package ai

import (
	"context"
	"errors"
	"fmt"
	"io"
	"net/http"
	"strings"

	qdrant "github.com/qdrant/go-client/qdrant"
	openai "github.com/sashabaranov/go-openai"
	"google.golang.org/grpc"
	"google.golang.org/grpc/credentials/insecure"
	"google.golang.org/grpc/metadata"

	"sundynix-micro-go/app/plant/api/internal/svc"
	plantModel "sundynix-micro-go/app/plant/model"
	plantPb "sundynix-micro-go/app/plant/rpc/plant"
)

func getActiveAiConfig(svcCtx *svc.ServiceContext) (*plantModel.SysAiConfig, error) {
	var cfg plantModel.SysAiConfig
	err := svcCtx.DB.Where("is_active = 1").First(&cfg).Error
	if err != nil {
		return nil, errors.New("AI/RAG 问答服务暂未激活或数据库配置缺失")
	}
	return &cfg, nil
}

func chatModel(dbCfg *plantModel.SysAiConfig) string {
	if dbCfg.ChatModelName != "" {
		return dbCfg.ChatModelName
	}
	return "gpt-4o-mini"
}

func newQdrantConn(cfg *plantModel.SysAiConfig) (*grpc.ClientConn, context.Context, error) {
	addr := strings.TrimPrefix(cfg.QdrantUrl, "http://")
	addr = strings.TrimPrefix(addr, "https://")
	conn, err := grpc.NewClient(addr, grpc.WithTransportCredentials(insecure.NewCredentials()))
	if err != nil {
		return nil, nil, fmt.Errorf("qdrant grpc dial failed: %w", err)
	}
	ctx := context.Background()
	if cfg.QdrantApiKey != "" {
		ctx = metadata.NewOutgoingContext(ctx, metadata.Pairs("api-key", cfg.QdrantApiKey))
	}
	return conn, ctx, nil
}

func retrieveRAGContext(ctx context.Context, svcCtx *svc.ServiceContext, dbCfg *plantModel.SysAiConfig, question string) string {
	if dbCfg.EmbeddingApiUrl == "" || dbCfg.EmbeddingApiKey == "" || dbCfg.QdrantUrl == "" || dbCfg.QdrantCollection == "" {
		return ""
	}

	config := openai.DefaultConfig(dbCfg.EmbeddingApiKey)
	if dbCfg.EmbeddingApiUrl != "" {
		config.BaseURL = dbCfg.EmbeddingApiUrl
	}
	client := openai.NewClientWithConfig(config)
	embResp, err := client.CreateEmbeddings(ctx, openai.EmbeddingRequest{
		Input: []string{question},
		Model: openai.EmbeddingModel(dbCfg.EmbeddingModelName),
	})
	if err != nil {
		return ""
	}

	conn, qdCtx, connErr := newQdrantConn(dbCfg)
	if connErr != nil {
		return ""
	}
	defer conn.Close()

	ptsClient := qdrant.NewPointsClient(conn)
	searchRes, searchErr := ptsClient.Search(qdCtx, &qdrant.SearchPoints{
		CollectionName: dbCfg.QdrantCollection,
		Vector:         embResp.Data[0].Embedding,
		Limit:          3,
		WithPayload: &qdrant.WithPayloadSelector{
			SelectorOptions: &qdrant.WithPayloadSelector_Enable{Enable: true},
		},
	})
	if searchErr != nil {
		return ""
	}

	var b strings.Builder
	for _, pt := range searchRes.GetResult() {
		if txt, ok := pt.GetPayload()["full_text"]; ok {
			b.WriteString(txt.GetStringValue())
			b.WriteString("\n")
		}
	}
	return b.String()
}

func SaveHistory(ctx context.Context, svcCtx *svc.ServiceContext, userID, question, answer string) {
	if userID == "" || question == "" || answer == "" {
		return
	}
	_, _ = svcCtx.PlantRpc.SaveAiChatHistory(ctx, &plantPb.SaveAiChatHistoryReq{
		UserId: userID, Question: question, Answer: answer,
	})
}

func ChatCompletion(ctx context.Context, svcCtx *svc.ServiceContext, userID, question string) (string, error) {
	dbCfg, err := getActiveAiConfig(svcCtx)
	if err != nil {
		return "", err
	}
	if err := ensureQuota(ctx, svcCtx, userID, dbCfg); err != nil {
		return "", err
	}

	systemPrompt := "你是一个专业的植物百科助手。回答规则：基于知识库信息回答，不够则结合通用知识；不要使用 Markdown；用纯文本分段；回答简洁专业、条理清晰。"
	if ctxText := retrieveRAGContext(ctx, svcCtx, dbCfg, question); ctxText != "" {
		systemPrompt += "\n--- 知识库 ---\n" + ctxText + "\n--------------"
	}

	config := openai.DefaultConfig(dbCfg.ChatApiKey)
	if dbCfg.ChatApiUrl != "" {
		config.BaseURL = dbCfg.ChatApiUrl
	}
	client := openai.NewClientWithConfig(config)
	resp, err := client.CreateChatCompletion(ctx, openai.ChatCompletionRequest{
		Model: chatModel(dbCfg),
		Messages: []openai.ChatCompletionMessage{
			{Role: openai.ChatMessageRoleSystem, Content: systemPrompt},
			{Role: openai.ChatMessageRoleUser, Content: question},
		},
	})
	if err != nil {
		return "", err
	}
	if len(resp.Choices) == 0 || resp.Choices[0].Message.Content == "" {
		return "", errors.New("AI 响应为空")
	}
	answer := resp.Choices[0].Message.Content
	SaveHistory(ctx, svcCtx, userID, question, answer)
	return answer, nil
}

func StreamChat(ctx context.Context, svcCtx *svc.ServiceContext, userID, question string, w io.Writer) error {
	dbCfg, err := getActiveAiConfig(svcCtx)
	if err != nil {
		return err
	}
	if err := ensureQuota(ctx, svcCtx, userID, dbCfg); err != nil {
		return err
	}

	systemPrompt := "你是一个专业的植物百科助手。回答规则：基于知识库信息回答，不够则结合通用知识；不要使用 Markdown；用纯文本分段；回答简洁专业、条理清晰。"
	if ctxText := retrieveRAGContext(ctx, svcCtx, dbCfg, question); ctxText != "" {
		systemPrompt += "\n--- 知识库 ---\n" + ctxText + "\n--------------"
	}

	config := openai.DefaultConfig(dbCfg.ChatApiKey)
	if dbCfg.ChatApiUrl != "" {
		config.BaseURL = dbCfg.ChatApiUrl
	}
	client := openai.NewClientWithConfig(config)
	stream, err := client.CreateChatCompletionStream(ctx, openai.ChatCompletionRequest{
		Model: chatModel(dbCfg),
		Messages: []openai.ChatCompletionMessage{
			{Role: openai.ChatMessageRoleSystem, Content: systemPrompt},
			{Role: openai.ChatMessageRoleUser, Content: question},
		},
		Stream: true,
	})
	if err != nil {
		return err
	}
	defer stream.Close()

	var answer strings.Builder
	for {
		resp, recvErr := stream.Recv()
		if errors.Is(recvErr, io.EOF) {
			break
		}
		if recvErr != nil {
			return recvErr
		}
		if len(resp.Choices) > 0 {
			content := resp.Choices[0].Delta.Content
			if content != "" {
				_, _ = fmt.Fprintf(w, "data: %s\n\n", content)
				if flusher, ok := w.(http.Flusher); ok {
					flusher.Flush()
				}
				answer.WriteString(content)
			}
		}
	}

	SaveHistory(ctx, svcCtx, userID, question, answer.String())
	return nil
}

func ensureQuota(ctx context.Context, svcCtx *svc.ServiceContext, userID string, dbCfg *plantModel.SysAiConfig) error {
	if userID == "" {
		return nil
	}
	quota, err := svcCtx.PlantRpc.GetAiChatQuota(ctx, &plantPb.GetProfileReq{UserId: userID})
	if err != nil {
		return err
	}
	limit := int64(dbCfg.DailyQueryLimit)
	if limit > 0 && quota.Remaining <= 0 {
		return fmt.Errorf("今日问答次数已达上限（%d次），明天再来吧", limit)
	}
	return nil
}