AgentSkillsCN

m15-anti-pattern

当您需要审查代码是否存在反模式时,本指南将为您提供专业指导。关键词:反模式、常见错误、陷阱、代码异味、不良实践、代码审查、这是否属于反模式、还有更好的做法吗、避免常见的错误、为什么这样做不好、地道的写法、初学者常犯的错误、与借用检查器斗智斗勇、到处都用 clone、在生产环境中使用 unwrap、我是否应该重构?本指南将助您轻松应对各种反模式问题,教您如何规避常见错误、避免不良实践,以更优雅的方式编写代码。

SKILL.md
--- frontmatter
name: m15-anti-pattern
description: "Use when reviewing code for anti-patterns. Keywords: anti-pattern, common mistake, pitfall, code smell, bad practice, code review, is this an anti-pattern, better way to do this, common mistake to avoid, why is this bad, idiomatic way, beginner mistake, fighting borrow checker, clone everywhere, unwrap in production, should I refactor, 反模式, 常见错误, 代码异味, 最佳实践, 地道写法"
user-invocable: false

Anti-Patterns

Layer 2: Design Choices

Core Question

Is this pattern hiding a design problem?

When reviewing code:

  • Is this solving the symptom or the cause?
  • Is there a more idiomatic approach?
  • Does this fight or flow with Rust?

Anti-Pattern → Better Pattern

Anti-PatternWhy BadBetter
.clone() everywhereHides ownership issuesProper references or ownership
.unwrap() in productionRuntime panics?, expect, or handling
Rc when single ownerUnnecessary overheadSimple ownership
unsafe for convenienceUB riskFind safe pattern
OOP via DerefMisleading APIComposition, traits
Giant match armsUnmaintainableExtract to methods
String everywhereAllocation waste&str, Cow<str>
Ignoring #[must_use]Lost errorsHandle or let _ =

Thinking Prompt

When seeing suspicious code:

  1. Is this symptom or cause?

    • Clone to avoid borrow? → Ownership design issue
    • Unwrap "because it won't fail"? → Unhandled case
  2. What would idiomatic code look like?

    • References instead of clones
    • Iterators instead of index loops
    • Pattern matching instead of flags
  3. Does this fight Rust?

    • Fighting borrow checker → restructure
    • Excessive unsafe → find safe pattern

Trace Up ↑

To design understanding:

code
"Why does my code have so many clones?"
    ↑ Ask: Is the ownership model correct?
    ↑ Check: m09-domain (data flow design)
    ↑ Check: m01-ownership (reference patterns)
Anti-PatternTrace ToQuestion
Clone everywherem01-ownershipWho should own this data?
Unwrap everywherem06-error-handlingWhat's the error strategy?
Rc everywherem09-domainIs ownership clear?
Fighting lifetimesm09-domainShould data structure change?

Trace Down ↓

To implementation (Layer 1):

code
"Replace clone with proper ownership"
    ↓ m01-ownership: Reference patterns
    ↓ m02-resource: Smart pointer if needed

"Replace unwrap with proper handling"
    ↓ m06-error-handling: ? operator
    ↓ m06-error-handling: expect with message

Top 5 Beginner Mistakes

RankMistakeFix
1Clone to escape borrow checkerUse references
2Unwrap in productionPropagate with ?
3String for everythingUse &str
4Index loopsUse iterators
5Fighting lifetimesRestructure to own data

Code Smell → Refactoring

SmellIndicatesRefactoring
Many .clone()Ownership unclearClarify data flow
Many .unwrap()Error handling missingAdd proper handling
Many pub fieldsEncapsulation brokenPrivate + accessors
Deep nestingComplex logicExtract methods
Long functionsMultiple responsibilitiesSplit
Giant enumsMissing abstractionTrait + types

Common Error Patterns

ErrorAnti-Pattern CauseFix
E0382 use after moveCloning vs ownershipProper references
Panic in productionUnwrap everywhere?, matching
Slow performanceString for all text&str, Cow
Borrow checker fightsWrong structureRestructure
Memory bloatRc/Arc everywhereSimple ownership

Deprecated → Better

DeprecatedBetter
Index-based loops.iter(), .enumerate()
collect::<Vec<_>>() then iterateChain iterators
Manual unsafe cellCell, RefCell
mem::transmute for castsas or TryFrom
Custom linked listVec, VecDeque
lazy_static!std::sync::OnceLock

Quick Review Checklist

  • No .clone() without justification
  • No .unwrap() in library code
  • No pub fields with invariants
  • No index loops when iterator works
  • No String where &str suffices
  • No ignored #[must_use] warnings
  • No unsafe without SAFETY comment
  • No giant functions (>50 lines)

Related Skills

WhenSee
Ownership patternsm01-ownership
Error handlingm06-error-handling
Mental modelsm14-mental-model
Performancem10-performance