diff --git a/client/agent.go b/client/agent.go index 5260b381..91f8f5b1 100644 --- a/client/agent.go +++ b/client/agent.go @@ -150,7 +150,11 @@ func (as *AgentSession) Ask(ctx context.Context, query string, callback AgentEve } // AskWithRequest sends a customized agent request for this session. -func (as *AgentSession) AskWithRequest(ctx context.Context, request *AgentQARequest, callback AgentEventCallback) error { +func (as *AgentSession) AskWithRequest( + ctx context.Context, + request *AgentQARequest, + callback AgentEventCallback, +) error { return as.client.AgentQAStreamWithRequest(ctx, as.sessionID, request, callback) } diff --git a/client/cmd/agent_test/main.go b/client/cmd/agent_test/main.go index 6e89457f..ddf0d72d 100644 --- a/client/cmd/agent_test/main.go +++ b/client/cmd/agent_test/main.go @@ -250,7 +250,12 @@ func (cli *CLI) askQuestion(query string) { case client.AgentResponseTypeReferences: if resp.KnowledgeReferences != nil { references = append(references, resp.KnowledgeReferences...) - fmt.Printf("%s📚 Knowledge References:%s Found %d reference(s)\n", ColorCyan, ColorReset, len(resp.KnowledgeReferences)) + fmt.Printf( + "%s📚 Knowledge References:%s Found %d reference(s)\n", + ColorCyan, + ColorReset, + len(resp.KnowledgeReferences), + ) for i, ref := range resp.KnowledgeReferences { fmt.Printf(" %d. [Score: %.3f] %s\n", i+1, ref.Score, truncateString(ref.Content, 80)) fmt.Printf(" Knowledge: %s (Chunk: %d)\n", ref.KnowledgeTitle, ref.ChunkIndex) diff --git a/client/knowledge.go b/client/knowledge.go index 3a3ea9dc..29edad97 100644 --- a/client/knowledge.go +++ b/client/knowledge.go @@ -172,7 +172,12 @@ func (c *Client) CreateKnowledgeFromFile(ctx context.Context, } // CreateKnowledgeFromURL creates a knowledge entry from a web URL -func (c *Client) CreateKnowledgeFromURL(ctx context.Context, knowledgeBaseID string, url string, enableMultimodel *bool) (*Knowledge, error) { +func (c *Client) CreateKnowledgeFromURL( + ctx context.Context, + knowledgeBaseID string, + url string, + enableMultimodel *bool, +) (*Knowledge, error) { path := fmt.Sprintf("/api/v1/knowledge-bases/%s/knowledge/url", knowledgeBaseID) reqBody := struct { diff --git a/client/message.go b/client/message.go index 28ae55ac..b59d691f 100644 --- a/client/message.go +++ b/client/message.go @@ -19,7 +19,7 @@ type Message struct { RequestID string `json:"request_id"` Content string `json:"content"` Role string `json:"role"` - KnowledgeReferences []*SearchResult `json:"knowledge_references" ` + KnowledgeReferences []*SearchResult `json:"knowledge_references"` IsCompleted bool `json:"is_completed"` CreatedAt time.Time `json:"created_at"` UpdatedAt time.Time `json:"updated_at"` @@ -32,7 +32,12 @@ type MessageListResponse struct { } // LoadMessages loads session messages, supports pagination and time filtering -func (c *Client) LoadMessages(ctx context.Context, sessionID string, limit int, beforeTime *time.Time) ([]Message, error) { +func (c *Client) LoadMessages( + ctx context.Context, + sessionID string, + limit int, + beforeTime *time.Time, +) ([]Message, error) { path := fmt.Sprintf("/api/v1/messages/%s/load", sessionID) queryParams := url.Values{} @@ -61,7 +66,12 @@ func (c *Client) GetRecentMessages(ctx context.Context, sessionID string, limit } // GetMessagesBefore gets messages before a specified time -func (c *Client) GetMessagesBefore(ctx context.Context, sessionID string, beforeTime time.Time, limit int) ([]Message, error) { +func (c *Client) GetMessagesBefore( + ctx context.Context, + sessionID string, + beforeTime time.Time, + limit int, +) ([]Message, error) { return c.LoadMessages(ctx, sessionID, limit, &beforeTime) } diff --git a/client/session.go b/client/session.go index 4408547d..b89c49b1 100644 --- a/client/session.go +++ b/client/session.go @@ -236,7 +236,12 @@ type StreamResponse struct { } // KnowledgeQAStream knowledge Q&A streaming API -func (c *Client) KnowledgeQAStream(ctx context.Context, sessionID string, query string, callback func(*StreamResponse) error) error { +func (c *Client) KnowledgeQAStream( + ctx context.Context, + sessionID string, + query string, + callback func(*StreamResponse) error, +) error { path := fmt.Sprintf("/api/v1/knowledge-chat/%s", sessionID) fmt.Printf("Starting KnowledgeQAStream request, session ID: %s, query: %s\n", sessionID, query) @@ -315,7 +320,12 @@ func (c *Client) KnowledgeQAStream(ctx context.Context, sessionID string, query } // ContinueStream continues to receive an active stream for a session -func (c *Client) ContinueStream(ctx context.Context, sessionID string, messageID string, callback func(*StreamResponse) error) error { +func (c *Client) ContinueStream( + ctx context.Context, + sessionID string, + messageID string, + callback func(*StreamResponse) error, +) error { path := fmt.Sprintf("/api/v1/sessions/continue-stream/%s", sessionID) queryParams := url.Values{} diff --git a/client/tenant.go b/client/tenant.go index 4f80cbcd..1e22327e 100644 --- a/client/tenant.go +++ b/client/tenant.go @@ -24,23 +24,23 @@ type RetrieverEngineParams struct { // Tenant represents tenant information in the system type Tenant struct { - ID uint `yaml:"id" json:"id" gorm:"primaryKey"` + ID uint `yaml:"id" json:"id" gorm:"primaryKey"` // Tenant name - Name string `yaml:"name" json:"name"` + Name string `yaml:"name" json:"name"` // Tenant description - Description string `yaml:"description" json:"description"` + Description string `yaml:"description" json:"description"` // API key for authentication - APIKey string `yaml:"api_key" json:"api_key"` + APIKey string `yaml:"api_key" json:"api_key"` // Tenant status (active, inactive) - Status string `yaml:"status" json:"status" gorm:"default:'active'"` + Status string `yaml:"status" json:"status" gorm:"default:'active'"` // Configured retrieval engines RetrieverEngines RetrieverEngines `yaml:"retriever_engines" json:"retriever_engines" gorm:"type:json"` // Business/department information - Business string `yaml:"business" json:"business"` + Business string `yaml:"business" json:"business"` // Creation timestamp - CreatedAt time.Time `yaml:"created_at" json:"created_at"` + CreatedAt time.Time `yaml:"created_at" json:"created_at"` // Last update timestamp - UpdatedAt time.Time `yaml:"updated_at" json:"updated_at"` + UpdatedAt time.Time `yaml:"updated_at" json:"updated_at"` } // TenantResponse represents the API response structure for tenant operations diff --git a/internal/agent/engine.go b/internal/agent/engine.go index 023d4d4e..c6480879 100644 --- a/internal/agent/engine.go +++ b/internal/agent/engine.go @@ -71,7 +71,11 @@ func NewAgentEngine( // Execute executes the agent with conversation history and streaming output // All events are emitted to EventBus and handled by subscribers (like Handler layer) -func (e *AgentEngine) Execute(ctx context.Context, sessionID, messageID, query string, llmContext []chat.Message) (*types.AgentState, error) { +func (e *AgentEngine) Execute( + ctx context.Context, + sessionID, messageID, query string, + llmContext []chat.Message, +) (*types.AgentState, error) { logger.Infof(ctx, "========== Agent Execution Started ==========") logger.Infof(ctx, "[Agent] SessionID: %s, MessageID: %s", sessionID, messageID) logger.Infof(ctx, "[Agent] User Query: %s", query) @@ -92,7 +96,11 @@ func (e *AgentEngine) Execute(ctx context.Context, sessionID, messageID, query s } // Build system prompt using progressive RAG prompt - systemPrompt := BuildProgressiveRAGSystemPrompt(e.knowledgeBasesInfo, e.config.WebSearchEnabled, e.systemPromptTemplate) + systemPrompt := BuildProgressiveRAGSystemPrompt( + e.knowledgeBasesInfo, + e.config.WebSearchEnabled, + e.systemPromptTemplate, + ) logger.Debugf(ctx, "[Agent] SystemPrompt Length: %d characters", len(systemPrompt)) logger.Debugf(ctx, "[Agent] SystemPrompt (stream)\n----\n%s\n----", systemPrompt) @@ -193,7 +201,12 @@ func (e *AgentEngine) executeLoop( // Debug: log finish reason and tool call count from LLM logger.Infof(ctx, "[Agent][Round-%d] LLM response received: finish_reason=%s, tool_calls=%d, content_length=%d", state.CurrentRound+1, response.FinishReason, len(response.ToolCalls), len(response.Content)) - logger.Debugf(ctx, "[Agent] LLM response finish=%s, toolCalls=%d", response.FinishReason, len(response.ToolCalls)) + logger.Debugf( + ctx, + "[Agent] LLM response finish=%s, toolCalls=%d", + response.FinishReason, + len(response.ToolCalls), + ) if response.Content != "" { logger.Debugf(ctx, "[Agent][Round-%d] LLM thought content:\n%s", state.CurrentRound+1, response.Content) } @@ -229,13 +242,23 @@ func (e *AgentEngine) executeLoop( Done: true, }, }) - logger.Infof(ctx, "[Agent][Round-%d] Duration: %dms", state.CurrentRound+1, time.Since(roundStart).Milliseconds()) + logger.Infof( + ctx, + "[Agent][Round-%d] Duration: %dms", + state.CurrentRound+1, + time.Since(roundStart).Milliseconds(), + ) break } // 3. Act: Execute tool calls if any if len(response.ToolCalls) > 0 { - logger.Infof(ctx, "[Agent][Round-%d] Executing %d tool calls...", state.CurrentRound+1, len(response.ToolCalls)) + logger.Infof( + ctx, + "[Agent][Round-%d] Executing %d tool calls...", + state.CurrentRound+1, + len(response.ToolCalls), + ) for i, tc := range response.ToolCalls { logger.Infof(ctx, "[Agent][Round-%d][Tool-%d/%d] Tool: %s, ID: %s", @@ -483,7 +506,11 @@ func (e *AgentEngine) buildToolsForLLM() []chat.Tool { // appendToolResults adds tool results to the message history following OpenAI's tool calling format // Also writes these messages to the context manager for persistence -func (e *AgentEngine) appendToolResults(ctx context.Context, messages []chat.Message, step types.AgentStep) []chat.Message { +func (e *AgentEngine) appendToolResults( + ctx context.Context, + messages []chat.Message, + step types.AgentStep, +) []chat.Message { // Add assistant message with tool calls (if any) if step.Thought != "" || len(step.ToolCalls) > 0 { assistantMsg := chat.Message{ @@ -744,7 +771,11 @@ func (e *AgentEngine) streamFinalAnswerToEventBus( }) // Build messages with all context - systemPrompt := BuildProgressiveRAGSystemPrompt(e.knowledgeBasesInfo, e.config.WebSearchEnabled, e.systemPromptTemplate) + systemPrompt := BuildProgressiveRAGSystemPrompt( + e.knowledgeBasesInfo, + e.config.WebSearchEnabled, + e.systemPromptTemplate, + ) messages := []chat.Message{ {Role: "system", Content: systemPrompt}, @@ -837,7 +868,10 @@ func countTotalToolCalls(steps []types.AgentStep) int { } // buildMessagesWithLLMContext builds the message array with LLM context -func (e *AgentEngine) buildMessagesWithLLMContext(systemPrompt, currentQuery string, llmContext []chat.Message) []chat.Message { +func (e *AgentEngine) buildMessagesWithLLMContext( + systemPrompt, currentQuery string, + llmContext []chat.Message, +) []chat.Message { messages := []chat.Message{ {Role: "system", Content: systemPrompt}, } diff --git a/internal/agent/prompts.go b/internal/agent/prompts.go index 917c1c5b..2f0bb9b2 100644 --- a/internal/agent/prompts.go +++ b/internal/agent/prompts.go @@ -189,7 +189,12 @@ func renderPromptPlaceholders(template string, knowledgeBases []*KnowledgeBaseIn // - {{knowledge_bases}} // - {{web_search_status}} -> "Enabled" or "Disabled" // - {{current_time}} -> current time string -func renderPromptPlaceholdersWithStatus(template string, knowledgeBases []*KnowledgeBaseInfo, webSearchEnabled bool, currentTime string) string { +func renderPromptPlaceholdersWithStatus( + template string, + knowledgeBases []*KnowledgeBaseInfo, + webSearchEnabled bool, + currentTime string, +) string { result := renderPromptPlaceholders(template, knowledgeBases) status := "Disabled" if webSearchEnabled { @@ -205,7 +210,10 @@ func renderPromptPlaceholdersWithStatus(template string, knowledgeBases []*Knowl } // BuildProgressiveRAGSystemPromptWithWeb builds the progressive RAG system prompt with web search enabled -func BuildProgressiveRAGSystemPromptWithWeb(knowledgeBases []*KnowledgeBaseInfo, systemPromptTemplate ...string) string { +func BuildProgressiveRAGSystemPromptWithWeb( + knowledgeBases []*KnowledgeBaseInfo, + systemPromptTemplate ...string, +) string { var template string if len(systemPromptTemplate) > 0 && systemPromptTemplate[0] != "" { template = systemPromptTemplate[0] @@ -217,7 +225,10 @@ func BuildProgressiveRAGSystemPromptWithWeb(knowledgeBases []*KnowledgeBaseInfo, } // BuildProgressiveRAGSystemPromptWithoutWeb builds the progressive RAG system prompt without web search -func BuildProgressiveRAGSystemPromptWithoutWeb(knowledgeBases []*KnowledgeBaseInfo, systemPromptTemplate ...string) string { +func BuildProgressiveRAGSystemPromptWithoutWeb( + knowledgeBases []*KnowledgeBaseInfo, + systemPromptTemplate ...string, +) string { var template string if len(systemPromptTemplate) > 0 && systemPromptTemplate[0] != "" { template = systemPromptTemplate[0] @@ -230,7 +241,11 @@ func BuildProgressiveRAGSystemPromptWithoutWeb(knowledgeBases []*KnowledgeBaseIn // BuildProgressiveRAGSystemPrompt builds the progressive RAG system prompt based on web search status // This is the main function to use - it automatically selects the appropriate version -func BuildProgressiveRAGSystemPrompt(knowledgeBases []*KnowledgeBaseInfo, webSearchEnabled bool, systemPromptTemplate ...string) string { +func BuildProgressiveRAGSystemPrompt( + knowledgeBases []*KnowledgeBaseInfo, + webSearchEnabled bool, + systemPromptTemplate ...string, +) string { if webSearchEnabled { return BuildProgressiveRAGSystemPromptWithWeb(knowledgeBases, systemPromptTemplate...) } diff --git a/internal/agent/tools/database_query.go b/internal/agent/tools/database_query.go index 464d8c16..41f27fb4 100644 --- a/internal/agent/tools/database_query.go +++ b/internal/agent/tools/database_query.go @@ -373,7 +373,11 @@ func (t *DatabaseQueryTool) validateAndSecureSQL(sqlQuery string) (string, error } // formatQueryResults formats query results into readable text -func (t *DatabaseQueryTool) formatQueryResults(columns []string, results []map[string]interface{}, query string) string { +func (t *DatabaseQueryTool) formatQueryResults( + columns []string, + results []map[string]interface{}, + query string, +) string { output := "=== 查询结果 ===\n\n" output += fmt.Sprintf("执行的SQL: %s\n\n", query) output += fmt.Sprintf("返回 %d 行数据\n\n", len(results)) diff --git a/internal/agent/tools/get_document_info.go b/internal/agent/tools/get_document_info.go index 0ba069c6..da8b4c94 100644 --- a/internal/agent/tools/get_document_info.go +++ b/internal/agent/tools/get_document_info.go @@ -133,10 +133,11 @@ func (t *GetDocumentInfoTool) Execute(ctx context.Context, args map[string]inter } // Get chunk count - _, total, err := t.chunkService.GetRepository().ListPagedChunksByKnowledgeID(ctx, t.tenantID, id, &types.Pagination{ - Page: 1, - PageSize: 1000, - }, []types.ChunkType{"text"}, "") + _, total, err := t.chunkService.GetRepository(). + ListPagedChunksByKnowledgeID(ctx, t.tenantID, id, &types.Pagination{ + Page: 1, + PageSize: 1000, + }, []types.ChunkType{"text"}, "") if err != nil { mu.Lock() results[id] = &docInfo{ diff --git a/internal/agent/tools/grep_chunks.go b/internal/agent/tools/grep_chunks.go index 8bcf3f5b..f23a8c12 100644 --- a/internal/agent/tools/grep_chunks.go +++ b/internal/agent/tools/grep_chunks.go @@ -203,7 +203,12 @@ func (t *GrepChunksTool) Execute(ctx context.Context, args map[string]interface{ if maxResults > 0 && mmrK > maxResults { mmrK = maxResults } - logger.Debugf(ctx, "[Tool][GrepChunks] Applying MMR: k=%d, lambda=0.7, input=%d results", mmrK, len(scoredResults)) + logger.Debugf( + ctx, + "[Tool][GrepChunks] Applying MMR: k=%d, lambda=0.7, input=%d results", + mmrK, + len(scoredResults), + ) mmrResults := t.applyMMR(ctx, scoredResults, patterns, mmrK, 0.7) if len(mmrResults) > 0 { finalResults = mmrResults @@ -252,9 +257,9 @@ func (t *GrepChunksTool) Execute(ctx context.Context, args map[string]interface{ type chunkWithTitle struct { types.Chunk - KnowledgeTitle string `json:"knowledge_title" gorm:"column:knowledge_title"` - MatchScore float64 `json:"match_score" gorm:"column:match_score"` // Score based on match count and position - MatchedPatterns int `json:"matched_patterns"` // Number of unique patterns matched + KnowledgeTitle string `json:"knowledge_title" gorm:"column:knowledge_title"` + MatchScore float64 `json:"match_score" gorm:"column:match_score"` // Score based on match count and position + MatchedPatterns int `json:"matched_patterns"` // Number of unique patterns matched TotalChunkCount int `json:"total_chunk_count" gorm:"column:total_chunk_count"` } @@ -344,14 +349,16 @@ func (t *GrepChunksTool) formatOutput( patternSummaries = append(patternSummaries, fmt.Sprintf("%s=%d", pattern, count)) } - output.WriteString(fmt.Sprintf("%d) knowledge_id=%s | title=%s | chunk_hits=%d | chunk_total=%d | pattern_hits=[%s]\n", - idx+1, - result.KnowledgeID, - result.KnowledgeTitle, - result.ChunkHitCount, - result.TotalChunkCount, - strings.Join(patternSummaries, ", "), - )) + output.WriteString( + fmt.Sprintf("%d) knowledge_id=%s | title=%s | chunk_hits=%d | chunk_total=%d | pattern_hits=[%s]\n", + idx+1, + result.KnowledgeID, + result.KnowledgeTitle, + result.ChunkHitCount, + result.TotalChunkCount, + strings.Join(patternSummaries, ", "), + ), + ) } return output.String() } @@ -553,7 +560,11 @@ func (t *GrepChunksTool) buildContentSignature(content string) string { } // scoreChunks calculates match scores for chunks based on pattern matches -func (t *GrepChunksTool) scoreChunks(ctx context.Context, results []chunkWithTitle, patterns []string) []chunkWithTitle { +func (t *GrepChunksTool) scoreChunks( + ctx context.Context, + results []chunkWithTitle, + patterns []string, +) []chunkWithTitle { scored := make([]chunkWithTitle, len(results)) for i := range results { scored[i] = results[i] diff --git a/internal/agent/tools/knowledge_search.go b/internal/agent/tools/knowledge_search.go index fd9ff38b..7e852c60 100644 --- a/internal/agent/tools/knowledge_search.go +++ b/internal/agent/tools/knowledge_search.go @@ -246,8 +246,14 @@ func (t *KnowledgeSearchTool) Execute(ctx context.Context, args map[string]inter minScore = 0.3 } - logger.Infof(ctx, "[Tool][KnowledgeSearch] Search params: top_k=%d, vector_threshold=%.2f, keyword_threshold=%.2f, min_score=%.2f", - topK, vectorThreshold, keywordThreshold, minScore) + logger.Infof( + ctx, + "[Tool][KnowledgeSearch] Search params: top_k=%d, vector_threshold=%.2f, keyword_threshold=%.2f, min_score=%.2f", + topK, + vectorThreshold, + keywordThreshold, + minScore, + ) // Execute concurrent search (hybrid search handles both vector and keyword) logger.Infof(ctx, "[Tool][KnowledgeSearch] Starting concurrent search across %d KBs", len(kbIDs)) @@ -280,8 +286,13 @@ func (t *KnowledgeSearchTool) Execute(ctx context.Context, args map[string]inter } if t.chatModel != nil && len(deduplicatedBeforeRerank) > 0 && rerankQuery != "" { - logger.Infof(ctx, "[Tool][KnowledgeSearch] Applying LLM-based rerank with model: %s, input: %d results, queries: %v", - t.chatModel.GetModelName(), len(deduplicatedBeforeRerank), queries) + logger.Infof( + ctx, + "[Tool][KnowledgeSearch] Applying LLM-based rerank with model: %s, input: %d results, queries: %v", + t.chatModel.GetModelName(), + len(deduplicatedBeforeRerank), + queries, + ) rerankedResults, err := t.rerankResults(ctx, rerankQuery, deduplicatedBeforeRerank) if err != nil { logger.Warnf(ctx, "[Tool][KnowledgeSearch] LLM rerank failed, using original results: %v", err) @@ -320,7 +331,12 @@ func (t *KnowledgeSearchTool) Execute(ctx context.Context, args map[string]inter mmrK = 1 } // Apply MMR with lambda=0.7 (balance between relevance and diversity) - logger.Debugf(ctx, "[Tool][KnowledgeSearch] Applying MMR: k=%d, lambda=0.7, input=%d results", mmrK, len(filteredResults)) + logger.Debugf( + ctx, + "[Tool][KnowledgeSearch] Applying MMR: k=%d, lambda=0.7, input=%d results", + mmrK, + len(filteredResults), + ) mmrResults := t.applyMMR(ctx, filteredResults, mmrK, 0.7) if len(mmrResults) > 0 { filteredResults = mmrResults @@ -659,7 +675,8 @@ func (t *KnowledgeSearchTool) rerankWithLLM( } // Optimized prompt focused on retrieval matching and reranking - prompt := fmt.Sprintf(`You are a search result reranking expert. Your task is to evaluate how well each retrieved passage matches the user's search query and information need. + prompt := fmt.Sprintf( + `You are a search result reranking expert. Your task is to evaluate how well each retrieved passage matches the user's search query and information need. User Query: %s @@ -690,7 +707,12 @@ Passage 3: X.XX ... Passage %d: X.XX -Output only the scores, no explanations or additional text.`, query, passagesBuilder.String(), len(batch), len(batch)) +Output only the scores, no explanations or additional text.`, + query, + passagesBuilder.String(), + len(batch), + len(batch), + ) messages := []chat.Message{ { @@ -727,8 +749,13 @@ Output only the scores, no explanations or additional text.`, query, passagesBui // Parse scores from response batchScores, err := t.parseScoresFromResponse(response.Content, len(batch)) if err != nil { - logger.Warnf(ctx, "[Tool][KnowledgeSearch] Failed to parse LLM scores for batch %d-%d: %v, using original scores", - batchStart+1, batchEnd, err) + logger.Warnf( + ctx, + "[Tool][KnowledgeSearch] Failed to parse LLM scores for batch %d-%d: %v, using original scores", + batchStart+1, + batchEnd, + err, + ) // Use original scores for this batch on parsing error for i := batchStart; i < batchEnd; i++ { allScores[i] = results[i].Score @@ -862,7 +889,12 @@ func (t *KnowledgeSearchTool) rerankWithModel( } } - logger.Infof(ctx, "[Tool][KnowledgeSearch] Reranked %d results from %d original results", len(reranked), len(results)) + logger.Infof( + ctx, + "[Tool][KnowledgeSearch] Reranked %d results from %d original results", + len(reranked), + len(results), + ) return reranked, nil } @@ -1027,7 +1059,12 @@ func (t *KnowledgeSearchTool) formatOutput( if result.KnowledgeBaseType == types.KnowledgeBaseTypeFAQ { meta, err := t.getFAQMetadata(ctx, result.ID, faqMetadataCache) if err != nil { - logger.Warnf(ctx, "[Tool][KnowledgeSearch] Failed to load FAQ metadata for chunk %s: %v", result.ID, err) + logger.Warnf( + ctx, + "[Tool][KnowledgeSearch] Failed to load FAQ metadata for chunk %s: %v", + result.ID, + err, + ) } else { faqMeta = meta } @@ -1055,7 +1092,12 @@ func (t *KnowledgeSearchTool) formatOutput( &types.Pagination{Page: 1, PageSize: 1}, []types.ChunkType{types.ChunkTypeText}, "") if err != nil { - logger.Warnf(ctx, "[Tool][KnowledgeSearch] Failed to get total chunks for knowledge %s: %v", result.KnowledgeID, err) + logger.Warnf( + ctx, + "[Tool][KnowledgeSearch] Failed to get total chunks for knowledge %s: %v", + result.KnowledgeID, + err, + ) knowledgeTotalMap[result.KnowledgeID] = 0 } else { knowledgeTotalMap[result.KnowledgeID] = total @@ -1065,7 +1107,12 @@ func (t *KnowledgeSearchTool) formatOutput( // relevanceLevel := GetRelevanceLevel(result.Score) output += fmt.Sprintf("\nResult #%d:\n", i+1) - output += fmt.Sprintf(" [chunk_id: %s][chunk_index: %d]\nContent: %s\n", result.ID, result.ChunkIndex, result.Content) + output += fmt.Sprintf( + " [chunk_id: %s][chunk_index: %d]\nContent: %s\n", + result.ID, + result.ChunkIndex, + result.Content, + ) if faqMeta != nil { if faqMeta.StandardQuestion != "" { @@ -1134,7 +1181,11 @@ func (t *KnowledgeSearchTool) formatOutput( if len(missingRanges) == 0 { // No gaps found (shouldn't happen if remaining > 0, but handle it) - output += fmt.Sprintf(" - 获取全部内容: list_knowledge_chunks(knowledge_id=\"%s\", offset=0, limit=%d)\n", knowledgeID, totalChunks) + output += fmt.Sprintf( + " - 获取全部内容: list_knowledge_chunks(knowledge_id=\"%s\", offset=0, limit=%d)\n", + knowledgeID, + totalChunks, + ) } else if len(missingRanges) == 1 && missingRanges[0].start == 0 && missingRanges[0].end == int(totalChunks)-1 { // All chunks are missing (shouldn't happen, but handle it) output += fmt.Sprintf(" - 获取全部内容: list_knowledge_chunks(knowledge_id=\"%s\", offset=0, limit=%d)\n", knowledgeID, totalChunks) @@ -1284,8 +1335,12 @@ func (t *KnowledgeSearchTool) normalizeKeywordSearchResults(ctx context.Context, for _, r := range keywordResults { r.Score = 1.0 } - logger.Infof(ctx, "[Tool][KnowledgeSearch] Keyword scores have no variance, all set to 1.0: count=%d, score=%.3f", - len(keywordResults), minS) + logger.Infof( + ctx, + "[Tool][KnowledgeSearch] Keyword scores have no variance, all set to 1.0: count=%d, score=%.3f", + len(keywordResults), + minS, + ) return } @@ -1336,8 +1391,15 @@ func (t *KnowledgeSearchTool) normalizeKeywordSearchResults(ctx context.Context, r.Score = ns } - logger.Infof(ctx, "[Tool][KnowledgeSearch] Normalized keyword scores: count=%d, raw_min=%.3f, raw_max=%.3f, normalize_min=%.3f, normalize_max=%.3f", - len(keywordResults), minS, maxS, normalizeMin, normalizeMax) + logger.Infof( + ctx, + "[Tool][KnowledgeSearch] Normalized keyword scores: count=%d, raw_min=%.3f, raw_max=%.3f, normalize_min=%.3f, normalize_max=%.3f", + len(keywordResults), + minS, + maxS, + normalizeMin, + normalizeMax, + ) } else { // Fallback: all scores are the same after percentile filtering for _, r := range keywordResults { diff --git a/internal/agent/tools/list_knowledge_chunks.go b/internal/agent/tools/list_knowledge_chunks.go index c40c00a5..c6e8d183 100644 --- a/internal/agent/tools/list_knowledge_chunks.go +++ b/internal/agent/tools/list_knowledge_chunks.go @@ -203,7 +203,13 @@ func (t *ListKnowledgeChunksTool) buildOutput( } return builder.String() } - fmt.Fprintf(builder, "本次拉取: %d 条， 检索范围: %d - %d\n\n", fetched, chunks[0].ChunkIndex, chunks[len(chunks)-1].ChunkIndex) + fmt.Fprintf( + builder, + "本次拉取: %d 条， 检索范围: %d - %d\n\n", + fetched, + chunks[0].ChunkIndex, + chunks[len(chunks)-1].ChunkIndex, + ) builder.WriteString("=== 分块内容预览 ===\n\n") for idx, c := range chunks { diff --git a/internal/agent/tools/mcp_tool.go b/internal/agent/tools/mcp_tool.go index 17892a16..d757d624 100644 --- a/internal/agent/tools/mcp_tool.go +++ b/internal/agent/tools/mcp_tool.go @@ -177,7 +177,12 @@ func sanitizeName(name string) string { } // RegisterMCPTools registers MCP tools from given services -func RegisterMCPTools(ctx context.Context, registry *ToolRegistry, services []*types.MCPService, mcpManager *mcp.MCPManager) error { +func RegisterMCPTools( + ctx context.Context, + registry *ToolRegistry, + services []*types.MCPService, + mcpManager *mcp.MCPManager, +) error { if len(services) == 0 { return nil } @@ -239,7 +244,11 @@ func RegisterMCPTools(ctx context.Context, registry *ToolRegistry, services []*t } // GetMCPToolsInfo returns information about available MCP tools -func GetMCPToolsInfo(ctx context.Context, services []*types.MCPService, mcpManager *mcp.MCPManager) (map[string][]string, error) { +func GetMCPToolsInfo( + ctx context.Context, + services []*types.MCPService, + mcpManager *mcp.MCPManager, +) (map[string][]string, error) { result := make(map[string][]string) // Use provided context with timeout diff --git a/internal/agent/tools/query_knowledge_graph.go b/internal/agent/tools/query_knowledge_graph.go index 025e3335..21013610 100644 --- a/internal/agent/tools/query_knowledge_graph.go +++ b/internal/agent/tools/query_knowledge_graph.go @@ -369,7 +369,10 @@ func (t *QueryKnowledgeGraphTool) Execute(ctx context.Context, args map[string]i } // buildGraphVisualizationData builds structured data for graph visualization -func buildGraphVisualizationData(results []*types.SearchResult, graphConfigs map[string]map[string]interface{}) map[string]interface{} { +func buildGraphVisualizationData( + results []*types.SearchResult, + graphConfigs map[string]map[string]interface{}, +) map[string]interface{} { // Build a simple graph structure for frontend visualization nodes := make([]map[string]interface{}, 0) edges := make([]map[string]interface{}, 0) diff --git a/internal/agent/tools/registry.go b/internal/agent/tools/registry.go index 0e21ed67..53bce94c 100644 --- a/internal/agent/tools/registry.go +++ b/internal/agent/tools/registry.go @@ -69,7 +69,11 @@ func (r *ToolRegistry) GetFunctionDefinitions() []types.FunctionDefinition { } // ExecuteTool executes a tool by name with the given arguments -func (r *ToolRegistry) ExecuteTool(ctx context.Context, name string, args map[string]interface{}) (*types.ToolResult, error) { +func (r *ToolRegistry) ExecuteTool( + ctx context.Context, + name string, + args map[string]interface{}, +) (*types.ToolResult, error) { common.PipelineInfo(ctx, "AgentTool", "execute_start", map[string]interface{}{ "tool": name, "args": args, diff --git a/internal/agent/tools/sequentialthinking.go b/internal/agent/tools/sequentialthinking.go index b389a3ad..5158f845 100644 --- a/internal/agent/tools/sequentialthinking.go +++ b/internal/agent/tools/sequentialthinking.go @@ -196,7 +196,8 @@ func (t *SequentialThinkingTool) Execute(ctx context.Context, args map[string]in branchKeys = append(branchKeys, k) } - incomplete := thoughtData.NextThoughtNeeded || thoughtData.NeedsMoreThoughts || thoughtData.ThoughtNumber < thoughtData.TotalThoughts + incomplete := thoughtData.NextThoughtNeeded || thoughtData.NeedsMoreThoughts || + thoughtData.ThoughtNumber < thoughtData.TotalThoughts responseData := map[string]interface{}{ "thought_number": thoughtData.ThoughtNumber, @@ -209,7 +210,12 @@ func (t *SequentialThinkingTool) Execute(ctx context.Context, args map[string]in "incomplete_steps": incomplete, } - logger.Infof(ctx, "[Tool][SequentialThinking] Execute completed - Thought %d/%d", thoughtData.ThoughtNumber, thoughtData.TotalThoughts) + logger.Infof( + ctx, + "[Tool][SequentialThinking] Execute completed - Thought %d/%d", + thoughtData.ThoughtNumber, + thoughtData.TotalThoughts, + ) outputMsg := "Thought process recorded" if incomplete { diff --git a/internal/agent/tools/web_fetch.go b/internal/agent/tools/web_fetch.go index 17018563..7bdd952c 100644 --- a/internal/agent/tools/web_fetch.go +++ b/internal/agent/tools/web_fetch.go @@ -252,7 +252,10 @@ func (t *WebFetchTool) validateParams(p webFetchParams) error { return nil } -func (t *WebFetchTool) executeFetch(ctx context.Context, params webFetchParams) (string, map[string]interface{}, error) { +func (t *WebFetchTool) executeFetch( + ctx context.Context, + params webFetchParams, +) (string, map[string]interface{}, error) { logger.Infof(ctx, "[Tool][WebFetch] Fetching URL: %s", params.URL) finalURL := t.normalizeGitHubURL(params.URL) @@ -375,14 +378,17 @@ func (t *WebFetchTool) fetchHTMLContent(ctx context.Context, targetURL string) ( func (t *WebFetchTool) fetchWithChromedp(ctx context.Context, targetURL string) (string, error) { logger.Debugf(ctx, "[Tool][WebFetch] Chromedp 抓取开始 url=%s", targetURL) - opts := append(chromedp.DefaultExecAllocatorOptions[:], + opts := append( + chromedp.DefaultExecAllocatorOptions[:], chromedp.Flag("headless", true), chromedp.Flag("disable-setuid-sandbox", true), chromedp.Flag("disable-dev-shm-usage", true), chromedp.Flag("disable-gpu", true), chromedp.Flag("disable-blink-features", "AutomationControlled"), chromedp.Flag("disable-features", "VizDisplayCompositor"), - chromedp.UserAgent("Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/120.0.0.0 Safari/537.36"), + chromedp.UserAgent( + "Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/120.0.0.0 Safari/537.36", + ), ) allocCtx, cancel := chromedp.NewExecAllocator(ctx, opts...) @@ -435,7 +441,10 @@ func (t *WebFetchTool) fetchWithTimeout(ctx context.Context, targetURL string) ( } req.Header.Set("User-Agent", "Mozilla/5.0 (compatible; WebFetchTool/1.0)") - req.Header.Set("Accept", "text/html,application/xhtml+xml,application/xml;q=0.9,image/avif,image/webp,image/apng,*/*;q=0.8") + req.Header.Set( + "Accept", + "text/html,application/xhtml+xml,application/xml;q=0.9,image/avif,image/webp,image/apng,*/*;q=0.8", + ) req.Header.Set("Accept-Language", "zh-CN,zh;q=0.9,en;q=0.8") req.Header.Set("Cache-Control", "no-cache") diff --git a/internal/agent/tools/web_search.go b/internal/agent/tools/web_search.go index 3fe776de..b5ec41ff 100644 --- a/internal/agent/tools/web_search.go +++ b/internal/agent/tools/web_search.go @@ -57,7 +57,10 @@ func NewWebSearchTool( **Parameters**: - query (required): Search query string -**Returns**: Web search results with title, URL, snippet, and content (up to ` + fmt.Sprintf("%d", maxResults) + ` results) +**Returns**: Web search results with title, URL, snippet, and content (up to ` + fmt.Sprintf( + "%d", + maxResults, + ) + ` results) ## Examples @@ -80,7 +83,10 @@ func NewWebSearchTool( - Use this tool when knowledge bases don't have the information you need - Results include URL, title, snippet, and content snippet (may be truncated) - **CRITICAL**: If content is truncated or you need full details, use **web_fetch** to fetch complete page content -- Maximum ` + fmt.Sprintf("%d", maxResults) + ` results will be returned per search` +- Maximum ` + fmt.Sprintf( + "%d", + maxResults, + ) + ` results will be returned per search` return &WebSearchTool{ BaseTool: NewBaseTool("web_search", description), @@ -152,7 +158,12 @@ func (t *WebSearchTool) Execute(ctx context.Context, args map[string]interface{} searchConfig.MaxResults = t.maxResults // Perform web search - logger.Infof(ctx, "[Tool][WebSearch] Performing web search with provider: %s, maxResults: %d", searchConfig.Provider, searchConfig.MaxResults) + logger.Infof( + ctx, + "[Tool][WebSearch] Performing web search with provider: %s, maxResults: %d", + searchConfig.Provider, + searchConfig.MaxResults, + ) webResults, err := t.webSearchService.Search(ctx, &searchConfig, query) if err != nil { logger.Errorf(ctx, "[Tool][WebSearch] Web search failed: %v", err) @@ -165,7 +176,8 @@ func (t *WebSearchTool) Execute(ctx context.Context, args map[string]interface{} logger.Infof(ctx, "[Tool][WebSearch] Web search returned %d results", len(webResults)) // Apply RAG compression if configured - if len(webResults) > 0 && tenant.WebSearchConfig.CompressionMethod != "none" && tenant.WebSearchConfig.CompressionMethod != "" { + if len(webResults) > 0 && tenant.WebSearchConfig.CompressionMethod != "none" && + tenant.WebSearchConfig.CompressionMethod != "" { // Load session-scoped temp KB state from Redis using SessionService tempKBID, seen, ids := t.sessionService.GetWebSearchTempKBState(ctx, t.sessionID) diff --git a/internal/application/repository/chunk.go b/internal/application/repository/chunk.go index 5cc90f11..0402da6b 100644 --- a/internal/application/repository/chunk.go +++ b/internal/application/repository/chunk.go @@ -112,7 +112,11 @@ func (r *chunkRepository) ListPagedChunksByKnowledgeID( return chunks, total, nil } -func (r *chunkRepository) ListChunkByParentID(ctx context.Context, tenantID uint64, parentID string) ([]*types.Chunk, error) { +func (r *chunkRepository) ListChunkByParentID( + ctx context.Context, + tenantID uint64, + parentID string, +) ([]*types.Chunk, error) { var chunks []*types.Chunk if err := r.db.WithContext(ctx). Where("tenant_id = ? AND parent_chunk_id = ?", tenantID, parentID). @@ -166,7 +170,11 @@ func (r *chunkRepository) DeleteByKnowledgeList(ctx context.Context, tenantID ui } // CountChunksByKnowledgeBaseID counts the number of chunks in a knowledge base -func (r *chunkRepository) CountChunksByKnowledgeBaseID(ctx context.Context, tenantID uint64, kbID string) (int64, error) { +func (r *chunkRepository) CountChunksByKnowledgeBaseID( + ctx context.Context, + tenantID uint64, + kbID string, +) (int64, error) { var count int64 err := r.db.WithContext(ctx).Model(&types.Chunk{}). Where("tenant_id = ? AND knowledge_base_id = ?", tenantID, kbID). @@ -175,7 +183,11 @@ func (r *chunkRepository) CountChunksByKnowledgeBaseID(ctx context.Context, tena } // DeleteUnindexedChunks by knowledge id and chunk index range -func (r *chunkRepository) DeleteUnindexedChunks(ctx context.Context, tenantID uint64, knowledgeID string) ([]*types.Chunk, error) { +func (r *chunkRepository) DeleteUnindexedChunks( + ctx context.Context, + tenantID uint64, + knowledgeID string, +) ([]*types.Chunk, error) { var chunks []*types.Chunk if err := r.db.WithContext(ctx). Where("tenant_id = ? AND knowledge_id = ? AND status = ?", tenantID, knowledgeID, types.ChunkStatusStored). @@ -194,7 +206,11 @@ func (r *chunkRepository) DeleteUnindexedChunks(ctx context.Context, tenantID ui // ListAllFAQChunksByKnowledgeID lists all FAQ chunks for a knowledge ID (only essential fields for efficiency) // Uses batch query to handle large datasets -func (r *chunkRepository) ListAllFAQChunksByKnowledgeID(ctx context.Context, tenantID uint64, knowledgeID string) ([]*types.Chunk, error) { +func (r *chunkRepository) ListAllFAQChunksByKnowledgeID( + ctx context.Context, + tenantID uint64, + knowledgeID string, +) ([]*types.Chunk, error) { const batchSize = 1000 // 每批查询1000条 var allChunks []*types.Chunk offset := 0 diff --git a/internal/application/repository/knowledge.go b/internal/application/repository/knowledge.go index 3750f4e5..e6a01e4a 100644 --- a/internal/application/repository/knowledge.go +++ b/internal/application/repository/knowledge.go @@ -28,7 +28,11 @@ func (r *knowledgeRepository) CreateKnowledge(ctx context.Context, knowledge *ty } // GetKnowledgeByID gets knowledge -func (r *knowledgeRepository) GetKnowledgeByID(ctx context.Context, tenantID uint64, id string) (*types.Knowledge, error) { +func (r *knowledgeRepository) GetKnowledgeByID( + ctx context.Context, + tenantID uint64, + id string, +) (*types.Knowledge, error) { var knowledge types.Knowledge if err := r.db.WithContext(ctx).Where("tenant_id = ? AND id = ?", tenantID, id).First(&knowledge).Error; err != nil { if errors.Is(err, gorm.ErrRecordNotFound) { @@ -217,13 +221,22 @@ func (r *knowledgeRepository) AminusB( return knowledgeIDs, err } -func (r *knowledgeRepository) UpdateKnowledgeColumn(ctx context.Context, id string, column string, value interface{}) error { +func (r *knowledgeRepository) UpdateKnowledgeColumn( + ctx context.Context, + id string, + column string, + value interface{}, +) error { err := r.db.WithContext(ctx).Model(&types.Knowledge{}).Where("id = ?", id).Update(column, value).Error return err } // CountKnowledgeByKnowledgeBaseID counts the number of knowledge items in a knowledge base -func (r *knowledgeRepository) CountKnowledgeByKnowledgeBaseID(ctx context.Context, tenantID uint64, kbID string) (int64, error) { +func (r *knowledgeRepository) CountKnowledgeByKnowledgeBaseID( + ctx context.Context, + tenantID uint64, + kbID string, +) (int64, error) { var count int64 err := r.db.WithContext(ctx).Model(&types.Knowledge{}). Where("tenant_id = ? AND knowledge_base_id = ?", tenantID, kbID). diff --git a/internal/application/repository/mcp_service.go b/internal/application/repository/mcp_service.go index 5b95f939..9ee1a942 100644 --- a/internal/application/repository/mcp_service.go +++ b/internal/application/repository/mcp_service.go @@ -69,7 +69,11 @@ func (r *mcpServiceRepository) ListEnabled(ctx context.Context, tenantID uint64) } // ListByIDs retrieves MCP services by multiple IDs for a tenant -func (r *mcpServiceRepository) ListByIDs(ctx context.Context, tenantID uint64, ids []string) ([]*types.MCPService, error) { +func (r *mcpServiceRepository) ListByIDs( + ctx context.Context, + tenantID uint64, + ids []string, +) ([]*types.MCPService, error) { if len(ids) == 0 { return []*types.MCPService{}, nil } diff --git a/internal/application/repository/model.go b/internal/application/repository/model.go index 72c25bf3..26b256f5 100644 --- a/internal/application/repository/model.go +++ b/internal/application/repository/model.go @@ -79,7 +79,12 @@ func (r *modelRepository) Delete(ctx context.Context, tenantID uint64, id string // ClearDefaultByType clears the default flag for all models of a specific type // This is a batch operation that updates all matching records in one query -func (r *modelRepository) ClearDefaultByType(ctx context.Context, tenantID uint, modelType types.ModelType, excludeID string) error { +func (r *modelRepository) ClearDefaultByType( + ctx context.Context, + tenantID uint, + modelType types.ModelType, + excludeID string, +) error { query := r.db.WithContext(ctx).Model(&types.Model{}).Where( "tenant_id = ? AND type = ? AND is_default = ?", tenantID, modelType, true, ) diff --git a/internal/application/repository/retriever/elasticsearch/structs.go b/internal/application/repository/retriever/elasticsearch/structs.go index 08e1bcd0..05f1e8ab 100644 --- a/internal/application/repository/retriever/elasticsearch/structs.go +++ b/internal/application/repository/retriever/elasticsearch/structs.go @@ -9,14 +9,14 @@ import ( // VectorEmbedding defines the Elasticsearch document structure for vector embeddings type VectorEmbedding struct { - Content string `json:"content" gorm:"column:content;not null"` // Text content of the chunk - SourceID string `json:"source_id" gorm:"column:source_id;not null"` // ID of the source document - SourceType int `json:"source_type" gorm:"column:source_type;not null"` // Type of the source document - ChunkID string `json:"chunk_id" gorm:"column:chunk_id"` // Unique ID of the text chunk - KnowledgeID string `json:"knowledge_id" gorm:"column:knowledge_id"` // ID of the knowledge item - KnowledgeBaseID string `json:"knowledge_base_id" gorm:"column:knowledge_base_id"` // ID of the knowledge base - Embedding []float32 `json:"embedding" gorm:"column:embedding;not null"` // Vector embedding of the content - IsEnabled bool `json:"is_enabled"` // Whether the chunk is enabled + Content string `json:"content" gorm:"column:content;not null"` // Text content of the chunk + SourceID string `json:"source_id" gorm:"column:source_id;not null"` // ID of the source document + SourceType int `json:"source_type" gorm:"column:source_type;not null"` // Type of the source document + ChunkID string `json:"chunk_id" gorm:"column:chunk_id"` // Unique ID of the text chunk + KnowledgeID string `json:"knowledge_id" gorm:"column:knowledge_id"` // ID of the knowledge item + KnowledgeBaseID string `json:"knowledge_base_id" gorm:"column:knowledge_base_id"` // ID of the knowledge base + Embedding []float32 `json:"embedding" gorm:"column:embedding;not null"` // Vector embedding of the content + IsEnabled bool `json:"is_enabled"` // Whether the chunk is enabled } // VectorEmbeddingWithScore extends VectorEmbedding with similarity score diff --git a/internal/application/repository/retriever/elasticsearch/v7/repository.go b/internal/application/repository/retriever/elasticsearch/v7/repository.go index 1f311af2..01b84815 100644 --- a/internal/application/repository/retriever/elasticsearch/v7/repository.go +++ b/internal/application/repository/retriever/elasticsearch/v7/repository.go @@ -1094,7 +1094,10 @@ func (e *elasticsearchRepository) saveCopiedIndices(ctx context.Context, indexIn } // BatchUpdateChunkEnabledStatus updates the enabled status of chunks in batch -func (e *elasticsearchRepository) BatchUpdateChunkEnabledStatus(ctx context.Context, chunkStatusMap map[string]bool) error { +func (e *elasticsearchRepository) BatchUpdateChunkEnabledStatus( + ctx context.Context, + chunkStatusMap map[string]bool, +) error { log := logger.GetLogger(ctx) if len(chunkStatusMap) == 0 { log.Warnf("[ElasticsearchV7] Chunk status map is empty, skipping update") diff --git a/internal/application/repository/retriever/elasticsearch/v8/repository.go b/internal/application/repository/retriever/elasticsearch/v8/repository.go index 493120b8..378ecb0d 100644 --- a/internal/application/repository/retriever/elasticsearch/v8/repository.go +++ b/internal/application/repository/retriever/elasticsearch/v8/repository.go @@ -547,7 +547,10 @@ func (e *elasticsearchRepository) CopyIndices(ctx context.Context, } // BatchUpdateChunkEnabledStatus updates the enabled status of chunks in batch -func (e *elasticsearchRepository) BatchUpdateChunkEnabledStatus(ctx context.Context, chunkStatusMap map[string]bool) error { +func (e *elasticsearchRepository) BatchUpdateChunkEnabledStatus( + ctx context.Context, + chunkStatusMap map[string]bool, +) error { log := logger.GetLogger(ctx) if len(chunkStatusMap) == 0 { log.Warnf("[Elasticsearch] Chunk status map is empty, skipping update") diff --git a/internal/application/repository/retriever/neo4j/repository.go b/internal/application/repository/retriever/neo4j/repository.go index 63288ff3..20e9f7fb 100644 --- a/internal/application/repository/retriever/neo4j/repository.go +++ b/internal/application/repository/retriever/neo4j/repository.go @@ -152,7 +152,11 @@ func (n *Neo4jRepository) DelGraph(ctx context.Context, namespaces []types.NameS return nil } -func (n *Neo4jRepository) SearchNode(ctx context.Context, namespace types.NameSpace, nodes []string) (*types.GraphData, error) { +func (n *Neo4jRepository) SearchNode( + ctx context.Context, + namespace types.NameSpace, + nodes []string, +) (*types.GraphData, error) { if n.driver == nil { logger.Warnf(ctx, "NOT SUPPORT RETRIEVE GRAPH") return nil, nil diff --git a/internal/application/repository/retriever/postgres/repository.go b/internal/application/repository/retriever/postgres/repository.go index 5c5a1d68..a551112c 100644 --- a/internal/application/repository/retriever/postgres/repository.go +++ b/internal/application/repository/retriever/postgres/repository.go @@ -481,7 +481,8 @@ func (g *pgRepository) BatchUpdateChunkEnabledStatus(ctx context.Context, chunkS logger.GetLogger(ctx).Errorf("[Postgres] Failed to update enabled chunks: %v", result.Error) return result.Error } - logger.GetLogger(ctx).Infof("[Postgres] Updated %d chunks to enabled, rows affected: %d", len(enabledChunkIDs), result.RowsAffected) + logger.GetLogger(ctx). + Infof("[Postgres] Updated %d chunks to enabled, rows affected: %d", len(enabledChunkIDs), result.RowsAffected) } // Batch update disabled chunks @@ -493,7 +494,8 @@ func (g *pgRepository) BatchUpdateChunkEnabledStatus(ctx context.Context, chunkS logger.GetLogger(ctx).Errorf("[Postgres] Failed to update disabled chunks: %v", result.Error) return result.Error } - logger.GetLogger(ctx).Infof("[Postgres] Updated %d chunks to disabled, rows affected: %d", len(disabledChunkIDs), result.RowsAffected) + logger.GetLogger(ctx). + Infof("[Postgres] Updated %d chunks to disabled, rows affected: %d", len(disabledChunkIDs), result.RowsAffected) } logger.GetLogger(ctx).Infof("[Postgres] Successfully batch updated chunk enabled status") diff --git a/internal/application/repository/retriever/postgres/structs.go b/internal/application/repository/retriever/postgres/structs.go index 915d43b6..b673e555 100644 --- a/internal/application/repository/retriever/postgres/structs.go +++ b/internal/application/repository/retriever/postgres/structs.go @@ -13,35 +13,35 @@ import ( // pgVector defines the database model for vector embeddings storage type pgVector struct { - ID uint `json:"id" gorm:"primarykey"` - CreatedAt time.Time `json:"created_at" gorm:"column:created_at"` - UpdatedAt time.Time `json:"updated_at" gorm:"column:updated_at"` - SourceID string `json:"source_id" gorm:"column:source_id;not null"` - SourceType int `json:"source_type" gorm:"column:source_type;not null"` - ChunkID string `json:"chunk_id" gorm:"column:chunk_id"` - KnowledgeID string `json:"knowledge_id" gorm:"column:knowledge_id"` + ID uint `json:"id" gorm:"primarykey"` + CreatedAt time.Time `json:"created_at" gorm:"column:created_at"` + UpdatedAt time.Time `json:"updated_at" gorm:"column:updated_at"` + SourceID string `json:"source_id" gorm:"column:source_id;not null"` + SourceType int `json:"source_type" gorm:"column:source_type;not null"` + ChunkID string `json:"chunk_id" gorm:"column:chunk_id"` + KnowledgeID string `json:"knowledge_id" gorm:"column:knowledge_id"` KnowledgeBaseID string `json:"knowledge_base_id" gorm:"column:knowledge_base_id"` - Content string `json:"content" gorm:"column:content;not null"` - Dimension int `json:"dimension" gorm:"column:dimension;not null"` - Embedding pgvector.HalfVector `json:"embedding" gorm:"column:embedding;not null"` - IsEnabled bool `json:"is_enabled" gorm:"column:is_enabled;default:true;index"` + Content string `json:"content" gorm:"column:content;not null"` + Dimension int `json:"dimension" gorm:"column:dimension;not null"` + Embedding pgvector.HalfVector `json:"embedding" gorm:"column:embedding;not null"` + IsEnabled bool `json:"is_enabled" gorm:"column:is_enabled;default:true;index"` } // pgVectorWithScore extends pgVector with similarity score field type pgVectorWithScore struct { - ID uint `json:"id" gorm:"primarykey"` - CreatedAt time.Time `json:"created_at" gorm:"column:created_at"` - UpdatedAt time.Time `json:"updated_at" gorm:"column:updated_at"` - SourceID string `json:"source_id" gorm:"column:source_id;not null"` - SourceType int `json:"source_type" gorm:"column:source_type;not null"` - ChunkID string `json:"chunk_id" gorm:"column:chunk_id"` - KnowledgeID string `json:"knowledge_id" gorm:"column:knowledge_id"` + ID uint `json:"id" gorm:"primarykey"` + CreatedAt time.Time `json:"created_at" gorm:"column:created_at"` + UpdatedAt time.Time `json:"updated_at" gorm:"column:updated_at"` + SourceID string `json:"source_id" gorm:"column:source_id;not null"` + SourceType int `json:"source_type" gorm:"column:source_type;not null"` + ChunkID string `json:"chunk_id" gorm:"column:chunk_id"` + KnowledgeID string `json:"knowledge_id" gorm:"column:knowledge_id"` KnowledgeBaseID string `json:"knowledge_base_id" gorm:"column:knowledge_base_id"` - Content string `json:"content" gorm:"column:content;not null"` - Dimension int `json:"dimension" gorm:"column:dimension;not null"` - Embedding pgvector.HalfVector `json:"embedding" gorm:"column:embedding;not null"` - IsEnabled bool `json:"is_enabled" gorm:"column:is_enabled;default:true;index"` - Score float64 `json:"score" gorm:"column:score"` + Content string `json:"content" gorm:"column:content;not null"` + Dimension int `json:"dimension" gorm:"column:dimension;not null"` + Embedding pgvector.HalfVector `json:"embedding" gorm:"column:embedding;not null"` + IsEnabled bool `json:"is_enabled" gorm:"column:is_enabled;default:true;index"` + Score float64 `json:"score" gorm:"column:score"` } // TableName specifies the database table name for pgVectorWithScore diff --git a/internal/application/repository/tag.go b/internal/application/repository/tag.go index 7f1857fe..18af8524 100644 --- a/internal/application/repository/tag.go +++ b/internal/application/repository/tag.go @@ -35,7 +35,11 @@ func (r *knowledgeTagRepository) GetByID(ctx context.Context, tenantID uint64, i return &tag, nil } -func (r *knowledgeTagRepository) ListByKB(ctx context.Context, tenantID uint64, kbID string) ([]*types.KnowledgeTag, error) { +func (r *knowledgeTagRepository) ListByKB( + ctx context.Context, + tenantID uint64, + kbID string, +) ([]*types.KnowledgeTag, error) { var tags []*types.KnowledgeTag if err := r.db.WithContext(ctx). Where("tenant_id = ? AND knowledge_base_id = ?", tenantID, kbID). diff --git a/internal/application/service/agent_service.go b/internal/application/service/agent_service.go index 986d47e4..cd89ac5c 100644 --- a/internal/application/service/agent_service.go +++ b/internal/application/service/agent_service.go @@ -184,7 +184,13 @@ func (s *agentService) registerTools( if tid, ok := ctx.Value(types.TenantIDContextKey).(uint64); ok { tenantID = tid } - logger.Infof(ctx, "Registering tools: %v, tenant ID: %d, webSearchEnabled: %v", allowedTools, tenantID, config.WebSearchEnabled) + logger.Infof( + ctx, + "Registering tools: %v, tenant ID: %d, webSearchEnabled: %v", + allowedTools, + tenantID, + config.WebSearchEnabled, + ) // Register each allowed tool for _, toolName := range allowedTools { @@ -224,7 +230,12 @@ func (s *agentService) registerTools( sessionID, config.WebSearchMaxResults, )) - logger.Infof(ctx, "Registered web_search tool for session: %s, maxResults: %d", sessionID, config.WebSearchMaxResults) + logger.Infof( + ctx, + "Registered web_search tool for session: %s, maxResults: %d", + sessionID, + config.WebSearchMaxResults, + ) case "web_fetch": registry.RegisterTool(tools.NewWebFetchTool(chatModel)) diff --git a/internal/application/service/chat_pipline/merge.go b/internal/application/service/chat_pipline/merge.go index cc29236e..3abd3fe5 100644 --- a/internal/application/service/chat_pipline/merge.go +++ b/internal/application/service/chat_pipline/merge.go @@ -212,7 +212,11 @@ func mergeImageInfo(ctx context.Context, target *types.SearchResult, source *typ return nil } -func (p *PluginMerge) populateFAQAnswers(ctx context.Context, chatManage *types.ChatManage, results []*types.SearchResult) []*types.SearchResult { +func (p *PluginMerge) populateFAQAnswers( + ctx context.Context, + chatManage *types.ChatManage, + results []*types.SearchResult, +) []*types.SearchResult { if len(results) == 0 || p.chunkRepo == nil { return results } @@ -332,7 +336,11 @@ func buildFAQAnswerContent(meta *types.FAQChunkMetadata) string { return strings.TrimSpace(builder.String()) } -func (p *PluginMerge) expandShortContextWithNeighbors(ctx context.Context, chatManage *types.ChatManage, results []*types.SearchResult) []*types.SearchResult { +func (p *PluginMerge) expandShortContextWithNeighbors( + ctx context.Context, + chatManage *types.ChatManage, + results []*types.SearchResult, +) []*types.SearchResult { const ( minLen = 350 maxLen = 850 @@ -495,7 +503,8 @@ func (p *PluginMerge) expandShortContextWithNeighbors(ctx context.Context, chatM expanded := false if prevCursor != "" { p.fetchChunksIfMissing(ctx, tenantID, chunkMap, prevCursor) - if prevChunk := chunkMap[prevCursor]; prevChunk != nil && prevChunk.KnowledgeID == baseChunk.KnowledgeID { + if prevChunk := chunkMap[prevCursor]; prevChunk != nil && + prevChunk.KnowledgeID == baseChunk.KnowledgeID { prevContent = concatNoOverlap(prevChunk.Content, prevContent) prevIDs = append([]string{prevChunk.ID}, prevIDs...) prevCursor = prevChunk.PreChunkID @@ -512,7 +521,8 @@ func (p *PluginMerge) expandShortContextWithNeighbors(ctx context.Context, chatM if nextCursor != "" { p.fetchChunksIfMissing(ctx, tenantID, chunkMap, nextCursor) - if nextChunk := chunkMap[nextCursor]; nextChunk != nil && nextChunk.KnowledgeID == baseChunk.KnowledgeID { + if nextChunk := chunkMap[nextCursor]; nextChunk != nil && + nextChunk.KnowledgeID == baseChunk.KnowledgeID { nextContent = concatNoOverlap(nextContent, nextChunk.Content) nextIDs = append(nextIDs, nextChunk.ID) nextCursor = nextChunk.NextChunkID @@ -625,7 +635,12 @@ func containsID(ids []string, target string) bool { return false } -func (p *PluginMerge) fetchChunksIfMissing(ctx context.Context, tenantID uint64, chunkMap map[string]*types.Chunk, chunkIDs ...string) { +func (p *PluginMerge) fetchChunksIfMissing( + ctx context.Context, + tenantID uint64, + chunkMap map[string]*types.Chunk, + chunkIDs ...string, +) { missing := make([]string, 0, len(chunkIDs)) for _, id := range chunkIDs { if id == "" { diff --git a/internal/application/service/chat_pipline/preprocess.go b/internal/application/service/chat_pipline/preprocess.go index 99393fbb..029cc745 100644 --- a/internal/application/service/chat_pipline/preprocess.go +++ b/internal/application/service/chat_pipline/preprocess.go @@ -91,7 +91,12 @@ func (p *PluginPreprocess) ActivationEvents() []types.EventType { } // OnEvent Process events -func (p *PluginPreprocess) OnEvent(ctx context.Context, eventType types.EventType, chatManage *types.ChatManage, next func() *PluginError) *PluginError { +func (p *PluginPreprocess) OnEvent( + ctx context.Context, + eventType types.EventType, + chatManage *types.ChatManage, + next func() *PluginError, +) *PluginError { rawQuery := strings.TrimSpace(chatManage.RewriteQuery) if rawQuery == "" { return next() @@ -270,15 +275,30 @@ type intentResp struct { func (p *PluginPreprocess) detectIntentLLM(ctx context.Context, chatManage *types.ChatManage, text string) string { if p.modelService == nil || chatManage.ChatModelID == "" { - pipelineWarn(ctx, "IntentDetect", "skip", map[string]interface{}{"reason": "no_model", "session_id": chatManage.SessionID}) + pipelineWarn( + ctx, + "IntentDetect", + "skip", + map[string]interface{}{"reason": "no_model", "session_id": chatManage.SessionID}, + ) return "general" } chatModel, err := p.modelService.GetChatModel(ctx, chatManage.ChatModelID) if err != nil { - pipelineWarn(ctx, "IntentDetect", "get_model_failed", map[string]interface{}{"error": err.Error(), "model_id": chatManage.ChatModelID}) + pipelineWarn( + ctx, + "IntentDetect", + "get_model_failed", + map[string]interface{}{"error": err.Error(), "model_id": chatManage.ChatModelID}, + ) return "general" } - pipelineInfo(ctx, "IntentDetect", "start", map[string]interface{}{"session_id": chatManage.SessionID, "model_id": chatManage.ChatModelID}) + pipelineInfo( + ctx, + "IntentDetect", + "start", + map[string]interface{}{"session_id": chatManage.SessionID, "model_id": chatManage.ChatModelID}, + ) sys := "You are a query intent classifier. Classify the user's query into one of: definition, howto, compare, qa, general. Respond ONLY with a JSON object {\"intent\": \"...\", \"confidence\": 0.0 } inside a markdown fenced block." usr := text think := false @@ -296,7 +316,12 @@ func (p *PluginPreprocess) detectIntentLLM(ctx context.Context, chatManage *type pipelineWarn(ctx, "IntentDetect", "parse_failed", map[string]interface{}{"body": body, "error": err.Error()}) return "general" } - pipelineInfo(ctx, "IntentDetect", "result", map[string]interface{}{"intent": ir.Intent, "confidence": ir.Confidence}) + pipelineInfo( + ctx, + "IntentDetect", + "result", + map[string]interface{}{"intent": ir.Intent, "confidence": ir.Confidence}, + ) switch strings.ToLower(strings.TrimSpace(ir.Intent)) { case "definition", "howto", "compare", "qa", "general": return strings.ToLower(ir.Intent) diff --git a/internal/application/service/chat_pipline/rerank.go b/internal/application/service/chat_pipline/rerank.go index 48bd7d4e..30df6896 100644 --- a/internal/application/service/chat_pipline/rerank.go +++ b/internal/application/service/chat_pipline/rerank.go @@ -222,7 +222,13 @@ func compositeScore(sr *types.SearchResult, modelScore, baseScore float64, chatM return composite } -func applyMMR(ctx context.Context, results []*types.SearchResult, chatManage *types.ChatManage, k int, lambda float64) []*types.SearchResult { +func applyMMR( + ctx context.Context, + results []*types.SearchResult, + chatManage *types.ChatManage, + k int, + lambda float64, +) []*types.SearchResult { if k <= 0 || len(results) == 0 { return nil } diff --git a/internal/application/service/chat_pipline/search.go b/internal/application/service/chat_pipline/search.go index d7ce4213..8b2750ba 100644 --- a/internal/application/service/chat_pipline/search.go +++ b/internal/application/service/chat_pipline/search.go @@ -299,7 +299,11 @@ func buildContentSignature(content string) string { } // searchKnowledgeBases performs KB searches for rewrite and processed queries across KB IDs -func (p *PluginSearch) searchKnowledgeBases(ctx context.Context, knowledgeBaseIDs []string, chatManage *types.ChatManage) []*types.SearchResult { +func (p *PluginSearch) searchKnowledgeBases( + ctx context.Context, + knowledgeBaseIDs []string, + chatManage *types.ChatManage, +) []*types.SearchResult { // Build base params for rewrite query baseParams := types.SearchParams{ QueryText: strings.TrimSpace(chatManage.RewriteQuery), diff --git a/internal/application/service/chat_pipline/search_entity.go b/internal/application/service/chat_pipline/search_entity.go index 3f7ffb2f..62a2dbd4 100644 --- a/internal/application/service/chat_pipline/search_entity.go +++ b/internal/application/service/chat_pipline/search_entity.go @@ -77,7 +77,13 @@ func (p *PluginSearchEntity) OnEvent(ctx context.Context, return } - logger.Infof(ctx, "KB %s entity search result count: %d nodes, %d relations", knowledgeBaseID, len(graph.Node), len(graph.Relation)) + logger.Infof( + ctx, + "KB %s entity search result count: %d nodes, %d relations", + knowledgeBaseID, + len(graph.Node), + len(graph.Relation), + ) mu.Lock() allNodes = append(allNodes, graph.Node...) @@ -109,7 +115,11 @@ func (p *PluginSearchEntity) OnEvent(ctx context.Context, for _, chunk := range chunks { knowledgeIDs = append(knowledgeIDs, chunk.KnowledgeID) } - knowledges, err := p.knowledgeRepo.GetKnowledgeBatch(ctx, ctx.Value(types.TenantIDContextKey).(uint64), knowledgeIDs) + knowledges, err := p.knowledgeRepo.GetKnowledgeBatch( + ctx, + ctx.Value(types.TenantIDContextKey).(uint64), + knowledgeIDs, + ) if err != nil { logger.Errorf(ctx, "Failed to list knowledge, session_id: %s, error: %v", chatManage.SessionID, err) return next() @@ -129,7 +139,12 @@ func (p *PluginSearchEntity) OnEvent(ctx context.Context, logger.Infof(ctx, "No new search result, session_id: %s", chatManage.SessionID) return ErrSearchNothing } - logger.Infof(ctx, "search entity result count: %d, session_id: %s", len(chatManage.SearchResult), chatManage.SessionID) + logger.Infof( + ctx, + "search entity result count: %d, session_id: %s", + len(chatManage.SearchResult), + chatManage.SessionID, + ) return next() } diff --git a/internal/application/service/chunk.go b/internal/application/service/chunk.go index 91a0cff8..b91f1250 100644 --- a/internal/application/service/chunk.go +++ b/internal/application/service/chunk.go @@ -138,7 +138,14 @@ func (s *chunkService) ListPagedChunksByKnowledgeID(ctx context.Context, knowledgeID string, page *types.Pagination, chunkType []types.ChunkType, ) (*types.PageResult, error) { tenantID := ctx.Value(types.TenantIDContextKey).(uint64) - chunks, total, err := s.chunkRepository.ListPagedChunksByKnowledgeID(ctx, tenantID, knowledgeID, page, chunkType, "") + chunks, total, err := s.chunkRepository.ListPagedChunksByKnowledgeID( + ctx, + tenantID, + knowledgeID, + page, + chunkType, + "", + ) if err != nil { logger.ErrorWithFields(ctx, err, map[string]interface{}{ "knowledge_id": knowledgeID, @@ -298,7 +305,11 @@ func (s *chunkService) DeleteByKnowledgeList(ctx context.Context, ids []string) return nil } -func (s *chunkService) ListChunkByParentID(ctx context.Context, tenantID uint64, parentID string) ([]*types.Chunk, error) { +func (s *chunkService) ListChunkByParentID( + ctx context.Context, + tenantID uint64, + parentID string, +) ([]*types.Chunk, error) { logger.Info(ctx, "Start listing chunk by parent ID") logger.Infof(ctx, "Parent ID: %s", parentID) diff --git a/internal/application/service/extract.go b/internal/application/service/extract.go index ff35479d..405a818f 100644 --- a/internal/application/service/extract.go +++ b/internal/application/service/extract.go @@ -16,7 +16,13 @@ import ( "github.com/hibiken/asynq" ) -func NewChunkExtractTask(ctx context.Context, client *asynq.Client, tenantID uint64, chunkID string, modelID string) error { +func NewChunkExtractTask( + ctx context.Context, + client *asynq.Client, + tenantID uint64, + chunkID string, + modelID string, +) error { if strings.ToLower(os.Getenv("NEO4J_ENABLE")) != "true" { logger.Warn(ctx, "NEO4J is not enabled, skip chunk extract task") return nil diff --git a/internal/application/service/knowledge.go b/internal/application/service/knowledge.go index aa4c9e7a..e5b8c57b 100644 --- a/internal/application/service/knowledge.go +++ b/internal/application/service/knowledge.go @@ -307,7 +307,13 @@ func (s *knowledgeService) CreateKnowledgeFromFile(ctx context.Context, // 即使入队失败，也返回knowledge，因为文件已保存 return knowledge, nil } - logger.Infof(ctx, "Enqueued document process task: id=%s queue=%s knowledge_id=%s", info.ID, info.Queue, knowledge.ID) + logger.Infof( + ctx, + "Enqueued document process task: id=%s queue=%s knowledge_id=%s", + info.ID, + info.Queue, + knowledge.ID, + ) logger.Infof(ctx, "Knowledge from file created successfully, ID: %s", knowledge.ID) return knowledge, nil @@ -664,7 +670,10 @@ func (s *knowledgeService) DeleteKnowledge(ctx context.Context, id string) error // Delete knowledge embeddings from vector store wg.Go(func() error { tenantInfo := ctx.Value(types.TenantInfoContextKey).(*types.Tenant) - retrieveEngine, err := retriever.NewCompositeRetrieveEngine(s.retrieveEngine, tenantInfo.RetrieverEngines.Engines) + retrieveEngine, err := retriever.NewCompositeRetrieveEngine( + s.retrieveEngine, + tenantInfo.RetrieverEngines.Engines, + ) if err != nil { logger.GetLogger(ctx).WithField("error", err).Errorf("DeleteKnowledge delete knowledge embedding failed") return err @@ -738,7 +747,10 @@ func (s *knowledgeService) DeleteKnowledgeList(ctx context.Context, ids []string // 2. Delete knowledge embeddings from vector store wg.Go(func() error { tenantInfo := ctx.Value(types.TenantInfoContextKey).(*types.Tenant) - retrieveEngine, err := retriever.NewCompositeRetrieveEngine(s.retrieveEngine, tenantInfo.RetrieverEngines.Engines) + retrieveEngine, err := retriever.NewCompositeRetrieveEngine( + s.retrieveEngine, + tenantInfo.RetrieverEngines.Engines, + ) if err != nil { logger.GetLogger(ctx).WithField("error", err).Errorf("DeleteKnowledge delete knowledge embedding failed") return err @@ -754,7 +766,9 @@ func (s *knowledgeService) DeleteKnowledgeList(ctx context.Context, ids []string return err } if err := retrieveEngine.DeleteByKnowledgeIDList(ctx, knowledgeList, embeddingModel.GetDimensions()); err != nil { - logger.GetLogger(ctx).WithField("error", err).Errorf("DeleteKnowledge delete knowledge embedding failed") + logger.GetLogger(ctx). + WithField("error", err). + Errorf("DeleteKnowledge delete knowledge embedding failed") return err } } @@ -792,7 +806,10 @@ func (s *knowledgeService) DeleteKnowledgeList(ctx context.Context, ids []string wg.Go(func() error { namespaces := []types.NameSpace{} for _, knowledge := range knowledgeList { - namespaces = append(namespaces, types.NameSpace{KnowledgeBase: knowledge.KnowledgeBaseID, Knowledge: knowledge.ID}) + namespaces = append( + namespaces, + types.NameSpace{KnowledgeBase: knowledge.KnowledgeBaseID, Knowledge: knowledge.ID}, + ) } if err := s.graphEngine.DelGraph(ctx, namespaces); err != nil { logger.GetLogger(ctx).WithField("error", err).Errorf("DeleteKnowledge delete knowledge graph failed") @@ -808,7 +825,11 @@ func (s *knowledgeService) DeleteKnowledgeList(ctx context.Context, ids []string return s.repo.DeleteKnowledgeList(ctx, tenantInfo.ID, ids) } -func (s *knowledgeService) cloneKnowledge(ctx context.Context, src *types.Knowledge, targetKB *types.KnowledgeBase) (err error) { +func (s *knowledgeService) cloneKnowledge( + ctx context.Context, + src *types.Knowledge, + targetKB *types.KnowledgeBase, +) (err error) { if src.ParseStatus != "completed" { logger.GetLogger(ctx).WithField("knowledge_id", src.ID).Errorf("MoveKnowledge parse status is not completed") return nil @@ -1657,7 +1678,12 @@ func (s *knowledgeService) updateChunkVector(ctx context.Context, kbID string, c return nil } -func (s *knowledgeService) UpdateImageInfo(ctx context.Context, knowledgeID string, chunkID string, imageInfo string) error { +func (s *knowledgeService) UpdateImageInfo( + ctx context.Context, + knowledgeID string, + chunkID string, + imageInfo string, +) error { var images []*types.ImageInfo if err := json.Unmarshal([]byte(imageInfo), &images); err != nil { logger.Errorf(ctx, "Failed to unmarshal image info: %v", err) @@ -2165,7 +2191,12 @@ func (s *knowledgeService) executeFAQImport(ctx context.Context, taskID string, if payload.Mode == types.FAQBatchModeReplace { // Replace模式：计算需要删除、创建、更新的条目 - entriesToProcess, chunksToDelete, skippedCount, err = s.calculateReplaceOperations(ctx, tenantID, faqKnowledge.ID, payload.Entries) + entriesToProcess, chunksToDelete, skippedCount, err = s.calculateReplaceOperations( + ctx, + tenantID, + faqKnowledge.ID, + payload.Entries, + ) if err != nil { return fmt.Errorf("failed to calculate replace operations: %w", err) } @@ -2193,7 +2224,14 @@ func (s *knowledgeService) executeFAQImport(ctx context.Context, taskID string, } } - logger.Infof(ctx, "FAQ import task %s: total entries: %d, to process: %d, skipped: %d", taskID, len(payload.Entries), len(entriesToProcess), skippedCount) + logger.Infof( + ctx, + "FAQ import task %s: total entries: %d, to process: %d, skipped: %d", + taskID, + len(payload.Entries), + len(entriesToProcess), + skippedCount, + ) // 如果没有需要处理的条目，直接返回 if len(entriesToProcess) == 0 { @@ -2206,7 +2244,14 @@ func (s *knowledgeService) executeFAQImport(ctx context.Context, taskID string, totalStartTime := time.Now() actualProcessed := skippedCount + processedCount - logger.Infof(ctx, "FAQ import task %s: starting batch processing, remaining entries: %d, total entries: %d, batch size: %d", taskID, remainingEntries, totalEntries, faqImportBatchSize) + logger.Infof( + ctx, + "FAQ import task %s: starting batch processing, remaining entries: %d, total entries: %d, batch size: %d", + taskID, + remainingEntries, + totalEntries, + faqImportBatchSize, + ) for i := 0; i < remainingEntries; i += faqImportBatchSize { batchStartTime := time.Now() @@ -2263,7 +2308,15 @@ func (s *knowledgeService) executeFAQImport(ctx context.Context, taskID string, return fmt.Errorf("failed to create chunks: %w", err) } createDuration := time.Since(createStartTime) - logger.Infof(ctx, "FAQ import task %s: batch %d-%d created %d chunks in %v", taskID, i+1, end, len(chunks), createDuration) + logger.Infof( + ctx, + "FAQ import task %s: batch %d-%d created %d chunks in %v", + taskID, + i+1, + end, + len(chunks), + createDuration, + ) // 索引chunks indexStartTime := time.Now() @@ -2272,7 +2325,15 @@ func (s *knowledgeService) executeFAQImport(ctx context.Context, taskID string, return fmt.Errorf("failed to index chunks: %w", err) } indexDuration := time.Since(indexStartTime) - logger.Infof(ctx, "FAQ import task %s: batch %d-%d indexed %d chunks in %v", taskID, i+1, end, len(chunks), indexDuration) + logger.Infof( + ctx, + "FAQ import task %s: batch %d-%d indexed %d chunks in %v", + taskID, + i+1, + end, + len(chunks), + indexDuration, + ) // 更新chunks的Status为已索引 chunksToUpdate := make([]*types.Chunk, 0, len(chunks)) @@ -2292,13 +2353,32 @@ func (s *knowledgeService) executeFAQImport(ctx context.Context, taskID string, } batchDuration := time.Since(batchStartTime) - logger.Infof(ctx, "FAQ import task %s: batch %d-%d completed in %v (build: %v, create: %v, index: %v), total progress: %d/%d (%d%%)", - taskID, i+1, end, batchDuration, buildDuration, createDuration, indexDuration, actualProcessed, totalEntries, progress) + logger.Infof( + ctx, + "FAQ import task %s: batch %d-%d completed in %v (build: %v, create: %v, index: %v), total progress: %d/%d (%d%%)", + taskID, + i+1, + end, + batchDuration, + buildDuration, + createDuration, + indexDuration, + actualProcessed, + totalEntries, + progress, + ) } totalDuration := time.Since(totalStartTime) - logger.Infof(ctx, "FAQ import task %s: all batches completed, processed: %d entries (skipped: %d) in %v, avg: %v per entry", - taskID, actualProcessed, skippedCount, totalDuration, totalDuration/time.Duration(actualProcessed)) + logger.Infof( + ctx, + "FAQ import task %s: all batches completed, processed: %d entries (skipped: %d) in %v, avg: %v per entry", + taskID, + actualProcessed, + skippedCount, + totalDuration, + totalDuration/time.Duration(actualProcessed), + ) return nil } @@ -2358,7 +2438,10 @@ func (s *knowledgeService) UpdateFAQEntry(ctx context.Context, if isEnabledUpdated { chunkStatusMap := map[string]bool{chunk.ID: chunk.IsEnabled} tenantInfo := ctx.Value(types.TenantInfoContextKey).(*types.Tenant) - retrieveEngine, err := retriever.NewCompositeRetrieveEngine(s.retrieveEngine, tenantInfo.RetrieverEngines.Engines) + retrieveEngine, err := retriever.NewCompositeRetrieveEngine( + s.retrieveEngine, + tenantInfo.RetrieverEngines.Engines, + ) if err != nil { return err } @@ -2464,7 +2547,10 @@ func (s *knowledgeService) UpdateFAQEntryStatusBatch(ctx context.Context, // Sync update to retriever engines if len(chunkStatusMap) > 0 { tenantInfo := ctx.Value(types.TenantInfoContextKey).(*types.Tenant) - retrieveEngine, err := retriever.NewCompositeRetrieveEngine(s.retrieveEngine, tenantInfo.RetrieverEngines.Engines) + retrieveEngine, err := retriever.NewCompositeRetrieveEngine( + s.retrieveEngine, + tenantInfo.RetrieverEngines.Engines, + ) if err != nil { return err } @@ -2839,7 +2925,11 @@ func (s *knowledgeService) validateFAQKnowledgeBase(ctx context.Context, kbID st return kb, nil } -func (s *knowledgeService) findFAQKnowledge(ctx context.Context, tenantID uint64, kbID string) (*types.Knowledge, error) { +func (s *knowledgeService) findFAQKnowledge( + ctx context.Context, + tenantID uint64, + kbID string, +) (*types.Knowledge, error) { knowledges, err := s.repo.ListKnowledgeByKnowledgeBaseID(ctx, tenantID, kbID) if err != nil { return nil, err @@ -2852,7 +2942,11 @@ func (s *knowledgeService) findFAQKnowledge(ctx context.Context, tenantID uint64 return nil, nil } -func (s *knowledgeService) ensureFAQKnowledge(ctx context.Context, tenantID uint64, kb *types.KnowledgeBase) (*types.Knowledge, error) { +func (s *knowledgeService) ensureFAQKnowledge( + ctx context.Context, + tenantID uint64, + kb *types.KnowledgeBase, +) (*types.Knowledge, error) { existing, err := s.findFAQKnowledge(ctx, tenantID, kb.ID) if err != nil { return nil, err @@ -2879,15 +2973,23 @@ func (s *knowledgeService) ensureFAQKnowledge(ctx context.Context, tenantID uint return knowledge, nil } -func (s *knowledgeService) updateFAQImportStatus(ctx context.Context, knowledgeID string, status types.FAQImportTaskStatus, - progress, total, processed int, errorMsg string, +func (s *knowledgeService) updateFAQImportStatus( + ctx context.Context, + knowledgeID string, + status types.FAQImportTaskStatus, + progress, total, processed int, + errorMsg string, ) error { return s.updateFAQImportStatusWithRanges(ctx, knowledgeID, status, progress, total, processed, errorMsg) } // updateFAQImportStatusWithRanges 更新FAQ Knowledge的导入任务状态，包含NextChunkIndex -func (s *knowledgeService) updateFAQImportStatusWithRanges(ctx context.Context, knowledgeID string, status types.FAQImportTaskStatus, - progress, total, processed int, errorMsg string, +func (s *knowledgeService) updateFAQImportStatusWithRanges( + ctx context.Context, + knowledgeID string, + status types.FAQImportTaskStatus, + progress, total, processed int, + errorMsg string, ) error { tenantID := ctx.Value(types.TenantIDContextKey).(uint64) knowledge, err := s.repo.GetKnowledgeByID(ctx, tenantID, knowledgeID) @@ -2925,7 +3027,11 @@ func (s *knowledgeService) updateFAQImportStatusWithRanges(ctx context.Context, } // getRunningFAQImportTask 获取指定知识库的进行中导入任务 -func (s *knowledgeService) getRunningFAQImportTask(ctx context.Context, kbID string, tenantID uint64) (*types.Knowledge, error) { +func (s *knowledgeService) getRunningFAQImportTask( + ctx context.Context, + kbID string, + tenantID uint64, +) (*types.Knowledge, error) { faqKnowledge, err := s.findFAQKnowledge(ctx, tenantID, kbID) if err != nil { return nil, err @@ -3023,7 +3129,11 @@ func buildFAQIndexContent(meta *types.FAQChunkMetadata, mode types.FAQIndexMode) } // buildFAQIndexInfoList 构建FAQ索引信息列表，支持分别索引模式 -func (s *knowledgeService) buildFAQIndexInfoList(ctx context.Context, kb *types.KnowledgeBase, chunk *types.Chunk) ([]*types.IndexInfo, error) { +func (s *knowledgeService) buildFAQIndexInfoList( + ctx context.Context, + kb *types.KnowledgeBase, + chunk *types.Chunk, +) ([]*types.IndexInfo, error) { indexMode := types.FAQIndexModeQuestionAnswer questionIndexMode := types.FAQQuestionIndexModeCombined if kb.FAQConfig != nil { @@ -3137,7 +3247,13 @@ func (s *knowledgeService) indexFAQChunks(ctx context.Context, chunkIDs = append(chunkIDs, chunk.ID) } buildIndexInfoDuration := time.Since(buildIndexInfoStartTime) - logger.Debugf(ctx, "indexFAQChunks: built %d index info entries for %d chunks in %v", len(indexInfo), len(chunks), buildIndexInfoDuration) + logger.Debugf( + ctx, + "indexFAQChunks: built %d index info entries for %d chunks in %v", + len(indexInfo), + len(chunks), + buildIndexInfoDuration, + ) var size int64 if adjustStorage { @@ -3195,8 +3311,16 @@ func (s *knowledgeService) indexFAQChunks(ctx context.Context, } totalDuration := time.Since(indexStartTime) - logger.Debugf(ctx, "indexFAQChunks: completed indexing %d chunks in %v (build: %v, delete: %v, batchIndex: %v, update: %v)", - len(chunks), totalDuration, buildIndexInfoDuration, deleteDuration, batchIndexDuration, updateDuration) + logger.Debugf( + ctx, + "indexFAQChunks: completed indexing %d chunks in %v (build: %v, delete: %v, batchIndex: %v, update: %v)", + len(chunks), + totalDuration, + buildIndexInfoDuration, + deleteDuration, + batchIndexDuration, + updateDuration, + ) return err } @@ -3311,7 +3435,9 @@ func (s *knowledgeService) triggerManualProcessing(ctx context.Context, Separators: kb.ChunkingConfig.Separators, EnableMultimodal: enableMultimodel, StorageConfig: &proto.StorageConfig{ - Provider: proto.StorageProvider(proto.StorageProvider_value[strings.ToUpper(kb.StorageConfig.Provider)]), + Provider: proto.StorageProvider( + proto.StorageProvider_value[strings.ToUpper(kb.StorageConfig.Provider)], + ), Region: kb.StorageConfig.Region, BucketName: kb.StorageConfig.BucketName, AccessKeyId: kb.StorageConfig.SecretID, @@ -3354,7 +3480,10 @@ func (s *knowledgeService) cleanupKnowledgeResources(ctx context.Context, knowle tenantInfo := ctx.Value(types.TenantInfoContextKey).(*types.Tenant) if knowledge.EmbeddingModelID != "" { - retrieveEngine, err := retriever.NewCompositeRetrieveEngine(s.retrieveEngine, tenantInfo.RetrieverEngines.Engines) + retrieveEngine, err := retriever.NewCompositeRetrieveEngine( + s.retrieveEngine, + tenantInfo.RetrieverEngines.Engines, + ) if err != nil { logger.GetLogger(ctx).WithField("error", err).Error("Failed to init retrieve engine during cleanup") cleanupErr = errors.Join(cleanupErr, err) @@ -3470,12 +3599,18 @@ func (s *knowledgeService) ProcessDocument(ctx context.Context, t *asynq.Task) e if knowledge.ParseStatus == "failed" { // 检查是否可恢复（例如：超时、临时错误等） // 对于不可恢复的错误，直接返回 - logger.Warnf(ctx, "Document processing previously failed: %s, error: %s", payload.KnowledgeID, knowledge.ErrorMessage) + logger.Warnf( + ctx, + "Document processing previously failed: %s, error: %s", + payload.KnowledgeID, + knowledge.ErrorMessage, + ) // 这里可以根据错误类型判断是否可恢复，暂时允许重试 } // 检查是否有部分处理（有chunks但状态不是completed） - if knowledge.ParseStatus != "completed" && knowledge.ParseStatus != "pending" && knowledge.ParseStatus != "processing" { + if knowledge.ParseStatus != "completed" && knowledge.ParseStatus != "pending" && + knowledge.ParseStatus != "processing" { // 状态异常，记录日志但继续处理 logger.Warnf(ctx, "Unexpected parse status: %s for knowledge: %s", knowledge.ParseStatus, payload.KnowledgeID) } @@ -3546,7 +3681,9 @@ func (s *knowledgeService) ProcessDocument(ctx context.Context, t *asynq.Task) e Separators: kb.ChunkingConfig.Separators, EnableMultimodal: payload.EnableMultimodel, StorageConfig: &proto.StorageConfig{ - Provider: proto.StorageProvider(proto.StorageProvider_value[strings.ToUpper(kb.StorageConfig.Provider)]), + Provider: proto.StorageProvider( + proto.StorageProvider_value[strings.ToUpper(kb.StorageConfig.Provider)], + ), Region: kb.StorageConfig.Region, BucketName: kb.StorageConfig.BucketName, AccessKeyId: kb.StorageConfig.SecretID, @@ -3721,7 +3858,10 @@ func (s *knowledgeService) ProcessFAQImport(ctx context.Context, t *asynq.Task) // 删除索引数据 embeddingModel, err := s.modelService.GetEmbeddingModel(ctx, kb.EmbeddingModelID) if err == nil { - retrieveEngine, err := retriever.NewCompositeRetrieveEngine(s.retrieveEngine, tenantInfo.RetrieverEngines.Engines) + retrieveEngine, err := retriever.NewCompositeRetrieveEngine( + s.retrieveEngine, + tenantInfo.RetrieverEngines.Engines, + ) if err == nil { chunkIDs := make([]string, 0, len(chunksDeleted)) for _, chunk := range chunksDeleted { @@ -3740,7 +3880,12 @@ func (s *knowledgeService) ProcessFAQImport(ctx context.Context, t *asynq.Task) payload.Entries = payload.Entries[processedCount:] } // Replace 模式使用hash去重，不截断payload.Entries - logger.Infof(ctx, "Continuing FAQ import from entry %d, remaining: %d entries", processedCount, len(payload.Entries)) + logger.Infof( + ctx, + "Continuing FAQ import from entry %d, remaining: %d entries", + processedCount, + len(payload.Entries), + ) } // 更新任务状态为运行中 diff --git a/internal/application/service/knowledgebase.go b/internal/application/service/knowledgebase.go index ba590510..647c0cc4 100644 --- a/internal/application/service/knowledgebase.go +++ b/internal/application/service/knowledgebase.go @@ -248,7 +248,10 @@ func (s *knowledgeBaseService) DeleteKnowledgeBase(ctx context.Context, id strin // Delete embeddings from vector store logger.Infof(ctx, "Deleting embeddings from vector store") tenantInfo := ctx.Value(types.TenantInfoContextKey).(*types.Tenant) - retrieveEngine, err := retriever.NewCompositeRetrieveEngine(s.retrieveEngine, tenantInfo.RetrieverEngines.Engines) + retrieveEngine, err := retriever.NewCompositeRetrieveEngine( + s.retrieveEngine, + tenantInfo.RetrieverEngines.Engines, + ) if err != nil { logger.Warnf(ctx, "Failed to create retrieve engine: %v", err) } else { @@ -489,7 +492,8 @@ func (s *knowledgeBaseService) HybridSearch(ctx context.Context, } // Add keyword retrieval params if supported and not FAQ - if retrieveEngine.SupportRetriever(types.KeywordsRetrieverType) && !params.DisableKeywordsMatch && kb.Type != types.KnowledgeBaseTypeFAQ { + if retrieveEngine.SupportRetriever(types.KeywordsRetrieverType) && !params.DisableKeywordsMatch && + kb.Type != types.KnowledgeBaseTypeFAQ { logger.Info(ctx, "Keyword retrieval supported, preparing keyword retrieval parameters") retrieveParams = append(retrieveParams, types.RetrieveParams{ Query: params.QueryText, @@ -537,7 +541,9 @@ func (s *knowledgeBaseService) HybridSearch(ctx context.Context, logger.Infof(ctx, "Result count before deduplication: %d", len(matchResults)) // First, try standard deduplication - deduplicatedChunks := common.DeduplicateWithScore(func(r *types.IndexWithScore) string { return r.ChunkID }, matchResults...) + deduplicatedChunks := common.DeduplicateWithScore( + func(r *types.IndexWithScore) string { return r.ChunkID }, + matchResults...) logger.Infof(ctx, "Result count after deduplication: %d", len(deduplicatedChunks)) kb.EnsureDefaults() @@ -549,7 +555,13 @@ func (s *knowledgeBaseService) HybridSearch(ctx context.Context, if needsIterativeRetrieval { logger.Info(ctx, "Not enough unique chunks, using iterative retrieval for FAQ") // Use iterative retrieval to get more unique chunks (with negative question filtering inside) - deduplicatedChunks = s.iterativeRetrieveWithDeduplication(ctx, retrieveEngine, retrieveParams, params.MatchCount, params.QueryText) + deduplicatedChunks = s.iterativeRetrieveWithDeduplication( + ctx, + retrieveEngine, + retrieveParams, + params.MatchCount, + params.QueryText, + ) } else if kb.Type == types.KnowledgeBaseTypeFAQ { // Filter by negative questions if not using iterative retrieval deduplicatedChunks = s.filterByNegativeQuestions(ctx, deduplicatedChunks, params.QueryText) @@ -606,7 +618,12 @@ func (s *knowledgeBaseService) iterativeRetrieveWithDeduplication(ctx context.Co // Check if we got fewer results than requested - means no more results available totalRetrieved := len(iterationResults) if totalRetrieved < currentTopK { - logger.Infof(ctx, "Retrieved %d results (less than TopK %d), no more results available", totalRetrieved, currentTopK) + logger.Infof( + ctx, + "Retrieved %d results (less than TopK %d), no more results available", + totalRetrieved, + currentTopK, + ) } // Deduplicate and merge (keep highest score for each chunk) @@ -631,8 +648,14 @@ func (s *knowledgeBaseService) iterativeRetrieveWithDeduplication(ctx context.Co uniqueChunks[chunk.ChunkID] = chunk } - logger.Infof(ctx, "After iteration %d: retrieved %d results, found %d unique chunks after filtering (target: %d)", - i+1, totalRetrieved, len(uniqueChunks), matchCount) + logger.Infof( + ctx, + "After iteration %d: retrieved %d results, found %d unique chunks after filtering (target: %d)", + i+1, + totalRetrieved, + len(uniqueChunks), + matchCount, + ) // Early stop: Check if we have enough unique chunks after deduplication and filtering if len(uniqueChunks) >= matchCount { diff --git a/internal/application/service/llmcontext/compression_strategies.go b/internal/application/service/llmcontext/compression_strategies.go index 249b90f7..9372724d 100644 --- a/internal/application/service/llmcontext/compression_strategies.go +++ b/internal/application/service/llmcontext/compression_strategies.go @@ -24,7 +24,11 @@ func NewSlidingWindowStrategy(recentMessageCount int) interfaces.CompressionStra // Compress implements the sliding window compression // Keeps system messages and the most recent N messages -func (s *slidingWindowStrategy) Compress(ctx context.Context, messages []chat.Message, maxTokens int) ([]chat.Message, error) { +func (s *slidingWindowStrategy) Compress( + ctx context.Context, + messages []chat.Message, + maxTokens int, +) ([]chat.Message, error) { if len(messages) <= s.recentMessageCount { return messages, nil } @@ -83,7 +87,11 @@ type smartCompressionStrategy struct { } // NewSmartCompressionStrategy creates a new smart compression strategy -func NewSmartCompressionStrategy(recentMessageCount int, chatModel chat.Chat, summarizeThreshold int) interfaces.CompressionStrategy { +func NewSmartCompressionStrategy( + recentMessageCount int, + chatModel chat.Chat, + summarizeThreshold int, +) interfaces.CompressionStrategy { return &smartCompressionStrategy{ recentMessageCount: recentMessageCount, chatModel: chatModel, @@ -93,7 +101,11 @@ func NewSmartCompressionStrategy(recentMessageCount int, chatModel chat.Chat, su // Compress implements smart compression with LLM summarization // Summarizes old messages and keeps recent messages intact -func (s *smartCompressionStrategy) Compress(ctx context.Context, messages []chat.Message, maxTokens int) ([]chat.Message, error) { +func (s *smartCompressionStrategy) Compress( + ctx context.Context, + messages []chat.Message, + maxTokens int, +) ([]chat.Message, error) { if len(messages) <= s.recentMessageCount { return messages, nil } @@ -156,8 +168,15 @@ func (s *smartCompressionStrategy) Compress(ctx context.Context, messages []chat }) result = append(result, recentMessages...) - logger.Infof(ctx, "[SmartCompression] Compressed %d messages to %d messages (summarized %d old + kept %d recent + %d system)", - len(messages), len(result), len(oldMessages), len(recentMessages), len(systemMessages)) + logger.Infof( + ctx, + "[SmartCompression] Compressed %d messages to %d messages (summarized %d old + kept %d recent + %d system)", + len(messages), + len(result), + len(oldMessages), + len(recentMessages), + len(systemMessages), + ) return result, nil } diff --git a/internal/application/service/llmcontext/context_manager.go b/internal/application/service/llmcontext/context_manager.go index 8e6bf1af..08fbb228 100644 --- a/internal/application/service/llmcontext/context_manager.go +++ b/internal/application/service/llmcontext/context_manager.go @@ -18,7 +18,11 @@ type contextManager struct { } // NewContextManager creates a new context manager with the specified storage and compression strategy -func NewContextManager(storage ContextStorage, compressionStrategy interfaces.CompressionStrategy, maxTokens int) interfaces.ContextManager { +func NewContextManager( + storage ContextStorage, + compressionStrategy interfaces.CompressionStrategy, + maxTokens int, +) interfaces.ContextManager { return &contextManager{ storage: storage, compressionStrategy: compressionStrategy, @@ -27,7 +31,10 @@ func NewContextManager(storage ContextStorage, compressionStrategy interfaces.Co } // NewContextManagerWithMemory creates a context manager with in-memory storage (for backward compatibility) -func NewContextManagerWithMemory(compressionStrategy interfaces.CompressionStrategy, maxTokens int) interfaces.ContextManager { +func NewContextManagerWithMemory( + compressionStrategy interfaces.CompressionStrategy, + maxTokens int, +) interfaces.ContextManager { return &contextManager{ storage: NewMemoryStorage(), compressionStrategy: compressionStrategy, @@ -86,7 +93,12 @@ func (cm *contextManager) AddMessage(ctx context.Context, sessionID string, mess return fmt.Errorf("failed to save context: %w", err) } - logger.Infof(ctx, "[ContextManager][Session-%s] Successfully added message (total: %d messages)", sessionID, len(messages)) + logger.Infof( + ctx, + "[ContextManager][Session-%s] Successfully added message (total: %d messages)", + sessionID, + len(messages), + ) return nil } diff --git a/internal/application/service/llmcontext/context_manager_factory.go b/internal/application/service/llmcontext/context_manager_factory.go index 8f2f85b1..195de1f5 100644 --- a/internal/application/service/llmcontext/context_manager_factory.go +++ b/internal/application/service/llmcontext/context_manager_factory.go @@ -22,7 +22,11 @@ const ( ) // NewContextManagerFromConfig creates a ContextManager based on configuration -func NewContextManagerFromConfig(contextCfg *types.ContextConfig, storage ContextStorage, chatModel chat.Chat) interfaces.ContextManager { +func NewContextManagerFromConfig( + contextCfg *types.ContextConfig, + storage ContextStorage, + chatModel chat.Chat, +) interfaces.ContextManager { // Use default values if config is nil if contextCfg == nil { logger.Info(context.TODO(), "ContextManager config not found, using default memory-based context manager") diff --git a/internal/application/service/mcp_service.go b/internal/application/service/mcp_service.go index 9fb1c32a..7bafe7d7 100644 --- a/internal/application/service/mcp_service.go +++ b/internal/application/service/mcp_service.go @@ -49,7 +49,11 @@ func (s *mcpServiceService) CreateMCPService(ctx context.Context, service *types } // GetMCPServiceByID retrieves an MCP service by ID -func (s *mcpServiceService) GetMCPServiceByID(ctx context.Context, tenantID uint64, id string) (*types.MCPService, error) { +func (s *mcpServiceService) GetMCPServiceByID( + ctx context.Context, + tenantID uint64, + id string, +) (*types.MCPService, error) { service, err := s.mcpServiceRepo.GetByID(ctx, tenantID, id) if err != nil { logger.GetLogger(ctx).Errorf("Failed to get MCP service: %v", err) @@ -80,7 +84,11 @@ func (s *mcpServiceService) ListMCPServices(ctx context.Context, tenantID uint64 } // ListMCPServicesByIDs retrieves multiple MCP services by IDs -func (s *mcpServiceService) ListMCPServicesByIDs(ctx context.Context, tenantID uint64, ids []string) ([]*types.MCPService, error) { +func (s *mcpServiceService) ListMCPServicesByIDs( + ctx context.Context, + tenantID uint64, + ids []string, +) ([]*types.MCPService, error) { if len(ids) == 0 { return []*types.MCPService{}, nil } @@ -221,7 +229,11 @@ func (s *mcpServiceService) DeleteMCPService(ctx context.Context, tenantID uint6 } // TestMCPService tests the connection to an MCP service and returns available tools/resources -func (s *mcpServiceService) TestMCPService(ctx context.Context, tenantID uint64, id string) (*types.MCPTestResult, error) { +func (s *mcpServiceService) TestMCPService( + ctx context.Context, + tenantID uint64, + id string, +) (*types.MCPTestResult, error) { // Get service service, err := s.mcpServiceRepo.GetByID(ctx, tenantID, id) if err != nil { @@ -280,15 +292,23 @@ func (s *mcpServiceService) TestMCPService(ctx context.Context, tenantID uint64, } return &types.MCPTestResult{ - Success: true, - Message: fmt.Sprintf("Connected successfully to %s v%s", initResult.ServerInfo.Name, initResult.ServerInfo.Version), + Success: true, + Message: fmt.Sprintf( + "Connected successfully to %s v%s", + initResult.ServerInfo.Name, + initResult.ServerInfo.Version, + ), Tools: tools, Resources: resources, }, nil } // GetMCPServiceTools retrieves the list of tools from an MCP service -func (s *mcpServiceService) GetMCPServiceTools(ctx context.Context, tenantID uint64, id string) ([]*types.MCPTool, error) { +func (s *mcpServiceService) GetMCPServiceTools( + ctx context.Context, + tenantID uint64, + id string, +) ([]*types.MCPTool, error) { // Get service service, err := s.mcpServiceRepo.GetByID(ctx, tenantID, id) if err != nil { @@ -314,7 +334,11 @@ func (s *mcpServiceService) GetMCPServiceTools(ctx context.Context, tenantID uin } // GetMCPServiceResources retrieves the list of resources from an MCP service -func (s *mcpServiceService) GetMCPServiceResources(ctx context.Context, tenantID uint64, id string) ([]*types.MCPResource, error) { +func (s *mcpServiceService) GetMCPServiceResources( + ctx context.Context, + tenantID uint64, + id string, +) ([]*types.MCPResource, error) { // Get service service, err := s.mcpServiceRepo.GetByID(ctx, tenantID, id) if err != nil { diff --git a/internal/application/service/metric/rouge_score.go b/internal/application/service/metric/rouge_score.go index 542ccfab..03ccc084 100644 --- a/internal/application/service/metric/rouge_score.go +++ b/internal/application/service/metric/rouge_score.go @@ -198,7 +198,10 @@ func unionLcs(evaluatedSentences []string, referenceSentence string, prevUnion * return newLcsCount, lcsUnion } -func rougeLSummaryLevel(evaluatedSentences, referenceSentences []string, rawResults, exclusive bool) map[string]float64 { +func rougeLSummaryLevel( + evaluatedSentences, referenceSentences []string, + rawResults, exclusive bool, +) map[string]float64 { referenceNgrams := NewNgrams(exclusive) referenceNgrams.BatchAdd(splitIntoWords(referenceSentences)) m := referenceNgrams.Len() diff --git a/internal/application/service/retriever/composite.go b/internal/application/service/retriever/composite.go index 03527c2a..a6f828f9 100644 --- a/internal/application/service/retriever/composite.go +++ b/internal/application/service/retriever/composite.go @@ -62,7 +62,10 @@ func (c *CompositeRetrieveEngine) Retrieve(ctx context.Context, } // NewCompositeRetrieveEngine creates a new composite retrieve engine with the given parameters -func NewCompositeRetrieveEngine(registry interfaces.RetrieveEngineRegistry, engineParams []types.RetrieverEngineParams) (*CompositeRetrieveEngine, error) { +func NewCompositeRetrieveEngine( + registry interfaces.RetrieveEngineRegistry, + engineParams []types.RetrieverEngineParams, +) (*CompositeRetrieveEngine, error) { engineInfos := make(map[types.RetrieverEngineType]*engineInfo) for _, engineParam := range engineParams { repo, err := registry.GetRetrieveEngineService(engineParam.RetrieverEngineType) @@ -100,7 +103,10 @@ func (c *CompositeRetrieveEngine) SupportRetriever(r types.RetrieverType) bool { } // BatchUpdateChunkEnabledStatus updates the enabled status of chunks in batch -func (c *CompositeRetrieveEngine) BatchUpdateChunkEnabledStatus(ctx context.Context, chunkStatusMap map[string]bool) error { +func (c *CompositeRetrieveEngine) BatchUpdateChunkEnabledStatus( + ctx context.Context, + chunkStatusMap map[string]bool, +) error { return c.concurrentExecWithError(ctx, func(ctx context.Context, engineInfo *engineInfo) error { if err := engineInfo.retrieveEngine.BatchUpdateChunkEnabledStatus(ctx, chunkStatusMap); err != nil { return err diff --git a/internal/application/service/retriever/keywords_vector_hybrid_indexer.go b/internal/application/service/retriever/keywords_vector_hybrid_indexer.go index 03279533..c8ec2da0 100644 --- a/internal/application/service/retriever/keywords_vector_hybrid_indexer.go +++ b/internal/application/service/retriever/keywords_vector_hybrid_indexer.go @@ -165,6 +165,9 @@ func (v *KeywordsVectorHybridRetrieveEngineService) CopyIndices( } // BatchUpdateChunkEnabledStatus updates the enabled status of chunks in batch -func (v *KeywordsVectorHybridRetrieveEngineService) BatchUpdateChunkEnabledStatus(ctx context.Context, chunkStatusMap map[string]bool) error { +func (v *KeywordsVectorHybridRetrieveEngineService) BatchUpdateChunkEnabledStatus( + ctx context.Context, + chunkStatusMap map[string]bool, +) error { return v.indexRepository.BatchUpdateChunkEnabledStatus(ctx, chunkStatusMap) } diff --git a/internal/application/service/session.go b/internal/application/service/session.go index aebbfb3b..8c4da289 100644 --- a/internal/application/service/session.go +++ b/internal/application/service/session.go @@ -326,7 +326,12 @@ func (s *sessionService) GenerateTitle(ctx context.Context, // GenerateTitleAsync generates a title for the session asynchronously // This method clones the session and generates the title in a goroutine // It emits an event when the title is generated -func (s *sessionService) GenerateTitleAsync(ctx context.Context, session *types.Session, userQuery string, eventBus *event.EventBus) { +func (s *sessionService) GenerateTitleAsync( + ctx context.Context, + session *types.Session, + userQuery string, + eventBus *event.EventBus, +) { // Extract values from context before cloning tenantID := ctx.Value(types.TenantIDContextKey) requestID := ctx.Value(types.RequestIDContextKey) @@ -384,8 +389,23 @@ func (s *sessionService) GenerateTitleAsync(ctx context.Context, session *types. // KnowledgeQA performs knowledge base question answering with LLM summarization // Events are emitted through eventBus (references, answer chunks, completion) -func (s *sessionService) KnowledgeQA(ctx context.Context, session *types.Session, query string, knowledgeBaseIDs []string, assistantMessageID string, summaryModelID string, webSearchEnabled bool, eventBus *event.EventBus) error { - logger.Infof(ctx, "Knowledge base question answering parameters, session ID: %s, query: %s, webSearchEnabled: %v", session.ID, query, webSearchEnabled) +func (s *sessionService) KnowledgeQA( + ctx context.Context, + session *types.Session, + query string, + knowledgeBaseIDs []string, + assistantMessageID string, + summaryModelID string, + webSearchEnabled bool, + eventBus *event.EventBus, +) error { + logger.Infof( + ctx, + "Knowledge base question answering parameters, session ID: %s, query: %s, webSearchEnabled: %v", + session.ID, + query, + webSearchEnabled, + ) // If no knowledge base IDs provided, fall back to session's default if len(knowledgeBaseIDs) == 0 { @@ -492,7 +512,12 @@ func (s *sessionService) KnowledgeQA(ctx context.Context, session *types.Session } // Create chat management object with session settings - logger.Infof(ctx, "Creating chat manage object, knowledge base IDs: %v, chat model ID: %s", knowledgeBaseIDs, chatModelID) + logger.Infof( + ctx, + "Creating chat manage object, knowledge base IDs: %v, chat model ID: %s", + knowledgeBaseIDs, + chatModelID, + ) chatManage := &types.ChatManage{ Query: query, RewriteQuery: query, @@ -563,13 +588,23 @@ func (s *sessionService) KnowledgeQA(ctx context.Context, session *types.Session // 3. First knowledge base with a Remote model // 4. Session's SummaryModelID (if not Remote) // 5. First knowledge base's SummaryModelID -func (s *sessionService) selectChatModelIDWithOverride(ctx context.Context, session *types.Session, knowledgeBaseIDs []string, summaryModelID string) (string, error) { +func (s *sessionService) selectChatModelIDWithOverride( + ctx context.Context, + session *types.Session, + knowledgeBaseIDs []string, + summaryModelID string, +) (string, error) { // First, check if request has summaryModelID override if summaryModelID != "" { // Validate that the model exists model, err := s.modelService.GetModelByID(ctx, summaryModelID) if err != nil { - logger.Warnf(ctx, "Request provided invalid summary model ID %s: %v, falling back to default selection", summaryModelID, err) + logger.Warnf( + ctx, + "Request provided invalid summary model ID %s: %v, falling back to default selection", + summaryModelID, + err, + ) } else if model != nil { logger.Infof(ctx, "Using request's summary model override: %s", summaryModelID) return summaryModelID, nil @@ -586,7 +621,11 @@ func (s *sessionService) selectChatModelIDWithOverride(ctx context.Context, sess // 2. First knowledge base with a Remote model // 3. Session's SummaryModelID (if not Remote) // 4. First knowledge base's SummaryModelID -func (s *sessionService) selectChatModelID(ctx context.Context, session *types.Session, knowledgeBaseIDs []string) (string, error) { +func (s *sessionService) selectChatModelID( + ctx context.Context, + session *types.Session, + knowledgeBaseIDs []string, +) (string, error) { // First, check if session has a SummaryModelID and if it's a Remote model if session.SummaryModelID != "" { model, err := s.modelService.GetModelByID(ctx, session.SummaryModelID) @@ -630,7 +669,12 @@ func (s *sessionService) selectChatModelID(ctx context.Context, session *types.S return "", fmt.Errorf("failed to get knowledge base %s: %w", knowledgeBaseIDs[0], err) } if kb != nil && kb.SummaryModelID != "" { - logger.Infof(ctx, "Using summary model from first knowledge base %s: %s", knowledgeBaseIDs[0], kb.SummaryModelID) + logger.Infof( + ctx, + "Using summary model from first knowledge base %s: %s", + knowledgeBaseIDs[0], + kb.SummaryModelID, + ) return kb.SummaryModelID, nil } else { logger.Errorf(ctx, "Knowledge base %s has no summary model ID", knowledgeBaseIDs[0]) @@ -639,7 +683,9 @@ func (s *sessionService) selectChatModelID(ctx context.Context, session *types.S } logger.Error(ctx, "No chat model ID available") - return "", errors.New("no chat model ID available: session has no SummaryModelID and knowledge bases have no SummaryModelID") + return "", errors.New( + "no chat model ID available: session has no SummaryModelID and knowledge bases have no SummaryModelID", + ) } // KnowledgeQAByEvent processes knowledge QA through a series of events in the pipeline @@ -674,7 +720,12 @@ func (s *sessionService) KnowledgeQAByEvent(ctx context.Context, // Handle case where search returns no results if err == chatpipline.ErrSearchNothing { - logger.Warnf(ctx, "Event %v triggered, search result is empty, using fallback response, strategy: %v", eventType, chatManage.FallbackStrategy) + logger.Warnf( + ctx, + "Event %v triggered, search result is empty, using fallback response, strategy: %v", + eventType, + chatManage.FallbackStrategy, + ) s.handleFallbackResponse(ctx, chatManage) return nil } @@ -800,7 +851,13 @@ func (s *sessionService) SearchKnowledge(ctx context.Context, } // AgentQA performs agent-based question answering with conversation history and streaming support -func (s *sessionService) AgentQA(ctx context.Context, session *types.Session, query string, assistantMessageID string, eventBus *event.EventBus) error { +func (s *sessionService) AgentQA( + ctx context.Context, + session *types.Session, + query string, + assistantMessageID string, + eventBus *event.EventBus, +) error { sessionID := session.ID tenantID := ctx.Value(types.TenantIDContextKey).(uint64) sessionJSON, err := json.Marshal(session) @@ -940,7 +997,16 @@ func (s *sessionService) AgentQA(ctx context.Context, session *types.Session, qu // Create agent engine with EventBus and ContextManager logger.Info(ctx, "Creating agent engine") - engine, err := s.agentService.CreateAgentEngine(ctx, agentConfig, summaryModel, rerankModel, eventBus, contextManager, session.ID, s) + engine, err := s.agentService.CreateAgentEngine( + ctx, + agentConfig, + summaryModel, + rerankModel, + eventBus, + contextManager, + session.ID, + s, + ) if err != nil { logger.Errorf(ctx, "Failed to create agent engine: %v", err) return err @@ -968,7 +1034,11 @@ func (s *sessionService) AgentQA(ctx context.Context, session *types.Session, qu // getContextManagerForSession creates a context manager for the session based on configuration // Returns the configured context manager (tenant-level or session-level) or default -func (s *sessionService) getContextManagerForSession(ctx context.Context, session *types.Session, chatModel chat.Chat) interfaces.ContextManager { +func (s *sessionService) getContextManagerForSession( + ctx context.Context, + session *types.Session, + chatModel chat.Chat, +) interfaces.ContextManager { // Get tenant to access global context configuration tenant, _ := ctx.Value(types.TenantInfoContextKey).(*types.Tenant) // Determine which context config to use: session-specific or tenant-level @@ -995,7 +1065,11 @@ func (s *sessionService) getContextManagerForSession(ctx context.Context, sessio } // getContextForSession retrieves LLM context for a session -func (s *sessionService) getContextForSession(ctx context.Context, contextManager interfaces.ContextManager, sessionID string) ([]chat.Message, error) { +func (s *sessionService) getContextForSession( + ctx context.Context, + contextManager interfaces.ContextManager, + sessionID string, +) ([]chat.Message, error) { history, err := contextManager.GetContext(ctx, sessionID) if err != nil { return nil, fmt.Errorf("failed to get context: %w", err) @@ -1019,7 +1093,10 @@ func (s *sessionService) ClearContext(ctx context.Context, sessionID string) err } // GetWebSearchTempKBState retrieves the temporary KB state for web search from Redis -func (s *sessionService) GetWebSearchTempKBState(ctx context.Context, sessionID string) (tempKBID string, seenURLs map[string]bool, knowledgeIDs []string) { +func (s *sessionService) GetWebSearchTempKBState( + ctx context.Context, + sessionID string, +) (tempKBID string, seenURLs map[string]bool, knowledgeIDs []string) { stateKey := fmt.Sprintf("tempkb:%s", sessionID) if raw, getErr := s.redisClient.Get(ctx, stateKey).Bytes(); getErr == nil && len(raw) > 0 { var state struct { @@ -1042,7 +1119,13 @@ func (s *sessionService) GetWebSearchTempKBState(ctx context.Context, sessionID } // SaveWebSearchTempKBState saves the temporary KB state for web search to Redis -func (s *sessionService) SaveWebSearchTempKBState(ctx context.Context, sessionID string, tempKBID string, seenURLs map[string]bool, knowledgeIDs []string) { +func (s *sessionService) SaveWebSearchTempKBState( + ctx context.Context, + sessionID string, + tempKBID string, + seenURLs map[string]bool, + knowledgeIDs []string, +) { stateKey := fmt.Sprintf("tempkb:%s", sessionID) state := struct { KBID string `json:"kbID"` @@ -1207,7 +1290,11 @@ func (s *sessionService) renderFallbackPrompt(ctx context.Context, chatManage *t } // consumeFallbackStream consumes the streaming response and emits events -func (s *sessionService) consumeFallbackStream(ctx context.Context, chatManage *types.ChatManage, responseChan <-chan types.StreamResponse) { +func (s *sessionService) consumeFallbackStream( + ctx context.Context, + chatManage *types.ChatManage, + responseChan <-chan types.StreamResponse, +) { fallbackID := generateEventID("fallback") eventBus := chatManage.EventBus var finalContent string diff --git a/internal/application/service/tag.go b/internal/application/service/tag.go index 69844d1d..6b40f55a 100644 --- a/internal/application/service/tag.go +++ b/internal/application/service/tag.go @@ -69,7 +69,13 @@ func (s *knowledgeTagService) ListTags(ctx context.Context, kbID string) ([]*typ } // CreateTag creates a new tag under a KB. -func (s *knowledgeTagService) CreateTag(ctx context.Context, kbID string, name string, color string, sortOrder int) (*types.KnowledgeTag, error) { +func (s *knowledgeTagService) CreateTag( + ctx context.Context, + kbID string, + name string, + color string, + sortOrder int, +) (*types.KnowledgeTag, error) { name = strings.TrimSpace(name) if kbID == "" || name == "" { return nil, werrors.NewBadRequestError("知识库ID和标签名称不能为空") @@ -96,7 +102,13 @@ func (s *knowledgeTagService) CreateTag(ctx context.Context, kbID string, name s } // UpdateTag updates tag basic information. -func (s *knowledgeTagService) UpdateTag(ctx context.Context, id string, name *string, color *string, sortOrder *int) (*types.KnowledgeTag, error) { +func (s *knowledgeTagService) UpdateTag( + ctx context.Context, + id string, + name *string, + color *string, + sortOrder *int, +) (*types.KnowledgeTag, error) { if id == "" { return nil, werrors.NewBadRequestError("标签ID不能为空") } diff --git a/internal/application/service/user.go b/internal/application/service/user.go index 0c7c9f54..c33c215d 100644 --- a/internal/application/service/user.go +++ b/internal/application/service/user.go @@ -52,7 +52,11 @@ type userService struct { } // NewUserService creates a new user service instance -func NewUserService(userRepo interfaces.UserRepository, tokenRepo interfaces.AuthTokenRepository, tenantService interfaces.TenantService) interfaces.UserService { +func NewUserService( + userRepo interfaces.UserRepository, + tokenRepo interfaces.AuthTokenRepository, + tenantService interfaces.TenantService, +) interfaces.UserService { return &userService{ userRepo: userRepo, tokenRepo: tokenRepo, @@ -294,7 +298,10 @@ func (s *userService) ValidatePassword(ctx context.Context, userID string, passw } // GenerateTokens generates access and refresh tokens for user -func (s *userService) GenerateTokens(ctx context.Context, user *types.User) (accessToken, refreshToken string, err error) { +func (s *userService) GenerateTokens( + ctx context.Context, + user *types.User, +) (accessToken, refreshToken string, err error) { // Generate access token (expires in 24 hours) accessClaims := jwt.MapClaims{ "user_id": user.ID, @@ -385,7 +392,10 @@ func (s *userService) ValidateToken(ctx context.Context, tokenString string) (*t } // RefreshToken refreshes access token using refresh token -func (s *userService) RefreshToken(ctx context.Context, refreshTokenString string) (accessToken, newRefreshToken string, err error) { +func (s *userService) RefreshToken( + ctx context.Context, + refreshTokenString string, +) (accessToken, newRefreshToken string, err error) { token, err := jwt.Parse(refreshTokenString, func(token *jwt.Token) (interface{}, error) { if _, ok := token.Method.(*jwt.SigningMethodHMAC); !ok { return nil, fmt.Errorf("unexpected signing method: %v", token.Header["alg"]) diff --git a/internal/application/service/web_search.go b/internal/application/service/web_search.go index 3a28bec9..043a5cf3 100644 --- a/internal/application/service/web_search.go +++ b/internal/application/service/web_search.go @@ -55,7 +55,10 @@ func (s *WebSearchService) CompressWithRAG( } createdKB, err = kbSvc.CreateKnowledgeBase(ctx, kb) if err != nil { - return nil, tempKBID, seenURLs, knowledgeIDs, fmt.Errorf("failed to create temporary knowledge base: %w", err) + return nil, tempKBID, seenURLs, knowledgeIDs, fmt.Errorf( + "failed to create temporary knowledge base: %w", + err, + ) } tempKBID = createdKB.ID } @@ -240,7 +243,11 @@ func stripMarker(content string) string { // Search performs web search using the specified provider // This method implements the interface expected by PluginSearch -func (s *WebSearchService) Search(ctx context.Context, config *types.WebSearchConfig, query string) ([]*types.WebSearchResult, error) { +func (s *WebSearchService) Search( + ctx context.Context, + config *types.WebSearchConfig, + query string, +) ([]*types.WebSearchResult, error) { if config == nil { return nil, fmt.Errorf("web search config is required") } @@ -310,7 +317,10 @@ func NewWebSearchService(cfg *config.Config) (interfaces.WebSearchService, error } // filterBlacklist filters results based on blacklist rules -func (s *WebSearchService) filterBlacklist(results []*types.WebSearchResult, blacklist []string) []*types.WebSearchResult { +func (s *WebSearchService) filterBlacklist( + results []*types.WebSearchResult, + blacklist []string, +) []*types.WebSearchResult { if len(blacklist) == 0 { return results } diff --git a/internal/application/service/web_search/duckduckgo.go b/internal/application/service/web_search/duckduckgo.go index d85b837c..296554bb 100644 --- a/internal/application/service/web_search/duckduckgo.go +++ b/internal/application/service/web_search/duckduckgo.go @@ -38,7 +38,12 @@ func (p *DuckDuckGoProvider) Name() string { } // Search performs a web search using DuckDuckGo HTML endpoint with API fallback -func (p *DuckDuckGoProvider) Search(ctx context.Context, query string, maxResults int, includeDate bool) ([]*types.WebSearchResult, error) { +func (p *DuckDuckGoProvider) Search( + ctx context.Context, + query string, + maxResults int, + includeDate bool, +) ([]*types.WebSearchResult, error) { if maxResults <= 0 { maxResults = 5 } @@ -58,7 +63,11 @@ func (p *DuckDuckGoProvider) Search(ctx context.Context, query string, maxResult return nil, fmt.Errorf("duckduckgo API search failed: %w", apiErr) } -func (p *DuckDuckGoProvider) searchHTML(ctx context.Context, query string, maxResults int) ([]*types.WebSearchResult, error) { +func (p *DuckDuckGoProvider) searchHTML( + ctx context.Context, + query string, + maxResults int, +) ([]*types.WebSearchResult, error) { baseURL := "https://html.duckduckgo.com/html/" params := url.Values{} params.Set("q", query) @@ -71,10 +80,16 @@ func (p *DuckDuckGoProvider) searchHTML(ctx context.Context, query string, maxRe return nil, fmt.Errorf("failed to create request: %w", err) } // Use a realistic UA to avoid blocks - req.Header.Set("User-Agent", "Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/120.0.0.0 Safari/537.36") + req.Header.Set( + "User-Agent", + "Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/120.0.0.0 Safari/537.36", + ) // print curl of request - curlCommand := fmt.Sprintf("curl -X GET '%s' -H 'User-Agent: Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/120.0.0.0 Safari/537.36'", req.URL.String()) + curlCommand := fmt.Sprintf( + "curl -X GET '%s' -H 'User-Agent: Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/120.0.0.0 Safari/537.36'", + req.URL.String(), + ) logger.Infof(ctx, "Curl of request: %s", secutils.SanitizeForLog(curlCommand)) resp, err := p.client.Do(req) @@ -119,7 +134,11 @@ func (p *DuckDuckGoProvider) searchHTML(ctx context.Context, query string, maxRe return results, nil } -func (p *DuckDuckGoProvider) searchAPI(ctx context.Context, query string, maxResults int) ([]*types.WebSearchResult, error) { +func (p *DuckDuckGoProvider) searchAPI( + ctx context.Context, + query string, + maxResults int, +) ([]*types.WebSearchResult, error) { baseURL := "https://api.duckduckgo.com/" params := url.Values{} params.Set("q", query) diff --git a/internal/application/service/web_search/duckduckgo_test.go b/internal/application/service/web_search/duckduckgo_test.go index 6d77bc36..c1c8c539 100644 --- a/internal/application/service/web_search/duckduckgo_test.go +++ b/internal/application/service/web_search/duckduckgo_test.go @@ -95,10 +95,12 @@ func TestDuckDuckGoProvider_Search_HTMLSuccess(t *testing.T) { if len(results) != 2 { t.Fatalf("expected 2 results, got %d", len(results)) } - if results[0].Title != "Example One" || !strings.HasPrefix(results[0].URL, "https://example.com/") || results[0].Snippet != "Snippet one" { + if results[0].Title != "Example One" || !strings.HasPrefix(results[0].URL, "https://example.com/") || + results[0].Snippet != "Snippet one" { t.Fatalf("unexpected first result: %+v", results[0]) } - if results[1].Title != "Example Two" || !strings.HasPrefix(results[1].URL, "https://example.org/") || results[1].Snippet != "Snippet two" { + if results[1].Title != "Example Two" || !strings.HasPrefix(results[1].URL, "https://example.org/") || + results[1].Snippet != "Snippet two" { t.Fatalf("unexpected second result: %+v", results[1]) } } diff --git a/internal/config/config.go b/internal/config/config.go index 38689b18..a89402b6 100644 --- a/internal/config/config.go +++ b/internal/config/config.go @@ -14,16 +14,16 @@ import ( // Config 应用程序总配置 type Config struct { - Conversation *ConversationConfig `yaml:"conversation" json:"conversation"` - Server *ServerConfig `yaml:"server" json:"server"` - KnowledgeBase *KnowledgeBaseConfig `yaml:"knowledge_base" json:"knowledge_base"` - Tenant *TenantConfig `yaml:"tenant" json:"tenant"` - Models []ModelConfig `yaml:"models" json:"models"` + Conversation *ConversationConfig `yaml:"conversation" json:"conversation"` + Server *ServerConfig `yaml:"server" json:"server"` + KnowledgeBase *KnowledgeBaseConfig `yaml:"knowledge_base" json:"knowledge_base"` + Tenant *TenantConfig `yaml:"tenant" json:"tenant"` + Models []ModelConfig `yaml:"models" json:"models"` VectorDatabase *VectorDatabaseConfig `yaml:"vector_database" json:"vector_database"` - DocReader *DocReaderConfig `yaml:"docreader" json:"docreader"` - StreamManager *StreamManagerConfig `yaml:"stream_manager" json:"stream_manager"` - ExtractManager *ExtractManagerConfig `yaml:"extract" json:"extract"` - WebSearch *WebSearchConfig `yaml:"web_search" json:"web_search"` + DocReader *DocReaderConfig `yaml:"docreader" json:"docreader"` + StreamManager *StreamManagerConfig `yaml:"stream_manager" json:"stream_manager"` + ExtractManager *ExtractManagerConfig `yaml:"extract" json:"extract"` + WebSearch *WebSearchConfig `yaml:"web_search" json:"web_search"` } type DocReaderConfig struct { @@ -36,59 +36,59 @@ type VectorDatabaseConfig struct { // ConversationConfig 对话服务配置 type ConversationConfig struct { - MaxRounds int `yaml:"max_rounds" json:"max_rounds"` - KeywordThreshold float64 `yaml:"keyword_threshold" json:"keyword_threshold"` - EmbeddingTopK int `yaml:"embedding_top_k" json:"embedding_top_k"` - VectorThreshold float64 `yaml:"vector_threshold" json:"vector_threshold"` - RerankTopK int `yaml:"rerank_top_k" json:"rerank_top_k"` - RerankThreshold float64 `yaml:"rerank_threshold" json:"rerank_threshold"` - FallbackStrategy string `yaml:"fallback_strategy" json:"fallback_strategy"` - FallbackResponse string `yaml:"fallback_response" json:"fallback_response"` - FallbackPrompt string `yaml:"fallback_prompt" json:"fallback_prompt"` - EnableRewrite bool `yaml:"enable_rewrite" json:"enable_rewrite"` - EnableQueryExpansion bool `yaml:"enable_query_expansion" json:"enable_query_expansion"` - EnableRerank bool `yaml:"enable_rerank" json:"enable_rerank"` - Summary *SummaryConfig `yaml:"summary" json:"summary"` + MaxRounds int `yaml:"max_rounds" json:"max_rounds"` + KeywordThreshold float64 `yaml:"keyword_threshold" json:"keyword_threshold"` + EmbeddingTopK int `yaml:"embedding_top_k" json:"embedding_top_k"` + VectorThreshold float64 `yaml:"vector_threshold" json:"vector_threshold"` + RerankTopK int `yaml:"rerank_top_k" json:"rerank_top_k"` + RerankThreshold float64 `yaml:"rerank_threshold" json:"rerank_threshold"` + FallbackStrategy string `yaml:"fallback_strategy" json:"fallback_strategy"` + FallbackResponse string `yaml:"fallback_response" json:"fallback_response"` + FallbackPrompt string `yaml:"fallback_prompt" json:"fallback_prompt"` + EnableRewrite bool `yaml:"enable_rewrite" json:"enable_rewrite"` + EnableQueryExpansion bool `yaml:"enable_query_expansion" json:"enable_query_expansion"` + EnableRerank bool `yaml:"enable_rerank" json:"enable_rerank"` + Summary *SummaryConfig `yaml:"summary" json:"summary"` GenerateSessionTitlePrompt string `yaml:"generate_session_title_prompt" json:"generate_session_title_prompt"` - GenerateSummaryPrompt string `yaml:"generate_summary_prompt" json:"generate_summary_prompt"` - RewritePromptSystem string `yaml:"rewrite_prompt_system" json:"rewrite_prompt_system"` - RewritePromptUser string `yaml:"rewrite_prompt_user" json:"rewrite_prompt_user"` - SimplifyQueryPrompt string `yaml:"simplify_query_prompt" json:"simplify_query_prompt"` - SimplifyQueryPromptUser string `yaml:"simplify_query_prompt_user" json:"simplify_query_prompt_user"` - ExtractEntitiesPrompt string `yaml:"extract_entities_prompt" json:"extract_entities_prompt"` - ExtractRelationshipsPrompt string `yaml:"extract_relationships_prompt" json:"extract_relationships_prompt"` + GenerateSummaryPrompt string `yaml:"generate_summary_prompt" json:"generate_summary_prompt"` + RewritePromptSystem string `yaml:"rewrite_prompt_system" json:"rewrite_prompt_system"` + RewritePromptUser string `yaml:"rewrite_prompt_user" json:"rewrite_prompt_user"` + SimplifyQueryPrompt string `yaml:"simplify_query_prompt" json:"simplify_query_prompt"` + SimplifyQueryPromptUser string `yaml:"simplify_query_prompt_user" json:"simplify_query_prompt_user"` + ExtractEntitiesPrompt string `yaml:"extract_entities_prompt" json:"extract_entities_prompt"` + ExtractRelationshipsPrompt string `yaml:"extract_relationships_prompt" json:"extract_relationships_prompt"` } // SummaryConfig 摘要配置 type SummaryConfig struct { - MaxTokens int `yaml:"max_tokens" json:"max_tokens"` - RepeatPenalty float64 `yaml:"repeat_penalty" json:"repeat_penalty"` - TopK int `yaml:"top_k" json:"top_k"` - TopP float64 `yaml:"top_p" json:"top_p"` - FrequencyPenalty float64 `yaml:"frequency_penalty" json:"frequency_penalty"` - PresencePenalty float64 `yaml:"presence_penalty" json:"presence_penalty"` - Prompt string `yaml:"prompt" json:"prompt"` - ContextTemplate string `yaml:"context_template" json:"context_template"` - Temperature float64 `yaml:"temperature" json:"temperature"` - Seed int `yaml:"seed" json:"seed"` + MaxTokens int `yaml:"max_tokens" json:"max_tokens"` + RepeatPenalty float64 `yaml:"repeat_penalty" json:"repeat_penalty"` + TopK int `yaml:"top_k" json:"top_k"` + TopP float64 `yaml:"top_p" json:"top_p"` + FrequencyPenalty float64 `yaml:"frequency_penalty" json:"frequency_penalty"` + PresencePenalty float64 `yaml:"presence_penalty" json:"presence_penalty"` + Prompt string `yaml:"prompt" json:"prompt"` + ContextTemplate string `yaml:"context_template" json:"context_template"` + Temperature float64 `yaml:"temperature" json:"temperature"` + Seed int `yaml:"seed" json:"seed"` MaxCompletionTokens int `yaml:"max_completion_tokens" json:"max_completion_tokens"` - NoMatchPrefix string `yaml:"no_match_prefix" json:"no_match_prefix"` + NoMatchPrefix string `yaml:"no_match_prefix" json:"no_match_prefix"` } // ServerConfig 服务器配置 type ServerConfig struct { - Port int `yaml:"port" json:"port"` - Host string `yaml:"host" json:"host"` - LogPath string `yaml:"log_path" json:"log_path"` + Port int `yaml:"port" json:"port"` + Host string `yaml:"host" json:"host"` + LogPath string `yaml:"log_path" json:"log_path"` ShutdownTimeout time.Duration `yaml:"shutdown_timeout" json:"shutdown_timeout" default:"30s"` } // KnowledgeBaseConfig 知识库配置 type KnowledgeBaseConfig struct { - ChunkSize int `yaml:"chunk_size" json:"chunk_size"` - ChunkOverlap int `yaml:"chunk_overlap" json:"chunk_overlap"` - SplitMarkers []string `yaml:"split_markers" json:"split_markers"` - KeepSeparator bool `yaml:"keep_separator" json:"keep_separator"` + ChunkSize int `yaml:"chunk_size" json:"chunk_size"` + ChunkOverlap int `yaml:"chunk_overlap" json:"chunk_overlap"` + SplitMarkers []string `yaml:"split_markers" json:"split_markers"` + KeepSeparator bool `yaml:"keep_separator" json:"keep_separator"` ImageProcessing *ImageProcessingConfig `yaml:"image_processing" json:"image_processing"` } @@ -99,44 +99,44 @@ type ImageProcessingConfig struct { // TenantConfig 租户配置 type TenantConfig struct { - DefaultSessionName string `yaml:"default_session_name" json:"default_session_name"` - DefaultSessionTitle string `yaml:"default_session_title" json:"default_session_title"` + DefaultSessionName string `yaml:"default_session_name" json:"default_session_name"` + DefaultSessionTitle string `yaml:"default_session_title" json:"default_session_title"` DefaultSessionDescription string `yaml:"default_session_description" json:"default_session_description"` } // ModelConfig 模型配置 type ModelConfig struct { - Type string `yaml:"type" json:"type"` - Source string `yaml:"source" json:"source"` + Type string `yaml:"type" json:"type"` + Source string `yaml:"source" json:"source"` ModelName string `yaml:"model_name" json:"model_name"` Parameters map[string]interface{} `yaml:"parameters" json:"parameters"` } // StreamManagerConfig 流管理器配置 type StreamManagerConfig struct { - Type string `yaml:"type" json:"type"` // 类型: "memory" 或 "redis" - Redis RedisConfig `yaml:"redis" json:"redis"` // Redis配置 + Type string `yaml:"type" json:"type"` // 类型: "memory" 或 "redis" + Redis RedisConfig `yaml:"redis" json:"redis"` // Redis配置 CleanupTimeout time.Duration `yaml:"cleanup_timeout" json:"cleanup_timeout"` // 清理超时，单位秒 } // RedisConfig Redis配置 type RedisConfig struct { - Address string `yaml:"address" json:"address"` // Redis地址 + Address string `yaml:"address" json:"address"` // Redis地址 Password string `yaml:"password" json:"password"` // Redis密码 - DB int `yaml:"db" json:"db"` // Redis数据库 - Prefix string `yaml:"prefix" json:"prefix"` // 键前缀 - TTL time.Duration `yaml:"ttl" json:"ttl"` // 过期时间(小时) + DB int `yaml:"db" json:"db"` // Redis数据库 + Prefix string `yaml:"prefix" json:"prefix"` // 键前缀 + TTL time.Duration `yaml:"ttl" json:"ttl"` // 过期时间(小时) } // ExtractManagerConfig 抽取管理器配置 type ExtractManagerConfig struct { - ExtractGraph *types.PromptTemplateStructured `yaml:"extract_graph" json:"extract_graph"` + ExtractGraph *types.PromptTemplateStructured `yaml:"extract_graph" json:"extract_graph"` ExtractEntity *types.PromptTemplateStructured `yaml:"extract_entity" json:"extract_entity"` - FabriText *FebriText `yaml:"fabri_text" json:"fabri_text"` + FabriText *FebriText `yaml:"fabri_text" json:"fabri_text"` } type FebriText struct { - WithTag string `yaml:"with_tag" json:"with_tag"` + WithTag string `yaml:"with_tag" json:"with_tag"` WithNoTag string `yaml:"with_no_tag" json:"with_no_tag"` } @@ -194,25 +194,25 @@ func LoadConfig() (*Config, error) { // WebSearchConfig represents the web search configuration type WebSearchConfig struct { Providers []WebSearchProviderConfig `yaml:"providers" json:"providers"` - Default WebSearchDefaultConfig `yaml:"default" json:"default"` - Timeout int `yaml:"timeout" json:"timeout"` // 超时时间（秒） + Default WebSearchDefaultConfig `yaml:"default" json:"default"` + Timeout int `yaml:"timeout" json:"timeout"` // 超时时间（秒） } // WebSearchProviderConfig represents configuration for a web search provider type WebSearchProviderConfig struct { - ID string `yaml:"id" json:"id"` - Name string `yaml:"name" json:"name"` - Free bool `yaml:"free" json:"free"` - RequiresAPIKey bool `yaml:"requires_api_key" json:"requires_api_key"` + ID string `yaml:"id" json:"id"` + Name string `yaml:"name" json:"name"` + Free bool `yaml:"free" json:"free"` + RequiresAPIKey bool `yaml:"requires_api_key" json:"requires_api_key"` Description string `yaml:"description,omitempty" json:"description,omitempty"` - APIURL string `yaml:"api_url,omitempty" json:"api_url,omitempty"` + APIURL string `yaml:"api_url,omitempty" json:"api_url,omitempty"` } // WebSearchDefaultConfig represents the default web search configuration type WebSearchDefaultConfig struct { - Provider string `yaml:"provider" json:"provider"` - MaxResults int `yaml:"max_results" json:"max_results"` - IncludeDate bool `yaml:"include_date" json:"include_date"` + Provider string `yaml:"provider" json:"provider"` + MaxResults int `yaml:"max_results" json:"max_results"` + IncludeDate bool `yaml:"include_date" json:"include_date"` CompressionMethod string `yaml:"compression_method" json:"compression_method"` - Blacklist []string `yaml:"blacklist" json:"blacklist"` + Blacklist []string `yaml:"blacklist" json:"blacklist"` } diff --git a/internal/container/container.go b/internal/container/container.go index d4e67eb5..f1a024a1 100644 --- a/internal/container/container.go +++ b/internal/container/container.go @@ -282,7 +282,10 @@ func initDatabase(cfg *config.Config) (*gorm.DB, error) { if err := database.RunMigrations(migrateDSN); err != nil { // Log warning but don't fail startup - migrations might be handled externally logger.Warnf(context.Background(), "Database migration failed: %v", err) - logger.Warnf(context.Background(), "Continuing with application startup. Please run migrations manually if needed.") + logger.Warnf( + context.Background(), + "Continuing with application startup. Please run migrations manually if needed.", + ) } } else { logger.Infof(context.Background(), "Auto-migration is disabled (AUTO_MIGRATE=false)") diff --git a/internal/handler/chunk.go b/internal/handler/chunk.go index 474c4fa0..999c8b58 100644 --- a/internal/handler/chunk.go +++ b/internal/handler/chunk.go @@ -189,7 +189,10 @@ func (h *ChunkHandler) validateAndGetChunk(c *gin.Context) (*types.Chunk, string logger.Warnf( ctx, "Tenant has no permission to access chunk, knowledge ID: %s, chunk ID: %s, req tenant: %d, chunk tenant: %d", - knowledgeID, id, tenantID, chunk.TenantID, + knowledgeID, + id, + tenantID, + chunk.TenantID, ) return nil, knowledgeID, errors.NewForbiddenError("No permission to access this chunk") } diff --git a/internal/handler/initialization.go b/internal/handler/initialization.go index df19cf6c..3c867631 100644 --- a/internal/handler/initialization.go +++ b/internal/handler/initialization.go @@ -85,7 +85,7 @@ func NewInitializationHandler( // KBModelConfigRequest 知识库模型配置请求（简化版，只传模型ID） type KBModelConfigRequest struct { - LLMModelID string `json:"llmModelId" binding:"required"` + LLMModelID string `json:"llmModelId" binding:"required"` EmbeddingModelID string `json:"embeddingModelId" binding:"required"` VLMConfig *types.VLMConfig `json:"vlm_config"` @@ -357,7 +357,12 @@ func (h *InitializationHandler) InitializeByKB(c *gin.Context) { c.Error(errors.NewBadRequestError(err.Error())) return } - logger.Infof(ctx, "Starting knowledge base configuration update, kbId: %s, request: %s", utils.SanitizeForLog(kbIdStr), utils.SanitizeForLog(utils.ToJSON(req))) + logger.Infof( + ctx, + "Starting knowledge base configuration update, kbId: %s, request: %s", + utils.SanitizeForLog(kbIdStr), + utils.SanitizeForLog(utils.ToJSON(req)), + ) // 获取指定知识库信息 kb, err := h.kbService.GetKnowledgeBaseByID(ctx, kbIdStr) @@ -1253,7 +1258,7 @@ func (h *InitializationHandler) buildConfigResponse(ctx context.Context, models // RemoteModelCheckRequest 远程模型检查请求结构 type RemoteModelCheckRequest struct { ModelName string `json:"modelName" binding:"required"` - BaseURL string `json:"baseUrl" binding:"required"` + BaseURL string `json:"baseUrl" binding:"required"` APIKey string `json:"apiKey"` } @@ -1764,8 +1769,8 @@ func (h *InitializationHandler) testMultimodalWithDocReader( // TextRelationExtractionRequest 文本关系提取请求结构 type TextRelationExtractionRequest struct { - Text string `json:"text" binding:"required"` - Tags []string `json:"tags" binding:"required"` + Text string `json:"text" binding:"required"` + Tags []string `json:"tags" binding:"required"` LLMConfig LLMConfig `json:"llmConfig"` } @@ -1824,7 +1829,12 @@ func (h *InitializationHandler) ExtractTextRelations(c *gin.Context) { } // extractRelationsFromText 从文本中提取关系 -func (h *InitializationHandler) extractRelationsFromText(ctx context.Context, text string, tags []string, llm LLMConfig) (*TextRelationExtractionResponse, error) { +func (h *InitializationHandler) extractRelationsFromText( + ctx context.Context, + text string, + tags []string, + llm LLMConfig, +) (*TextRelationExtractionResponse, error) { chatModel, err := chat.NewChat(&chat.ChatConfig{ ModelID: "initialization", APIKey: llm.ApiKey, diff --git a/internal/handler/knowledge.go b/internal/handler/knowledge.go index eb2446a7..5f73f375 100644 --- a/internal/handler/knowledge.go +++ b/internal/handler/knowledge.go @@ -155,7 +155,12 @@ func (h *KnowledgeHandler) CreateKnowledgeFromFile(c *gin.Context) { return } - logger.Infof(ctx, "Knowledge created successfully, ID: %s, title: %s", secutils.SanitizeForLog(knowledge.ID), secutils.SanitizeForLog(knowledge.Title)) + logger.Infof( + ctx, + "Knowledge created successfully, ID: %s, title: %s", + secutils.SanitizeForLog(knowledge.ID), + secutils.SanitizeForLog(knowledge.Title), + ) c.JSON(http.StatusOK, gin.H{ "success": true, "data": knowledge, @@ -186,7 +191,12 @@ func (h *KnowledgeHandler) CreateKnowledgeFromURL(c *gin.Context) { } logger.Infof(ctx, "Received URL request: %s", secutils.SanitizeForLog(req.URL)) - logger.Infof(ctx, "Creating knowledge from URL, knowledge base ID: %s, URL: %s", secutils.SanitizeForLog(kbID), secutils.SanitizeForLog(req.URL)) + logger.Infof( + ctx, + "Creating knowledge from URL, knowledge base ID: %s, URL: %s", + secutils.SanitizeForLog(kbID), + secutils.SanitizeForLog(req.URL), + ) // Create knowledge entry from the URL knowledge, err := h.kgService.CreateKnowledgeFromURL(ctx, kbID, req.URL, req.EnableMultimodel) @@ -200,7 +210,12 @@ func (h *KnowledgeHandler) CreateKnowledgeFromURL(c *gin.Context) { return } - logger.Infof(ctx, "Knowledge created successfully from URL, ID: %s, title: %s", secutils.SanitizeForLog(knowledge.ID), secutils.SanitizeForLog(knowledge.Title)) + logger.Infof( + ctx, + "Knowledge created successfully from URL, ID: %s, title: %s", + secutils.SanitizeForLog(knowledge.ID), + secutils.SanitizeForLog(knowledge.Title), + ) c.JSON(http.StatusCreated, gin.H{ "success": true, "data": knowledge, @@ -268,7 +283,12 @@ func (h *KnowledgeHandler) GetKnowledge(c *gin.Context) { return } - logger.Infof(ctx, "Knowledge retrieved successfully, ID: %s, title: %s", secutils.SanitizeForLog(knowledge.ID), secutils.SanitizeForLog(knowledge.Title)) + logger.Infof( + ctx, + "Knowledge retrieved successfully, ID: %s, title: %s", + secutils.SanitizeForLog(knowledge.ID), + secutils.SanitizeForLog(knowledge.Title), + ) c.JSON(http.StatusOK, gin.H{ "success": true, "data": knowledge, @@ -299,8 +319,14 @@ func (h *KnowledgeHandler) ListKnowledge(c *gin.Context) { tagID := c.Query("tag_id") - logger.Infof(ctx, "Retrieving knowledge list under knowledge base, knowledge base ID: %s, tag_id: %s, page: %d, page size: %d", - secutils.SanitizeForLog(kbID), secutils.SanitizeForLog(tagID), pagination.Page, pagination.PageSize) + logger.Infof( + ctx, + "Retrieving knowledge list under knowledge base, knowledge base ID: %s, tag_id: %s, page: %d, page size: %d", + secutils.SanitizeForLog(kbID), + secutils.SanitizeForLog(tagID), + pagination.Page, + pagination.PageSize, + ) // Retrieve paginated knowledge entries result, err := h.kgService.ListPagedKnowledgeByKnowledgeBaseID(ctx, kbID, &pagination, tagID) @@ -310,7 +336,12 @@ func (h *KnowledgeHandler) ListKnowledge(c *gin.Context) { return } - logger.Infof(ctx, "Knowledge list retrieved successfully, knowledge base ID: %s, total: %d", secutils.SanitizeForLog(kbID), result.Total) + logger.Infof( + ctx, + "Knowledge list retrieved successfully, knowledge base ID: %s, total: %d", + secutils.SanitizeForLog(kbID), + result.Total, + ) c.JSON(http.StatusOK, gin.H{ "success": true, "data": result.Data, @@ -374,7 +405,12 @@ func (h *KnowledgeHandler) DownloadKnowledgeFile(c *gin.Context) { } defer file.Close() - logger.Infof(ctx, "Knowledge file retrieved successfully, ID: %s, filename: %s", secutils.SanitizeForLog(id), secutils.SanitizeForLog(filename)) + logger.Infof( + ctx, + "Knowledge file retrieved successfully, ID: %s, filename: %s", + secutils.SanitizeForLog(id), + secutils.SanitizeForLog(filename), + ) // Set response headers for file download c.Header("Content-Description", "File Transfer") diff --git a/internal/handler/knowledgebase.go b/internal/handler/knowledgebase.go index b648a5fe..4ab66bdb 100644 --- a/internal/handler/knowledgebase.go +++ b/internal/handler/knowledgebase.go @@ -178,9 +178,9 @@ func (h *KnowledgeBaseHandler) ListKnowledgeBases(c *gin.Context) { // UpdateKnowledgeBaseRequest defines the request body structure for updating a knowledge base type UpdateKnowledgeBaseRequest struct { - Name string `json:"name" binding:"required"` + Name string `json:"name" binding:"required"` Description string `json:"description"` - Config *types.KnowledgeBaseConfig `json:"config" binding:"required"` + Config *types.KnowledgeBaseConfig `json:"config" binding:"required"` } // UpdateKnowledgeBase handles requests to update an existing knowledge base diff --git a/internal/handler/model.go b/internal/handler/model.go index 2666bb50..bae77569 100644 --- a/internal/handler/model.go +++ b/internal/handler/model.go @@ -61,11 +61,11 @@ func hideSensitiveInfo(model *types.Model) *types.Model { // CreateModelRequest defines the structure for model creation requests // Contains all fields required to create a new model in the system type CreateModelRequest struct { - Name string `json:"name" binding:"required"` - Type types.ModelType `json:"type" binding:"required"` - Source types.ModelSource `json:"source" binding:"required"` + Name string `json:"name" binding:"required"` + Type types.ModelType `json:"type" binding:"required"` + Source types.ModelSource `json:"source" binding:"required"` Description string `json:"description"` - Parameters types.ModelParameters `json:"parameters" binding:"required"` + Parameters types.ModelParameters `json:"parameters" binding:"required"` } // CreateModel handles the HTTP request to create a new model @@ -109,7 +109,12 @@ func (h *ModelHandler) CreateModel(c *gin.Context) { return } - logger.Infof(ctx, "Model created successfully, ID: %s, Name: %s", secutils.SanitizeForLog(model.ID), secutils.SanitizeForLog(model.Name)) + logger.Infof( + ctx, + "Model created successfully, ID: %s, Name: %s", + secutils.SanitizeForLog(model.ID), + secutils.SanitizeForLog(model.Name), + ) // Hide sensitive information for builtin models (though newly created models are unlikely to be builtin) responseModel := hideSensitiveInfo(model) diff --git a/internal/handler/session/qa.go b/internal/handler/session/qa.go index 5fee0a7a..a6f4db25 100644 --- a/internal/handler/session/qa.go +++ b/internal/handler/session/qa.go @@ -104,7 +104,12 @@ func (h *Handler) KnowledgeQA(c *gin.Context) { return } - logger.Infof(ctx, "Knowledge QA request, session ID: %s, query: %s", sessionID, secutils.SanitizeForLog(request.Query)) + logger.Infof( + ctx, + "Knowledge QA request, session ID: %s, query: %s", + sessionID, + secutils.SanitizeForLog(request.Query), + ) // Get session to prepare knowledge base IDs session, err := h.sessionService.GetSession(ctx, sessionID) @@ -118,7 +123,11 @@ func (h *Handler) KnowledgeQA(c *gin.Context) { knowledgeBaseIDs := request.KnowledgeBaseIDs if len(knowledgeBaseIDs) == 0 && session.KnowledgeBaseID != "" { knowledgeBaseIDs = []string{session.KnowledgeBaseID} - logger.Infof(ctx, "No knowledge base IDs in request, using session default: %s", secutils.SanitizeForLog(session.KnowledgeBaseID)) + logger.Infof( + ctx, + "No knowledge base IDs in request, using session default: %s", + secutils.SanitizeForLog(session.KnowledgeBaseID), + ) } // Use shared function to handle KnowledgeQA request @@ -284,21 +293,42 @@ func (h *Handler) AgentQA(c *gin.Context) { // If still empty, use session default knowledge base if len(knowledgeBaseIDs) == 0 && session.KnowledgeBaseID != "" { knowledgeBaseIDs = []string{session.KnowledgeBaseID} - logger.Infof(ctx, "Using session default knowledge base: %s", secutils.SanitizeForLog(session.KnowledgeBaseID)) + logger.Infof( + ctx, + "Using session default knowledge base: %s", + secutils.SanitizeForLog(session.KnowledgeBaseID), + ) } // Validate at least one knowledge base is available if len(knowledgeBaseIDs) == 0 { logger.Error(ctx, "No knowledge base available for delegation") - c.Error(errors.NewBadRequestError("No knowledge base available. Please configure at least one knowledge base.")) + c.Error( + errors.NewBadRequestError("No knowledge base available. Please configure at least one knowledge base."), + ) return } - logger.Infof(ctx, "Delegating to KnowledgeQA with knowledge bases: %s", secutils.SanitizeForLog(fmt.Sprintf("%v", knowledgeBaseIDs))) + logger.Infof( + ctx, + "Delegating to KnowledgeQA with knowledge bases: %s", + secutils.SanitizeForLog(fmt.Sprintf("%v", knowledgeBaseIDs)), + ) // Use shared function to handle KnowledgeQA request (no title generation for AgentQA fallback) - h.handleKnowledgeQARequest(ctx, c, session, secutils.SanitizeForLog(request.Query), - secutils.SanitizeForLogArray(knowledgeBaseIDs), assistantMessage, false, secutils.SanitizeForLog(request.SummaryModelID), request.WebSearchEnabled) + h.handleKnowledgeQARequest( + ctx, + c, + session, + secutils.SanitizeForLog(request.Query), + secutils.SanitizeForLogArray( + knowledgeBaseIDs, + ), + assistantMessage, + false, + secutils.SanitizeForLog(request.SummaryModelID), + request.WebSearchEnabled, + ) return } @@ -369,7 +399,13 @@ func (h *Handler) AgentQA(c *gin.Context) { h.completeAssistantMessage(asyncCtx, assistantMessage) logger.Infof(asyncCtx, "Agent QA service completed for session: %s", sessionID) }() - err := h.sessionService.AgentQA(asyncCtx, session, secutils.SanitizeForLog(request.Query), assistantMessage.ID, eventBus) + err := h.sessionService.AgentQA( + asyncCtx, + session, + secutils.SanitizeForLog(request.Query), + assistantMessage.ID, + eventBus, + ) if err != nil { logger.ErrorWithFields(asyncCtx, err, nil) // Emit error event to dedicated EventBus @@ -479,10 +515,23 @@ func (h *Handler) handleKnowledgeQARequest( if r := recover(); r != nil { buf := make([]byte, 10240) runtime.Stack(buf, true) - logger.ErrorWithFields(asyncCtx, errors.NewInternalServerError(fmt.Sprintf("Knowledge QA service panicked: %v\n%s", r, string(buf))), nil) + logger.ErrorWithFields( + asyncCtx, + errors.NewInternalServerError(fmt.Sprintf("Knowledge QA service panicked: %v\n%s", r, string(buf))), + nil, + ) } }() - err := h.sessionService.KnowledgeQA(asyncCtx, session, query, knowledgeBaseIDs, assistantMessage.ID, summaryModelID, webSearchEnabled, eventBus) + err := h.sessionService.KnowledgeQA( + asyncCtx, + session, + query, + knowledgeBaseIDs, + assistantMessage.ID, + summaryModelID, + webSearchEnabled, + eventBus, + ) if err != nil { logger.ErrorWithFields(asyncCtx, err, nil) // Emit error event to dedicated EventBus diff --git a/internal/handler/session/stream.go b/internal/handler/session/stream.go index a14e91c1..0a21bb47 100644 --- a/internal/handler/session/stream.go +++ b/internal/handler/session/stream.go @@ -290,7 +290,11 @@ func (h *Handler) handleAgentEventsForSSE( select { case <-c.Request.Context().Done(): // Connection closed, exit gracefully without panic - log.Infof("Client disconnected, stopping SSE streaming for session=%s, message=%s", sessionID, assistantMessageID) + log.Infof( + "Client disconnected, stopping SSE streaming for session=%s, message=%s", + sessionID, + assistantMessageID, + ) return case <-ticker.C: diff --git a/internal/handler/session/types.go b/internal/handler/session/types.go index cd0ae1aa..b4aa6f81 100644 --- a/internal/handler/session/types.go +++ b/internal/handler/session/types.go @@ -53,16 +53,16 @@ type GenerateTitleRequest struct { // CreateKnowledgeQARequest defines the request structure for knowledge QA type CreateKnowledgeQARequest struct { - Query string `json:"query" binding:"required"` // Query text for knowledge base search - KnowledgeBaseIDs []string `json:"knowledge_base_ids"` // Selected knowledge base ID for this request - AgentEnabled bool `json:"agent_enabled"` // Whether agent mode is enabled for this request - WebSearchEnabled bool `json:"web_search_enabled"` // Whether web search is enabled for this request - SummaryModelID string `json:"summary_model_id"` // Optional summary model ID for this request (overrides session default) + Query string `json:"query" binding:"required"` // Query text for knowledge base search + KnowledgeBaseIDs []string `json:"knowledge_base_ids"` // Selected knowledge base ID for this request + AgentEnabled bool `json:"agent_enabled"` // Whether agent mode is enabled for this request + WebSearchEnabled bool `json:"web_search_enabled"` // Whether web search is enabled for this request + SummaryModelID string `json:"summary_model_id"` // Optional summary model ID for this request (overrides session default) } // SearchKnowledgeRequest defines the request structure for searching knowledge without LLM summarization type SearchKnowledgeRequest struct { - Query string `json:"query" binding:"required"` // Query text to search for + Query string `json:"query" binding:"required"` // Query text to search for KnowledgeBaseID string `json:"knowledge_base_id" binding:"required"` // ID of the knowledge base to search } diff --git a/internal/handler/tag.go b/internal/handler/tag.go index 7b2209b8..3d719616 100644 --- a/internal/handler/tag.go +++ b/internal/handler/tag.go @@ -40,7 +40,7 @@ func (h *TagHandler) ListTags(c *gin.Context) { } type createTagRequest struct { - Name string `json:"name" binding:"required"` + Name string `json:"name" binding:"required"` Color string `json:"color"` SortOrder int `json:"sort_order"` } diff --git a/internal/handler/tenant.go b/internal/handler/tenant.go index 47cd34e5..7999d337 100644 --- a/internal/handler/tenant.go +++ b/internal/handler/tenant.go @@ -70,7 +70,12 @@ func (h *TenantHandler) CreateTenant(c *gin.Context) { return } - logger.Infof(ctx, "Tenant created successfully, ID: %d, name: %s", createdTenant.ID, secutils.SanitizeForLog(createdTenant.Name)) + logger.Infof( + ctx, + "Tenant created successfully, ID: %d, name: %s", + createdTenant.ID, + secutils.SanitizeForLog(createdTenant.Name), + ) c.JSON(http.StatusCreated, gin.H{ "success": true, "data": createdTenant, @@ -151,7 +156,12 @@ func (h *TenantHandler) UpdateTenant(c *gin.Context) { return } - logger.Infof(ctx, "Tenant updated successfully, ID: %d, Name: %s", updatedTenant.ID, secutils.SanitizeForLog(updatedTenant.Name)) + logger.Infof( + ctx, + "Tenant updated successfully, ID: %d, Name: %s", + updatedTenant.ID, + secutils.SanitizeForLog(updatedTenant.Name), + ) c.JSON(http.StatusOK, gin.H{ "success": true, "data": updatedTenant, diff --git a/internal/middleware/auth.go b/internal/middleware/auth.go index d09efa90..1069e716 100644 --- a/internal/middleware/auth.go +++ b/internal/middleware/auth.go @@ -38,7 +38,11 @@ func isNoAuthAPI(path string, method string) bool { } // Auth 认证中间件 -func Auth(tenantService interfaces.TenantService, userService interfaces.UserService, cfg *config.Config) gin.HandlerFunc { +func Auth( + tenantService interfaces.TenantService, + userService interfaces.UserService, + cfg *config.Config, +) gin.HandlerFunc { return func(c *gin.Context) { // ignore OPTIONS request if c.Request.Method == "OPTIONS" { diff --git a/internal/models/chat/remote_api.go b/internal/models/chat/remote_api.go index 0cf8831b..fc21ddf0 100644 --- a/internal/models/chat/remote_api.go +++ b/internal/models/chat/remote_api.go @@ -263,7 +263,11 @@ func (c *RemoteAPIChat) Chat(ctx context.Context, messages []Message, opts *Chat } // chatWithQwen 使用自定义请求处理 qwen 模型 -func (c *RemoteAPIChat) chatWithQwen(ctx context.Context, messages []Message, opts *ChatOptions) (*types.ChatResponse, error) { +func (c *RemoteAPIChat) chatWithQwen( + ctx context.Context, + messages []Message, + opts *ChatOptions, +) (*types.ChatResponse, error) { // 构建 qwen 请求参数 req := c.buildQwenChatCompletionRequest(messages, opts, false) diff --git a/internal/models/embedding/openai.go b/internal/models/embedding/openai.go index 37e56f26..b1db6e94 100644 --- a/internal/models/embedding/openai.go +++ b/internal/models/embedding/openai.go @@ -103,7 +103,8 @@ func (e *OpenAIEmbedder) doRequestWithRetry(ctx context.Context, jsonData []byte if backoffTime > 10*time.Second { backoffTime = 10 * time.Second } - logger.GetLogger(ctx).Infof("OpenAIEmbedder retrying request (%d/%d), waiting %v", i, e.maxRetries, backoffTime) + logger.GetLogger(ctx). + Infof("OpenAIEmbedder retrying request (%d/%d), waiting %v", i, e.maxRetries, backoffTime) select { case <-time.After(backoffTime): diff --git a/internal/models/rerank/reranker.go b/internal/models/rerank/reranker.go index a236d199..4ff6071a 100644 --- a/internal/models/rerank/reranker.go +++ b/internal/models/rerank/reranker.go @@ -88,7 +88,10 @@ type RerankerConfig struct { // NewReranker creates a reranker func NewReranker(config *RerankerConfig) (Reranker, error) { // 根据URL判断模型来源，而不是依赖Source字段 - if strings.Contains(config.BaseURL, "https://dashscope.aliyuncs.com/api/v1/services/rerank/text-rerank/text-rerank") { + if strings.Contains( + config.BaseURL, + "https://dashscope.aliyuncs.com/api/v1/services/rerank/text-rerank/text-rerank", + ) { return NewAliyunReranker(config) } else { return NewOpenAIReranker(config) diff --git a/internal/models/utils/ollama/ollama.go b/internal/models/utils/ollama/ollama.go index 2e5abd9f..5665aacf 100644 --- a/internal/models/utils/ollama/ollama.go +++ b/internal/models/utils/ollama/ollama.go @@ -186,7 +186,8 @@ func (s *OllamaService) EnsureModelAvailable(ctx context.Context, modelName stri available, err := s.IsModelAvailable(ctx, modelName) if err != nil { if s.isOptional { - logger.GetLogger(ctx).Warnf("Failed to check model %s availability, but Ollama is set as optional", modelName) + logger.GetLogger(ctx). + Warnf("Failed to check model %s availability, but Ollama is set as optional", modelName) return nil } return err diff --git a/internal/stream/memory_manager.go b/internal/stream/memory_manager.go index c4485696..e30ac4f8 100644 --- a/internal/stream/memory_manager.go +++ b/internal/stream/memory_manager.go @@ -60,7 +60,11 @@ func (m *MemoryStreamManager) getStream(sessionID, messageID string) *memoryStre } // AppendEvent appends a single event to the stream -func (m *MemoryStreamManager) AppendEvent(ctx context.Context, sessionID, messageID string, event interfaces.StreamEvent) error { +func (m *MemoryStreamManager) AppendEvent( + ctx context.Context, + sessionID, messageID string, + event interfaces.StreamEvent, +) error { stream := m.getOrCreateStream(sessionID, messageID) stream.mu.Lock() @@ -80,7 +84,11 @@ func (m *MemoryStreamManager) AppendEvent(ctx context.Context, sessionID, messag // GetEvents gets events starting from offset // Returns: events slice, next offset, error -func (m *MemoryStreamManager) GetEvents(ctx context.Context, sessionID, messageID string, fromOffset int) ([]interfaces.StreamEvent, int, error) { +func (m *MemoryStreamManager) GetEvents( + ctx context.Context, + sessionID, messageID string, + fromOffset int, +) ([]interfaces.StreamEvent, int, error) { stream := m.getStream(sessionID, messageID) if stream == nil { // Stream doesn't exist yet diff --git a/internal/stream/redis_manager.go b/internal/stream/redis_manager.go index 72607982..bc6e1b36 100644 --- a/internal/stream/redis_manager.go +++ b/internal/stream/redis_manager.go @@ -54,7 +54,11 @@ func (r *RedisStreamManager) buildKey(sessionID, messageID string) string { } // AppendEvent appends a single event to the stream using Redis RPush -func (r *RedisStreamManager) AppendEvent(ctx context.Context, sessionID, messageID string, event interfaces.StreamEvent) error { +func (r *RedisStreamManager) AppendEvent( + ctx context.Context, + sessionID, messageID string, + event interfaces.StreamEvent, +) error { key := r.buildKey(sessionID, messageID) // Set timestamp if not already set @@ -83,7 +87,11 @@ func (r *RedisStreamManager) AppendEvent(ctx context.Context, sessionID, message // GetEvents gets events starting from offset using Redis LRange // Returns: events slice, next offset, error -func (r *RedisStreamManager) GetEvents(ctx context.Context, sessionID, messageID string, fromOffset int) ([]interfaces.StreamEvent, int, error) { +func (r *RedisStreamManager) GetEvents( + ctx context.Context, + sessionID, messageID string, + fromOffset int, +) ([]interfaces.StreamEvent, int, error) { key := r.buildKey(sessionID, messageID) // Get all events from offset to end using LRange diff --git a/internal/types/chunk.go b/internal/types/chunk.go index b64fdd22..99a16053 100644 --- a/internal/types/chunk.go +++ b/internal/types/chunk.go @@ -44,7 +44,7 @@ const ( // ImageInfo 表示与 Chunk 关联的图片信息 type ImageInfo struct { // 图片URL（COS） - URL string `json:"url" gorm:"type:text"` + URL string `json:"url" gorm:"type:text"` // 原始图片URL OriginalURL string `json:"original_url" gorm:"type:text"` // 图片在文本中的开始位置 @@ -65,7 +65,7 @@ type ImageInfo struct { // Chunks can be independently embedded as vectors and retrieved, supporting precise content localization type Chunk struct { // Unique identifier of the chunk, using UUID format - ID string `json:"id" gorm:"type:varchar(36);primaryKey"` + ID string `json:"id" gorm:"type:varchar(36);primaryKey"` // Tenant ID, used for multi-tenant isolation TenantID uint64 `json:"tenant_id"` // ID of the parent knowledge, associated with the Knowledge model @@ -73,17 +73,17 @@ type Chunk struct { // ID of the knowledge base, for quick location KnowledgeBaseID string `json:"knowledge_base_id"` // Optional tag ID for categorization within a knowledge base (used for FAQ) - TagID string `json:"tag_id" gorm:"type:varchar(36);index"` + TagID string `json:"tag_id" gorm:"type:varchar(36);index"` // Actual text content of the chunk Content string `json:"content"` // Index position of the chunk in the original document ChunkIndex int `json:"chunk_index"` // Whether the chunk is enabled, can be used to temporarily disable certain chunks - IsEnabled bool `json:"is_enabled" gorm:"default:true"` + IsEnabled bool `json:"is_enabled" gorm:"default:true"` // Status of the chunk - Status int `json:"status" gorm:"default:0"` + Status int `json:"status" gorm:"default:0"` // Starting character position in the original text - StartAt int `json:"start_at" ` + StartAt int `json:"start_at"` // Ending character position in the original text EndAt int `json:"end_at"` // Previous chunk ID @@ -91,23 +91,23 @@ type Chunk struct { // Next chunk ID NextChunkID string `json:"next_chunk_id"` // Chunk 类型，用于区分不同类型的 Chunk - ChunkType ChunkType `json:"chunk_type" gorm:"type:varchar(20);default:'text'"` + ChunkType ChunkType `json:"chunk_type" gorm:"type:varchar(20);default:'text'"` // 父 Chunk ID，用于关联图片 Chunk 和原始文本 Chunk - ParentChunkID string `json:"parent_chunk_id" gorm:"type:varchar(36);index"` + ParentChunkID string `json:"parent_chunk_id" gorm:"type:varchar(36);index"` // 关系 Chunk ID，用于关联关系 Chunk 和原始文本 Chunk - RelationChunks JSON `json:"relation_chunks" gorm:"type:json"` + RelationChunks JSON `json:"relation_chunks" gorm:"type:json"` // 间接关系 Chunk ID，用于关联间接关系 Chunk 和原始文本 Chunk IndirectRelationChunks JSON `json:"indirect_relation_chunks" gorm:"type:json"` // Metadata 存储 chunk 级别的扩展信息，例如 FAQ 元数据 - Metadata JSON `json:"metadata" gorm:"type:json"` + Metadata JSON `json:"metadata" gorm:"type:json"` // ContentHash 存储内容的 hash 值，用于快速匹配（主要用于 FAQ） - ContentHash string `json:"content_hash" gorm:"type:varchar(64);index"` + ContentHash string `json:"content_hash" gorm:"type:varchar(64);index"` // 图片信息，存储为 JSON - ImageInfo string `json:"image_info" gorm:"type:text"` + ImageInfo string `json:"image_info" gorm:"type:text"` // Chunk creation time CreatedAt time.Time `json:"created_at"` // Chunk last update time UpdatedAt time.Time `json:"updated_at"` // Soft delete marker, supports data recovery - DeletedAt gorm.DeletedAt `json:"deleted_at" gorm:"index"` + DeletedAt gorm.DeletedAt `json:"deleted_at" gorm:"index"` } diff --git a/internal/types/faq.go b/internal/types/faq.go index c7da4ca6..7c209c85 100644 --- a/internal/types/faq.go +++ b/internal/types/faq.go @@ -129,10 +129,10 @@ type FAQEntry struct { // FAQEntryPayload 用于创建/更新 FAQ 条目的 payload type FAQEntryPayload struct { - StandardQuestion string `json:"standard_question" binding:"required"` + StandardQuestion string `json:"standard_question" binding:"required"` SimilarQuestions []string `json:"similar_questions"` NegativeQuestions []string `json:"negative_questions"` - Answers []string `json:"answers" binding:"required"` + Answers []string `json:"answers" binding:"required"` TagID string `json:"tag_id"` IsEnabled *bool `json:"is_enabled,omitempty"` } @@ -144,14 +144,14 @@ const ( // FAQBatchUpsertPayload 批量导入 FAQ 条目 type FAQBatchUpsertPayload struct { - Entries []FAQEntryPayload `json:"entries" binding:"required"` - Mode string `json:"mode" binding:"oneof=append replace"` + Entries []FAQEntryPayload `json:"entries" binding:"required"` + Mode string `json:"mode" binding:"oneof=append replace"` KnowledgeID string `json:"knowledge_id"` } // FAQSearchRequest FAQ检索请求参数 type FAQSearchRequest struct { - QueryText string `json:"query_text" binding:"required"` + QueryText string `json:"query_text" binding:"required"` VectorThreshold float64 `json:"vector_threshold"` MatchCount int `json:"match_count"` } diff --git a/internal/types/interfaces/agent.go b/internal/types/interfaces/agent.go index f95d59b3..a063dcbb 100644 --- a/internal/types/interfaces/agent.go +++ b/internal/types/interfaces/agent.go @@ -21,13 +21,26 @@ type AgentStreamEvent struct { // AgentEngine defines the interface for agent execution engine type AgentEngine interface { // Execute executes the agent with conversation history and returns a stream of events - Execute(ctx context.Context, sessionID, messageID, query string, llmContext []chat.Message) (*types.AgentState, error) + Execute( + ctx context.Context, + sessionID, messageID, query string, + llmContext []chat.Message, + ) (*types.AgentState, error) } // AgentService defines the interface for agent-related operations type AgentService interface { // CreateAgentEngine creates an agent engine with the given configuration, EventBus, and ContextManager - CreateAgentEngine(ctx context.Context, config *types.AgentConfig, chatModel chat.Chat, rerankModel rerank.Reranker, eventBus *event.EventBus, contextManager ContextManager, sessionID string, sessionService SessionService) (AgentEngine, error) + CreateAgentEngine( + ctx context.Context, + config *types.AgentConfig, + chatModel chat.Chat, + rerankModel rerank.Reranker, + eventBus *event.EventBus, + contextManager ContextManager, + sessionID string, + sessionService SessionService, + ) (AgentEngine, error) // ValidateConfig validates an agent configuration ValidateConfig(config *types.AgentConfig) error diff --git a/internal/types/interfaces/knowledge.go b/internal/types/interfaces/knowledge.go index 5d6ae569..523eecc1 100644 --- a/internal/types/interfaces/knowledge.go +++ b/internal/types/interfaces/knowledge.go @@ -21,13 +21,22 @@ type KnowledgeService interface { customFileName string, ) (*types.Knowledge, error) // CreateKnowledgeFromURL creates knowledge from a URL. - CreateKnowledgeFromURL(ctx context.Context, kbID string, url string, enableMultimodel *bool) (*types.Knowledge, error) + CreateKnowledgeFromURL( + ctx context.Context, + kbID string, + url string, + enableMultimodel *bool, + ) (*types.Knowledge, error) // CreateKnowledgeFromPassage creates knowledge from text passages. CreateKnowledgeFromPassage(ctx context.Context, kbID string, passage []string) (*types.Knowledge, error) // CreateKnowledgeFromPassageSync creates knowledge from text passages and waits until chunks are indexed. CreateKnowledgeFromPassageSync(ctx context.Context, kbID string, passage []string) (*types.Knowledge, error) // CreateKnowledgeFromManual creates or saves manual Markdown knowledge content. - CreateKnowledgeFromManual(ctx context.Context, kbID string, payload *types.ManualKnowledgePayload) (*types.Knowledge, error) + CreateKnowledgeFromManual( + ctx context.Context, + kbID string, + payload *types.ManualKnowledgePayload, + ) (*types.Knowledge, error) // GetKnowledgeByID retrieves knowledge by ID. GetKnowledgeByID(ctx context.Context, id string) (*types.Knowledge, error) // GetKnowledgeBatch retrieves a batch of knowledge by IDs. @@ -49,7 +58,11 @@ type KnowledgeService interface { // UpdateKnowledge updates knowledge information. UpdateKnowledge(ctx context.Context, knowledge *types.Knowledge) error // UpdateManualKnowledge updates manual Markdown knowledge content. - UpdateManualKnowledge(ctx context.Context, knowledgeID string, payload *types.ManualKnowledgePayload) (*types.Knowledge, error) + UpdateManualKnowledge( + ctx context.Context, + knowledgeID string, + payload *types.ManualKnowledgePayload, + ) (*types.Knowledge, error) // CloneKnowledgeBase clones knowledge to another knowledge base. CloneKnowledgeBase(ctx context.Context, srcID, dstID string) error // UpdateImageInfo updates image information for a knowledge chunk. diff --git a/internal/types/interfaces/session.go b/internal/types/interfaces/session.go index a25c29d3..b0e0b6bc 100644 --- a/internal/types/interfaces/session.go +++ b/internal/types/interfaces/session.go @@ -40,13 +40,28 @@ type SessionService interface { SearchKnowledge(ctx context.Context, knowledgeBaseID, query string) ([]*types.SearchResult, error) // AgentQA performs agent-based question answering with conversation history and streaming support // eventBus is optional - if nil, uses service's default EventBus - AgentQA(ctx context.Context, session *types.Session, query string, assistantMessageID string, eventBus *event.EventBus) error + AgentQA( + ctx context.Context, + session *types.Session, + query string, + assistantMessageID string, + eventBus *event.EventBus, + ) error // ClearContext clears the LLM context for a session ClearContext(ctx context.Context, sessionID string) error // GetWebSearchTempKBState retrieves the temporary KB state for web search from Redis - GetWebSearchTempKBState(ctx context.Context, sessionID string) (tempKBID string, seenURLs map[string]bool, knowledgeIDs []string) + GetWebSearchTempKBState( + ctx context.Context, + sessionID string, + ) (tempKBID string, seenURLs map[string]bool, knowledgeIDs []string) // SaveWebSearchTempKBState saves the temporary KB state for web search to Redis - SaveWebSearchTempKBState(ctx context.Context, sessionID string, tempKBID string, seenURLs map[string]bool, knowledgeIDs []string) + SaveWebSearchTempKBState( + ctx context.Context, + sessionID string, + tempKBID string, + seenURLs map[string]bool, + knowledgeIDs []string, + ) // DeleteWebSearchTempKBState deletes the temporary KB state for web search from Redis and cleans up associated knowledge base and knowledge items DeleteWebSearchTempKBState(ctx context.Context, sessionID string) error } diff --git a/internal/types/interfaces/tag.go b/internal/types/interfaces/tag.go index cb0069a0..c08c5acf 100644 --- a/internal/types/interfaces/tag.go +++ b/internal/types/interfaces/tag.go @@ -26,5 +26,10 @@ type KnowledgeTagRepository interface { ListByKB(ctx context.Context, tenantID uint64, kbID string) ([]*types.KnowledgeTag, error) Delete(ctx context.Context, tenantID uint64, id string) error // CountReferences returns number of knowledges and chunks that reference the tag. - CountReferences(ctx context.Context, tenantID uint64, kbID string, tagID string) (knowledgeCount int64, chunkCount int64, err error) + CountReferences( + ctx context.Context, + tenantID uint64, + kbID string, + tagID string, + ) (knowledgeCount int64, chunkCount int64, err error) } diff --git a/internal/types/knowledge.go b/internal/types/knowledge.go index 185a1ed1..2dc82436 100644 --- a/internal/types/knowledge.go +++ b/internal/types/knowledge.go @@ -28,13 +28,13 @@ const ( // and references to the physical file if applicable. type Knowledge struct { // Unique identifier of the knowledge - ID string `json:"id" gorm:"type:varchar(36);primaryKey"` + ID string `json:"id" gorm:"type:varchar(36);primaryKey"` // Tenant ID TenantID uint64 `json:"tenant_id"` // ID of the knowledge base KnowledgeBaseID string `json:"knowledge_base_id"` // Optional tag ID for categorization within a knowledge base - TagID string `json:"tag_id" gorm:"type:varchar(36);index"` + TagID string `json:"tag_id" gorm:"type:varchar(36);index"` // Type of the knowledge Type string `json:"type"` // Title of the knowledge @@ -62,7 +62,7 @@ type Knowledge struct { // Storage size of the knowledge StorageSize int64 `json:"storage_size"` // Metadata of the knowledge - Metadata JSON `json:"metadata" gorm:"type:json"` + Metadata JSON `json:"metadata" gorm:"type:json"` // Creation time of the knowledge CreatedAt time.Time `json:"created_at"` // Last updated time of the knowledge @@ -72,7 +72,7 @@ type Knowledge struct { // Error message of the knowledge ErrorMessage string `json:"error_message"` // Deletion time of the knowledge - DeletedAt gorm.DeletedAt `json:"deleted_at" gorm:"index"` + DeletedAt gorm.DeletedAt `json:"deleted_at" gorm:"index"` } // GetMetadata returns the metadata as a map[string]string. diff --git a/internal/types/knowledgebase.go b/internal/types/knowledgebase.go index a7f3263f..cd542571 100644 --- a/internal/types/knowledgebase.go +++ b/internal/types/knowledgebase.go @@ -38,85 +38,85 @@ const ( // KnowledgeBase represents a knowledge base entity type KnowledgeBase struct { // Unique identifier of the knowledge base - ID string `yaml:"id" json:"id" gorm:"type:varchar(36);primaryKey"` + ID string `yaml:"id" json:"id" gorm:"type:varchar(36);primaryKey"` // Name of the knowledge base - Name string `yaml:"name" json:"name"` + Name string `yaml:"name" json:"name"` // Type of the knowledge base (document, faq, etc.) - Type string `yaml:"type" json:"type" gorm:"type:varchar(32);default:'document'"` + Type string `yaml:"type" json:"type" gorm:"type:varchar(32);default:'document'"` // Whether this knowledge base is temporary (ephemeral) and should be hidden from UI - IsTemporary bool `yaml:"is_temporary" json:"is_temporary" gorm:"default:false"` + IsTemporary bool `yaml:"is_temporary" json:"is_temporary" gorm:"default:false"` // Description of the knowledge base - Description string `yaml:"description" json:"description"` + Description string `yaml:"description" json:"description"` // Tenant ID - TenantID uint64 `yaml:"tenant_id" json:"tenant_id"` + TenantID uint64 `yaml:"tenant_id" json:"tenant_id"` // Chunking configuration - ChunkingConfig ChunkingConfig `yaml:"chunking_config" json:"chunking_config" gorm:"type:json"` + ChunkingConfig ChunkingConfig `yaml:"chunking_config" json:"chunking_config" gorm:"type:json"` // Image processing configuration ImageProcessingConfig ImageProcessingConfig `yaml:"image_processing_config" json:"image_processing_config" gorm:"type:json"` // ID of the embedding model - EmbeddingModelID string `yaml:"embedding_model_id" json:"embedding_model_id"` + EmbeddingModelID string `yaml:"embedding_model_id" json:"embedding_model_id"` // Summary model ID - SummaryModelID string `yaml:"summary_model_id" json:"summary_model_id"` + SummaryModelID string `yaml:"summary_model_id" json:"summary_model_id"` // VLM config - VLMConfig VLMConfig `yaml:"vlm_config" json:"vlm_config" gorm:"type:json"` + VLMConfig VLMConfig `yaml:"vlm_config" json:"vlm_config" gorm:"type:json"` // Storage config - StorageConfig StorageConfig `yaml:"cos_config" json:"cos_config" gorm:"column:cos_config;type:json"` + StorageConfig StorageConfig `yaml:"cos_config" json:"cos_config" gorm:"column:cos_config;type:json"` // Extract config - ExtractConfig *ExtractConfig `yaml:"extract_config" json:"extract_config" gorm:"column:extract_config;type:json"` + ExtractConfig *ExtractConfig `yaml:"extract_config" json:"extract_config" gorm:"column:extract_config;type:json"` // FAQConfig stores FAQ specific configuration such as indexing strategy - FAQConfig *FAQConfig `yaml:"faq_config" json:"faq_config" gorm:"column:faq_config;type:json"` + FAQConfig *FAQConfig `yaml:"faq_config" json:"faq_config" gorm:"column:faq_config;type:json"` // Creation time of the knowledge base - CreatedAt time.Time `yaml:"created_at" json:"created_at"` + CreatedAt time.Time `yaml:"created_at" json:"created_at"` // Last updated time of the knowledge base - UpdatedAt time.Time `yaml:"updated_at" json:"updated_at"` + UpdatedAt time.Time `yaml:"updated_at" json:"updated_at"` // Deletion time of the knowledge base - DeletedAt gorm.DeletedAt `yaml:"deleted_at" json:"deleted_at" gorm:"index"` + DeletedAt gorm.DeletedAt `yaml:"deleted_at" json:"deleted_at" gorm:"index"` // Knowledge count (not stored in database, calculated on query) - KnowledgeCount int64 `yaml:"knowledge_count" json:"knowledge_count" gorm:"-"` + KnowledgeCount int64 `yaml:"knowledge_count" json:"knowledge_count" gorm:"-"` // Chunk count (not stored in database, calculated on query) - ChunkCount int64 `yaml:"chunk_count" json:"chunk_count" gorm:"-"` + ChunkCount int64 `yaml:"chunk_count" json:"chunk_count" gorm:"-"` // IsProcessing indicates if there is a processing import task (for FAQ type knowledge bases) - IsProcessing bool `yaml:"is_processing" json:"is_processing" gorm:"-"` + IsProcessing bool `yaml:"is_processing" json:"is_processing" gorm:"-"` // ProcessingCount indicates the number of knowledge items being processed (for document type knowledge bases) - ProcessingCount int64 `yaml:"processing_count" json:"processing_count" gorm:"-"` + ProcessingCount int64 `yaml:"processing_count" json:"processing_count" gorm:"-"` } // KnowledgeBaseConfig represents the knowledge base configuration type KnowledgeBaseConfig struct { // Chunking configuration - ChunkingConfig ChunkingConfig `yaml:"chunking_config" json:"chunking_config"` + ChunkingConfig ChunkingConfig `yaml:"chunking_config" json:"chunking_config"` // Image processing configuration ImageProcessingConfig ImageProcessingConfig `yaml:"image_processing_config" json:"image_processing_config"` // FAQ configuration (only for FAQ type knowledge bases) - FAQConfig *FAQConfig `yaml:"faq_config" json:"faq_config"` + FAQConfig *FAQConfig `yaml:"faq_config" json:"faq_config"` } // ChunkingConfig represents the document splitting configuration type ChunkingConfig struct { // Chunk size - ChunkSize int `yaml:"chunk_size" json:"chunk_size"` + ChunkSize int `yaml:"chunk_size" json:"chunk_size"` // Chunk overlap ChunkOverlap int `yaml:"chunk_overlap" json:"chunk_overlap"` // Separators - Separators []string `yaml:"separators" json:"separators"` + Separators []string `yaml:"separators" json:"separators"` } // COSConfig represents the COS configuration type StorageConfig struct { // Secret ID - SecretID string `yaml:"secret_id" json:"secret_id"` + SecretID string `yaml:"secret_id" json:"secret_id"` // Secret Key - SecretKey string `yaml:"secret_key" json:"secret_key"` + SecretKey string `yaml:"secret_key" json:"secret_key"` // Region - Region string `yaml:"region" json:"region"` + Region string `yaml:"region" json:"region"` // Bucket Name BucketName string `yaml:"bucket_name" json:"bucket_name"` // App ID - AppID string `yaml:"app_id" json:"app_id"` + AppID string `yaml:"app_id" json:"app_id"` // Path Prefix PathPrefix string `yaml:"path_prefix" json:"path_prefix"` // Provider - Provider string `yaml:"provider" json:"provider"` + Provider string `yaml:"provider" json:"provider"` } func (c StorageConfig) Value() (driver.Value, error) { @@ -176,7 +176,7 @@ func (c *ImageProcessingConfig) Scan(value interface{}) error { // VLMConfig represents the VLM configuration type VLMConfig struct { - Enabled bool `yaml:"enabled" json:"enabled"` + Enabled bool `yaml:"enabled" json:"enabled"` ModelID string `yaml:"model_id" json:"model_id"` } @@ -199,10 +199,10 @@ func (c *VLMConfig) Scan(value interface{}) error { // ExtractConfig represents the extract configuration for a knowledge base type ExtractConfig struct { - Enabled bool `yaml:"enabled" json:"enabled"` - Text string `yaml:"text" json:"text,omitempty"` - Tags []string `yaml:"tags" json:"tags,omitempty"` - Nodes []*GraphNode `yaml:"nodes" json:"nodes,omitempty"` + Enabled bool `yaml:"enabled" json:"enabled"` + Text string `yaml:"text" json:"text,omitempty"` + Tags []string `yaml:"tags" json:"tags,omitempty"` + Nodes []*GraphNode `yaml:"nodes" json:"nodes,omitempty"` Relations []*GraphRelation `yaml:"relations" json:"relations,omitempty"` } @@ -225,7 +225,7 @@ func (e *ExtractConfig) Scan(value interface{}) error { // FAQConfig 存储 FAQ 知识库的特有配置 type FAQConfig struct { - IndexMode FAQIndexMode `yaml:"index_mode" json:"index_mode"` + IndexMode FAQIndexMode `yaml:"index_mode" json:"index_mode"` QuestionIndexMode FAQQuestionIndexMode `yaml:"question_index_mode" json:"question_index_mode"` } diff --git a/internal/types/mcp.go b/internal/types/mcp.go index 9f21de43..7d1fee36 100644 --- a/internal/types/mcp.go +++ b/internal/types/mcp.go @@ -20,21 +20,21 @@ const ( // MCPService represents an MCP (Model Context Protocol) service configuration type MCPService struct { - ID string `json:"id" gorm:"type:varchar(36);primaryKey"` - TenantID uint64 `json:"tenant_id" gorm:"index"` - Name string `json:"name" gorm:"type:varchar(255);not null"` - Description string `json:"description" gorm:"type:text"` - Enabled bool `json:"enabled" gorm:"default:true;index"` - TransportType MCPTransportType `json:"transport_type" gorm:"type:varchar(50);not null"` - URL *string `json:"url,omitempty" gorm:"type:varchar(512)"` // Optional: required for SSE/HTTP Streamable - Headers MCPHeaders `json:"headers" gorm:"type:json"` - AuthConfig *MCPAuthConfig `json:"auth_config" gorm:"type:json"` - AdvancedConfig *MCPAdvancedConfig `json:"advanced_config" gorm:"type:json"` + ID string `json:"id" gorm:"type:varchar(36);primaryKey"` + TenantID uint64 `json:"tenant_id" gorm:"index"` + Name string `json:"name" gorm:"type:varchar(255);not null"` + Description string `json:"description" gorm:"type:text"` + Enabled bool `json:"enabled" gorm:"default:true;index"` + TransportType MCPTransportType `json:"transport_type" gorm:"type:varchar(50);not null"` + URL *string `json:"url,omitempty" gorm:"type:varchar(512)"` // Optional: required for SSE/HTTP Streamable + Headers MCPHeaders `json:"headers" gorm:"type:json"` + AuthConfig *MCPAuthConfig `json:"auth_config" gorm:"type:json"` + AdvancedConfig *MCPAdvancedConfig `json:"advanced_config" gorm:"type:json"` StdioConfig *MCPStdioConfig `json:"stdio_config,omitempty" gorm:"type:json"` // Required for stdio transport - EnvVars MCPEnvVars `json:"env_vars,omitempty" gorm:"type:json"` // Environment variables for stdio + EnvVars MCPEnvVars `json:"env_vars,omitempty" gorm:"type:json"` // Environment variables for stdio CreatedAt time.Time `json:"created_at"` UpdatedAt time.Time `json:"updated_at"` - DeletedAt gorm.DeletedAt `json:"deleted_at" gorm:"index"` + DeletedAt gorm.DeletedAt `json:"deleted_at" gorm:"index"` } // MCPHeaders represents HTTP headers as a map diff --git a/internal/types/message.go b/internal/types/message.go index 56fc29da..9173c906 100644 --- a/internal/types/message.go +++ b/internal/types/message.go @@ -25,7 +25,7 @@ type History struct { // Messages can contain references to knowledge chunks used to generate responses type Message struct { // Unique identifier for the message - ID string `json:"id" gorm:"type:varchar(36);primaryKey"` + ID string `json:"id" gorm:"type:varchar(36);primaryKey"` // ID of the session this message belongs to SessionID string `json:"session_id"` // Request identifier for tracking API requests @@ -35,7 +35,7 @@ type Message struct { // Message role: "user", "assistant", "system" Role string `json:"role"` // References to knowledge chunks used in the response - KnowledgeReferences References `json:"knowledge_references" gorm:"type:json,column:knowledge_references"` + KnowledgeReferences References `json:"knowledge_references" gorm:"type:json,column:knowledge_references"` // Agent execution steps (only for assistant messages generated by agent) // This contains the detailed reasoning process and tool calls made by the agent // Stored for user history display, but NOT included in LLM context to avoid redundancy @@ -47,7 +47,7 @@ type Message struct { // Last update timestamp UpdatedAt time.Time `json:"updated_at"` // Soft delete timestamp - DeletedAt gorm.DeletedAt `json:"deleted_at" gorm:"index"` + DeletedAt gorm.DeletedAt `json:"deleted_at" gorm:"index"` } // AgentSteps represents a collection of agent execution steps diff --git a/internal/types/model.go b/internal/types/model.go index 2f3724cd..8d010df0 100644 --- a/internal/types/model.go +++ b/internal/types/model.go @@ -39,46 +39,46 @@ const ( // EmbeddingParameters represents the embedding parameters for a model type EmbeddingParameters struct { - Dimension int `yaml:"dimension" json:"dimension"` + Dimension int `yaml:"dimension" json:"dimension"` TruncatePromptTokens int `yaml:"truncate_prompt_tokens" json:"truncate_prompt_tokens"` } type ModelParameters struct { - BaseURL string `yaml:"base_url" json:"base_url"` - APIKey string `yaml:"api_key" json:"api_key"` - InterfaceType string `yaml:"interface_type" json:"interface_type"` + BaseURL string `yaml:"base_url" json:"base_url"` + APIKey string `yaml:"api_key" json:"api_key"` + InterfaceType string `yaml:"interface_type" json:"interface_type"` EmbeddingParameters EmbeddingParameters `yaml:"embedding_parameters" json:"embedding_parameters"` - ParameterSize string `yaml:"parameter_size" json:"parameter_size"` // Ollama model parameter size (e.g., "7B", "13B", "70B") + ParameterSize string `yaml:"parameter_size" json:"parameter_size"` // Ollama model parameter size (e.g., "7B", "13B", "70B") } // Model represents the AI model type Model struct { // Unique identifier of the model - ID string `yaml:"id" json:"id" gorm:"type:varchar(36);primaryKey"` + ID string `yaml:"id" json:"id" gorm:"type:varchar(36);primaryKey"` // Tenant ID - TenantID uint64 `yaml:"tenant_id" json:"tenant_id"` + TenantID uint64 `yaml:"tenant_id" json:"tenant_id"` // Name of the model - Name string `yaml:"name" json:"name"` + Name string `yaml:"name" json:"name"` // Type of the model - Type ModelType `yaml:"type" json:"type"` + Type ModelType `yaml:"type" json:"type"` // Source of the model - Source ModelSource `yaml:"source" json:"source"` + Source ModelSource `yaml:"source" json:"source"` // Description of the model Description string `yaml:"description" json:"description"` // Model parameters in JSON format - Parameters ModelParameters `yaml:"parameters" json:"parameters" gorm:"type:json"` + Parameters ModelParameters `yaml:"parameters" json:"parameters" gorm:"type:json"` // Whether the model is the default model - IsDefault bool `yaml:"is_default" json:"is_default"` + IsDefault bool `yaml:"is_default" json:"is_default"` // Whether the model is a builtin model (visible to all tenants) - IsBuiltin bool `yaml:"is_builtin" json:"is_builtin" gorm:"default:false"` + IsBuiltin bool `yaml:"is_builtin" json:"is_builtin" gorm:"default:false"` // Model status, default: active, possible: downloading, download_failed - Status ModelStatus `yaml:"status" json:"status"` + Status ModelStatus `yaml:"status" json:"status"` // Creation time of the model - CreatedAt time.Time `yaml:"created_at" json:"created_at"` + CreatedAt time.Time `yaml:"created_at" json:"created_at"` // Last updated time of the model - UpdatedAt time.Time `yaml:"updated_at" json:"updated_at"` + UpdatedAt time.Time `yaml:"updated_at" json:"updated_at"` // Deletion time of the model - DeletedAt gorm.DeletedAt `yaml:"deleted_at" json:"deleted_at" gorm:"index"` + DeletedAt gorm.DeletedAt `yaml:"deleted_at" json:"deleted_at" gorm:"index"` } // Value implements the driver.Valuer interface, used to convert ModelParameters to database value diff --git a/internal/types/retriever.go b/internal/types/retriever.go index 38990974..205d25bf 100644 --- a/internal/types/retriever.go +++ b/internal/types/retriever.go @@ -48,7 +48,7 @@ type RetrieverEngineParams struct { // Retriever engine type RetrieverEngineType RetrieverEngineType `yaml:"retriever_engine_type" json:"retriever_engine_type"` // Retriever type - RetrieverType RetrieverType `yaml:"retriever_type" json:"retriever_type"` + RetrieverType RetrieverType `yaml:"retriever_type" json:"retriever_type"` } // IndexWithScore represents the index with score diff --git a/internal/types/search.go b/internal/types/search.go index 1a2a9d69..292eaeb6 100644 --- a/internal/types/search.go +++ b/internal/types/search.go @@ -8,29 +8,29 @@ import ( // SearchResult represents the search result type SearchResult struct { // ID - ID string `gorm:"column:id" json:"id"` + ID string `gorm:"column:id" json:"id"` // Content - Content string `gorm:"column:content" json:"content"` + Content string `gorm:"column:content" json:"content"` // Knowledge ID - KnowledgeID string `gorm:"column:knowledge_id" json:"knowledge_id"` + KnowledgeID string `gorm:"column:knowledge_id" json:"knowledge_id"` // Chunk index - ChunkIndex int `gorm:"column:chunk_index" json:"chunk_index"` + ChunkIndex int `gorm:"column:chunk_index" json:"chunk_index"` // Knowledge title KnowledgeTitle string `gorm:"column:knowledge_title" json:"knowledge_title"` // Start at - StartAt int `gorm:"column:start_at" json:"start_at"` + StartAt int `gorm:"column:start_at" json:"start_at"` // End at - EndAt int `gorm:"column:end_at" json:"end_at"` + EndAt int `gorm:"column:end_at" json:"end_at"` // Seq - Seq int `gorm:"column:seq" json:"seq"` + Seq int `gorm:"column:seq" json:"seq"` // Score - Score float64 `json:"score"` + Score float64 ` json:"score"` // Match type - MatchType MatchType `json:"match_type"` + MatchType MatchType ` json:"match_type"` // SubChunkIndex - SubChunkID []string `json:"sub_chunk_id"` + SubChunkID []string ` json:"sub_chunk_id"` // Metadata - Metadata map[string]string `json:"metadata"` + Metadata map[string]string ` json:"metadata"` // Chunk 类型 ChunkType string `json:"chunk_type"` @@ -78,7 +78,7 @@ func (c *SearchResult) Scan(value interface{}) error { // Pagination represents the pagination parameters type Pagination struct { // Page - Page int `form:"page" json:"page" binding:"omitempty,min=1"` + Page int `form:"page" json:"page" binding:"omitempty,min=1"` // Page size PageSize int `form:"page_size" json:"page_size" binding:"omitempty,min=1,max=100"` } diff --git a/internal/types/session.go b/internal/types/session.go index f26b6b6c..d239563d 100644 --- a/internal/types/session.go +++ b/internal/types/session.go @@ -72,30 +72,30 @@ type ContextConfig struct { // Session represents the session type Session struct { // ID - ID string `json:"id" gorm:"type:varchar(36);primaryKey"` + ID string `json:"id" gorm:"type:varchar(36);primaryKey"` // Title Title string `json:"title"` // Description Description string `json:"description"` // Tenant ID - TenantID uint64 `json:"tenant_id" gorm:"index"` + TenantID uint64 `json:"tenant_id" gorm:"index"` // Strategy configuration - KnowledgeBaseID string `json:"knowledge_base_id"` // 关联的知识库ID - MaxRounds int `json:"max_rounds"` // 多轮保持轮数 - EnableRewrite bool `json:"enable_rewrite"` // 多轮改写开关 - FallbackStrategy FallbackStrategy `json:"fallback_strategy"` // 兜底策略 - FallbackResponse string `json:"fallback_response"` // 固定回复内容 - EmbeddingTopK int `json:"embedding_top_k"` // 向量召回TopK - KeywordThreshold float64 `json:"keyword_threshold"` // 关键词召回阈值 - VectorThreshold float64 `json:"vector_threshold"` // 向量召回阈值 - RerankModelID string `json:"rerank_model_id"` // 排序模型ID - RerankTopK int `json:"rerank_top_k"` // 排序TopK - RerankThreshold float64 `json:"rerank_threshold"` // 排序阈值 - SummaryModelID string `json:"summary_model_id"` // 总结模型ID - SummaryParameters *SummaryConfig `json:"summary_parameters" gorm:"type:json"` // 总结模型参数 - AgentConfig *SessionAgentConfig `json:"agent_config" gorm:"type:jsonb"` // Agent 配置（会话级别，仅存储enabled和knowledge_bases） - ContextConfig *ContextConfig `json:"context_config" gorm:"type:jsonb"` // 上下文管理配置（可选） + KnowledgeBaseID string `json:"knowledge_base_id"` // 关联的知识库ID + MaxRounds int `json:"max_rounds"` // 多轮保持轮数 + EnableRewrite bool `json:"enable_rewrite"` // 多轮改写开关 + FallbackStrategy FallbackStrategy `json:"fallback_strategy"` // 兜底策略 + FallbackResponse string `json:"fallback_response"` // 固定回复内容 + EmbeddingTopK int `json:"embedding_top_k"` // 向量召回TopK + KeywordThreshold float64 `json:"keyword_threshold"` // 关键词召回阈值 + VectorThreshold float64 `json:"vector_threshold"` // 向量召回阈值 + RerankModelID string `json:"rerank_model_id"` // 排序模型ID + RerankTopK int `json:"rerank_top_k"` // 排序TopK + RerankThreshold float64 `json:"rerank_threshold"` // 排序阈值 + SummaryModelID string `json:"summary_model_id"` // 总结模型ID + SummaryParameters *SummaryConfig `json:"summary_parameters" gorm:"type:json"` // 总结模型参数 + AgentConfig *SessionAgentConfig `json:"agent_config" gorm:"type:jsonb"` // Agent 配置（会话级别，仅存储enabled和knowledge_bases） + ContextConfig *ContextConfig `json:"context_config" gorm:"type:jsonb"` // 上下文管理配置（可选） CreatedAt time.Time `json:"created_at"` UpdatedAt time.Time `json:"updated_at"` diff --git a/internal/types/tag.go b/internal/types/tag.go index 5b9de754..c93e39b2 100644 --- a/internal/types/tag.go +++ b/internal/types/tag.go @@ -7,17 +7,17 @@ import "time" // Knowledge (documents) and FAQ Chunks. type KnowledgeTag struct { // Unique identifier of the tag (UUID) - ID string `json:"id" gorm:"type:varchar(36);primaryKey"` + ID string `json:"id" gorm:"type:varchar(36);primaryKey"` // Tenant ID TenantID uint64 `json:"tenant_id"` // Knowledge base ID that this tag belongs to KnowledgeBaseID string `json:"knowledge_base_id" gorm:"type:varchar(36);index"` // Tag name, unique within the same knowledge base - Name string `json:"name" gorm:"type:varchar(128);not null"` + Name string `json:"name" gorm:"type:varchar(128);not null"` // Optional display color - Color string `json:"color" gorm:"type:varchar(32)"` + Color string `json:"color" gorm:"type:varchar(32)"` // Sort order within the same knowledge base - SortOrder int `json:"sort_order" gorm:"default:0"` + SortOrder int `json:"sort_order" gorm:"default:0"` // Creation time CreatedAt time.Time `json:"created_at"` // Last updated time diff --git a/internal/types/tenant.go b/internal/types/tenant.go index 67e85287..0a597dc7 100644 --- a/internal/types/tenant.go +++ b/internal/types/tenant.go @@ -11,37 +11,37 @@ import ( // Tenant represents the tenant type Tenant struct { // ID - ID uint64 `yaml:"id" json:"id" gorm:"primaryKey"` + ID uint64 `yaml:"id" json:"id" gorm:"primaryKey"` // Name - Name string `yaml:"name" json:"name"` + Name string `yaml:"name" json:"name"` // Description - Description string `yaml:"description" json:"description"` + Description string `yaml:"description" json:"description"` // API key - APIKey string `yaml:"api_key" json:"api_key"` + APIKey string `yaml:"api_key" json:"api_key"` // Status - Status string `yaml:"status" json:"status" gorm:"default:'active'"` + Status string `yaml:"status" json:"status" gorm:"default:'active'"` // Retriever engines - RetrieverEngines RetrieverEngines `yaml:"retriever_engines" json:"retriever_engines" gorm:"type:json"` + RetrieverEngines RetrieverEngines `yaml:"retriever_engines" json:"retriever_engines" gorm:"type:json"` // Business - Business string `yaml:"business" json:"business"` + Business string `yaml:"business" json:"business"` // Storage quota (Bytes), default is 10GB, including vector, original file, text, index, etc. - StorageQuota int64 `yaml:"storage_quota" json:"storage_quota" gorm:"default:10737418240"` + StorageQuota int64 `yaml:"storage_quota" json:"storage_quota" gorm:"default:10737418240"` // Storage used (Bytes) - StorageUsed int64 `yaml:"storage_used" json:"storage_used" gorm:"default:0"` + StorageUsed int64 `yaml:"storage_used" json:"storage_used" gorm:"default:0"` // Global Agent configuration for this tenant (default for all sessions) - AgentConfig *AgentConfig `yaml:"agent_config" json:"agent_config" gorm:"type:jsonb"` + AgentConfig *AgentConfig `yaml:"agent_config" json:"agent_config" gorm:"type:jsonb"` // Global Context configuration for this tenant (default for all sessions) - ContextConfig *ContextConfig `yaml:"context_config" json:"context_config" gorm:"type:jsonb"` + ContextConfig *ContextConfig `yaml:"context_config" json:"context_config" gorm:"type:jsonb"` // Global WebSearch configuration for this tenant - WebSearchConfig *WebSearchConfig `yaml:"web_search_config" json:"web_search_config" gorm:"type:jsonb"` + WebSearchConfig *WebSearchConfig `yaml:"web_search_config" json:"web_search_config" gorm:"type:jsonb"` // Global Conversation configuration for this tenant (default for normal mode sessions) ConversationConfig *ConversationConfig `yaml:"conversation_config" json:"conversation_config" gorm:"type:jsonb"` // Creation time - CreatedAt time.Time `yaml:"created_at" json:"created_at"` + CreatedAt time.Time `yaml:"created_at" json:"created_at"` // Last updated time - UpdatedAt time.Time `yaml:"updated_at" json:"updated_at"` + UpdatedAt time.Time `yaml:"updated_at" json:"updated_at"` // Deletion time - DeletedAt gorm.DeletedAt `yaml:"deleted_at" json:"deleted_at" gorm:"index"` + DeletedAt gorm.DeletedAt `yaml:"deleted_at" json:"deleted_at" gorm:"index"` } // RetrieverEngines represents the retriever engines for a tenant diff --git a/internal/types/user.go b/internal/types/user.go index 09e64321..5d2f16c5 100644 --- a/internal/types/user.go +++ b/internal/types/user.go @@ -9,19 +9,19 @@ import ( // User represents a user in the system type User struct { // Unique identifier of the user - ID string `json:"id" gorm:"type:varchar(36);primaryKey"` + ID string `json:"id" gorm:"type:varchar(36);primaryKey"` // Username of the user - Username string `json:"username" gorm:"type:varchar(100);uniqueIndex;not null"` + Username string `json:"username" gorm:"type:varchar(100);uniqueIndex;not null"` // Email address of the user - Email string `json:"email" gorm:"type:varchar(255);uniqueIndex;not null"` + Email string `json:"email" gorm:"type:varchar(255);uniqueIndex;not null"` // Hashed password of the user - PasswordHash string `json:"-" gorm:"type:varchar(255);not null"` + PasswordHash string `json:"-" gorm:"type:varchar(255);not null"` // Avatar URL of the user - Avatar string `json:"avatar" gorm:"type:varchar(500)"` + Avatar string `json:"avatar" gorm:"type:varchar(500)"` // Tenant ID that the user belongs to - TenantID uint64 `json:"tenant_id" gorm:"index"` + TenantID uint64 `json:"tenant_id" gorm:"index"` // Whether the user is active - IsActive bool `json:"is_active" gorm:"default:true"` + IsActive bool `json:"is_active" gorm:"default:true"` // Creation time of the user CreatedAt time.Time `json:"created_at"` // Last updated time of the user @@ -36,11 +36,11 @@ type User struct { // AuthToken represents an authentication token type AuthToken struct { // Unique identifier of the token - ID string `json:"id" gorm:"type:varchar(36);primaryKey"` + ID string `json:"id" gorm:"type:varchar(36);primaryKey"` // User ID that owns this token - UserID string `json:"user_id" gorm:"type:varchar(36);index;not null"` + UserID string `json:"user_id" gorm:"type:varchar(36);index;not null"` // Token value (JWT or other format) - Token string `json:"token" gorm:"type:text;not null"` + Token string `json:"token" gorm:"type:text;not null"` // Token type (access_token, refresh_token) TokenType string `json:"token_type" gorm:"type:varchar(50);not null"` // Token expiration time @@ -58,14 +58,14 @@ type AuthToken struct { // LoginRequest represents a login request type LoginRequest struct { - Email string `json:"email" binding:"required,email"` + Email string `json:"email" binding:"required,email"` Password string `json:"password" binding:"required,min=6"` } // RegisterRequest represents a registration request type RegisterRequest struct { Username string `json:"username" binding:"required,min=3,max=50"` - Email string `json:"email" binding:"required,email"` + Email string `json:"email" binding:"required,email"` Password string `json:"password" binding:"required,min=6"` }