extension Types: Zero-Cost Inline Wrappers in Dart 3.3
36.Enhanced Enums with Fields, Methods & Patterns
37.Records: Anonymous Composite Tuples in Dart 3
38.Class Modifiers: sealed, final, base & interface
39.Sound Null Safety: Core Architecture
40.Nullable Types (Type?) & Non-Nullable Invariants
41.Null-Aware Operators: ?., ??, ??=, ?.. & ?[]
42.Null Assertion Operator (!) & Safe Invariants
43.The late Keyword: Lazy Initialization & Late Final
44.Type Promotion & Flow Analysis
45.Event Loop & Microtask Queue Architecture
46.Futures, async & await: Asynchronous Foundations
47.Future Combinators: wait, any, delayed & timeout
48.Streams: Asynchronous Data Flows & StreamControllers
49.Stream Transformers & Asynchronous Generators (async*)
50.Isolates: Multi-Core Concurrency & Isolate.run()
51.Packages & pubspec.yaml: Dependency Management
52.Libraries, Namespaces & Barrel Exports
53.Unit Testing with package:test & Matchers
54.Async Testing, Stream Tests & Fixtures
55.Effective Dart Guidelines & Linter Optimization
56.Comprehensive Dart Exercises & Algorithm Challenges
57.& Flutter Architecture Assessment & Knowledge Check
58.Capstone Project: Reactive State Machine & Stream Event Hub
59.Mastery Certification & Verified Graduate Credential
Course Certificate0%

Complete all lessons to earn

LearnRecords: Anonymous Composite Tuples in Dart 3
37 / 59 0%
Lesson 37 of 59
5 mins read

Records: Anonymous Composite Tuples in Dart 3

PreviousNext Lesson
In Plain English

Introduced in Dart 3, **Records** (`(TypeA, {TypeB name})`) are anonymous, aggregate, immutable tuple values. Records allow functions to return multiple values type-safely without having to create throwaway DTO classes. Records support positional fields (`$1`, `$2`) and named fields (`rec.name`).

The Mental Model

"A Record is like a partitioned bento box: you can hand someone a single box containing rice in compartment $1, salad in compartment $2, and salmon in the labeled "mainDish" compartment."

Deep Dive: How It Works

Anonymous & Strongly Typed: `(String, int)` is a concrete type with value equality out of the box.

Field Access: Positional fields are accessed via `$1`, `$2`, etc.; named fields are accessed by name (`record.fieldName`).

Value Equality: Two records with identical field values evaluate as equal (`==`) automatically.

Syntax Blueprint

// Return multiple values
(double lat, double lng) getCoordinates() {
  return (37.7749, -122.4194);
}

// Usage with destructuring
final (lat, lng) = getCoordinates();

Record return type and pattern destructuring.

PointCore Rules to Remember

Point star
Multiple Return Values: Eliminates boilerplate throwaway wrapper classes.
Point star
Built-In Value Equality: Records automatically implement `==` and `hashCode` based on contents.
Point star
Positional and Named: Mix positional elements and labeled named fields seamlessly.

Common Beginner Traps & How to Fix Them

Mistake:Single-element positional record syntax `(42)` instead of `(42,)`.

Why it happens: `(42)` is parsed as parenthesized integer expression 42.

How to fix: Include a trailing comma `(42,)` for single-element positional records.

Pro Tip: Combine records with pattern destructuring `final (name, :age) = getUser();` for clean JSON parsing.

Live Interactive Example

Hit Run Code to see it live
Returning and Destructuring Records
Python 3.12
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
Output Console
Click "Run Code" to view the rendered output.
How it works: Returned a record with positional and named fields, then destructured fields cleanly.

Your Turn: Micro Challenge

No pressure! Edit the starter code below and test your solution with instant feedback.

Micro Exercise

Return Min and Max as a Record

  • PointDefine `(int min, int max) getBounds(List<int> list) => (list.reduce((a, b) => a < b ? a : b), list.reduce((a, b) => a > b ? a : b));`.
  • PointIn `main()`, call `getBounds([10, 5, 20, 15])` and print `Min: ${b.$1}, Max: ${b.$2}`.
1
2
Sandbox Output
Click "Run & Check" to test your solution.

Finished reading and practicing?

Mark this lesson as completed to update your course progress.

PreviousNext Lesson
Overview100 SheetLearnCertificates