✨ Add weather report parsing for METAR, SPECI, and TAF

Implement comprehensive weather parsing capabilities following Clean Architecture principles with composite parser pattern for routing between aviation and weather messages. ## Features Added - Weather report parsing (METAR, SPECI, TAF) - Composite parser pattern for message routing - Lenient parsing with warnings for unrecognized tokens - Support for PROB and RMK sections in TAF - Rich domain modeling with typed weather elements ## Architecture **Domain Layer** (internal/domain/weather/): - WeatherMessage interface with Metar and Taf implementations - Weather elements: Wind, Visibility, Cloud, Temperature, Altimeter, Phenomenon - Domain errors: ErrInvalidFormat, ErrMissingStation, ErrMissingTime **Port Layer** (internal/port/weather_parser.go): - WeatherParser interface with CanParse and Parse methods **Adapter Layer** (internal/adapter/parser/weather/): - WeatherParserImpl with classification and parsing logic - Comprehensive regex patterns for weather elements - METAR/SPECI parser with element extraction - TAF parser with period handling (FM, TEMPO, BECMG, PROB) - Helper functions for time parsing and unit conversions **Composite Parser** (internal/adapter/parser/composite.go): - Routes weather reports to weather parser - Falls back to aviation parser for telegrams - Converts WeatherMessage to ParsedTelegram format ## Integration - Updated ProvideParser to create composite parser with weather parser - Added weather parser to Wire DI configuration - Updated processor_bench_test.go for weather parser integration - Documentation added in docs/weather-parser.md ## Testing - 29 comprehensive tests for weather parsing (all passing) - Tests for classification, METAR, SPECI, TAF, and composite routing - Benchmark compatibility maintained ## Fixes Applied - TAF PROB parsing: Include PROB/RMK in special section detection - Composite test: Updated to use properly formatted AFTN telegram - Linter issues: Switch statement refactor, removed unused patterns - Ineffective break statement fixed in TAF parser ## Coverage ~1,743 lines of new code with: - Complete METAR/SPECI parsing - TAF parsing with period support - Lenient error handling with warnings - Unit conversions and time utilities 🤖 Generated with [Claude Code](https://claude.com/claude-code) Co-Authored-By: Claude Sonnet 4.5 <noreply@anthropic.com>
2025-12-24 16:38:03 +08:00
parent c1808c0523
commit d0fd461e38
24 changed files with 1873 additions and 11 deletions
@@ -0,0 +1,105 @@
+package weather
+
+import "strings"
+
+// Tokenize splits the raw text into tokens by whitespace
+func Tokenize(raw string) []string {
+	raw = strings.TrimSpace(raw)
+	// Remove trailing '=' if present
+	if strings.HasSuffix(raw, "=") {
+		raw = raw[:len(raw)-1]
+		raw = strings.TrimSpace(raw)
+	}
+
+	tokens := strings.Fields(raw)
+	return tokens
+}
+
+// Section represents a special section in the report (RMK, TEMPO, BECMG, FM)
+type Section struct {
+	Type      string   // "RMK", "TEMPO", "BECMG", "FM"
+	StartIdx  int      // Starting token index
+	EndIdx    int      // Ending token index (exclusive)
+	Tokens    []string // Tokens in this section
+}
+
+// FindSpecialSections identifies special sections in the tokenized report
+func FindSpecialSections(tokens []string) []Section {
+	var sections []Section
+	var currentSection *Section
+
+	for i, token := range tokens {
+		upperToken := strings.ToUpper(token)
+
+		switch {
+		case strings.HasPrefix(upperToken, "RMK"):
+			if currentSection != nil {
+				currentSection.EndIdx = i
+				sections = append(sections, *currentSection)
+			}
+			currentSection = &Section{
+				Type:     "RMK",
+				StartIdx: i,
+				Tokens:   []string{token},
+			}
+
+		case strings.HasPrefix(upperToken, "TEMPO"):
+			if currentSection != nil && currentSection.Type != "RMK" {
+				currentSection.EndIdx = i
+				sections = append(sections, *currentSection)
+			}
+			currentSection = &Section{
+				Type:     "TEMPO",
+				StartIdx: i,
+				Tokens:   []string{token},
+			}
+
+		case strings.HasPrefix(upperToken, "BECMG"):
+			if currentSection != nil && currentSection.Type != "RMK" {
+				currentSection.EndIdx = i
+				sections = append(sections, *currentSection)
+			}
+			currentSection = &Section{
+				Type:     "BECMG",
+				StartIdx: i,
+				Tokens:   []string{token},
+			}
+
+		case strings.HasPrefix(upperToken, "FM"):
+			if currentSection != nil && currentSection.Type != "RMK" {
+				currentSection.EndIdx = i
+				sections = append(sections, *currentSection)
+			}
+			currentSection = &Section{
+				Type:     "FM",
+				StartIdx: i,
+				Tokens:   []string{token},
+			}
+
+		default:
+			if currentSection != nil {
+				currentSection.Tokens = append(currentSection.Tokens, token)
+			}
+		}
+	}
+
+	// Close the last section
+	if currentSection != nil {
+		currentSection.EndIdx = len(tokens)
+		sections = append(sections, *currentSection)
+	}
+
+	return sections
+}
+
+// ExtractSectionTokens extracts tokens for a specific section type
+func ExtractSectionTokens(tokens []string, sectionType string) []string {
+	sections := FindSpecialSections(tokens)
+	for _, section := range sections {
+		if section.Type == sectionType {
+			return section.Tokens
+		}
+	}
+	return nil
+}
+