Merge 8097610baf into e226645bc7

feat(search): add optional systemInstructions to API request body

docs/API/SEARCH.md

@@ -33,6 +33,7 @@ The API accepts a JSON object in the request body, where you define the focus mo
     ["human", "Hi, how are you?"],
     ["assistant", "I am doing well, how can I help you today?"]
   ],
+  "systemInstructions": "Focus on providing technical details about Perplexica's architecture.",
   "stream": false
 }
 ```
@@ -63,6 +64,8 @@ The API accepts a JSON object in the request body, where you define the focus mo
 
 - **`query`** (string, required): The search query or question.
 
+- **`systemInstructions`** (string, optional): Custom instructions provided by the user to guide the AI's response. These instructions are treated as user preferences and have lower priority than the system's core instructions. For example, you can specify a particular writing style, format, or focus area.
+
 - **`history`** (array, optional): An array of message pairs representing the conversation history. Each pair consists of a role (either 'human' or 'assistant') and the message content. This allows the system to use the context of the conversation to refine results. Example:
 
   ```json

sample.config.toml

@@ -22,5 +22,8 @@ MODEL_NAME = ""
 [MODELS.OLLAMA]
 API_URL = "" # Ollama API URL - http://host.docker.internal:11434
 
+[MODELS.DEEPSEEK]
+API_KEY = ""
+
 [API_ENDPOINTS]
 SEARXNG = "" # SearxNG API URL - http://localhost:32768

src/app/api/chat/route.ts

@@ -29,6 +29,7 @@ type Message = {
   messageId: string;
   chatId: string;
   content: string;
+  userSessionId: string;
 };
 
 type ChatModel = {
@@ -138,6 +139,7 @@ const handleHistorySave = async (
     where: eq(chats.id, message.chatId),
   });
 
+  let currentDate = new Date();
   if (!chat) {
     await db
       .insert(chats)
@@ -147,6 +149,8 @@ const handleHistorySave = async (
         createdAt: new Date().toString(),
         focusMode: focusMode,
         files: files.map(getFileDetails),
+        userSessionId: message.userSessionId,
+        timestamp: currentDate.toISOString(),
       })
       .execute();
   }

src/app/api/chats/route.ts

@@ -1,10 +1,47 @@
 import db from '@/lib/db';
+import { chats } from '@/lib/db/schema';
+import { eq, sql} from 'drizzle-orm';
 
 export const GET = async (req: Request) => {
   try {
-    let chats = await db.query.chats.findMany();
-    chats = chats.reverse();
-    return Response.json({ chats: chats }, { status: 200 });
+    // get header from request
+    const headers = await req.headers;
+    let userSessionId = headers.get('user-session-id')?.toString() ?? '';
+
+    if (userSessionId == '') {
+      return Response.json({ chats: {} }, { status: 200 });
+    }
+    
+    let chatsRes = await db.query.chats.findMany({
+      where: eq(chats.userSessionId, userSessionId),
+    });
+    
+    chatsRes = chatsRes.reverse();
+    // Keep only the latest 20 records in the database. Delete older records.
+    let maxRecordLimit = 20;
+    if (chatsRes.length > maxRecordLimit) {
+      const deleteChatsQuery = sql`DELETE FROM chats
+        WHERE userSessionId = ${userSessionId} AND (
+          timestamp IS NULL OR
+          timestamp NOT in (
+            SELECT timestamp FROM chats
+            WHERE userSessionId = ${userSessionId}
+            ORDER BY timestamp DESC
+            LIMIT ${maxRecordLimit}
+          )
+        )
+      `;
+      await db.run(deleteChatsQuery);
+      // Delete messages that no longer link with the chat from the database.
+      const deleteMessagesQuery = sql`DELETE FROM messages
+        WHERE chatId NOT IN (
+          SELECT id FROM chats
+        )
+      `;
+      await db.run(deleteMessagesQuery);
+    }
+
+    return Response.json({ chats: chatsRes }, { status: 200 });
   } catch (err) {
     console.error('Error in getting chats: ', err);
     return Response.json(

src/app/api/config/route.ts

@@ -7,6 +7,7 @@ import {
   getGroqApiKey,
   getOllamaApiEndpoint,
   getOpenaiApiKey,
+  getDeepseekApiKey,
   updateConfig,
 } from '@/lib/config';
 import {
@@ -53,6 +54,7 @@ export const GET = async (req: Request) => {
     config['anthropicApiKey'] = getAnthropicApiKey();
     config['groqApiKey'] = getGroqApiKey();
     config['geminiApiKey'] = getGeminiApiKey();
+    config['deepseekApiKey'] = getDeepseekApiKey();
     config['customOpenaiApiUrl'] = getCustomOpenaiApiUrl();
     config['customOpenaiApiKey'] = getCustomOpenaiApiKey();
     config['customOpenaiModelName'] = getCustomOpenaiModelName();
@@ -88,6 +90,9 @@ export const POST = async (req: Request) => {
         OLLAMA: {
           API_URL: config.ollamaApiUrl,
         },
+        DEEPSEEK: {
+          API_KEY: config.deepseekApiKey,
+        },
         CUSTOM_OPENAI: {
           API_URL: config.customOpenaiApiUrl,
           API_KEY: config.customOpenaiApiKey,

src/app/api/search/route.ts

@@ -34,6 +34,7 @@ interface ChatRequestBody {
   query: string;
   history: Array<[string, string]>;
   stream?: boolean;
+  systemInstructions?: string;
 }
 
 export const POST = async (req: Request) => {
@@ -125,7 +126,7 @@ export const POST = async (req: Request) => {
       embeddings,
       body.optimizationMode,
       [],
-      "",
+      body.systemInstructions || '',
     );
 
     if (!body.stream) {

src/app/library/page.tsx

@@ -1,5 +1,6 @@
 'use client';
 
+import crypto from 'crypto';
 import DeleteChat from '@/components/DeleteChat';
 import { cn, formatTimeDifference } from '@/lib/utils';
 import { BookOpenText, ClockIcon, Delete, ScanEye } from 'lucide-react';
@@ -21,10 +22,17 @@ const Page = () => {
     const fetchChats = async () => {
       setLoading(true);
 
+      let userSessionId = localStorage.getItem('userSessionId');
+      if (!userSessionId) {
+        userSessionId = crypto.randomBytes(20).toString('hex');
+        localStorage.setItem('userSessionId', userSessionId)
+      }
+
       const res = await fetch(`/api/chats`, {
         method: 'GET',
         headers: {
           'Content-Type': 'application/json',
+          'user-session-id': userSessionId!,
         },
       });
 

src/app/settings/page.tsx

@@ -20,6 +20,7 @@ interface SettingsType {
   anthropicApiKey: string;
   geminiApiKey: string;
   ollamaApiUrl: string;
+  deepseekApiKey: string;
   customOpenaiApiKey: string;
   customOpenaiApiUrl: string;
   customOpenaiModelName: string;
@@ -838,6 +839,25 @@ const Page = () => {
                     onSave={(value) => saveConfig('geminiApiKey', value)}
                   />
                 </div>
+
+                <div className="flex flex-col space-y-1">
+                  <p className="text-black/70 dark:text-white/70 text-sm">
+                    Deepseek API Key
+                  </p>
+                  <Input
+                    type="text"
+                    placeholder="Deepseek API Key"
+                    value={config.deepseekApiKey}
+                    isSaving={savingStates['deepseekApiKey']}
+                    onChange={(e) => {
+                      setConfig((prev) => ({
+                        ...prev!,
+                        deepseekApiKey: e.target.value,
+                      }));
+                    }}
+                    onSave={(value) => saveConfig('deepseekApiKey', value)}
+                  />
+                </div>
               </div>
             </SettingsSection>
           </div>

src/components/ChatWindow.tsx

@@ -95,6 +95,12 @@ const checkConfig = async (
       if (!embeddingModel || !embeddingModelProvider) {
         const embeddingModelProviders = providers.embeddingModelProviders;
 
+        let userSessionId = localStorage.getItem('userSessionId');
+        if (!userSessionId) {
+          userSessionId = crypto.randomBytes(20).toString('hex');
+          localStorage.setItem('userSessionId', userSessionId!)
+        }
+
         if (
           !embeddingModelProviders ||
           Object.keys(embeddingModelProviders).length === 0
@@ -342,6 +348,7 @@ const ChatWindow = ({ id }: { id?: string }) => {
     let added = false;
 
     messageId = messageId ?? crypto.randomBytes(7).toString('hex');
+    let userSessionId = localStorage.getItem('userSessionId');
 
     setMessages((prevMessages) => [
       ...prevMessages,
@@ -466,6 +473,7 @@ const ChatWindow = ({ id }: { id?: string }) => {
           messageId: messageId,
           chatId: chatId!,
           content: message,
+          userSessionId: userSessionId,
         },
         chatId: chatId!,
         files: fileIds,

src/components/MessageBox.tsx

@@ -48,6 +48,7 @@ const MessageBox = ({
   const [speechMessage, setSpeechMessage] = useState(message.content);
 
   useEffect(() => {
+    const citationRegex = /\[([^\]]+)\]/g;
     const regex = /\[(\d+)\]/g;
     let processedMessage = message.content;
 
@@ -67,11 +68,33 @@ const MessageBox = ({
     ) {
       setParsedMessage(
         processedMessage.replace(
-          regex,
-          (_, number) =>
-            `<a href="${
-              message.sources?.[number - 1]?.metadata?.url
-            }" target="_blank" className="bg-light-secondary dark:bg-dark-secondary px-1 rounded ml-1 no-underline text-xs text-black/70 dark:text-white/70 relative">${number}</a>`,
+          citationRegex,
+          (_, capturedContent: string) => {
+            const numbers = capturedContent
+              .split(',')
+              .map((numStr) => numStr.trim());
+
+            const linksHtml = numbers
+              .map((numStr) => {
+                const number = parseInt(numStr);
+
+                if (isNaN(number) || number <= 0) {
+                  return `[${numStr}]`;
+                }
+
+                const source = message.sources?.[number - 1];
+                const url = source?.metadata?.url;
+
+                if (url) {
+                  return `<a href="${url}" target="_blank" className="bg-light-secondary dark:bg-dark-secondary px-1 rounded ml-1 no-underline text-xs text-black/70 dark:text-white/70 relative">${numStr}</a>`;
+                } else {
+                  return `[${numStr}]`;
+                }
+              })
+              .join('');
+
+            return linksHtml;
+          },
         ),
       );
       return;

src/lib/config.ts

@@ -25,6 +25,9 @@ interface Config {
     OLLAMA: {
       API_URL: string;
     };
+    DEEPSEEK: {
+      API_KEY: string;
+    };
     CUSTOM_OPENAI: {
       API_URL: string;
       API_KEY: string;
@@ -63,6 +66,8 @@ export const getSearxngApiEndpoint = () =>
 
 export const getOllamaApiEndpoint = () => loadConfig().MODELS.OLLAMA.API_URL;
 
+export const getDeepseekApiKey = () => loadConfig().MODELS.DEEPSEEK.API_KEY;
+
 export const getCustomOpenaiApiKey = () =>
   loadConfig().MODELS.CUSTOM_OPENAI.API_KEY;
 

src/lib/db/schema.ts

@@ -25,4 +25,6 @@ export const chats = sqliteTable('chats', {
   files: text('files', { mode: 'json' })
     .$type<File[]>()
     .default(sql`'[]'`),
+  userSessionId: text('userSessionId'),
+  timestamp: text('timestamp'),
 });

src/lib/prompts/webSearch.ts

@@ -1,6 +1,6 @@
 export const webSearchRetrieverPrompt = `
 You are an AI question rephraser. You will be given a conversation and a follow-up question,  you will have to rephrase the follow up question so it is a standalone question and can be used by another LLM to search the web for information to answer it.
-If it is a smple writing task or a greeting (unless the greeting contains a question after it) like Hi, Hello, How are you, etc. than a question then you need to return \`not_needed\` as the response (This is because the LLM won't need to search the web for finding information on this topic).
+If it is a simple writing task or a greeting (unless the greeting contains a question after it) like Hi, Hello, How are you, etc. than a question then you need to return \`not_needed\` as the response (This is because the LLM won't need to search the web for finding information on this topic).
 If the user asks some question from some URL or wants you to summarize a PDF or a webpage (via URL) you need to return the links inside the \`links\` XML block and the question inside the \`question\` XML block. If the user wants to you to summarize the webpage or the PDF you need to return \`summarize\` inside the \`question\` XML block in place of a question and the link to summarize in the \`links\` XML block.
 You must always return the rephrased question inside the \`question\` XML block, if there are no links in the follow-up question then don't insert a \`links\` XML block in your response.
 

src/lib/providers/deepseek.ts

@@ -0,0 +1,44 @@
+import { ChatOpenAI } from '@langchain/openai';
+import { getDeepseekApiKey } from '../config';
+import { ChatModel } from '.';
+import { BaseChatModel } from '@langchain/core/language_models/chat_models';
+
+const deepseekChatModels: Record<string, string>[] = [
+  {
+    displayName: 'Deepseek Chat (Deepseek V3)',
+    key: 'deepseek-chat',
+  },
+  {
+    displayName: 'Deepseek Reasoner (Deepseek R1)',
+    key: 'deepseek-reasoner',
+  },
+];
+
+export const loadDeepseekChatModels = async () => {
+  const deepseekApiKey = getDeepseekApiKey();
+
+  if (!deepseekApiKey) return {};
+
+  try {
+    const chatModels: Record<string, ChatModel> = {};
+
+    deepseekChatModels.forEach((model) => {
+      chatModels[model.key] = {
+        displayName: model.displayName,
+        model: new ChatOpenAI({
+          openAIApiKey: deepseekApiKey,
+          modelName: model.key,
+          temperature: 0.7,
+          configuration: {
+            baseURL: 'https://api.deepseek.com',
+          },
+        }) as unknown as BaseChatModel,
+      };
+    });
+
+    return chatModels;
+  } catch (err) {
+    console.error(`Error loading Deepseek models: ${err}`);
+    return {};
+  }
+};

src/lib/providers/index.ts

@@ -12,6 +12,7 @@ import { loadGroqChatModels } from './groq';
 import { loadAnthropicChatModels } from './anthropic';
 import { loadGeminiChatModels, loadGeminiEmbeddingModels } from './gemini';
 import { loadTransformersEmbeddingsModels } from './transformers';
+import { loadDeepseekChatModels } from './deepseek';
 
 export interface ChatModel {
   displayName: string;
@@ -32,6 +33,7 @@ export const chatModelProviders: Record<
   groq: loadGroqChatModels,
   anthropic: loadAnthropicChatModels,
   gemini: loadGeminiChatModels,
+  deepseek: loadDeepseekChatModels,
 };
 
 export const embeddingModelProviders: Record<