feat(clickhouse): integrate goose migrations and analytics repository

Add a separate ClickHouse OLAP pipeline with goose/clickhouse DDL as the
sole schema source, model/analytics for ORM mapping, and repository/analytics
for reads (ChDB/GORM) and batch writes (ChConn). Refactor risk_control and
admin/logs to use the repository layer instead of inline SQL. Remove the
manual support-files DDL and document the workflow in database-migration skill.
This commit is contained in:
ryan
2026-06-19 11:56:58 +08:00
parent 13e9fead35
commit 00593742c3
22 changed files with 1033 additions and 352 deletions
@@ -0,0 +1,96 @@
// Copyright 2026 Arctel.net
// SPDX-License-Identifier: Apache-2.0
// Package analytics provides ClickHouse data access for analytics tables.
package analytics
import (
"context"
"fmt"
"github.com/Rain-kl/Wavelet/internal/db"
analyticsmodel "github.com/Rain-kl/Wavelet/internal/model/analytics"
"gorm.io/gorm"
)
// CountAccessLogs returns the number of access logs matching filter.
func CountAccessLogs(ctx context.Context, filter AccessLogFilter) (uint64, error) {
ch := db.ChDB(ctx)
if ch == nil {
return 0, fmt.Errorf("clickhouse gorm connection is not initialized")
}
var count int64
query := applyFilter(ch.Model(&analyticsmodel.UserAccessLog{}), filter)
if err := query.Count(&count).Error; err != nil {
return 0, fmt.Errorf("count access logs: %w", err)
}
return safeUint64Count(count), nil
}
// ListAccessLogs returns paginated access logs and the total match count.
func ListAccessLogs(ctx context.Context, filter AccessLogFilter, page, pageSize int) ([]analyticsmodel.UserAccessLog, uint64, error) {
ch := db.ChDB(ctx)
if ch == nil {
return nil, 0, fmt.Errorf("clickhouse gorm connection is not initialized")
}
if filter.UserIDs != nil && len(filter.UserIDs) == 0 {
return []analyticsmodel.UserAccessLog{}, 0, nil
}
var total int64
baseQuery := applyFilter(ch.Model(&analyticsmodel.UserAccessLog{}), filter)
if err := baseQuery.Count(&total).Error; err != nil {
return nil, 0, fmt.Errorf("count access logs: %w", err)
}
if total == 0 {
return []analyticsmodel.UserAccessLog{}, 0, nil
}
if page < 1 {
page = 1
}
if pageSize < 1 {
pageSize = 20
}
offset := (page - 1) * pageSize
var logs []analyticsmodel.UserAccessLog
err := applyFilter(ch.Model(&analyticsmodel.UserAccessLog{}), filter).
Order("created_at DESC, id DESC").
Limit(pageSize).
Offset(offset).
Find(&logs).Error
if err != nil {
return nil, 0, fmt.Errorf("list access logs: %w", err)
}
return logs, safeUint64Count(total), nil
}
func safeUint64Count(count int64) uint64 {
if count < 0 {
return 0
}
return uint64(count)
}
func applyFilter(query *gorm.DB, filter AccessLogFilter) *gorm.DB {
if filter.UserIDs != nil {
if len(filter.UserIDs) == 0 {
return query.Where("1 = 0")
}
query = query.Where("user_id IN ?", filter.UserIDs)
}
if filter.Path != "" {
query = query.Where("path LIKE ?", "%"+filter.Path+"%")
}
if filter.StartTime != nil {
query = query.Where("created_at >= ?", *filter.StartTime)
}
if filter.EndTime != nil {
query = query.Where("created_at <= ?", *filter.EndTime)
}
return query
}
@@ -0,0 +1,17 @@
// Copyright 2026 Arctel.net
// SPDX-License-Identifier: Apache-2.0
package analytics
import "time"
// AccessLogFilter scopes ClickHouse user access log queries.
type AccessLogFilter struct {
// UserIDs filters by user IDs. nil means no user filter; an empty slice means no matches.
UserIDs []uint64
Path string
// StartTime filters created_at >= StartTime when non-nil.
StartTime *time.Time
// EndTime filters created_at <= EndTime when non-nil.
EndTime *time.Time
}
@@ -0,0 +1,159 @@
// Copyright 2026 Arctel.net
// SPDX-License-Identifier: Apache-2.0
package analytics
import (
"context"
"fmt"
"sort"
"time"
"github.com/Rain-kl/Wavelet/internal/db"
analyticsmodel "github.com/Rain-kl/Wavelet/internal/model/analytics"
)
const hoursInDay = 24
// DailyTrend is a single day's access count.
type DailyTrend struct {
Date string
Count uint64
}
// BrowserShare is a browser group's share of access logs.
type BrowserShare struct {
Browser string
Count uint64
}
// TopUser is an active user ranked by access count.
type TopUser struct {
UserID uint64
Count uint64
}
// GetDailyTrend returns per-day access counts for the last days days (inclusive of today).
func GetDailyTrend(ctx context.Context, days int) ([]DailyTrend, error) {
if days < 1 {
days = 7
}
ch := db.ChDB(ctx)
if ch == nil {
return nil, fmt.Errorf("clickhouse gorm connection is not initialized")
}
startTime := time.Now().AddDate(0, 0, -(days - 1)).Truncate(hoursInDay * time.Hour)
tableName := analyticsmodel.UserAccessLog{}.TableName()
query := fmt.Sprintf(`
SELECT toDate(created_at) AS date, count() AS count
FROM %s
WHERE created_at >= ?
GROUP BY date
ORDER BY date ASC
`, tableName)
type trendRow struct {
Date time.Time
Count uint64
}
var rows []trendRow
if err := ch.Raw(query, startTime).Scan(&rows).Error; err != nil {
return nil, fmt.Errorf("get daily trend: %w", err)
}
trendMap := make(map[string]uint64, days)
for i := 0; i < days; i++ {
dateStr := time.Now().AddDate(0, 0, -i).Format("2006-01-02")
trendMap[dateStr] = 0
}
for _, row := range rows {
dateStr := row.Date.Format("2006-01-02")
trendMap[dateStr] = row.Count
}
result := make([]DailyTrend, 0, days)
for i := days - 1; i >= 0; i-- {
dateStr := time.Now().AddDate(0, 0, -i).Format("2006-01-02")
result = append(result, DailyTrend{
Date: dateStr,
Count: trendMap[dateStr],
})
}
return result, nil
}
// GetBrowserDistribution returns browser-grouped access counts since startTime.
func GetBrowserDistribution(ctx context.Context, startTime time.Time) ([]BrowserShare, error) {
ch := db.ChDB(ctx)
if ch == nil {
return nil, fmt.Errorf("clickhouse gorm connection is not initialized")
}
tableName := analyticsmodel.UserAccessLog{}.TableName()
query := fmt.Sprintf(`
SELECT user_agent, count() AS count
FROM %s
WHERE created_at >= ?
GROUP BY user_agent
`, tableName)
type uaRow struct {
UserAgent string
Count uint64
}
var rows []uaRow
if err := ch.Raw(query, startTime).Scan(&rows).Error; err != nil {
return nil, fmt.Errorf("get browser distribution: %w", err)
}
browserCounts := make(map[string]uint64)
for _, row := range rows {
browser := ParseBrowserName(row.UserAgent)
browserCounts[browser] += row.Count
}
result := make([]BrowserShare, 0, len(browserCounts))
for browser, count := range browserCounts {
result = append(result, BrowserShare{
Browser: browser,
Count: count,
})
}
sort.Slice(result, func(i, j int) bool {
return result[i].Count > result[j].Count
})
return result, nil
}
// GetTopActiveUsers returns the most active users since startTime.
func GetTopActiveUsers(ctx context.Context, startTime time.Time, limit int) ([]TopUser, error) {
if limit < 1 {
limit = 10
}
ch := db.ChDB(ctx)
if ch == nil {
return nil, fmt.Errorf("clickhouse gorm connection is not initialized")
}
tableName := analyticsmodel.UserAccessLog{}.TableName()
query := fmt.Sprintf(`
SELECT user_id, count() AS count
FROM %s
WHERE created_at >= ? AND user_id > 0
GROUP BY user_id
ORDER BY count DESC
LIMIT ?
`, tableName)
var users []TopUser
if err := ch.Raw(query, startTime, limit).Scan(&users).Error; err != nil {
return nil, fmt.Errorf("get top active users: %w", err)
}
return users, nil
}
@@ -0,0 +1,207 @@
// Copyright 2026 Arctel.net
// SPDX-License-Identifier: Apache-2.0
package analytics
import (
"context"
"testing"
"time"
"github.com/ClickHouse/clickhouse-go/v2/lib/column"
"github.com/ClickHouse/clickhouse-go/v2/lib/driver"
"github.com/Rain-kl/Wavelet/internal/db"
analyticsmodel "github.com/Rain-kl/Wavelet/internal/model/analytics"
"github.com/glebarez/sqlite"
"github.com/stretchr/testify/assert"
"github.com/stretchr/testify/require"
"gorm.io/gorm"
)
func setupChGormDB(t *testing.T) *gorm.DB {
t.Helper()
gormDB, err := gorm.Open(sqlite.Open(":memory:"), &gorm.Config{
DisableForeignKeyConstraintWhenMigrating: true,
})
require.NoError(t, err)
require.NoError(t, gormDB.AutoMigrate(&analyticsmodel.UserAccessLog{}))
db.SetChDBForTest(gormDB)
return gormDB
}
func TestParseBrowserName(t *testing.T) {
tests := []struct {
name string
ua string
want string
}{
{name: "chrome", ua: "Mozilla/5.0 Chrome/120.0.0.0", want: "Chrome"},
{name: "firefox", ua: "Mozilla/5.0 Firefox/121.0", want: "Firefox"},
{name: "safari", ua: "Mozilla/5.0 Safari/605.1.15", want: "Safari"},
{name: "edge", ua: "Mozilla/5.0 Edg/120.0.0.0", want: "Edge"},
{name: "wechat", ua: "MicroMessenger/8.0", want: "WeChat"},
{name: "postman", ua: "PostmanRuntime/7.36.0", want: "Postman"},
{name: "other", ua: "curl/8.0", want: "Other"},
}
for _, tt := range tests {
t.Run(tt.name, func(t *testing.T) {
assert.Equal(t, tt.want, ParseBrowserName(tt.ua))
})
}
}
func TestCountAccessLogs_EmptyUserIDs(t *testing.T) {
setupChGormDB(t)
t.Cleanup(func() { db.SetChDBForTest(nil) })
count, err := CountAccessLogs(context.Background(), AccessLogFilter{UserIDs: []uint64{}})
require.NoError(t, err)
assert.Equal(t, uint64(0), count)
}
func TestListAccessLogs_EmptyUserIDs(t *testing.T) {
setupChGormDB(t)
t.Cleanup(func() { db.SetChDBForTest(nil) })
logs, total, err := ListAccessLogs(context.Background(), AccessLogFilter{UserIDs: []uint64{}}, 1, 20)
require.NoError(t, err)
assert.Equal(t, uint64(0), total)
assert.Empty(t, logs)
}
func TestListAccessLogs_WithFilters(t *testing.T) {
gormDB := setupChGormDB(t)
t.Cleanup(func() { db.SetChDBForTest(nil) })
now := time.Now().UTC().Truncate(time.Second)
logs := []analyticsmodel.UserAccessLog{
{ID: 1, UserID: 10, Path: "/api/v1/users", Method: "GET", Status: 200, CreatedAt: now},
{ID: 2, UserID: 20, Path: "/api/v1/admin/logs", Method: "GET", Status: 200, CreatedAt: now},
{ID: 3, UserID: 10, Path: "/api/v1/other", Method: "POST", Status: 201, CreatedAt: now},
}
require.NoError(t, gormDB.Create(&logs).Error)
start := now.Add(-time.Hour)
filter := AccessLogFilter{
UserIDs: []uint64{10},
Path: "users",
StartTime: &start,
}
count, err := CountAccessLogs(context.Background(), filter)
require.NoError(t, err)
assert.Equal(t, uint64(1), count)
result, total, err := ListAccessLogs(context.Background(), filter, 1, 10)
require.NoError(t, err)
assert.Equal(t, uint64(1), total)
require.Len(t, result, 1)
assert.Equal(t, uint64(1), result[0].ID)
assert.Equal(t, "/api/v1/users", result[0].Path)
}
func TestBatchInsert_Empty(t *testing.T) {
err := BatchInsert(context.Background(), nil)
require.NoError(t, err)
}
func TestBatchInsert_UsesModelBatchSQL(t *testing.T) {
ctx := context.Background()
mockBatch := &mockBatch{}
mockConn := &mockConn{
batch: mockBatch,
batchQuery: analyticsmodel.UserAccessLog{}.BatchInsertSQL(),
}
db.SetChConnForTest(mockConn)
t.Cleanup(func() { db.SetChConnForTest(nil) })
createdAt := time.Now().UTC()
err := BatchInsert(ctx, []analyticsmodel.UserAccessLog{
{
ID: 1,
UserID: 42,
Path: "/api/v1/test",
Method: "GET",
IP: "127.0.0.1",
UserAgent: "test-agent",
Headers: "{}",
Status: 200,
Latency: 12,
CreatedAt: createdAt,
},
})
require.NoError(t, err)
assert.True(t, mockConn.prepareCalled)
assert.Equal(t, analyticsmodel.UserAccessLog{}.BatchInsertSQL(), mockConn.preparedQuery)
assert.True(t, mockBatch.sendCalled)
require.Len(t, mockBatch.rows, 1)
assert.Equal(t, uint64(42), mockBatch.rows[0][1])
}
type mockConn struct {
batch driver.Batch
batchQuery string
prepareCalled bool
preparedQuery string
}
func (m *mockConn) Contributors() []string { return nil }
func (m *mockConn) ServerVersion() (*driver.ServerVersion, error) { return nil, nil }
func (m *mockConn) Select(_ context.Context, _ any, _ string, _ ...any) error { return nil }
func (m *mockConn) Query(_ context.Context, _ string, _ ...any) (driver.Rows, error) {
return nil, nil
}
func (m *mockConn) QueryRow(_ context.Context, _ string, _ ...any) driver.Row { return nil }
func (m *mockConn) PrepareBatch(_ context.Context, query string, _ ...driver.PrepareBatchOption) (driver.Batch, error) {
m.prepareCalled = true
m.preparedQuery = query
return m.batch, nil
}
func (m *mockConn) Exec(_ context.Context, _ string, _ ...any) error { return nil }
func (m *mockConn) AsyncInsert(_ context.Context, _ string, _ bool, _ ...any) error { return nil }
func (m *mockConn) Ping(_ context.Context) error { return nil }
func (m *mockConn) Stats() driver.Stats { return driver.Stats{} }
func (m *mockConn) Close() error { return nil }
type mockBatch struct {
rows [][]any
sendCalled bool
}
func (m *mockBatch) Abort() error { return nil }
func (m *mockBatch) Append(v ...any) error {
m.rows = append(m.rows, v)
return nil
}
func (m *mockBatch) AppendStruct(_ any) error { return nil }
func (m *mockBatch) Column(_ int) driver.BatchColumn { return nil }
func (m *mockBatch) Flush() error { return nil }
func (m *mockBatch) Send() error {
m.sendCalled = true
return nil
}
func (m *mockBatch) IsSent() bool { return m.sendCalled }
func (m *mockBatch) Rows() int { return len(m.rows) }
func (m *mockBatch) Columns() []column.Interface { return nil }
func (m *mockBatch) Close() error { return nil }
@@ -0,0 +1,49 @@
// Copyright 2026 Arctel.net
// SPDX-License-Identifier: Apache-2.0
package analytics
import (
"context"
"fmt"
"github.com/Rain-kl/Wavelet/internal/db"
analyticsmodel "github.com/Rain-kl/Wavelet/internal/model/analytics"
)
// BatchInsert writes access logs to ClickHouse using the native batch API.
func BatchInsert(ctx context.Context, logs []analyticsmodel.UserAccessLog) error {
if len(logs) == 0 {
return nil
}
if db.ChConn == nil {
return fmt.Errorf("clickhouse connection is not initialized")
}
batch, err := db.ChConn.PrepareBatch(ctx, analyticsmodel.UserAccessLog{}.BatchInsertSQL())
if err != nil {
return fmt.Errorf("prepare clickhouse batch: %w", err)
}
for _, logItem := range logs {
if err := batch.Append(
logItem.ID,
logItem.UserID,
logItem.Path,
logItem.Method,
logItem.IP,
logItem.UserAgent,
logItem.Headers,
logItem.Status,
logItem.Latency,
logItem.CreatedAt,
); err != nil {
return fmt.Errorf("append access log to batch: %w", err)
}
}
if err := batch.Send(); err != nil {
return fmt.Errorf("send clickhouse batch: %w", err)
}
return nil
}
+30
View File
@@ -0,0 +1,30 @@
// Copyright 2026 Arctel.net
// SPDX-License-Identifier: Apache-2.0
package analytics
import "strings"
// ParseBrowserName performs lightweight User-Agent browser identification.
func ParseBrowserName(ua string) string {
uaLower := strings.ToLower(ua)
if strings.Contains(uaLower, "micromessenger") {
return "WeChat"
}
if strings.Contains(uaLower, "postman") {
return "Postman"
}
if strings.Contains(uaLower, "edg/") || strings.Contains(uaLower, "edge") {
return "Edge"
}
if strings.Contains(uaLower, "firefox") {
return "Firefox"
}
if strings.Contains(uaLower, "chrome") {
return "Chrome"
}
if strings.Contains(uaLower, "safari") {
return "Safari"
}
return "Other"
}