improvements nov2025 (#10)

* Add benchmark and httplog modules, update UI for modals artefacts

internal/benchmark/results.go

@@ -0,0 +1,130 @@
+package benchmark
+
+import (
+	"sort"
+	"time"
+)
+
+// Results holds the aggregated results of a benchmark run
+type Results struct {
+	ForwardID     string          `json:"forward_id"`
+	URL           string          `json:"url"`
+	Method        string          `json:"method"`
+	StartTime     time.Time       `json:"start_time"`
+	EndTime       time.Time       `json:"end_time"`
+	TotalRequests int             `json:"total_requests"`
+	Successful    int             `json:"successful"`
+	Failed        int             `json:"failed"`
+	Latencies     []time.Duration `json:"-"` // Raw latencies for percentile calculation
+	StatusCodes   map[int]int     `json:"status_codes"`
+	Errors        map[string]int  `json:"errors,omitempty"`
+	BytesRead     int64           `json:"bytes_read"`
+	BytesWritten  int64           `json:"bytes_written"`
+}
+
+// Stats holds calculated statistics
+type Stats struct {
+	MinLatency time.Duration `json:"min_latency_ms"`
+	MaxLatency time.Duration `json:"max_latency_ms"`
+	AvgLatency time.Duration `json:"avg_latency_ms"`
+	P50Latency time.Duration `json:"p50_latency_ms"`
+	P95Latency time.Duration `json:"p95_latency_ms"`
+	P99Latency time.Duration `json:"p99_latency_ms"`
+	Throughput float64       `json:"throughput_rps"`
+	Duration   time.Duration `json:"duration"`
+}
+
+// NewResults creates a new Results instance
+func NewResults(forwardID, url, method string) *Results {
+	return &Results{
+		ForwardID:   forwardID,
+		URL:         url,
+		Method:      method,
+		StartTime:   time.Now(),
+		Latencies:   make([]time.Duration, 0),
+		StatusCodes: make(map[int]int),
+		Errors:      make(map[string]int),
+	}
+}
+
+// RecordSuccess records a successful HTTP request (transport succeeded)
+// Note: only 2xx status codes are counted as successful for statistics
+func (r *Results) RecordSuccess(statusCode int, latency time.Duration, bytesRead, bytesWritten int64) {
+	r.TotalRequests++
+	// Only count 2xx as successful
+	if statusCode >= 200 && statusCode < 300 {
+		r.Successful++
+	} else {
+		r.Failed++
+	}
+	r.Latencies = append(r.Latencies, latency)
+	r.StatusCodes[statusCode]++
+	r.BytesRead += bytesRead
+	r.BytesWritten += bytesWritten
+}
+
+// RecordFailure records a failed request
+func (r *Results) RecordFailure(err error, latency time.Duration) {
+	r.TotalRequests++
+	r.Failed++
+	r.Latencies = append(r.Latencies, latency)
+	r.Errors[err.Error()]++
+}
+
+// Finalize marks the benchmark as complete
+func (r *Results) Finalize() {
+	r.EndTime = time.Now()
+}
+
+// CalculateStats calculates statistics from the results
+func (r *Results) CalculateStats() Stats {
+	stats := Stats{
+		Duration: r.EndTime.Sub(r.StartTime),
+	}
+
+	if len(r.Latencies) == 0 {
+		return stats
+	}
+
+	// Sort latencies for percentile calculation
+	sorted := make([]time.Duration, len(r.Latencies))
+	copy(sorted, r.Latencies)
+	sort.Slice(sorted, func(i, j int) bool {
+		return sorted[i] < sorted[j]
+	})
+
+	// Calculate min, max, avg
+	var total time.Duration
+	stats.MinLatency = sorted[0]
+	stats.MaxLatency = sorted[len(sorted)-1]
+
+	for _, lat := range sorted {
+		total += lat
+	}
+	stats.AvgLatency = total / time.Duration(len(sorted))
+
+	// Calculate percentiles
+	stats.P50Latency = percentile(sorted, 50)
+	stats.P95Latency = percentile(sorted, 95)
+	stats.P99Latency = percentile(sorted, 99)
+
+	// Calculate throughput
+	if stats.Duration > 0 {
+		stats.Throughput = float64(r.TotalRequests) / stats.Duration.Seconds()
+	}
+
+	return stats
+}
+
+// percentile calculates the p-th percentile of sorted durations
+func percentile(sorted []time.Duration, p int) time.Duration {
+	if len(sorted) == 0 {
+		return 0
+	}
+
+	idx := (p * len(sorted)) / 100
+	if idx >= len(sorted) {
+		idx = len(sorted) - 1
+	}
+	return sorted[idx]
+}