Skip to content

phsl / new-api

Go to a project
Commit c9ce1210 by tbphp
Options

feat: 优化代码，去除多余注释和修改

parent b32ed556
Showing with 39 additions and 119 deletions
.old/option.go 0 → 100644
This diff is collapsed. Click to expand it.
middleware/model-rate-limit.go
......@@ -19,25 +19,20 @@ const (
ModelRequestRateLimitSuccessCountMark = "MRRLS"
)
// 检查Redis中的请求限制
func checkRedisRateLimit(ctx context.Context, rdb *redis.Client, key string, maxCount int, duration int64) (bool, error) {
// 如果maxCount为0，表示不限制
if maxCount == 0 {
return true, nil
}
// 获取当前计数
length, err := rdb.LLen(ctx, key).Result()
if err != nil {
return false, err
}
// 如果未达到限制，允许请求
if length < int64(maxCount) {
return true, nil
}
// 检查时间窗口
oldTimeStr, _ := rdb.LIndex(ctx, key, -1).Result()
oldTime, err := time.Parse(timeFormat, oldTimeStr)
if err != nil {
......@@ -49,7 +44,6 @@ func checkRedisRateLimit(ctx context.Context, rdb *redis.Client, key string, max
if err != nil {
return false, err
}
// 如果在时间窗口内已达到限制，拒绝请求
subTime := nowTime.Sub(oldTime).Seconds()
if int64(subTime) < duration {
rdb.Expire(ctx, key, time.Duration(setting.ModelRequestRateLimitDurationMinutes)*time.Minute)
......@@ -59,9 +53,7 @@ func checkRedisRateLimit(ctx context.Context, rdb *redis.Client, key string, max
return true, nil
}
// 记录Redis请求
func recordRedisRequest(ctx context.Context, rdb *redis.Client, key string, maxCount int) {
// 如果maxCount为0，不记录请求
if maxCount == 0 {
return
}
......@@ -72,14 +64,12 @@ func recordRedisRequest(ctx context.Context, rdb *redis.Client, key string, maxC
rdb.Expire(ctx, key, time.Duration(setting.ModelRequestRateLimitDurationMinutes)*time.Minute)
}
// Redis限流处理器
func redisRateLimitHandler(duration int64, totalMaxCount, successMaxCount int) gin.HandlerFunc {
return func(c *gin.Context) {
userId := strconv.Itoa(c.GetInt("id"))
ctx := context.Background()
rdb := common.RDB
// 1. 检查成功请求数限制
successKey := fmt.Sprintf("rateLimit:%s:%s", ModelRequestRateLimitSuccessCountMark, userId)
allowed, err := checkRedisRateLimit(ctx, rdb, successKey, successMaxCount, duration)
if err != nil {
......@@ -92,9 +82,7 @@ func redisRateLimitHandler(duration int64, totalMaxCount, successMaxCount int) g
return
}
//2.检查总请求数限制并记录总请求（当totalMaxCount为0时会自动跳过，使用令牌桶限流器
totalKey := fmt.Sprintf("rateLimit:%s", userId)
// 初始化
tb := limiter.New(ctx, rdb)
allowed, err = tb.Allow(
ctx,
......@@ -114,17 +102,14 @@ func redisRateLimitHandler(duration int64, totalMaxCount, successMaxCount int) g
abortWithOpenAiMessage(c, http.StatusTooManyRequests, fmt.Sprintf("您已达到总请求数限制：%d分钟内最多请求%d次，包括失败次数，请检查您的请求是否正确", setting.ModelRequestRateLimitDurationMinutes, totalMaxCount))
}
// 4. 处理请求
c.Next()
// 5. 如果请求成功，记录成功请求
if c.Writer.Status() < 400 {
recordRedisRequest(ctx, rdb, successKey, successMaxCount)
}
}
}
// 内存限流处理器
func memoryRateLimitHandler(duration int64, totalMaxCount, successMaxCount int) gin.HandlerFunc {
inMemoryRateLimiter.Init(time.Duration(setting.ModelRequestRateLimitDurationMinutes) * time.Minute)
......@@ -133,15 +118,12 @@ func memoryRateLimitHandler(duration int64, totalMaxCount, successMaxCount int)
totalKey := ModelRequestRateLimitCountMark + userId
successKey := ModelRequestRateLimitSuccessCountMark + userId
// 1. 检查总请求数限制（当totalMaxCount为0时跳过）
if totalMaxCount > 0 && !inMemoryRateLimiter.Request(totalKey, totalMaxCount, duration) {
c.Status(http.StatusTooManyRequests)
c.Abort()
return
}
// 2. 检查成功请求数限制
// 使用一个临时key来检查限制，这样可以避免实际记录
checkKey := successKey + "_check"
if !inMemoryRateLimiter.Request(checkKey, successMaxCount, duration) {
c.Status(http.StatusTooManyRequests)
......@@ -149,54 +131,47 @@ func memoryRateLimitHandler(duration int64, totalMaxCount, successMaxCount int)
return
}
// 3. 处理请求
c.Next()
// 4. 如果请求成功，记录到实际的成功请求计数中
if c.Writer.Status() < 400 {
inMemoryRateLimiter.Request(successKey, successMaxCount, duration)
}
}
}
// ModelRequestRateLimit 模型请求限流中间件
func ModelRequestRateLimit() func(c *gin.Context) {
return func(c *gin.Context) {
// 在每个请求时检查是否启用限流
if !setting.ModelRequestRateLimitEnabled {
c.Next()
return
}
// 计算通用限流参数
duration := int64(setting.ModelRequestRateLimitDurationMinutes * 60)
// 获取用户组
group := c.GetString("token_group")
if group == "" {
group = c.GetString("group")
}
if group == "" {
group = "default" // 默认组
group = "default"
}
// 尝试获取用户组特定的限制
groupTotalCount, groupSuccessCount, found := setting.GetGroupRateLimit(group)
// 确定最终的限制值
finalTotalCount := setting.ModelRequestRateLimitCount // 默认使用全局总次数限制
finalSuccessCount := setting.ModelRequestRateLimitSuccessCount // 默认使用全局成功次数限制
finalTotalCount := setting.ModelRequestRateLimitCount
finalSuccessCount := setting.ModelRequestRateLimitSuccessCount
foundGroupLimit := false
groupTotalCount, groupSuccessCount, found := setting.GetGroupRateLimit(group)
if found {
// 如果找到用户组特定限制，则使用它们
finalTotalCount = groupTotalCount
finalSuccessCount = groupSuccessCount
foundGroupLimit = true
common.LogWarn(c.Request.Context(), fmt.Sprintf("Using rate limit for group '%s': total=%d, success=%d", group, finalTotalCount, finalSuccessCount))
} else {
}
if !foundGroupLimit {
common.LogInfo(c.Request.Context(), fmt.Sprintf("No specific rate limit found for group '%s', using global limits: total=%d, success=%d", group, finalTotalCount, finalSuccessCount))
}
// 根据存储类型选择并执行限流处理器，传入最终确定的限制值
if common.RedisEnabled {
redisRateLimitHandler(duration, finalTotalCount, finalSuccessCount)(c)
} else {
......
model/option.go
......@@ -94,11 +94,12 @@ func InitOptionMap() {
common.OptionMap["ModelRequestRateLimitCount"] = strconv.Itoa(setting.ModelRequestRateLimitCount)
common.OptionMap["ModelRequestRateLimitDurationMinutes"] = strconv.Itoa(setting.ModelRequestRateLimitDurationMinutes)
common.OptionMap["ModelRequestRateLimitSuccessCount"] = strconv.Itoa(setting.ModelRequestRateLimitSuccessCount)
jsonBytes, _ := json.Marshal(map[string][2]int{})
common.OptionMap["ModelRequestRateLimitGroup"] = string(jsonBytes)
common.OptionMap["ModelRatio"] = operation_setting.ModelRatio2JSONString()
common.OptionMap["ModelPrice"] = operation_setting.ModelPrice2JSONString()
common.OptionMap["CacheRatio"] = operation_setting.CacheRatio2JSONString()
common.OptionMap["GroupRatio"] = setting.GroupRatio2JSONString()
common.OptionMap[setting.ModelRequestRateLimitGroupKey] = "{}" // 添加用户组速率限制默认值
common.OptionMap["UserUsableGroups"] = setting.UserUsableGroups2JSONString()
common.OptionMap["CompletionRatio"] = operation_setting.CompletionRatio2JSONString()
common.OptionMap["TopUpLink"] = common.TopUpLink
......@@ -153,47 +154,31 @@ func SyncOptions(frequency int) {
}
func UpdateOption(key string, value string) error {
originalValue := value // 保存原始值以备后用
originalValue := value
// Validate and format specific keys before saving
if key == setting.ModelRequestRateLimitGroupKey {
if key == "ModelRequestRateLimitGroup" {
var cfg map[string][2]int
// Validate the JSON structure first using the original value
err := json.Unmarshal([]byte(originalValue), &cfg)
if err != nil {
// 提供更具体的错误信息
return fmt.Errorf("无效的 JSON 格式 for %s: %w", key, err)
}
// TODO: 可以添加更细致的结构验证，例如检查数组长度是否为2，值是否为非负数等。
// if !isValidModelRequestRateLimitGroupConfig(cfg) {
// return fmt.Errorf("无效的配置值 for %s", key)
// }
// If valid, format the JSON before saving
formattedValueBytes, marshalErr := json.MarshalIndent(cfg, "", " ")
if marshalErr != nil {
// This should ideally not happen if validation passed, but handle defensively
return fmt.Errorf("failed to marshal validated %s config: %w", key, marshalErr)
}
value = string(formattedValueBytes) // Use formatted JSON for saving and memory update
value = string(formattedValueBytes)
}
// Save to database
option := Option{
Key: key,
}
// https://gorm.io/docs/update.html#Save-All-Fields
DB.FirstOrCreate(&option, Option{Key: key})
option.Value = value
// Save is a combination function.
// If save value does not contain primary key, it will execute Create,
// otherwise it will execute Update (with all fields).
if err := DB.Save(&option).Error; err != nil {
return fmt.Errorf("保存选项 %s 到数据库失败: %w", key, err) // 添加错误上下文
return fmt.Errorf("保存选项 %s 到数据库失败: %w", key, err)
}
// Update OptionMap in memory using the potentially formatted value
// updateOptionMap 会处理内存中 setting.ModelRequestRateLimitGroupConfig 的更新
return updateOptionMap(key, value)
}
......@@ -370,6 +355,8 @@ func updateOptionMap(key string, value string) (err error) {
setting.ModelRequestRateLimitDurationMinutes, _ = strconv.Atoi(value)
case "ModelRequestRateLimitSuccessCount":
setting.ModelRequestRateLimitSuccessCount, _ = strconv.Atoi(value)
case "ModelRequestRateLimitGroup":
err = setting.UpdateModelRequestRateLimitGroup(value)
case "RetryTimes":
common.RetryTimes, _ = strconv.Atoi(value)
case "DataExportInterval":
......@@ -404,15 +391,6 @@ func updateOptionMap(key string, value string) (err error) {
operation_setting.AutomaticDisableKeywordsFromString(value)
case "StreamCacheQueueLength":
setting.StreamCacheQueueLength, _ = strconv.Atoi(value)
case setting.ModelRequestRateLimitGroupKey:
// Use the (potentially formatted) value passed from UpdateOption
// to update the actual configuration in memory.
// This is the single point where the memory state for this specific setting is updated.
err = setting.UpdateModelRequestRateLimitGroupConfig(value)
if err != nil {
// 添加错误上下文
err = fmt.Errorf("更新内存中的 %s 配置失败: %w", key, err)
}
}
return err
}
......
setting/rate_limit.go
......@@ -11,24 +11,17 @@ var ModelRequestRateLimitEnabled = false
var ModelRequestRateLimitDurationMinutes = 1
var ModelRequestRateLimitCount = 0
var ModelRequestRateLimitSuccessCount = 1000
var ModelRequestRateLimitGroup map[string][2]int
// ModelRequestRateLimitGroupKey 定义了模型请求按组速率限制的配置键
const ModelRequestRateLimitGroupKey = "ModelRequestRateLimitGroup"
// ModelRequestRateLimitGroupConfig 存储按用户组解析后的速率限制配置
// map[groupName][2]int{totalCount, successCount}
var ModelRequestRateLimitGroupConfig map[string][2]int
var ModelRequestRateLimitGroupMutex sync.RWMutex
// UpdateModelRequestRateLimitGroupConfig 解析、校验并更新内存中的用户组速率限制配置
func UpdateModelRequestRateLimitGroupConfig(jsonStr string) error {
func UpdateModelRequestRateLimitGroup(jsonStr string) error {
ModelRequestRateLimitGroupMutex.Lock()
defer ModelRequestRateLimitGroupMutex.Unlock()
var newConfig map[string][2]int
if jsonStr == "" || jsonStr == "{}" {
// 如果配置为空或空JSON对象，则清空内存配置
ModelRequestRateLimitGroupConfig = make(map[string][2]int)
ModelRequestRateLimitGroup = make(map[string][2]int)
common.SysLog("Model request rate limit group config cleared")
return nil
}
......@@ -38,37 +31,19 @@ func UpdateModelRequestRateLimitGroupConfig(jsonStr string) error {
return fmt.Errorf("failed to unmarshal ModelRequestRateLimitGroup config: %w", err)
}
// 校验配置值
for group, limits := range newConfig {
if len(limits) != 2 {
return fmt.Errorf("invalid config for group '%s': limits array length must be 2", group)
}
if limits[1] <= 0 { // successCount must be greater than 0
return fmt.Errorf("invalid config for group '%s': successCount (limits[1]) must be greater than 0", group)
}
if limits[0] < 0 { // totalCount can be 0 (no limit) or positive
return fmt.Errorf("invalid config for group '%s': totalCount (limits[0]) cannot be negative", group)
}
if limits[0] > 0 && limits[0] < limits[1] { // If totalCount is set, it must be >= successCount
return fmt.Errorf("invalid config for group '%s': totalCount (limits[0]) must be greater than or equal to successCount (limits[1]) when totalCount > 0", group)
}
}
ModelRequestRateLimitGroupConfig = newConfig
common.SysLog("Model request rate limit group config updated")
ModelRequestRateLimitGroup = newConfig
return nil
}
// GetGroupRateLimit 安全地获取指定用户组的速率限制值
func GetGroupRateLimit(group string) (totalCount, successCount int, found bool) {
ModelRequestRateLimitGroupMutex.RLock()
defer ModelRequestRateLimitGroupMutex.RUnlock()
if ModelRequestRateLimitGroupConfig == nil {
return 0, 0, false // 配置尚未初始化
if ModelRequestRateLimitGroup == nil {
return 0, 0, false
}
limits, found := ModelRequestRateLimitGroupConfig[group]
limits, found := ModelRequestRateLimitGroup[group]
if !found {
return 0, 0, false
}
......
web/src/components/RateLimitSetting.js
......@@ -13,7 +13,7 @@ const RateLimitSetting = () => {
ModelRequestRateLimitCount: 0,
ModelRequestRateLimitSuccessCount: 1000,
ModelRequestRateLimitDurationMinutes: 1,
ModelRequestRateLimitGroup: {},
ModelRequestRateLimitGroup: '{}',
});
let [loading, setLoading] = useState(false);
......@@ -24,13 +24,15 @@ const RateLimitSetting = () => {
if (success) {
let newInputs = {};
data.forEach((item) => {
// 检查 key 是否在初始 inputs 中定义
if (Object.prototype.hasOwnProperty.call(inputs, item.key)) {
if (item.key.endsWith('Enabled')) {
newInputs[item.key] = item.value === 'true' ? true : false;
newInputs[item.key] = item.value === 'true';
} else {
newInputs[item.key] = item.value;
}
}
});
setInputs(newInputs);
} else {
showError(message);
......@@ -40,7 +42,6 @@ const RateLimitSetting = () => {
try {
setLoading(true);
await getOptions();
// showSuccess('刷新成功');
} catch (error) {
showError('刷新失败');
} finally {
......@@ -55,13 +56,12 @@ const RateLimitSetting = () => {
return (
<>
<Spin spinning={loading} size='large'>
{/* AI请求速率限制 */}
<Card style={{ marginTop: '10px' }}>
<RequestRateLimit options={inputs} refresh={onRefresh} />
</Card>
</Spin>
</>
);
};
};
export default RateLimitSetting;
export default RateLimitSetting;
web/src/pages/Setting/RateLimit/SettingsRequestRateLimit.js
......@@ -18,7 +18,7 @@ export default function RequestRateLimit(props) {
ModelRequestRateLimitCount: -1,
ModelRequestRateLimitSuccessCount: 1000,
ModelRequestRateLimitDurationMinutes: 1,
ModelRequestRateLimitGroup: '{}', // 添加新字段并设置默认值
ModelRequestRateLimitGroup: '{}',
});
const refForm = useRef();
const [inputsRow, setInputsRow] = useState(inputs);
......@@ -33,13 +33,11 @@ export default function RequestRateLimit(props) {
} else {
value = inputs[item.key];
}
// 校验 ModelRequestRateLimitGroup 是否为有效的 JSON 对象字符串
if (item.key === 'ModelRequestRateLimitGroup') {
try {
JSON.parse(value);
} catch (e) {
showError(t('用户组速率限制配置不是有效的 JSON 格式！'));
// 阻止请求发送
return Promise.reject('Invalid JSON format');
}
}
......@@ -49,11 +47,9 @@ export default function RequestRateLimit(props) {
});
});
// 过滤掉无效的请求（例如，无效的 JSON）
const validRequests = requestQueue.filter(req => req !== null && req !== undefined && typeof req.then === 'function');
if (validRequests.length === 0 && requestQueue.length > 0) {
// 如果所有请求都被过滤掉了（因为 JSON 无效），则不继续执行
return;
}
......@@ -68,11 +64,9 @@ export default function RequestRateLimit(props) {
}
showSuccess(t('保存成功'));
props.refresh();
// 更新 inputsRow 以反映保存后的状态
setInputsRow(structuredClone(inputs));
})
.catch((error) => {
// 检查是否是由于无效 JSON 导致的错误
if (error !== 'Invalid JSON format') {
showError(t('保存失败，请重试'));
}
......@@ -85,17 +79,16 @@ export default function RequestRateLimit(props) {
useEffect(() => {
const currentInputs = {};
for (let key in props.options) {
if (Object.keys(inputs).includes(key)) {
if (Object.prototype.hasOwnProperty.call(inputs, key)) { // 使用 hasOwnProperty 检查
currentInputs[key] = props.options[key];
}
}
setInputs(currentInputs);
setInputsRow(structuredClone(currentInputs));
// 检查 refForm.current 是否存在
if (refForm.current) {
refForm.current.setValues(currentInputs);
}
}, [props.options]); // 依赖项保持不变，因为 inputs 状态的结构已固定
}, [props.options]);
return (
<>
......@@ -175,16 +168,15 @@ export default function RequestRateLimit(props) {
/>
</Col>
</Row>
{/* 用户组速率限制配置项 */}
<Row>
<Row style={{ marginTop: 15 }}>
<Col span={24}>
<Form.TextArea
label={t('用户组速率限制 (JSON)')}
field={'ModelRequestRateLimitGroup'}
placeholder={t( // 更新 placeholder
placeholder={t(
'请输入 JSON 格式的用户组限制，例如：\n{\n "default": [200, 100],\n "vip": [1000, 500]\n}',
)}
extraText={ // 更新 extraText
extraText={
<div>
<p>{t('说明:')}</p>
<ul>
......@@ -203,7 +195,7 @@ export default function RequestRateLimit(props) {
onChange={(value) => {
setInputs({
...inputs,
ModelRequestRateLimitGroup: value, // 直接更新字符串值
ModelRequestRateLimitGroup: value,
});
}}
/>
......@@ -219,4 +211,4 @@ export default function RequestRateLimit(props) {
</Spin>
</>
);
}
}
Markdown is supported
0% or
You are about to add 0 people to the discussion. Proceed with caution.
Finish editing this message first!
Please register or sign in to comment