Rustの技術面接では、言語固有の所有権システム、ガベージコレクタを使用しないメモリ管理、安全な並行コードの記述能力が評価されます。本ガイドでは、所有権の基礎から高度なasync・並行処理パターンまで、必須の質問を網羅しています。面接官はRustのメモリ安全性保証に対する理解を示す説明を重視します。コンパイラがコンパイル時にバグを防止する仕組みを説明できることが合否を分けます。 ## 所有権と借用 ### 質問1：Rustの所有権システムについて説明してください所有権はRustの中核概念であり、ガベージコレクタを使用せずにコンパイル時のメモリ安全性を保証するメモリ管理を実現します。 ```rust // ownership_basics.rs // The three fundamental rules of ownership fn main() { // Rule 1: Each value has a single owner let s1 = String::from("hello"); // s1 is the owner // Rule 2: Only one variable can own a value at a time let s2 = s1; // s1 is MOVED to s2 // println!("{}", s1); // ERROR: s1 is no longer valid println!("{}", s2); // OK: s2 is now the owner // Rule 3: When the owner goes out of scope, the value is dropped { let s3 = String::from("world"); // s3 is valid here } // s3 goes out of scope, memory is automatically freed // Copy types: simple types are copied, not moved let x = 5; let y = x; // x is COPIED, not moved println!("x = {}, y = {}", x, y); // Both are valid } // Move in action with functions fn take_ownership(s: String) { // s takes ownership of the String println!("{}", s); } // s is dropped here, memory freed fn makes_copy(i: i32) { // i is a copy of the argument println!("{}", i); } // i goes out of scope, nothing special (Copy type) fn ownership_with_functions() { let s = String::from("hello"); take_ownership(s); // s is moved into the function // println!("{}", s); // ERROR: s is no longer valid let x = 5; makes_copy(x); // x is copied println!("{}", x); // OK: x is still valid } ``` 所有権により、use-after-free、double-free、メモリリークといった一般的なメモリバグが排除されます。コンパイラがこれらの特性をコンパイル時に保証します。 ### 質問2：不変借用と可変借用の違いは何ですか借用は所有権を取得せずに値を使用する仕組みであり、データ競合を防ぐための厳格なルールがあります。 ```rust // borrowing_rules.rs // Immutable and mutable references fn main() { let mut s = String::from("hello"); // IMMUTABLE REFERENCES (&T) // Can coexist in unlimited numbers let r1 = &s; // immutable reference let r2 = &s; // another immutable reference println!("{} and {}", r1, r2); // OK // MUTABLE REFERENCE (&mut T) // Only one at a time, and no simultaneous immutable references let r3 = &mut s; // mutable reference // let r4 = &s; // ERROR: cannot have both immutable and mutable // let r5 = &mut s; // ERROR: only one mutable reference allowed r3.push_str(" world"); println!("{}", r3); // Reference scopes are limited to their last use let r6 = &s; // OK because r3 is no longer used println!("{}", r6); } // Practical example: modifying a struct struct User { name: String, age: u32, } impl User { // &self: read-only access fn get_name(&self) -> &str { &self.name } // &mut self: modification access fn set_name(&mut self, name: String) { self.name = name; } // self: takes ownership (consumes the instance) fn into_name(self) -> String { self.name // The User instance no longer exists after this } } fn borrowing_with_structs() { let mut user = User { name: String::from("Alice"), age: 30, }; // Reading println!("Name: {}", user.get_name()); // Modifying user.set_name(String::from("Bob")); // Consuming let name = user.into_name(); // user.age; // ERROR: user has been consumed } ``` これらのルールにより、コンパイル時にデータ競合の不在が保証されます。パフォーマンスを犠牲にせずにこの保証を提供する言語は他にありません。 Rust 2018以降、コンパイラはNLL（Non-Lexical Lifetimes）を使用して参照が使用されなくなるタイミングをより正確に判定し、柔軟性が向上しています。 ### 質問3：ライフタイムとは何か、いつアノテーションが必要ですかライフタイムは参照の有効期間をコンパイラに伝えるアノテーションであり、ダングリング参照を防止します。 ```rust // lifetimes.rs // Understanding and annotating lifetimes // ERROR: dangling reference // fn dangling() -> &String { // let s = String::from("hello"); // &s // s is dropped at function end, reference invalid // } // The compiler often infers lifetimes automatically fn first_word(s: &str) -> &str { // Elided lifetime: compiler understands the return // has the same lifetime as the input match s.find(' ') { Some(i) => &s[..i], None => s, } } // Explicit annotation needed with multiple references fn longest<'a>(x: &'a str, y: &'a str) -> &'a str { // 'a means: the return lives at least as long // as the shorter of the two inputs if x.len() > y.len() { x } else { y } } fn lifetime_example() { let string1 = String::from("long string"); let result; { let string2 = String::from("xyz"); result = longest(&string1, &string2); println!("Longest: {}", result); // OK here } // println!("{}", result); // ERROR: string2 is dropped } // Lifetimes in structs struct ImportantExcerpt<'a> { part: &'a str, // Struct cannot outlive part } impl<'a> ImportantExcerpt<'a> { // Method returning a reference with the same lifetime fn level(&self) -> i32 { 3 } // Elided lifetime for &self returning a new reference fn announce_and_return_part(&self, announcement: &str) -> &str { println!("Attention: {}", announcement); self.part // Returns with lifetime 'a } } // Static lifetime: lives for the entire program duration fn static_lifetime() { let s: &'static str = "hello"; // Stored in the binary // Constants have implicit 'static lifetime const MAX_POINTS: u32 = 100_000; } // Combining lifetimes and generics fn longest_with_announcement<'a, T>( x: &'a str, y: &'a str, ann: T, ) -> &'a str where T: std::fmt::Display, { println!("Announcement: {}", ann); if x.len() > y.len() { x } else { y } } ``` ライフタイムはコンパイル時に検証されます。コードがコンパイルされれば、参照の有効性が保証されます。 ## トレイトとジェネリクス ### 質問4：Rustのトレイトはどのように機能しますかトレイトは異なる型間で共有される振る舞いを定義するもので、インターフェースに類似しつつ追加機能を持ちます。 ```rust // traits_basics.rs // Defining and implementing traits // Trait definition trait Summary { // Required method (no body) fn summarize(&self) -> String; // Method with default implementation fn summarize_author(&self) -> String { String::from("(Anonymous)") } // Default method that calls a required method fn full_summary(&self) -> String { format!("By {} - {}", self.summarize_author(), self.summarize()) } } // Implementation for different types struct NewsArticle { headline: String, location: String, author: String, content: String, } impl Summary for NewsArticle { fn summarize(&self) -> String { format!("{}, by {} ({})", self.headline, self.author, self.location) } fn summarize_author(&self) -> String { format!("@{}", self.author) } } struct Tweet { username: String, content: String, reply: bool, retweet: bool, } impl Summary for Tweet { fn summarize(&self) -> String { format!("{}: {}", self.username, self.content) } } // Trait bounds: constraining generics fn notify(item: &T) { println!("Breaking news! {}", item.summarize()); } // Alternative syntax with where fn notify_verbose(item: &T) where T: Summary, { println!("Breaking news! {}", item.summarize()); } // Multiple trait bounds fn notify_complex(item: &T) { println!("{}", item); } // Return a type that implements a trait fn create_summarizable() -> impl Summary { Tweet { username: String::from("rust_lang"), content: String::from("Rust 2026 is amazing!"), reply: false, retweet: false, } } ``` トレイトはクラス継承を用いずにポリモーフィズムを実現し、継承よりもコンポジションを重視します。 ### 質問5：静的ディスパッチと動的ディスパッチの違いを説明してください Rustはポリモーフィズムに対して、モノモーフィゼーション（静的）とトレイトオブジェクト（動的）の2つのアプローチを提供しています。 ```rust // static_vs_dynamic_dispatch.rs // Static vs dynamic dispatch trait Animal { fn speak(&self) -> String; fn name(&self) -> &str; } struct Dog { name: String } struct Cat { name: String } impl Animal for Dog { fn speak(&self) -> String { String::from("Woof!") } fn name(&self) -> &str { &self.name } } impl Animal for Cat { fn speak(&self) -> String { String::from("Meow!") } fn name(&self) -> &str { &self.name } } // STATIC DISPATCH (monomorphization) // Compiler generates a version for each concrete type fn make_speak_static(animal: &T) { // At compile time, becomes make_speak_Dog and make_speak_Cat println!("{} says {}", animal.name(), animal.speak()); } // Advantages: inlining possible, no runtime overhead // Disadvantages: larger binary, type must be known at compile time // DYNAMIC DISPATCH (trait objects) // Uses a vtable to resolve methods at runtime fn make_speak_dynamic(animal: &dyn Animal) { // Resolved via a pointer table (vtable) at runtime println!("{} says {}", animal.name(), animal.speak()); } // Advantages: can store different types, smaller binary // Disadvantages: indirection overhead, no inlining fn main() { let dog = Dog { name: String::from("Rex") }; let cat = Cat { name: String::from("Whiskers") }; // Static: type is known at compile time make_speak_static(&dog); make_speak_static(&cat); // Dynamic: type is resolved at runtime make_speak_dynamic(&dog); make_speak_dynamic(&cat); // Heterogeneous collection (requires dynamic dispatch) let animals: Vec> = vec![ Box::new(Dog { name: String::from("Buddy") }), Box::new(Cat { name: String::from("Luna") }), ]; for animal in animals.iter() { println!("{} says {}", animal.name(), animal.speak()); } } // Object safety: not all traits can become trait objects trait ObjectSafe { fn method(&self); // No Self in return type // No generic parameters } // NOT object safe (cannot be dyn NotObjectSafe) trait NotObjectSafe { fn create() -> Self; // Self in return fn generic(&self, t: T); // Generic } ``` パフォーマンスを重視する場合は静的ディスパッチが適しています。異種コレクションや柔軟性が必要な場合は動的ディスパッチが有用です。 ## エラーハンドリング ### 質問6：ResultとOptionによるエラー処理の方法は Rustには例外がありません。エラー処理は`Result`と`Option`型をパターンマッチングで行います。 ```rust // error_handling.rs // Idiomatic error handling in Rust use std::fs::File; use std::io::{self, Read}; // Option: presence or absence of a value fn find_user(id: u32) -> Option { match id { 1 => Some(String::from("Alice")), 2 => Some(String::from("Bob")), _ => None, // No user found } } // Result: success or error fn divide(a: f64, b: f64) -> Result { if b == 0.0 { Err(String::from("Division by zero")) } else { Ok(a / b) } } fn option_combinators() { let user = find_user(1); // Pattern matching match user { Some(name) => println!("Found: {}", name), None => println!("Not found"), } // unwrap_or: default value let name = find_user(99).unwrap_or(String::from("Unknown")); // map: transform the value if present let upper = find_user(1).map(|n| n.to_uppercase()); // and_then (flatMap): chain Options let first_char = find_user(1).and_then(|n| n.chars().next()); // if let: simplified pattern matching if let Some(name) = find_user(2) { println!("User 2 is {}", name); } } fn result_handling() -> Result<(), Box> { // The ? operator propagates errors automatically let result = divide(10.0, 2.0)?; println!("Result: {}", result); // Equivalent to: // let result = match divide(10.0, 2.0) { // Ok(v) => v, // Err(e) => return Err(e.into()), // }; Ok(()) } // File reading with error propagation fn read_file_contents(path: &str) -> Result { let mut file = File::open(path)?; // Propagates error if failure let mut contents = String::new(); file.read_to_string(&mut contents)?; Ok(contents) } // Custom errors #[derive(Debug)] enum AppError { IoError(io::Error), ParseError(String), NotFound(String), } impl std::fmt::Display for AppError { fn fmt(&self, f: &mut std::fmt::Formatter) -> std::fmt::Result { match self { AppError::IoError(e) => write!(f, "IO error: {}", e), AppError::ParseError(s) => write!(f, "Parse error: {}", s), AppError::NotFound(s) => write!(f, "Not found: {}", s), } } } impl std::error::Error for AppError {} // Automatic conversion with From impl From for AppError { fn from(error: io::Error) -> Self { AppError::IoError(error) } } fn complex_operation() -> Result { let contents = std::fs::read_to_string("config.txt")?; // Auto-convert if contents.is_empty() { return Err(AppError::NotFound(String::from("Config is empty"))); } Ok(contents) } ``` `?`演算子によりコードが簡潔になりつつ、明示的なエラー処理が強制されます。実行時に予期しないエラーが発生することはありません。 `unwrap()`と`expect()`はNoneまたはErrの場合にパニックを起こします。プロトタイピングや失敗が不可能な場合にのみ使用してください。本番環境では`?`によるプロパゲーションまたはコンビネータの使用が推奨されます。 ### 質問7：thiserrorを使ったカスタムエラーの作成方法は `thiserror`クレートはエルゴノミックなカスタムエラーの作成を簡素化します。 ```rust // custom_errors.rs // Custom errors with thiserror use thiserror::Error; // Error definition with derive macro #[derive(Error, Debug)] pub enum DataStoreError { #[error("connection failed: {0}")] ConnectionFailed(String), #[error("query failed: {query}")] QueryFailed { query: String, source: std::io::Error }, #[error("record not found: id={id}")] NotFound { id: u64 }, #[error("invalid data: {0}")] InvalidData(#[from] serde_json::Error), #[error(transparent)] // Delegates Display to source Other(#[from] anyhow::Error), } // Implementation with rich context pub struct DataStore { connection_string: String, } impl DataStore { pub fn connect(conn_str: &str) -> Result { if conn_str.is_empty() { return Err(DataStoreError::ConnectionFailed( "Empty connection string".into() )); } Ok(Self { connection_string: conn_str.to_string() }) } pub fn get_record(&self, id: u64) -> Result { // Query simulation if id == 0 { return Err(DataStoreError::NotFound { id }); } Ok(Record { id, data: format!("Record {}", id) }) } } pub struct Record { pub id: u64, pub data: String, } // Usage with anyhow for applications use anyhow::{Context, Result}; fn application_code() -> Result<()> { let store = DataStore::connect("postgres://localhost/db") .context("Failed to connect to database")?; let record = store.get_record(42) .context("Failed to fetch user record")?; println!("Got: {}", record.data); Ok(()) } // Pattern: converting errors with context fn read_config() -> Result { let contents = std::fs::read_to_string("config.toml") .context("Failed to read config file")?; let config: Config = toml::from_str(&contents) .context("Failed to parse config file")?; Ok(config) } #[derive(Debug)] struct Config { // ... } ``` `thiserror`はライブラリ向け（型付きエラー）に最適であり、`anyhow`はアプリケーション向け（最大限の柔軟性）に適しています。 ## スマートポインタ ### 質問8：Box、Rc、Arc、RefCellについて説明してくださいスマートポインタはヒープメモリを管理し、単純な所有権では実現できないパターンを可能にします。 ```rust // smart_pointers.rs // Main smart pointers in Rust use std::rc::Rc; use std::sync::Arc; use std::cell::RefCell; // BOX: heap allocation // Used for: recursive types, large types, trait objects fn box_example() { // Simple heap allocation let b = Box::new(5); println!("b = {}", b); // Recursive type (impossible without Box) #[derive(Debug)] enum List { Cons(i32, Box), Nil, } let list = List::Cons(1, Box::new(List::Cons(2, Box::new(List::Cons(3, Box::new(List::Nil)))))); println!("{:?}", list); } // RC: Reference Counting (single-threaded) // Multiple owners for the same data fn rc_example() { let data = Rc::new(vec![1, 2, 3]); // Clone increments the reference counter let data_clone1 = Rc::clone(&data); // count = 2 let data_clone2 = Rc::clone(&data); // count = 3 println!("Reference count: {}", Rc::strong_count(&data)); // 3 // Each clone can read the data println!("data_clone1: {:?}", data_clone1); // Data is freed when the last Rc is dropped } // ARC: Atomic Reference Counting (thread-safe) // Like Rc but usable across threads fn arc_example() { use std::thread; let data = Arc::new(vec![1, 2, 3, 4, 5]); let mut handles = vec![]; for i in 0..3 { let data_clone = Arc::clone(&data); let handle = thread::spawn(move || { // Each thread has its own Arc println!("Thread {}: {:?}", i, data_clone); }); handles.push(handle); } for handle in handles { handle.join().unwrap(); } } // REFCELL: Interior Mutability // Allows mutation even with an immutable reference fn refcell_example() { let data = RefCell::new(5); // borrow() returns an immutable reference println!("Value: {}", *data.borrow()); // borrow_mut() returns a mutable reference *data.borrow_mut() += 1; println!("After mutation: {}", *data.borrow()); // Borrowing rules are checked at RUNTIME // Panics if rules are violated // let r1 = data.borrow(); // let r2 = data.borrow_mut(); // PANIC: already borrowed } // Common combination: Rc> // Multiple owners with possible mutation fn rc_refcell_example() { #[derive(Debug)] struct Node { value: i32, children: Vec>>, } let node1 = Rc::new(RefCell::new(Node { value: 1, children: vec![], })); let node2 = Rc::new(RefCell::new(Node { value: 2, children: vec![Rc::clone(&node1)], // node1 is child of node2 })); // Modify node1 from anywhere node1.borrow_mut().value = 10; println!("node2 child value: {}", node2.borrow().children[0].borrow().value); // 10 } // For threads: Arc> or Arc> fn arc_mutex_example() { use std::sync::Mutex; use std::thread; let counter = Arc::new(Mutex::new(0)); let mut handles = vec![]; for _ in 0..10 { let counter = Arc::clone(&counter); let handle = thread::spawn(move || { let mut num = counter.lock().unwrap(); *num += 1; }); handles.push(handle); } for handle in handles { handle.join().unwrap(); } println!("Final count: {}", *counter.lock().unwrap()); // 10 } ``` 状況に応じて適切なスマートポインタを選択します。単純なヒープ割り当てには`Box`、共有には`Rc`/`Arc`、内部可変性には`RefCell`/`Mutex`が適しています。 ## 並行処理 ### 質問9：Rustはどのようにスレッド安全性を保証しますか Rustの型システムは`Send`トレイトと`Sync`トレイトを通じて、コンパイル時にデータ競合を防止します。 ```rust // thread_safety.rs // Concurrent safety guarantees use std::thread; use std::sync::{Arc, Mutex, mpsc}; // SEND: a type can be transferred to another thread // SYNC: a type can be shared between threads via references // Most types are Send and Sync automatically // Exceptions: Rc (not Send/Sync), RefCell (not Sync), raw pointers fn send_example() { let data = vec![1, 2, 3]; // Vec is Send, so it can be moved to another thread let handle = thread::spawn(move || { println!("Data in thread: {:?}", data); }); handle.join().unwrap(); } // The compiler prevents concurrency errors fn compile_time_safety() { // This would NOT compile: // let data = std::rc::Rc::new(5); // thread::spawn(move || { // println!("{}", data); // ERROR: Rc is not Send // }); // Solution: use Arc let data = Arc::new(5); let data_clone = Arc::clone(&data); thread::spawn(move || { println!("{}", data_clone); // OK: Arc is Send }); } // Mutex for thread-safe shared mutation fn mutex_pattern() { let counter = Arc::new(Mutex::new(0)); let mut handles = vec![]; for _ in 0..10 { let counter = Arc::clone(&counter); let handle = thread::spawn(move || { // lock() blocks until exclusive access is obtained let mut num = counter.lock().unwrap(); *num += 1; // MutexGuard is dropped here, releasing the lock }); handles.push(handle); } for handle in handles { handle.join().unwrap(); } println!("Result: {}", *counter.lock().unwrap()); } // RwLock for multiple reads / exclusive write fn rwlock_example() { use std::sync::RwLock; let data = Arc::new(RwLock::new(vec![1, 2, 3])); let mut handles = vec![]; // Multiple simultaneous readers for i in 0..3 { let data = Arc::clone(&data); handles.push(thread::spawn(move || { let read = data.read().unwrap(); println!("Reader {}: {:?}", i, *read); })); } // Only one writer at a time { let data = Arc::clone(&data); handles.push(thread::spawn(move || { let mut write = data.write().unwrap(); write.push(4); println!("Writer added 4"); })); } for handle in handles { handle.join().unwrap(); } } // Channels for inter-thread communication fn channel_example() { let (tx, rx) = mpsc::channel(); // Multi-producer, single-consumer // Clone the sender for multiple producers let tx1 = tx.clone(); thread::spawn(move || { tx1.send("from thread 1").unwrap(); }); thread::spawn(move || { tx.send("from thread 2").unwrap(); }); // Receive messages for received in rx { println!("Got: {}", received); } } ``` 「恐れのない並行処理」：コードがコンパイルされれば、データ競合は存在しません。コンパイラが最初の防御線となります。スレッドがMutexを保持した状態でパニックすると、Mutexは「ポイズン」状態になります。その後の`lock()`呼び出しはエラーを返しますが、`into_inner()`で回復できます。 ### 質問10：Rustのasync/awaitはどのように機能しますか Rustの非同期処理はゼロコストFutureに基づいており、言語にランタイムが組み込まれていません。 ```rust // async_await.rs // Asynchronous programming in Rust use tokio::time::{sleep, Duration}; // async fn returns a Future that must be executed async fn fetch_data(url: &str) -> Result { // await suspends execution without blocking the thread let response = reqwest::get(url).await?; let body = response.text().await?; Ok(body) } // Futures are lazy: nothing executes without await or poll async fn lazy_example() { let future = async { println!("This won't print yet"); }; // Nothing happened future.await; // Now it executes } // Parallel execution of futures async fn parallel_execution() { // join! executes multiple futures in parallel let (result1, result2) = tokio::join!( fetch_data("https://api.example.com/1"), fetch_data("https://api.example.com/2"), ); println!("Results: {:?}, {:?}", result1, result2); } // select! for the first completed future async fn race_example() { tokio::select! { result = fetch_data("https://api1.example.com") => { println!("API 1 responded first: {:?}", result); } result = fetch_data("https://api2.example.com") => { println!("API 2 responded first: {:?}", result); } _ = sleep(Duration::from_secs(5)) => { println!("Timeout!"); } } } // Streams: asynchronous iterators use tokio_stream::StreamExt; async fn stream_example() { let mut stream = tokio_stream::iter(vec![1, 2, 3, 4, 5]); while let Some(value) = stream.next().await { println!("Got: {}", value); } } // Spawn for background tasks async fn spawn_tasks() { let handle = tokio::spawn(async { sleep(Duration::from_secs(1)).await; "Task completed" }); println!("Task spawned, doing other work..."); let result = handle.await.unwrap(); println!("Result: {}", result); } // Entry point with tokio #[tokio::main] async fn main() { // The tokio runtime executes futures parallel_execution().await; } // Alternative: multi-threaded or single-threaded runtime #[tokio::main(flavor = "current_thread")] async fn main_single_thread() { // Everything runs on a single thread } #[tokio::main(flavor = "multi_thread", worker_threads = 4)] async fn main_multi_thread() { // Pool of 4 worker threads } ``` Rustの非同期処理は「ランタイムは自分で用意する」方式です。tokio、async-std、smolなど、ユースケースに応じた最適化が可能です。 ## 高度なパターン ### 質問11：RustにおけるBuilderパターンについて説明してください Builderパターンは、多数のオプションフィールドを持つ複雑な構造体を構築するためのRustのイディオマティックなパターンです。 ```rust // builder_pattern.rs // Idiomatic Builder pattern in Rust #[derive(Debug, Clone)] pub struct Server { host: String, port: u16, max_connections: usize, timeout_seconds: u64, tls_enabled: bool, tls_cert_path: Option, } // Builder with consuming approach (ownership) #[derive(Default)] pub struct ServerBuilder { host: String, port: u16, max_connections: usize, timeout_seconds: u64, tls_enabled: bool, tls_cert_path: Option, } impl ServerBuilder { pub fn new() -> Self { Self { host: String::from("localhost"), port: 8080, max_connections: 100, timeout_seconds: 30, tls_enabled: false, tls_cert_path: None, } } // Each method takes self and returns Self for chaining pub fn host(mut self, host: impl Into) -> Self { self.host = host.into(); self } pub fn port(mut self, port: u16) -> Self { self.port = port; self } pub fn max_connections(mut self, max: usize) -> Self { self.max_connections = max; self } pub fn timeout(mut self, seconds: u64) -> Self { self.timeout_seconds = seconds; self } pub fn enable_tls(mut self, cert_path: impl Into) -> Self { self.tls_enabled = true; self.tls_cert_path = Some(cert_path.into()); self } // build() consumes the builder and creates the final structure pub fn build(self) -> Result { if self.tls_enabled && self.tls_cert_path.is_none() { return Err("TLS enabled but no certificate path provided".into()); } Ok(Server { host: self.host, port: self.port, max_connections: self.max_connections, timeout_seconds: self.timeout_seconds, tls_enabled: self.tls_enabled, tls_cert_path: self.tls_cert_path, }) } } // Fluent usage fn create_server() -> Result { ServerBuilder::new() .host("0.0.0.0") .port(443) .max_connections(1000) .timeout(60) .enable_tls("/etc/ssl/cert.pem") .build() } // Alternative with derive macro (typed-builder crate) // #[derive(TypedBuilder)] // pub struct Config { // #[builder(default = "localhost".to_string())] // host: String, // #[builder(default = 8080)] // port: u16, // } // Pattern with type-level validation (typestate pattern) pub struct Unvalidated; pub struct Validated; pub struct Request { url: String, method: String, headers: Vec<(String, String)>, _state: std::marker::PhantomData, } impl Request { pub fn new(url: &str) -> Self { Self { url: url.to_string(), method: "GET".to_string(), headers: vec![], _state: std::marker::PhantomData, } } pub fn method(mut self, method: &str) -> Self { self.method = method.to_string(); self } // validate() changes the state type pub fn validate(self) -> Result, String> { if self.url.is_empty() { return Err("URL cannot be empty".into()); } Ok(Request { url: self.url, method: self.method, headers: self.headers, _state: std::marker::PhantomData, }) } } impl Request { // send() is only available on validated requests pub async fn send(self) -> Result { // Implementation... todo!() } } struct Response; ``` Typestateパターンにより、特定の操作が正しい状態でのみ呼び出されることがコンパイル時に保証されます。 ### 質問12：外部型にトレイトを実装する方法は「オーファンルール」は外部トレイトを外部型に実装することを禁止していますが、回避策が存在します。 ```rust // newtype_pattern.rs // The Newtype pattern to work around the orphan rule use std::fmt; // ORPHAN RULE: cannot implement Display (std) for Vec (std) // impl fmt::Display for Vec { ... } // ERROR // SOLUTION 1: Newtype wrapper struct Wrapper(Vec); impl fmt::Display for Wrapper { fn fmt(&self, f: &mut fmt::Formatter) -> fmt::Result { write!(f, "[{}]", self.0.join(", ")) } } fn newtype_example() { let w = Wrapper(vec![ String::from("hello"), String::from("world"), ]); println!("{}", w); // [hello, world] } // Transparent access with Deref use std::ops::Deref; impl Deref for Wrapper { type Target = Vec; fn deref(&self) -> &Self::Target { &self.0 } } fn deref_example() { let w = Wrapper(vec![String::from("test")]); println!("Length: {}", w.len()); // Calls Vec::len via Deref } // SOLUTION 2: Extension trait (to add methods) trait VecExt { fn first_or_default(&self) -> Option<&T>; } impl VecExt for Vec { fn first_or_default(&self) -> Option<&T> { self.first() } } fn extension_trait_example() { let v = vec![1, 2, 3]; println!("First: {:?}", v.first_or_default()); } // Newtype with domain semantics #[derive(Debug, Clone, PartialEq, Eq, Hash)] struct Email(String); impl Email { pub fn new(email: &str) -> Result { if email.contains('@') && email.contains('.') { Ok(Self(email.to_string())) } else { Err("Invalid email format") } } pub fn as_str(&self) -> &str { &self.0 } } impl fmt::Display for Email { fn fmt(&self, f: &mut fmt::Formatter) -> fmt::Result { write!(f, "{}", self.0) } } #[derive(Debug, Clone, PartialEq, Eq, PartialOrd, Ord)] struct UserId(u64); impl UserId { pub fn new(id: u64) -> Self { Self(id) } } // Newtypes add type safety without runtime overhead fn process_user(id: UserId, email: Email) { println!("Processing user {} with email {}", id.0, email); } fn type_safety_example() { let id = UserId::new(42); let email = Email::new("user@example.com").unwrap(); process_user(id, email); // This would not compile: // process_user(email, id); // Types reversed // process_user(UserId::new(42), "string"); // String instead of Email } ``` Newtypeはメモリ表現の同一性保証により、実行時コストがゼロです。 ### 質問13：手続き型マクロの使い方は手続き型マクロはコンパイル時にコードを生成する機能であり、カスタムderive等で利用されます。 ```rust // procedural_macros.rs // Understanding procedural macros // Proc macros are defined in a separate crate with proc-macro = true // Crate: my_derive (Cargo.toml: proc-macro = true) use proc_macro::TokenStream; use quote::quote; use syn::{parse_macro_input, DeriveInput}; // DERIVE MACRO: #[derive(MyTrait)] #[proc_macro_derive(MyDebug)] pub fn my_debug_derive(input: TokenStream) -> TokenStream { // Parse input as a type definition let input = parse_macro_input!(input as DeriveInput); let name = input.ident; // Generate implementation code let expanded = quote! { impl std::fmt::Debug for #name { fn fmt(&self, f: &mut std::fmt::Formatter) -> std::fmt::Result { write!(f, stringify!(#name)) } } }; TokenStream::from(expanded) } // ATTRIBUTE MACRO: #[my_attribute] #[proc_macro_attribute] pub fn route(attr: TokenStream, item: TokenStream) -> TokenStream { // attr contains the attribute arguments // item contains the annotated element (function, struct, etc.) let method_path = attr.to_string(); // "GET, /users" let input = parse_macro_input!(item as syn::ItemFn); let fn_name = &input.sig.ident; let expanded = quote! { #input // Additionally generated code fn register_#fn_name() { println!("Registered route: {}", #method_path); } }; TokenStream::from(expanded) } // FUNCTION-LIKE MACRO: my_macro!(...) #[proc_macro] pub fn make_answer(_input: TokenStream) -> TokenStream { "fn answer() -> u32 { 42 }".parse().unwrap() } // --- Usage in client code --- // Derive macro #[derive(MyDebug)] struct Point { x: i32, y: i32, } // Attribute macro #[route("GET", "/users")] fn get_users() -> Vec { vec![] } // Function-like macro make_answer!(); // Generates fn answer() -> u32 { 42 } fn main() { let p = Point { x: 1, y: 2 }; println!("{:?}", p); // Uses our MyDebug println!("Answer: {}", answer()); // 42 } struct User; ``` 手続き型マクロは繰り返しコードに対して強力な機能を発揮します。シリアライゼーション、Webルーティング、バリデーションなどが典型的な用途です。 ## メモリ安全性とunsafe ### 質問14：unsafeを使用するタイミングと方法は `unsafe`ブロックは低レベルコードに対してコンパイラの特定のチェックをバイパスすることを許可します。 ```rust // unsafe_rust.rs // Understanding unsafe and its guarantees // The 5 superpowers of unsafe: // 1. Dereference raw pointers // 2. Call unsafe functions // 3. Access/modify mutable static variables // 4. Implement unsafe traits // 5. Access union fields // RAW POINTERS fn raw_pointers() { let mut num = 5; // Creating raw pointers is safe let r1 = &num as *const i32; let r2 = &mut num as *mut i32; // Dereferencing requires unsafe unsafe { println!("r1 is: {}", *r1); *r2 = 10; println!("r2 is: {}", *r2); } } // UNSAFE FUNCTION // The function guarantees safety IF preconditions are met unsafe fn dangerous() { // Code that assumes the caller verified invariants } fn call_dangerous() { // Must be in an unsafe block unsafe { dangerous(); } } // SAFE ABSTRACTION over unsafe code fn split_at_mut(values: &mut [i32], mid: usize) -> (&mut [i32], &mut [i32]) { let len = values.len(); let ptr = values.as_mut_ptr(); assert!(mid <= len); // Runtime check unsafe { // We know the two slices don't overlap ( std::slice::from_raw_parts_mut(ptr, mid), std::slice::from_raw_parts_mut(ptr.add(mid), len - mid), ) } } // FFI: calling C code extern "C" { fn abs(input: i32) -> i32; } fn call_c_function() { unsafe { println!("Absolute value: {}", abs(-3)); } } // Export a function for C #[no_mangle] pub extern "C" fn call_from_c() { println!("Called from C!"); } // MUTABLE STATIC static mut COUNTER: u32 = 0; fn increment_counter() { unsafe { COUNTER += 1; println!("COUNTER: {}", COUNTER); } } // UNSAFE TRAIT unsafe trait Dangerous { // Implementers guarantee invariants } unsafe impl Dangerous for i32 { // Implementer asserts respecting the trait's invariants } // Practical example: structure with internal pointer pub struct MyVec { ptr: *mut T, len: usize, capacity: usize, } impl MyVec { pub fn new() -> Self { Self { ptr: std::ptr::null_mut(), len: 0, capacity: 0, } } pub fn push(&mut self, value: T) { if self.len == self.capacity { self.grow(); } unsafe { std::ptr::write(self.ptr.add(self.len), value); } self.len += 1; } fn grow(&mut self) { // Unsafe allocation/reallocation... } } impl Drop for MyVec { fn drop(&mut self) { unsafe { // Properly free memory for i in 0..self.len { std::ptr::drop_in_place(self.ptr.add(i)); } if self.capacity > 0 { let layout = std::alloc::Layout::array::(self.capacity).unwrap(); std::alloc::dealloc(self.ptr as *mut u8, layout); } } } } ``` unsafeコードの範囲を最小化してください。unsafeコードは不変条件を保証する安全な抽象の中にカプセル化します。unsafeコードが周囲の安全なメモリを破壊してはなりません。 ### 質問15：借用チェッカーはどのように機能しますか借用チェッカーはRustコンパイラの中核であり、所有権と借用のルールを検証します。 ```rust // borrow_checker.rs // Understanding how the borrow checker works fn borrow_checker_basics() { let mut v = vec![1, 2, 3]; // RULE 1: Either multiple immutable references or one mutable let r1 = &v; let r2 = &v; println!("{:?} {:?}", r1, r2); // OK: multiple immutable references // From here, r1 and r2 are no longer used (NLL) let r3 = &mut v; // OK thanks to Non-Lexical Lifetimes r3.push(4); } // The borrow checker tracks lifetimes fn lifetime_tracking() { let mut data = String::from("hello"); let slice = &data[..]; // Immutable borrow starts // data.push_str(" world"); // ERROR: cannot mutate during borrow println!("{}", slice); // Last use of slice data.push_str(" world"); // OK: borrow ended } // Common problems and solutions mod common_patterns { // Problem: borrowing two mutable fields struct Data { field1: Vec, field2: Vec, } fn problem(data: &mut Data) { // This sometimes doesn't compile directly: // let f1 = &mut data.field1; // let f2 = &mut data.field2; // Solution: destructuring let Data { field1, field2 } = data; field1.push(1); field2.push(2); } // Problem: iterate and modify fn iterate_and_modify() { let mut v = vec![1, 2, 3, 4, 5]; // Does not compile: // for &x in &v { // if x % 2 == 0 { // v.push(x * 2); // ERROR: borrowed by iterator // } // } // Solution 1: collect indices first let to_add: Vec = v.iter() .filter(|&&x| x % 2 == 0) .map(|&x| x * 2) .collect(); v.extend(to_add); // Solution 2: use explicit indices let len = v.len(); for i in 0..len { if v[i] % 2 == 0 { let new_val = v[i] * 2; v.push(new_val); } } } // Problem: self-referential struct // struct SelfRef { // data: String, // slice: &str, // Reference to data - IMPOSSIBLE // } // Solution: use indices or crates like ouroboros struct SafeSelfRef { data: String, slice_start: usize, slice_end: usize, } impl SafeSelfRef { fn get_slice(&self) -> &str { &self.data[self.slice_start..self.slice_end] } } } // Patterns to work around limitations mod workarounds { use std::cell::RefCell; // Interior mutability when borrow checker is too restrictive struct Graph { nodes: RefCell>, } struct Node { value: i32, } impl Graph { fn add_node(&self, value: i32) { // Mutation possible despite &self self.nodes.borrow_mut().push(Node { value }); } fn get_node(&self, index: usize) -> Option { self.nodes.borrow().get(index).map(|n| n.value) } } } ``` 借用チェッカーは最初は制約が厳しく感じられますが、これらの制約により、他の言語で発生するバグのカテゴリ全体が排除されます。 ## まとめ Rustの技術面接では、所有権システムの深い理解、メモリ安全性の保証、データ競合のない並行コードの記述能力が評価されます。これらの概念を習得することが、Rustの固有の利点を活用できる開発者を見分ける鍵となります。 ### 面接準備チェックリスト - 所有権、借用、3つの基本ルールの理解 - ライフタイムのアノテーションが必要なタイミングと方法の把握 - トレイトと静的・動的ディスパッチの違いの習得 - ResultとOptionによるイディオマティックなエラー処理 - 状況に応じた適切なスマートポインタの選択 - Arc、Mutex、チャネルを使った並行コードの記述 - async/awaitとtokio等のランタイムの理解 - unsafeの安全な使用タイミングと方法の把握 Rust面接の準備には、言語固有の所有権概念の練習が不可欠です。Exercismでの演習、個人プロジェクト、Rustエコシステムへの貢献が、最も要求の厳しい技術面接に向けた知識の定着に有効です。