windyboyandClaude Sonnet 4.5
c0a66cf845
This commit implements a complete refactoring of the ICAO aviation parser, addressing 15 identified issues across security, performance, code quality, and documentation. Security Enhancements (P0 - Critical): - Add input size validation (max 1800 chars per AFTN standard) - Implement ReDoS protection with 100ms regex timeout mechanism - Add field validation to prevent nil pointer dereferences - Document intentional error handling pattern for audit compliance Performance & Design Improvements (P1 - Important): - Remove unnecessary mutex from BodyParser (eliminates serialization) - Fix tokenizer slash handling logic - Remove global logger dependencies (zap.S() calls) Code Quality Improvements (P2): - Refactor parseRemainingLines with clear helper functions - Document all regex patterns with ICAO format specifications - Replace magic numbers with named constants (5 new constants) - Add error message sanitization to prevent data leakage Documentation & Polish (P3): - Create comprehensive package documentation (doc.go) - Verify naming consistency across all functions - Add 54 comprehensive tests (all passing) - Verify performance with benchmarks (~10µs for simple messages) New Files: - validation.go: Input validation utilities with AFTN limits - validation_test.go: Comprehensive validation tests - regex_timeout.go: ReDoS protection mechanism - regex_timeout_test.go: Timeout protection tests - suite_test.go: Ginkgo test suite registration - doc.go: Package-level documentation All changes maintain backward compatibility and existing architecture while significantly enhancing security, maintainability, and code quality. 🤖 Generated with [Claude Code](https://claude.com/claude-code) Co-Authored-By: Claude Sonnet 4.5 <noreply@anthropic.com>
56 lines
1019 B
Go
56 lines
1019 B
Go
package aviation
|
|
|
|
import "strings"
|
|
|
|
// Token represents a lexeme in the body with its byte offsets.
|
|
type Token struct {
|
|
Text string
|
|
Start int
|
|
End int
|
|
}
|
|
|
|
// Tokenizer splits text into tokens using a whitespace set.
|
|
// Whitespace characters split tokens but are not emitted.
|
|
// All other characters (including '/') are included in tokens.
|
|
type Tokenizer struct {
|
|
Whitespace string
|
|
}
|
|
|
|
// Tokenize tokenizes input and returns tokens with byte offsets.
|
|
func (t Tokenizer) Tokenize(input string) []Token {
|
|
if t.Whitespace == "" {
|
|
t.Whitespace = " \n\t\r"
|
|
}
|
|
|
|
var tokens []Token
|
|
start := -1
|
|
|
|
for idx, r := range input {
|
|
if strings.ContainsRune(t.Whitespace, r) {
|
|
if start != -1 {
|
|
tokens = append(tokens, Token{
|
|
Text: input[start:idx],
|
|
Start: start,
|
|
End: idx,
|
|
})
|
|
start = -1
|
|
}
|
|
continue
|
|
}
|
|
|
|
if start == -1 {
|
|
start = idx
|
|
}
|
|
}
|
|
|
|
if start != -1 {
|
|
tokens = append(tokens, Token{
|
|
Text: input[start:],
|
|
Start: start,
|
|
End: len(input),
|
|
})
|
|
}
|
|
|
|
return tokens
|
|
}
|