java-code-review（Java 代码审查）

Java Code Review Skill

Systematic code review checklist for Java projects.

When to Use

• User says "review this code" / "check this PR" / "code review"
• Before merging a PR
• After implementing a feature

Review Strategy

1. Quick scan - Understand intent, identify scope
2. Checklist pass - Go through each category below
3. Summary - List findings by severity (Critical → Minor)

Output Format

## Code Review: [file/feature name]

### Critical
- [Issue description + line reference + suggestion]

### Improvements
- [Suggestion + rationale]

### Minor/Style
- [Nitpicks, optional improvements]

### Good Practices Observed
- [Positive feedback - important for morale]

Review Checklist

1. Null Safety

Check for:

// ❌ NPE risk
String name = user.getName().toUpperCase();

// ✅ Safe
String name = Optional.ofNullable(user.getName())
    .map(String::toUpperCase)
    .orElse("");

// ✅ Also safe (early return)
if (user.getName() == null) {
    return "";
}
return user.getName().toUpperCase();

Flags:

• Chained method calls without null checks
• Missing @Nullable / @NonNull annotations on public APIs
• Optional.get() without isPresent() check
• Returning null from methods that could return Optional or empty collection

Suggest:

• Use Optional for return types that may be absent
• Use Objects.requireNonNull() for constructor/method params
• Return empty collections instead of null: Collections.emptyList()

2. Exception Handling

Check for:

// ❌ Swallowing exceptions
try {
    process();
} catch (Exception e) {
    // silently ignored
}

// ❌ Catching too broad
catch (Exception e) { }
catch (Throwable t) { }

// ❌ Losing stack trace
catch (IOException e) {
    throw new RuntimeException(e.getMessage());
}

// ✅ Proper handling
catch (IOException e) {
    log.error("Failed to process file: {}", filename, e);
    throw new ProcessingException("File processing failed", e);
}

Flags:

• Empty catch blocks
• Catching Exception or Throwable broadly
• Losing original exception (not chaining)
• Using exceptions for flow control
• Checked exceptions leaking through API boundaries

Suggest:

• Log with context AND stack trace
• Use specific exception types
• Chain exceptions with cause
• Consider custom exceptions for domain errors

3. Collections & Streams

Check for:

// ❌ Modifying while iterating
for (Item item : items) {
    if (item.isExpired()) {
        items.remove(item);  // ConcurrentModificationException
    }
}

// ✅ Use removeIf
items.removeIf(Item::isExpired);

// ❌ Stream for simple operations
list.stream().forEach(System.out::println);

// ✅ Simple loop is cleaner
for (Item item : list) {
    System.out.println(item);
}

// ❌ Collecting to modify
List<String> names = users.stream()
    .map(User::getName)
    .collect(Collectors.toList());
names.add("extra");  // Might be immutable!

// ✅ Explicit mutable list
List<String> names = users.stream()
    .map(User::getName)
    .collect(Collectors.toCollection(ArrayList::new));

Flags:

• Modifying collections during iteration
• Overusing streams for simple operations
• Assuming Collectors.toList() returns mutable list
• Not using List.of(), Set.of(), Map.of() for immutable collections
• Parallel streams without understanding implications

Suggest:

• List.copyOf() for defensive copies
• removeIf() instead of iterator removal
• Streams for transformations, loops for side effects

4. Concurrency

Check for:

// ❌ Not thread-safe
private Map<String, User> cache = new HashMap<>();

// ✅ Thread-safe
private Map<String, User> cache = new ConcurrentHashMap<>();

// ❌ Check-then-act race condition
if (!map.containsKey(key)) {
    map.put(key, computeValue());
}

// ✅ Atomic operation
map.computeIfAbsent(key, k -> computeValue());

// ❌ Double-checked locking (broken without volatile)
if (instance == null) {
    synchronized(this) {
        if (instance == null) {
            instance = new Instance();
        }
    }
}

Flags:

• Shared mutable state without synchronization
• Check-then-act patterns without atomicity
• Missing volatile on shared variables
• Synchronized on non-final objects
• Thread-unsafe lazy initialization

Suggest:

• Prefer immutable objects
• Use java.util.concurrent classes
• AtomicReference, AtomicInteger for simple cases
• Consider @ThreadSafe / @NotThreadSafe annotations

5. Java Idioms

equals/hashCode:

// ❌ Only equals without hashCode
@Override
public boolean equals(Object o) { ... }
// Missing hashCode!

// ❌ Mutable fields in hashCode
@Override
public int hashCode() {
    return Objects.hash(id, mutableField);  // Breaks HashMap
}

// ✅ Use immutable fields, implement both
@Override
public boolean equals(Object o) {
    if (this == o) return true;
    if (!(o instanceof User user)) return false;
    return Objects.equals(id, user.id);
}

@Override
public int hashCode() {
    return Objects.hash(id);
}

toString:

// ❌ Missing - hard to debug
// No toString()

// ❌ Including sensitive data
return "User{password='" + password + "'}";

// ✅ Useful for debugging
@Override
public String toString() {
    return "User{id=" + id + ", name='" + name + "'}";
}

Builders:

// ✅ For classes with many optional parameters
User user = User.builder()
    .name("John")
    .email("john@example.com")
    .build();

Flags:

• equals without hashCode
• Mutable fields in hashCode
• Missing toString on domain objects
• Constructors with > 3-4 parameters (suggest builder)
• Not using instanceof pattern matching (Java 16+)

6. Resource Management

Check for:

// ❌ Resource leak
FileInputStream fis = new FileInputStream(file);
// ... might throw before close

// ✅ Try-with-resources
try (FileInputStream fis = new FileInputStream(file)) {
    // ...
}

// ❌ Multiple resources, wrong order
try (BufferedWriter writer = new BufferedWriter(new FileWriter(file))) {
    // FileWriter might not be closed if BufferedWriter fails
}

// ✅ Separate declarations
try (FileWriter fw = new FileWriter(file);
     BufferedWriter writer = new BufferedWriter(fw)) {
    // Both properly closed
}

Flags:

• Not using try-with-resources for Closeable/AutoCloseable
• Resources opened but not in try-with-resources
• Database connections/statements not properly closed

7. API Design

Check for:

// ❌ Boolean parameters
process(data, true, false);  // What do these mean?

// ✅ Use enums or builder
process(data, ProcessMode.ASYNC, ErrorHandling.STRICT);

// ❌ Returning null for "not found"
public User findById(Long id) {
    return users.get(id);  // null if not found
}

// ✅ Return Optional
public Optional<User> findById(Long id) {
    return Optional.ofNullable(users.get(id));
}

// ❌ Accepting null collections
public void process(List<Item> items) {
    if (items == null) items = Collections.emptyList();
}

// ✅ Require non-null, accept empty
public void process(List<Item> items) {
    Objects.requireNonNull(items, "items must not be null");
}

Flags:

• Boolean parameters (prefer enums)
• Methods with > 3 parameters (consider parameter object)
• Inconsistent null handling across similar methods
• Missing validation on public API inputs

8. Performance Considerations

Check for:

// ❌ String concatenation in loop
String result = "";
for (String s : strings) {
    result += s;  // Creates new String each iteration
}

// ✅ StringBuilder
StringBuilder sb = new StringBuilder();
for (String s : strings) {
    sb.append(s);
}

// ❌ Regex compilation in loop
for (String line : lines) {
    if (line.matches("pattern.*")) { }  // Compiles regex each time
}

// ✅ Pre-compiled pattern
private static final Pattern PATTERN = Pattern.compile("pattern.*");
for (String line : lines) {
    if (PATTERN.matcher(line).matches()) { }
}

// ❌ N+1 in loops
for (User user : users) {
    List<Order> orders = orderRepo.findByUserId(user.getId());
}

// ✅ Batch fetch
Map<Long, List<Order>> ordersByUser = orderRepo.findByUserIds(userIds);

Flags:

• String concatenation in loops
• Regex compilation in loops
• N+1 query patterns
• Creating objects in tight loops that could be reused
• Not using primitive streams (IntStream, LongStream)

9. Testing Hints

Suggest tests for:

• Null inputs
• Empty collections
• Boundary values
• Exception cases
• Concurrent access (if applicable)

Severity Guidelines

Token Optimization

• Focus on changed lines (use git diff)
• Don't repeat obvious issues - group similar findings
• Reference line numbers, not full code quotes
• Skip files that are auto-generated or test fixtures

Quick Reference Card