huiali/rust-skills/rust-ownership

Solution Patterns

Pattern 1: Value Moved After Use

let s1 = String::from("hello");
let s2 = s1;
// println!("{}", s1); // Compile error!

Root Cause: Ownership transferred from s1 to s2, s1 is no longer valid.

Solutions:

• Need two copies → use clone()
• Only need to read → pass by reference &s1
• s2 is temporary → consider redesign

Pattern 2: Borrow Conflict

let mut s = String::from("hello");
let r1 = &s;
let r2 = &mut s; // Conflict!
// println!("{}", r1);

Root Cause: Immutable and mutable borrows coexist.

Solutions:

• Ensure mutable borrow completes before creating new borrows
• Restructure code organization

Pattern 3: Lifetime Mismatch

fn longest<'a>(s1: &'a str, s2: &'a str) -> &'a str {
    if s1.len() > s2.len() { s1 } else { s2 }
}

Root Cause: Return value lifetime must be tied to one of the inputs.

Key Solutions:

1. Clarify each reference's lifetime
2. Use named lifetimes to express relationships
3. Prefer returning owned types

Workflow

Step 1: Who Owns the Data?

Step 2: Is Borrowing Appropriate?

Step 3: Can Lifetimes Be Avoided?

Return String instead of &str
    ↓
Use owned collections instead of slices
    ↓
Use Arc/Rc for shared ownership
    ↓
Lifetimes aren't always necessary

Smart Pointer Selection

Common Pitfalls

Practical Guidance

Common Beginner Questions

1. "When should I use references vs ownership?"

• Function parameters: use references (unless consuming)
• Function returns: use references (if caller doesn't need ownership)
• Storage: consider lifetime complexity

2. "How do I add lifetime annotations?"

• Most cases: compiler can infer
• Need explicit: structs, trait impls, methods returning references
• Use meaningful names: 'connection, 'file

3. "Why doesn't this borrow work?"

• Mutable borrows make original inaccessible
• Check borrow scope ranges
• Consider code reorganization

Error Code Quick Reference

Thinking Process

When encountering ownership issues, follow these steps:

1. What's this data's role in the domain?

• Entity (unique identity) → owned
• Value object (interchangeable) → clone/copy acceptable
• Temporary computation result → consider refactor

2. Is ownership design intentional or accidental?

• Intentional → work within constraints
• Accidental → consider redesign

3. Fix symptom or redesign?

• If failed 3 times → escalate to design level

Trace Up (Design Analysis)

When ownership errors persist, trace to design level:

E0382 (moved value)
    ↑ Ask: What design choice led to this ownership pattern?
    ↑ Check: Is this an entity or value object?
    ↑ Check: Are there other constraints?
 
Persistent E0382 → rust-resource: Should use Arc/Rc for sharing?
Persistent E0597 → rust-type-driven: Are scope boundaries correct?
E0506/E0507 → rust-mutability: Should use interior mutability?

Trace Down (Implementation)

From design decisions to implementation:

"Data needs immutable sharing"
    ↓ Multi-threaded: Arc<T>
    ↓ Single-threaded: Rc<T>
 
"Data needs exclusive ownership"
    ↓ Return owned value
 
"Data is temporary use only"
    ↓ Use references within scope
 
"Need to pass data between functions"
    ↓ Consider lifetimes or return owned

Review Checklist

When reviewing ownership-related code:

• [ ] Each value has a clear owner
• [ ] Borrows don't outlive the borrowed data
• [ ] Mutable and immutable borrows don't overlap
• [ ] Lifetime annotations accurately reflect data relationships
• [ ] Smart pointer choice matches threading requirements
• [ ] .clone() is used intentionally, not to avoid compiler errors
• [ ] Lifetime elision is leveraged where possible
• [ ] Complex lifetime scenarios are documented

Verification Commands

# Check compilation
cargo check
 
# Run tests
cargo test
 
# Check for common mistakes
cargo clippy -- -W clippy::clone_on_copy -W clippy::unnecessary_clone
 
# Verify no memory leaks in tests
cargo test --features leak-check

Common Pitfalls

1. Clone Abuse

Symptom: .clone() everywhere to satisfy compiler

Fix: Understand actual ownership requirements, use references where possible

2. Lifetime Overuse

Symptom: Complex lifetime annotations everywhere

Fix: Return owned types, use smart pointers

3. Fighting Borrow Checker

Symptom: Constantly rewriting to satisfy compiler

Fix: Step back and redesign data flow

Related Skills

• rust-mutability - Interior mutability patterns (Cell, RefCell)
• rust-concurrency - Send/Sync and thread safety
• rust-unsafe - Raw pointers and manual memory management
• rust-lifetime-complex - Advanced lifetime patterns (HRTB, GAT)
• rust-resource - Resource management and RAII patterns
• rust-type-driven - Type-driven design

Localized Reference

• Chinese version: SKILL_ZH.md - 完整中文版本，包含所有内容