Part of duplicate code analysis: #2907

Summary

internal/logger/global_helpers.go contains 10 nearly-identical wrapper functions — 5 initGlobal* and 5 closeGlobal* pairs — one pair per logger type. Each wrapper is a 3-line delegation to the same generic initGlobalLogger / closeGlobalLogger helpers.

Duplication Details

Pattern: Repeated initGlobal* / closeGlobal* wrapper functions

• Severity: Medium

• Occurrences: 10 functions (5 pairs)

• Location: internal/logger/global_helpers.go (lines ~120–168)

• Code Sample (pattern repeats for all 5 logger types):

  func initGlobalFileLogger(logger *FileLogger) {
      initGlobalLogger(&globalLoggerMu, &globalFileLogger, logger)
  }
  func closeGlobalFileLogger() error {
      return closeGlobalLogger(&globalLoggerMu, &globalFileLogger)
  }
  
  func initGlobalJSONLLogger(logger *JSONLLogger) {
      initGlobalLogger(&globalJSONLMu, &globalJSONLLogger, logger)
  }
  func closeGlobalJSONLLogger() error {
      return closeGlobalLogger(&globalJSONLMu, &globalJSONLLogger)
  }
  // ... identical pattern for Markdown, ServerFile, Tools loggers

Impact Analysis

• Maintainability: Adding a 6th logger type requires adding another pair of wrappers in the same pattern
• Bug Risk: Low currently, but grows as more loggers are added
• Code Bloat: ~40 lines of boilerplate; the real logic lives in 2 generic functions

Refactoring Recommendations

1. Document the pattern explicitly — if the wrappers are intentional for type safety, add a comment explaining the design rationale so future contributors understand why the repetition exists.

2. Use go:generate to produce wrapper functions from a template, so the single-source-of-truth lives in the template rather than in hand-written copies.

3. Consider a logger registry for when a 6th logger type needs to be added:

   type globalLoggerEntry[T io.Closer] struct {
       mu     *sync.RWMutex
       logger *T
   }

   A registry approach scales without adding new wrapper pairs.

Implementation Checklist

• Decide: document-as-intentional vs. generate vs. registry approach
• Implement chosen approach
• Verify existing callers are unaffected
• Run make test to verify no regressions

Parent Issue

See parent analysis report: #2907
Related to #2907

Generated by Duplicate Code Detector · ◷

• expires on Apr 7, 2026, 6:02 AM UTC