Merge branch 'main' into dockerfile_prepare

.github/workflows/helm-docs.yaml

@@ -0,0 +1,35 @@
+name: "Helm Docs"
+
+on:
+  pull_request:
+    branches:
+      - "*"
+
+  push:
+
+jobs:
+
+  helm:
+    name: Helm Docs
+    runs-on: ubuntu-latest
+    steps:
+      - name: Checkout
+        uses: actions/checkout@v4
+        with:
+          fetch-depth: 1
+
+      - name: Setup Go
+        uses: actions/setup-go@v5
+        with:
+          go-version: '1.22.9'
+
+      - name: Run helm-docs
+        run: |
+          GOBIN=$PWD GO111MODULE=on go install github.com/norwoodj/helm-docs/cmd/helm-docs@v1.14.2
+          ./helm-docs -c ${GITHUB_WORKSPACE}/helm/higress -f ../core/values.yaml
+          DIFF=$(git diff ${GITHUB_WORKSPACE}/helm/higress/*md)
+          if [ ! -z "$DIFF" ]; then
+            echo "Please use helm-docs in your clone, of your fork, of the project, and commit a updated README.md for the chart."
+          fi
+          git diff --exit-code
+          rm -f ./helm-docs

.gitignore

@@ -16,4 +16,5 @@ helm/**/charts/**.tgz
 target/
 tools/hack/cluster.conf
 envoy/1.20
-istio/1.12
+istio/1.12
+Cargo.lock

helm/higress/README.md.gotmpl

@@ -0,0 +1,34 @@
+## Higress for Kubernetes
+
+Higress is a cloud-native api gateway based on Alibaba's internal gateway practices.
+
+Powered by Istio and Envoy, Higress realizes the integration of the triple gateway architecture of traffic gateway, microservice gateway and security gateway, thereby greatly reducing the costs of deployment, operation and maintenance.
+
+## Setup Repo Info
+
+```console
+helm repo add higress.io https://higress.io/helm-charts
+helm repo update
+```
+
+## Install
+
+To install the chart with the release name `higress`:
+
+```console
+helm install higress -n higress-system higress.io/higress --create-namespace --render-subchart-notes
+```
+
+## Uninstall
+
+To uninstall/delete the higress deployment:
+
+```console
+helm delete higress -n higress-system
+```
+
+The command removes all the Kubernetes components associated with the chart and deletes the release.
+
+## Parameters
+
+{{ template "chart.valuesSection" . }}

plugins/wasm-go/extensions/ai-agent/README.md

@@ -5,7 +5,7 @@ description: AI Agent插件配置参考
 ---
 
 ## 功能说明
-一个可定制化的 API AI Agent，支持配置 http method 类型为 GET 与 POST 的 API，支持多轮对话，支持流式与非流式模式。
+一个可定制化的 API AI Agent，支持配置 http method 类型为 GET 与 POST 的 API，支持多轮对话，支持流式与非流式模式，支持将结果格式化为自定义的 json。
 agent流程图如下：
 ![ai-agent](https://img.alicdn.com/imgextra/i1/O1CN01PGSDW31WQfEPm173u_!!6000000002783-0-tps-2733-1473.jpg)
 
@@ -21,6 +21,7 @@ agent流程图如下：
 | `llm`            | object    | 必填    | -      | 配置 AI 服务提供商的信息     |
 | `apis`           | object    | 必填    | -      | 配置外部 API 服务提供商的信息 |
 | `promptTemplate` | object    | 非必填  | -      | 配置 Agent ReAct 模板的信息  |
+| `jsonResp`       | object    | 非必填  | -      | 配置 json 格式化的相关信息   |
 
 `llm`的配置字段说明如下：
 
@@ -78,7 +79,14 @@ agent流程图如下：
 | `observation`   | string    | 非必填     | -      | Agent ReAct 模板的 observation 部分          |
 | `thought2`      | string    | 非必填     | -      | Agent ReAct 模板的 thought2 部分             |
 
-## 用法示例
+`jsonResp`的配置字段说明如下：
+
+| 名称               | 数据类型   | 填写要求 | 默认值 | 描述                               |
+|--------------------|-----------|---------|--------|-----------------------------------|
+| `enable`           | bool      | 非必填   | false  | 是否开启 json 格式化。             |
+| `jsonSchema`       | string    | 非必填   | -      | 自定义 json schema                |
+
+## 用法示例-不开启 json 格式化
 
 **配置信息**
 
@@ -293,7 +301,7 @@ deepl提供了一个工具，用于翻译给定的句子，支持多语言。。
 **请求示例**
 
 ```shell
-curl 'http://<这里换成网关公网IP>/api/openai/v1/chat/completions' \
+curl 'http://<这里换成网关地址>/api/openai/v1/chat/completions' \
 -H 'Accept: application/json, text/event-stream' \
 -H 'Content-Type: application/json' \
 --data-raw '{"model":"qwen","frequency_penalty":0,"max_tokens":800,"stream":false,"messages":[{"role":"user","content":"我想在济南市鑫盛大厦附近喝咖啡，给我推荐几个"}],"presence_penalty":0,"temperature":0,"top_p":0}'
@@ -308,7 +316,7 @@ curl 'http://<这里换成网关公网IP>/api/openai/v1/chat/completions' \
 **请求示例**
 
 ```shell
-curl 'http://<这里换成网关公网IP>/api/openai/v1/chat/completions' \
+curl 'http://<这里换成网关地址>/api/openai/v1/chat/completions' \
 -H 'Accept: application/json, text/event-stream' \
 -H 'Content-Type: application/json' \
 --data-raw '{"model":"qwen","frequency_penalty":0,"max_tokens":800,"stream":false,"messages":[{"role":"user","content":"济南市现在的天气情况如何？"}],"presence_penalty":0,"temperature":0,"top_p":0}'
@@ -323,7 +331,7 @@ curl 'http://<这里换成网关公网IP>/api/openai/v1/chat/completions' \
 **请求示例**
 
 ```shell
-curl 'http://<这里换成网关公网IP>/api/openai/v1/chat/completions' \
+curl 'http://<这里换成网关地址>/api/openai/v1/chat/completions' \
 -H 'Accept: application/json, text/event-stream' \
 -H 'Content-Type: application/json' \
 --data-raw '{"model":"qwen","frequency_penalty":0,"max_tokens":800,"stream":false,"messages":[{"role": "user","content": "济南的天气如何？"},{ "role": "assistant","content": "目前，济南市的天气为多云，气温为24℃，数据更新时间为2024年9月12日21时50分14秒。"},{"role": "user","content": "北京呢？"}],"presence_penalty":0,"temperature":0,"top_p":0}'
@@ -338,7 +346,7 @@ curl 'http://<这里换成网关公网IP>/api/openai/v1/chat/completions' \
 **请求示例**
 
 ```shell
-curl 'http://<这里换成网关公网IP>/api/openai/v1/chat/completions' \
+curl 'http://<这里换成网关地址>/api/openai/v1/chat/completions' \
 -H 'Accept: application/json, text/event-stream' \
 -H 'Content-Type: application/json' \
 --data-raw '{"model":"qwen","frequency_penalty":0,"max_tokens":800,"stream":false,"messages":[{"role":"user","content":"济南市现在的天气情况如何？用华氏度表示，用日语回答"}],"presence_penalty":0,"temperature":0,"top_p":0}'
@@ -353,7 +361,7 @@ curl 'http://<这里换成网关公网IP>/api/openai/v1/chat/completions' \
 **请求示例**
 
 ```shell
-curl 'http://<这里换成网关公网IP>/api/openai/v1/chat/completions' \
+curl 'http://<这里换成网关地址>/api/openai/v1/chat/completions' \
 -H 'Accept: application/json, text/event-stream' \
 -H 'Content-Type: application/json' \
 --data-raw '{"model":"qwen","frequency_penalty":0,"max_tokens":800,"stream":false,"messages":[{"role":"user","content":"帮我用德语翻译以下句子：九头蛇万岁!"}],"presence_penalty":0,"temperature":0,"top_p":0}'
@@ -364,3 +372,71 @@ curl 'http://<这里换成网关公网IP>/api/openai/v1/chat/completions' \
 ```json
 {"id":"65dcf12c-61ff-9e68-bffa-44fc9e6070d5","choices":[{"index":0,"message":{"role":"assistant","content":" “九头蛇万岁!”的德语翻译为“Hoch lebe Hydra!”。"},"finish_reason":"stop"}],"created":1724043865,"model":"qwen-max-0403","object":"chat.completion","usage":{"prompt_tokens":908,"completion_tokens":52,"total_tokens":960}}
 ```
+
+## 用法示例-开启 json 格式化
+
+**配置信息**
+在上述配置的基础上增加 jsonResp 配置
+```yaml
+jsonResp:
+  enable: true
+```
+
+**请求示例**
+
+```shell
+curl 'http://<这里换成网关地址>/api/openai/v1/chat/completions' \
+-H 'Accept: application/json, text/event-stream' \
+-H 'Content-Type: application/json' \
+--data-raw '{"model":"qwen","frequency_penalty":0,"max_tokens":800,"stream":false,"messages":[{"role":"user","content":"北京市现在的天气情况如何？"}],"presence_penalty":0,"temperature":0,"top_p":0}'
+```
+
+**响应示例**
+
+```json
+{"id":"ebd6ea91-8e38-9e14-9a5b-90178d2edea4","choices":[{"index":0,"message":{"role":"assistant","content": "{\"city\": \"北京市\", \"weather_condition\": \"多云\", \"temperature\": \"19℃\", \"data_update_time\": \"2024年10月9日16时37分53秒\"}"},"finish_reason":"stop"}],"created":1723187991,"model":"qwen-max-0403","object":"chat.completion","usage":{"prompt_tokens":890,"completion_tokens":56,"total_tokens":946}}
+```
+如果不自定义 json schema，大模型会自动生成一个 json 格式
+
+**配置信息**
+增加自定义 json schema 配置
+```yaml
+jsonResp:
+  enable: true
+  jsonSchema: |
+    title: WeatherSchema
+    type: object
+    properties:
+      location:
+        type: string
+        description: 城市名称.
+      weather:
+        type: string
+        description: 天气情况.
+      temperature:
+        type: string
+        description: 温度.
+      update_time:
+        type: string
+        description: 数据更新时间.
+    required:
+      - location
+      - weather
+      - temperature
+    additionalProperties: false
+```
+
+**请求示例**
+
+```shell
+curl 'http://<这里换成网关地址>/api/openai/v1/chat/completions' \
+-H 'Accept: application/json, text/event-stream' \
+-H 'Content-Type: application/json' \
+--data-raw '{"model":"qwen","frequency_penalty":0,"max_tokens":800,"stream":false,"messages":[{"role":"user","content":"北京市现在的天气情况如何？"}],"presence_penalty":0,"temperature":0,"top_p":0}'
+```
+
+**响应示例**
+
+```json
+{"id":"ebd6ea91-8e38-9e14-9a5b-90178d2edea4","choices":[{"index":0,"message":{"role":"assistant","content": "{\"location\": \"北京市\", \"weather\": \"多云\", \"temperature\": \"19℃\", \"update_time\": \"2024年10月9日16时37分53秒\"}"},"finish_reason":"stop"}],"created":1723187991,"model":"qwen-max-0403","object":"chat.completion","usage":{"prompt_tokens":890,"completion_tokens":56,"total_tokens":946}}
+```

plugins/wasm-go/extensions/ai-agent/README_EN.md

@@ -4,7 +4,7 @@ keywords: [ AI Gateway, AI Agent ]
 description: AI Agent plugin configuration reference
 ---
 ## Functional Description
-A customizable API AI Agent that supports configuring HTTP method types as GET and POST APIs. Supports multiple dialogue rounds, streaming and non-streaming modes.  
+A customizable API AI Agent that supports configuring HTTP method types as GET and POST APIs. Supports multiple dialogue rounds, streaming and non-streaming modes, support for formatting results as custom json.  
 The agent flow chart is as follows:  
 ![ai-agent](https://github.com/user-attachments/assets/b0761a0c-1afa-496c-a98e-bb9f38b340f8)  
 
@@ -20,6 +20,7 @@ Plugin execution priority: `200`
 | `llm`            | object    | Required    | -             | Configuration information for AI service provider  |
 | `apis`           | object    | Required    | -             | Configuration information for external API service provider  |
 | `promptTemplate` | object    | Optional    | -             | Configuration information for Agent ReAct template  |
+| `jsonResp`       | object    | Optional    | -             | Configuring json formatting information  |
 
 The configuration fields for `llm` are as follows:  
 | Name               | Data Type | Requirement | Default Value | Description                         |
@@ -71,7 +72,13 @@ The configuration fields for `chTemplate` and `enTemplate` are as follows:
 | `observation`   | string    | Optional    | -             | The observation part of the Agent ReAct template     |
 | `thought2`      | string    | Optional    | -             | The thought2 part of the Agent ReAct template       |
 
-## Usage Example
+The configuration fields for `jsonResp` are as follows:  
+| Name               | Data Type | Requirement | Default Value | Description                         |
+|--------------------|-----------|-------------|---------------|------------------------------------|
+| `enable`           | bool      | Optional    | -             | Whether to enable json formatting.  |
+| `jsonSchema`       | string    | Optional    | -             | Custom json schema               |
+
+## Usage Example-disable json formatting
 **Configuration Information**  
 ```yaml  
 llm:  
@@ -335,3 +342,68 @@ curl 'http://<replace with gateway public IP>/api/openai/v1/chat/completions' \
 {"id":"65dcf12c-61ff-9e68-bffa-44fc9e6070d5","choices":[{"index":0,"message":{"role":"assistant","content":" The German translation of \"Hail Hydra!\" is \"Hoch lebe Hydra!\"."},"finish_reason":"stop"}],"created":1724043865,"model":"qwen-max-0403","object":"chat.completion","usage":{"prompt_tokens":908,"completion_tokens":52,"total_tokens":960}}  
 ```  
 
+## Usage Example-enable json formatting
+**Configuration Information**  
+Add jsonResp configuration to the above configuration
+```yaml
+jsonResp:
+  enable: true
+```
+
+**Request Example**  
+```shell
+curl 'http://<replace with gateway public IP>/api/openai/v1/chat/completions' \
+-H 'Accept: application/json, text/event-stream' \
+-H 'Content-Type: application/json' \
+--data-raw '{"model":"qwen","frequency_penalty":0,"max_tokens":800,"stream":false,"messages":[{"role":"user","content":"What is the current weather in Beijing ?"}],"presence_penalty":0,"temperature":0,"top_p":0}'
+```
+
+**Response Example**
+
+```json
+{"id":"ebd6ea91-8e38-9e14-9a5b-90178d2edea4","choices":[{"index":0,"message":{"role":"assistant","content": "{\"city\": \"BeiJing\", \"weather_condition\": \"cloudy\", \"temperature\": \"19℃\", \"data_update_time\": \"Oct 9, 2024, at 16:37\"}"},"finish_reason":"stop"}],"created":1723187991,"model":"qwen-max-0403","object":"chat.completion","usage":{"prompt_tokens":890,"completion_tokens":56,"total_tokens":946}}
+```
+If you don't customise the json schema, the big model will automatically generate a json format
+
+**Configuration Information**
+Add custom json schema configuration
+```yaml
+jsonResp:
+  enable: true
+  jsonSchema:
+    title: WeatherSchema
+    type: object
+    properties:
+      location:
+        type: string
+        description: city name.
+      weather:
+        type: string
+        description: weather conditions.
+      temperature:
+        type: string
+        description: temperature.
+      update_time:
+        type: string
+        description: the update time of data.
+    required:
+      - location
+      - weather
+      - temperature
+    additionalProperties: false
+```
+
+**Request Example**
+
+```shell
+curl 'http://<replace with gateway public IP>/api/openai/v1/chat/completions' \
+-H 'Accept: application/json, text/event-stream' \
+-H 'Content-Type: application/json' \
+--data-raw '{"model":"qwen","frequency_penalty":0,"max_tokens":800,"stream":false,"messages":[{"role":"user","content":"What is the current weather in Beijing ?"}],"presence_penalty":0,"temperature":0,"top_p":0}'
+```
+
+**Response Example**
+
+```json
+{"id":"ebd6ea91-8e38-9e14-9a5b-90178d2edea4","choices":[{"index":0,"message":{"role":"assistant","content": "{\"location\": \"Beijing\", \"weather\": \"cloudy\", \"temperature\": \"19℃\", \"update_time\": \"Oct 9, 2024, at 16:37\"}"},"finish_reason":"stop"}],"created":1723187991,"model":"qwen-max-0403","object":"chat.completion","usage":{"prompt_tokens":890,"completion_tokens":56,"total_tokens":946}}
+```

plugins/wasm-go/extensions/ai-agent/config.go

@@ -211,6 +211,15 @@ type LLMInfo struct {
 	MaxTokens int64 `yaml:"maxToken" json:"maxTokens"`
 }
 
+type JsonResp struct {
+	// @Title zh-CN Enable
+	// @Description zh-CN 是否要启用json格式化输出
+	Enable bool `yaml:"enable" json:"enable"`
+	// @Title zh-CN Json Schema
+	// @Description zh-CN 用以验证响应json的Json Schema, 为空则只验证返回的响应是否为合法json
+	JsonSchema map[string]interface{} `required:"false" json:"jsonSchema" yaml:"jsonSchema"`
+}
+
 type PluginConfig struct {
 	// @Title zh-CN 返回 HTTP 响应的模版
 	// @Description zh-CN 用 %s 标记需要被 cache value 替换的部分
@@ -225,6 +234,7 @@ type PluginConfig struct {
 	LLMClient      wrapper.HttpClient `yaml:"-" json:"-"`
 	APIsParam      []APIsParam        `yaml:"-" json:"-"`
 	PromptTemplate PromptTemplate     `yaml:"promptTemplate" json:"promptTemplate"`
+	JsonResp       JsonResp           `yaml:"jsonResp" json:"jsonResp"`
 }
 
 func initResponsePromptTpl(gjson gjson.Result, c *PluginConfig) {
@@ -402,3 +412,15 @@ func initLLMClient(gjson gjson.Result, c *PluginConfig) {
 		Host: c.LLMInfo.Domain,
 	})
 }
+
+func initJsonResp(gjson gjson.Result, c *PluginConfig) {
+	c.JsonResp.Enable = false
+	if c.JsonResp.Enable = gjson.Get("jsonResp.enable").Bool(); c.JsonResp.Enable {
+		c.JsonResp.JsonSchema = nil
+		if jsonSchemaValue := gjson.Get("jsonResp.jsonSchema"); jsonSchemaValue.Exists() {
+			if schemaValue, ok := jsonSchemaValue.Value().(map[string]interface{}); ok {
+				c.JsonResp.JsonSchema = schemaValue
+			}
+		}
+	}
+}