[前端+后端+RAG] 检索范围切换(当前文件/整个知识库)；联网搜索功能(SearXNG)；搜索结果带网络链接；修复RAG检索source格式不匹配bug

chat_web_backend/src/main/java/com/inspur/llm/chat/gpt/controller/gpt/SmartChatController.java

@@ -247,6 +247,7 @@ public class SmartChatController extends BaseController {
         }
 
         talkDto.setKnowledgeBaseNameList(knowledgeBaseNameList);
+        talkDto.setWebSearch(smartChatQueryDto.getWebSearch());
 
         talkDto.setQuery(chatMessages.getContent().replaceAll("\n", ""));
         talkDto.setStream(true);

chat_web_backend/src/main/java/com/inspur/llm/chat/gpt/pojo/dto/SmartChatQueryDto.java

@@ -35,4 +35,7 @@ public class SmartChatQueryDto {
     /* 用户请求类型 **/
     private Integer chatType;
 
+    /* 是否开启联网搜索 **/
+    private Boolean webSearch;
+
 }

chat_web_backend/src/main/java/com/inspur/llm/chat/gpt/pojo/dto/SmartChatSelfDto.java

@@ -46,6 +46,10 @@ public class SmartChatSelfDto {
     /* 用户请求类型 **/
     private Integer chatType;
 
+    /* 是否开启联网搜索 **/
+    @JsonProperty("web_search")
+    private Boolean webSearch;
+
     public String toJsonString() {
         StringBuffer str = new StringBuffer();
         extracted(knowledgeBaseNameList, str);
@@ -63,7 +67,7 @@ public class SmartChatSelfDto {
                 ", \"fileNames\":" + fileNameList +
                 ", \"quote\":\"" + ReplaceUtils.replaceHiddenChars(quote) + '\"' +
                 ", \"prompt_name\":\"" + promptName + '\"' +
-//                ", \"use_model_self_response\":\"" + "False" + '\"' +
+                ", \"web_search\":" + (webSearch != null && webSearch ? "true" : "false") +
                 '}';
     }
 

chat_web_front/src/components/ReadingCreate.vue

@@ -9,16 +9,27 @@
     <!-- 文字窗口-->
     <div>
       <div class="tool-bar">
-        <div class="label">
+        <div class="label"></div>
-          <!--          <img src="../assets/images/writing/start.png">
-                    <div>AI写作助手</div>-->
-        </div>
         <div class="clean" @click="cleanChat">
           <img src="../assets/images/writing/brush.png">
           <div>清除对话</div>
         </div>
       </div>
 
+      <div class="search-scope" v-if="selectedFile">
+        <div class="scope-label">检索范围：</div>
+        <div class="scope-option" :class="{ active: searchScope === 'file' }" @click="searchScope = 'file'"
+             :title="'仅在「' + selectedFile.fileName + '」中检索'">
+          <span class="scope-icon">📄</span>
+          <span class="scope-name">{{ selectedFile.fileName }}</span>
+        </div>
+        <div class="scope-option" :class="{ active: searchScope === 'kb' }" @click="searchScope = 'kb'"
+             :title="'在「' + selectedFile.folderName + '」知识库的所有文件中检索'">
+          <span class="scope-icon">📁</span>
+          <span class="scope-name">{{ selectedFile.folderName }}</span>
+        </div>
+      </div>
+
       <div class="text-box">
         <div class="quote-box" v-if="quoteMsg">
           <div class="vertical-line"></div>
@@ -34,10 +45,17 @@
             @input="handleInput"
             @keydown.enter="keyDown"
             placeholder="请输入你想提的问题，字数不能超过1000字"/>
-        <div>
+        <div class="text-box-bottom">
-          <img v-if="textarea&&!sendStatus" style="width: 38px" src="../assets/images/writing/send-blue.png" @click="send('','0')">
+          <div class="web-search-toggle" :class="{ active: webSearchEnabled }" @click="webSearchEnabled = !webSearchEnabled"
-          <img v-if="!textarea&&!sendStatus" src="../assets/images/writing/send-gray.png">
+               :title="webSearchEnabled ? '联网搜索已开启，点击关闭' : '开启联网搜索，从互联网获取最新信息'">
-          <img v-if="sendStatus" src="../assets/images/chat/stopChat.png" @click="handleStop"></img>
+            <span class="ws-icon">🌐</span>
+            <span class="ws-text">联网搜索</span>
+          </div>
+          <div class="send-btn">
+            <img v-if="textarea&&!sendStatus" style="width: 38px" src="../assets/images/writing/send-blue.png" @click="send('','0')">
+            <img v-if="!textarea&&!sendStatus" src="../assets/images/writing/send-gray.png">
+            <img v-if="sendStatus" src="../assets/images/chat/stopChat.png" @click="handleStop">
+          </div>
         </div>
       </div>
 
@@ -76,6 +94,8 @@ const clearQuote = () => {
 }
 
 //const title = inject('aiboxTitle');
+const searchScope = ref<'file' | 'kb'>('file');
+const webSearchEnabled = ref(false);
 const textarea = ref("");
 const firstChat = ref(true);
 const sendStatus = ref(false);
@@ -198,12 +218,13 @@ const getFetchChatAPIProcess = async (type: string) => {
         headers: headers,
         signal: controller.signal,
         body: JSON.stringify({
-          fileNames: [selectedFile.value?.embeddingId],
+          fileNames: searchScope.value === 'file' ? [selectedFile.value?.embeddingId] : [],
           conversationId: conversationId.value,
           promptName: "default",
           knowledgeBaseIdList: [selectedFile.value?.folderId],
           chatType: type,
-          quote: quoteMsg.value
+          quote: quoteMsg.value,
+          webSearch: webSearchEnabled.value
         }),
       }
   );
@@ -372,6 +393,7 @@ const loadChatHistory = async () => {
 
 // 监听文件切换，重新加载对话历史
 watch(() => selectedFile.value?.fileId, () => {
+  searchScope.value = 'file';
   loadChatHistory();
 });
 
@@ -434,7 +456,7 @@ const handleStop = async () => {
 
 <style lang="less" scoped>
 .message-content {
-  height: calc(100% - 290px);
+  height: calc(100% - 320px);
   overflow-y: auto;
   padding: 20px;
 
@@ -505,28 +527,93 @@ const handleStop = async () => {
   }
 }
 
+.search-scope {
+  display: flex;
+  align-items: center;
+  padding: 4px 12px;
+  gap: 6px;
+
+  .scope-label {
+    font-size: 13px;
+    color: #333;
+    flex-shrink: 0;
+  }
+
+  .scope-option {
+    display: flex;
+    align-items: center;
+    gap: 5px;
+    padding: 6px 14px;
+    border-radius: 14px;
+    border: 1px solid #E0E0E0;
+    cursor: pointer;
+    font-size: 13px;
+    color: #666;
+    transition: all 0.2s;
+    max-width: 45%;
+    overflow: hidden;
+
+    &:hover { border-color: #004EA0; color: #004EA0; }
+    &.active { border-color: #004EA0; color: #fff; background: #004EA0; }
+
+    .scope-icon { font-size: 14px; flex-shrink: 0; }
+    .scope-name { overflow: hidden; text-overflow: ellipsis; white-space: nowrap; }
+  }
+}
+
 .text-box {
-  //width: 100%;
   height: 190px;
   background: #FFFFFF;
   border-radius: 8px;
   border: 1px solid #D5DDFF;
   margin: 12px 20px 12px 20px;
+  display: flex;
+  flex-direction: column;
+
   .box-textarea {
     outline: none;
     border: none;
     resize: none;
     width: 100%;
-    height: calc(100% - 54px);
+    flex: 1;
     padding: 16px;
     line-height: 24px;
     border-radius: 8px;
   }
 
+  .text-box-bottom {
+    display: flex;
+    justify-content: space-between;
+    align-items: center;
+    padding: 6px 12px;
+
+    .web-search-toggle {
+      display: flex;
+      align-items: center;
+      gap: 5px;
+      padding: 6px 14px;
+      border-radius: 14px;
+      border: 1px solid #E0E0E0;
+      cursor: pointer;
+      font-size: 13px;
+      color: #999;
+      transition: all 0.2s;
+      user-select: none;
+
+      &:hover { border-color: #10a37f; color: #10a37f; }
+      &.active { border-color: #10a37f; color: #fff; background: #10a37f; }
+
+      .ws-icon { font-size: 14px; }
+      .ws-text { font-size: 13px; }
+    }
+
+    .send-btn {
+      img { width: 38px; cursor: pointer; }
+    }
+  }
+
   img {
-    cursor:pointer;
+    cursor: pointer;
-    float: right;
-    margin-right: 16px;
   }
 
   .quote-box {

langchain-chat/server/chat/ZhipuSearchAPI.py

@@ -71,10 +71,7 @@ class ZhipuSearchAPIWrapper:
 
             logging.info(f"Zhipu检索内容:{search_query}")
             url = "http://ywk3hvt4d:01Jp2V1tR9PdTsYSz919779Rb9_@134.122.191.214/search"
-            if "天气" in search_query:
+            engines = "duckduckgo,bing"
-                engines = "google"
-            else:
-                engines = "baidu"
             data = {
                 "format":"json",
                 "q":search_query,

langchain-chat/server/chat/self_kb_chat.py

@@ -50,6 +50,7 @@ async def self_kb_chat(
                             "content": "虎头虎脑"}]]
                     ),
                     stream: bool = Body(True, description="流式输出"),
+                    web_search: bool = Body(False, description="是否开启联网搜索"),
                     ):
     """
         个人知识库对话api\n
@@ -149,7 +150,29 @@ async def self_kb_chat(
         except Exception as e:
             logger.error(f"个人知识库问答路由错误: {self_kb_route}", exc_info=True)
             docs = []
-        logger.info(f"个人知识库问答source_documents: {docs}")
+        logger.info(f"个人知识库问答source_documents: {len(docs)}条")
+
+        # 联网搜索
+        web_search_context = ""
+        web_search_results = []  # 保存搜索结果供后面引用
+        if web_search:
+            try:
+                from server.chat.ZhipuSearchAPI import ZhipuSearchAPIWrapper
+                searcher = ZhipuSearchAPIWrapper()
+                web_results = searcher.zhipu_search(search_query)
+                web_search_results = web_results[:5] if web_results else []
+                if web_results:
+                    web_parts = []
+                    for i, r in enumerate(web_results[:5], 1):
+                        title = r.get("title", "")
+                        content = r.get("content", "")[:300]
+                        url = r.get("url", "")
+                        web_parts.append(f"[{i}] {title}\n{content}\n来源: {url}")
+                    web_search_context = "\n\n【联网搜索结果】\n" + "\n\n".join(web_parts)
+                    logger.info(f"联网搜索获取到 {len(web_results)} 条结果")
+            except Exception as e:
+                logger.error(f"联网搜索失败: {e}")
+
         # if SELF_USE_RERANKER:
         #     reranker_model_path = MODEL_PATH["reranker"].get(RERANKER_MODEL,"BAAI/bge-reranker-large")
         #     print("-----------------model path------------------")
@@ -184,25 +207,28 @@ async def self_kb_chat(
         if '0' in self_kb_route:
             context = "\n".join([doc.page_content for doc in docs]).strip("xa0")
             logger.info(f"个人知识库问答 context 长度：{len(context)}")
-            # context_70 = context if len(context)<30000 else TextRank(context,num_sentences=70)
+            context = context[:30000] if len(context)>30000 else context
-            context = context[:40000] if len(context)>40000 else context
+            if web_search_context:
-            logger.info(f"截取后个人知识库问答 context 长度：{len(context)}")
+                context += web_search_context
+            logger.info(f"最终 context 长度：{len(context)}")
             if history:
                 history = history if len(history) < 20000 else TextRank(history,num_sentences=1)
-            # logger.info(f"个人知识库问答 context 长度超过 30000，使用 TextRank 算法进行降维得到 context 长度：{len(context)}")
             chain = LLMChain(prompt=chat_prompt, llm=model1, verbose=True)
             task = asyncio.create_task(wrap_done(
                 chain.acall({"context": context, "question": query, "history": history, "quote": quote, "fileName":fileNames}),
                 callback.done),
             )
         elif '1' in self_kb_route:
+            # 联网搜索结果作为额外文档加入
+            if web_search_context:
+                from langchain.docstore.document import Document as LCDocument
+                docs.append(LCDocument(page_content=web_search_context, metadata={"source": "web_search"}))
             chain = load_qa_chain(
-                model, 
+                model,
-                chain_type="stuff", 
+                chain_type="stuff",
-                prompt=chat_prompt, 
+                prompt=chat_prompt,
                 verbose=True
             )
-            # Begin a task that runs in the background.
             task = asyncio.create_task(wrap_done(
                 chain.acall({"input_documents": docs, "question": query, "history": history, "quote": quote, "fileName":fileNames}),
                 callback.done),
@@ -235,14 +261,18 @@ async def self_kb_chat(
             yield json.dumps(response, ensure_ascii=False)
         await task
         source_documents = []
-        if len(docs) == 0:  # 没有找到相关文档
+        if len(docs) == 0 and not web_search_context:
             source_documents.append(f"""暂未从本篇文献中找到答案,该回答为大模型自身能力解答！""")
         else:
-            # 去除文件扩展名
+            if len(docs) > 0:
-            # fileNames_without_ext = [name.rsplit('.', 1)[0] for name in fileNames]
+                source_documents.append(f"""[{len(source_documents) + 1}] [{docs[0].metadata.get("source")}]()\n""")
-            # 连接文件名（如果有多个文件名）
+            # 联网搜索结果链接
-            # joined_fileNames = ', '.join(fileNames_without_ext)
+            if web_search_results:
-            source_documents.append(f"""[{len(source_documents) + 1}] [{docs[0].metadata.get("source")}]()\n""")    
+                for r in web_search_results:
+                    title = r.get("title", "").replace("\n", "")
+                    url = r.get("url", "")
+                    if title and url:
+                        source_documents.append(f"""[{len(source_documents) + 1}] [{title}]({url})\n""")
         yield json.dumps({"docs": source_documents}, ensure_ascii=False) 
 
     return EventSourceResponse(knowledge_base_chat_iterator(query))