eta_data_analysis/services/liangyou/commodity_liangyou.go

package liangyou

import (
	"context"
	"encoding/json"
	"eta/eta_data_analysis/models"
	"eta/eta_data_analysis/services/alarm_msg"
	"eta/eta_data_analysis/utils"
	"fmt"
	"github.com/beego/beego/v2/core/logs"
	"github.com/chromedp/cdproto/cdp"
	"log"
	"os"
	"regexp"
	"strconv"
	"strings"
	"time"

	"github.com/chromedp/chromedp"
)

var (
	lyLoginPath = "https://www.fao.com.cn/"
)

func LyDataDeal(cont context.Context) (err error) {

	// 读取 JSON 文件
	configFile, err := os.ReadFile(utils.LY_JSON_PATH)
	if err != nil {
		fmt.Printf("读取配置文件错误: %v\n", err)
		return
	}

	// 定义通用的 map 结构体来解析 JSON
	var data map[string]map[string]map[string][]string

	// 解析 JSON 文件内容
	err = json.Unmarshal(configFile, &data)
	if err != nil {
		fmt.Printf("解析配置文件错误: %v\n", err)
		return
	}

	// 打印解析后的数据以验证
	fmt.Printf("%+v\n", data)

	// 创建 chromedp 执行上下文
	options := []chromedp.ExecAllocatorOption{
		chromedp.Flag("headless", false),
		chromedp.Flag("disable-blink-features", "AutomationControlled"),
		chromedp.UserAgent(`Mozilla/5.0 (Macintosh; Intel Mac OS X 10_15_7) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/98.0.4758.80 Safari/537.36`),
	}

	allocCtx, cancel := chromedp.NewExecAllocator(context.Background(), options...)
	defer cancel()
	ctx, cancel := chromedp.NewContext(allocCtx)
	defer cancel()

	// 登录操作
	err = login(ctx)
	if err != nil {
		fmt.Printf("登录错误: %v\n", err)
		return
	}

	// 遍历配置并爬取数据
	for product, productData := range data {
		for category, categoryData := range productData {
			for report, keywords := range categoryData {
				fmt.Printf("正在获取数据: %s -> %s -> %s\n", product, category, report)
				err = fetchReportData(ctx, product, category, report, keywords)
				if err != nil {
					fmt.Printf("获取数据错误: %s -> %s -> %s: %v\n", product, category, report, err)
					// 您看文章的速度太快了，歇一会再看吧
					if strings.Contains(err.Error(), "您看文章的速度太快了，歇一会再看吧") {
						return
					}
				}
			}
		}
	}
	return nil
}

func login(ctx context.Context) error {

	return chromedp.Run(ctx,
		chromedp.Navigate(lyLoginPath),
		chromedp.Sleep(5*time.Second),
		chromedp.Click(`a[id="btnLogin"]`, chromedp.ByQuery),
		chromedp.Sleep(2*time.Second),
		chromedp.SetValue(`input[id="userName"]`, utils.LY_USERNAME, chromedp.ByQuery),
		chromedp.SetValue(`input[id="pwd"]`, utils.LY_PASSWORD, chromedp.ByQuery),
		chromedp.Sleep(2*time.Second),
		chromedp.Click(`input[id="btn_Login"]`, chromedp.ByQuery),
		chromedp.Sleep(5*time.Second),
	)
}

func fetchReportData(ctx context.Context, product, category, report string, keywords []string) error {
	// Navigate to the main page
	err := chromedp.Run(ctx,
		chromedp.Navigate(lyLoginPath),
		chromedp.Sleep(5*time.Second),
	)
	if err != nil {
		return err
	}

	// Navigate to the product page
	productPageURL, err := fillProductPageURL(ctx, product, category)
	if err != nil {
		return err
	}

	// Navigate to the category page
	var categoryPageURL string
	err = chromedp.Run(ctx,
		chromedp.Navigate(productPageURL),
		chromedp.Sleep(5*time.Second),
		chromedp.Click(fmt.Sprintf(`//div[contains(@class, "newBox")]//a[contains(text(), '%s')]`, category), chromedp.BySearch),
		chromedp.Sleep(5*time.Second),
		chromedp.Location(&categoryPageURL),
	)
	if err != nil {
		return err
	}
	logs.Info("categoryPageURL: %s: %s: %s", product, category, categoryPageURL)

	//var allReportURLs []string
	var allReportURLMap = make(map[string]string)
	for {
		var htmlContent string
		err = chromedp.Run(ctx,
			chromedp.Navigate(categoryPageURL),
			chromedp.Sleep(5*time.Second),
			chromedp.OuterHTML("html", &htmlContent),
		)
		if err != nil {
			return err
		}
		fmt.Printf("页面内容: %s\n", htmlContent)

		// Extract report URLs containing the partial keyword
		reportURLMap := extractReportURLs(htmlContent, report)
		//allReportURLs = append(allReportURLs, reportURLs...)
		for key, value := range reportURLMap {
			allReportURLMap[key] = value
		}

		//  测试环境跑部分数据，上线放开
		//break
		// Check if next page button is disabled
		//  测试环境跑部分数据，上线放开
		var nextPageDisabled bool
		err = chromedp.Run(ctx,
			chromedp.Evaluate(`document.querySelector('div.my-page-next').classList.contains('my-page-forbid')`, &nextPageDisabled),
		)
		if err != nil {
			return err
		}

		if nextPageDisabled {
			break
		}

		// Click the next page button
		err = chromedp.Run(ctx,
			chromedp.Click(`div.my-page-next`, chromedp.ByQuery),
			chromedp.Sleep(5*time.Second),
			chromedp.Location(&categoryPageURL),
		)
		if err != nil {
			return err
		}

	}

	logs.Info("所有报告 URLs: %s: %s: %v", product, category, allReportURLMap)

	if len(allReportURLMap) == 0 {
		return fmt.Errorf("未找到报告 URL")
	}

	// 处理报告数据
	for key, value := range allReportURLMap {
		// 查询报告是否已经处理  这里只对近7天的数据进行处理
		paramsLib := make(map[string]interface{})
		paramsLib["Url"] = key
		postEdbLib, err := utils.PostEdbLibRequest(paramsLib, utils.GET_LY_INDEX_RECORD_BY_URL)
		if err != nil {
			// 有错误就不继续执行
			log.Printf("postEdbLib err: %v", err)
			continue
		}
		var requestResponse models.RequestResponse[models.BaseFromLyIndexRecord]
		err = json.Unmarshal(postEdbLib, &requestResponse)
		lyIndexRecord := requestResponse.Data
		if lyIndexRecord.DataTime != "" {
			toTime, err := utils.StringToTime(lyIndexRecord.DataTime + " 00:00:00")
			if err != nil {
				logs.Error("时间格式转换错误: %s: %s: %s: %s: %v", product, category, report, key, err)
				continue
			}

			if time.Now().Sub(toTime) > 7*24*time.Hour {
				logs.Info("报告已处理: %s: %s: %s: %s", product, category, report, key)
				continue
			}
		}

		// 随机睡眠
		rand := utils.RangeRand(20, 100)
		fmt.Println(report+";sleep:", strconv.Itoa(int(rand)))
		time.Sleep(time.Duration(rand) * time.Second)

		err = processReport(ctx, product, category, key, keywords)
		if err != nil {
			logs.Error("处理报告错误: %s: %s: %s: %s: %v", product, category, report, key, err)
			if strings.Contains(err.Error(), "您看文章的速度太快了，歇一会再看吧") {
				// 如果报告内容包含 “您看文章的速度太快了，歇一会再看吧” 则停止处理，发短信通知
				// 发送短信通知
				alarm_msg.SendAlarmMsg(fmt.Sprintf("粮油商务网-爬取指标数据被限制，请稍后重试, ErrMsg: %s", err.Error()), 1)
				return nil
			}
			continue
		}

		format, err := utils.ConvertTimeFormat(value)
		if err != nil {
			logs.Error("时间格式转换错误: %s, %s, %v: %v", product, category, value, err)
			continue
		}

		// 处理报告成功，将维护指标数据读取进度到数据库，避免后面重复读取
		record := &models.BaseFromLyIndexRecord{
			CreateTime: utils.GetCurrentTime(),
			ModifyTime: utils.GetCurrentTime(),
			Product:    product,
			Category:   category,
			Url:        key,
			DataTime:   format,
		}
		// 转换成json
		marshal, err := json.Marshal(record)
		if err != nil {
			logs.Error("维护指标数据读取进度错误: %s, %s, %v: %v", product, category, key, err)
			continue
		}
		// json 转 interface
		var result map[string]interface{}
		err = json.Unmarshal(marshal, &result)

		postEdbLib, err = utils.PostEdbLibRequest(result, utils.ADD_LY_INDEX_RECORD)
		if err != nil {
			// 有错误就不继续执行
			logs.Error("维护指标数据读取进度错误: %s, %s, %v: %v", product, category, key, err)
			continue
		}
		logs.Info("维护指标数据读取进度成功: %s, %s, %v", product, category, key)
	}

	return nil
}

func fillProductPageURL(ctx context.Context, product string, category string) (string, error) {
	// 选择 dl 标签下所有 a 标签的 XPath
	selector := `//dl[contains(@class, 'dl_hot')]//a`
	logs.Info("选择器表达式: %s", selector)

	var nodes []*cdp.Node
	var productPageURL string

	// 获取 dl 标签下的所有 a 标签节点
	err := chromedp.Run(ctx,
		chromedp.WaitReady(selector, chromedp.BySearch),
		chromedp.Nodes(selector, &nodes, chromedp.BySearch),
	)
	if err != nil {
		return "", err
	}

	// 提取并打印所有 a 标签的 OuterHTML
	var targetURL string
	for _, node := range nodes {
		var outerHTML string

		// 获取 a 标签的 OuterHTML
		err = chromedp.Run(ctx,
			chromedp.OuterHTML(node.FullXPath(), &outerHTML, chromedp.BySearch),
		)
		if err != nil {
			return "", err
		}

		// 打印获取的 OuterHTML 内容
		logs.Info("Link OuterHTML: %s", outerHTML)

		// 从 OuterHTML 中提取 href 和文本内容
		// 使用正则或字符串处理提取 href 和文本内容
		href, linkText := extractHrefAndText(outerHTML)

		// 打印提取的 href 和文本内容
		logs.Info("Link Text: %s, Href: %s", linkText, href)

		// 如果文本内容匹配目标产品
		if linkText == product {
			// 拼接完整的 URL
			/*if !strings.HasPrefix(href, "http") {
				href = lyLoginPath + href
			}*/
			targetURL = href
			break
		}
	}

	if targetURL == "" {
		return "", fmt.Errorf("未找到匹配的产品链接")
	}

	// 显示更多内容
	err = chromedp.Run(ctx,
		chromedp.Evaluate(`document.getElementById("moreSpeList").style.display = "block";`, nil),
	)
	if err != nil {
		return "", err
	}

	// 点击目标产品的链接
	clickSelector := fmt.Sprintf(`//a[@href='%s']`, targetURL)
	err = chromedp.Run(ctx,
		chromedp.WaitReady(clickSelector, chromedp.BySearch),
		chromedp.Click(clickSelector, chromedp.BySearch),
		chromedp.Sleep(5*time.Second),
		chromedp.Location(&productPageURL),
	)
	if err != nil {
		return "", err
	}

	// 返回点击后的页面URL
	logs.Info("productPageURL: %s", productPageURL)
	return productPageURL, nil
}

// extractHrefAndText 从 OuterHTML 提取 href 和文本内容的辅助函数
func extractHrefAndText(outerHTML string) (string, string) {
	// 使用正则表达式或其他字符串处理方法提取 href 和文本内容
	// 这里只是一个简单的例子，具体实现需要根据 HTML 结构来调整
	hrefRegex := `href="([^"]+)"`
	textRegex := `>([^<]+)<`

	hrefMatches := regexp.MustCompile(hrefRegex).FindStringSubmatch(outerHTML)
	textMatches := regexp.MustCompile(textRegex).FindStringSubmatch(outerHTML)

	href := ""
	linkText := ""
	if len(hrefMatches) > 1 {
		href = hrefMatches[1]
	}
	if len(textMatches) > 1 {
		linkText = textMatches[1]
	}

	return href, linkText
}

// Extract report URLs from the HTML content
func extractReportURLs(htmlContent, keyword string) map[string]string {
	//var reportURLs []string
	var reportURLMap = make(map[string]string)
	var reportURL string

	// Find all occurrences of the keyword and extract report URLs
	content := htmlContent
	for {
		startIdx := strings.Index(content, keyword)
		if startIdx == -1 {
			break
		}
		startIdx += len(keyword)

		// Extract the URL from the HTML content
		urlStartIdx := strings.LastIndex(content[:startIdx], `href="`) + len(`href="`)
		urlEndIdx := strings.Index(content[urlStartIdx:], `"`) + urlStartIdx
		if urlStartIdx > 0 && urlEndIdx > urlStartIdx {
			reportURL = content[urlStartIdx:urlEndIdx]
			//reportURLs = append(reportURLs, content[urlStartIdx:urlEndIdx])
		}

		content = content[startIdx:]

		// Now extract the content inside the first <div class="short_right">
		divStartIdx := strings.Index(content, `<div class="short_right">`)
		if divStartIdx != -1 {
			divStartIdx += len(`<div class="short_right">`)
			divEndIdx := strings.Index(content[divStartIdx:], `</div>`) + divStartIdx
			if divEndIdx > divStartIdx {
				shortRightContent := content[divStartIdx:divEndIdx]

				// Extract the first <div> content inside <div class="short_right">
				innerDivStartIdx := strings.Index(shortRightContent, `<div>`)
				if innerDivStartIdx != -1 {
					innerDivStartIdx += len(`<div>`)
					//innerDivEndIdx := strings.Index(shortRightContent[innerDivStartIdx:], `</div>`) + innerDivStartIdx
					innerDivContent := shortRightContent[innerDivStartIdx:]
					fmt.Println("Inner Div Content:", innerDivContent)
					reportURLMap[reportURL] = innerDivContent
				}
			}
		}
	}

	return reportURLMap
}

// Process the report data
func processReport(ctx context.Context, product string, category string, reportURL string, keywords []string) error {
	// Navigate to the report page
	var reportContent string
	err := chromedp.Run(ctx,
		chromedp.Navigate(lyLoginPath+reportURL),
		chromedp.WaitVisible("body", chromedp.ByQuery), // 等待 body 元素可见，确保页面已加载
		chromedp.Sleep(5*time.Second),                  // 等待额外时间，以确保动态内容加载
		chromedp.OuterHTML("html", &reportContent),     // 获取页面 HTML 内容
	)
	if err != nil {
		return err
	}

	// 如果文章内容包含 “您看文章的速度太快了，歇一会再看吧” 则返回指定错误
	if strings.Contains(reportContent, "您看文章的速度太快了，歇一会再看吧") {
		return fmt.Errorf("您看文章的速度太快了，歇一会再看吧")
	}

	var lyIndexDataList []models.BaseFromLyData
	var edbDataLyList []models.EdbDataLy
	// Process the data based on keywords
	for _, keyword := range keywords {
		partialKeyword := strings.Split(keyword, "：")
		// Select appropriate processor based on product and category
		processor, err := GetProcessor(product, category)
		if err != nil {
			return err
		}

		// 查询报告所属分类
		paramsLib := make(map[string]interface{})
		paramsLib["CategoryName"] = product
		postEdbLib, err := utils.PostEdbLibRequest(paramsLib, utils.GET_LY_CLASSIFY_BY_NAME)
		if err != nil {
			// 有错误就不继续执行
			log.Printf("postEdbLib err: %v", err)
			continue
		}
		var requestResponse models.RequestResponse[models.BaseFromLyClassify]
		err = json.Unmarshal(postEdbLib, &requestResponse)
		if err != nil {
			return err
		}
		classify := requestResponse.Data

		// Process the report content using the selected processor
		baseFromLyDataList, err := processor.Process(ctx, product, reportContent, partialKeyword, classify.BaseFromLyClassifyId)
		if err != nil {
			return err
		}
		if len(baseFromLyDataList) > 0 {
			for _, baseFromLyData := range baseFromLyDataList {
				if baseFromLyData.DataTime != "" && baseFromLyData.IndexCode != "" && baseFromLyData.IndexCode != "lysww" {
					baseFromLyData.CreateTime = utils.GetCurrentTime()
					baseFromLyData.ModifyTime = utils.GetCurrentTime()

					// 补充 判断是否存在于指标库
					paramsLib = make(map[string]interface{})
					paramsLib["IndexCode"] = baseFromLyData.IndexCode
					paramsLib["Source"] = utils.DATA_SOURCE_LY
					postEdbLib, err := httpRequestFill(paramsLib, utils.GET_EDB_INFO_BY_INDEX_CODE)
					if err != nil {
						// 有错误就不继续执行
						log.Printf("postEdbLib err: %v", err)
						continue
					}
					var requestResponse models.RequestResponse[models.EdbInfo]
					err = json.Unmarshal(postEdbLib, &requestResponse)
					if err != nil {
						log.Printf("postEdbLib err: %v", err)
						continue
					}

					lyIndexDataList = append(lyIndexDataList, baseFromLyData)

					edbIndexData := requestResponse.Data
					if edbIndexData.EdbCode == "" {
						// 不存在 不用维护指标库数据
						continue
					}

					// 存在 则同步新增指标库中的指标数据
					edbDataLy := models.EdbDataLy{
						CreateTime:    utils.GetCurrentTime(),
						ModifyTime:    utils.GetCurrentTime(),
						EdbInfoId:     edbIndexData.EdbInfoId,
						EdbCode:       edbIndexData.EdbCode,
						DataTime:      baseFromLyData.DataTime,
						Value:         baseFromLyData.Value,
						DataTimestamp: uint64(time.Now().UnixNano() / int64(time.Millisecond)),
					}
					edbDataLyList = append(edbDataLyList, edbDataLy)
				}
			}
		}
	}
	// 新增指标数据
	if len(lyIndexDataList) > 0 {
		// 转换成json
		marshal, err := json.Marshal(lyIndexDataList)
		if err != nil {
			log.Printf("postEdbLib err: %v", err)
			return err
		}
		_, err = utils.HttpPostRequest(utils.EDB_LIB_URL+utils.ADD_LY_DATA_LIST, string(marshal), "application/json")
		if err != nil {
			// 有错误就不继续执行
			log.Printf("postEdbLib err: %v", err)
			return err
		}
	}

	// 新增指标库数据
	if len(edbDataLyList) > 0 {
		// 转换成json
		marshal, err := json.Marshal(edbDataLyList)
		if err != nil {
			log.Printf("postEdbLib err: %v", err)
			return err
		}
		_, err = utils.HttpPostRequest(utils.EDB_LIB_URL+utils.ADD_BATCH_LY_EDB_DATA, string(marshal), "application/json")
		//_, err := httpRequestFill(edbDataLyList, utils.ADD_BATCH_LY_EDB_DATA)
		if err != nil {
			// 有错误就不继续执行
			log.Printf("postEdbLib err: %v", err)
			return err
		}
	}

	return nil
}

func addLyIndex(classifyId int, indexCode string, indexName string, unit string, frequency string) (int, error) {
	// 添加指标
	index := &models.BaseFromLyIndex{
		CreateTime:           utils.GetCurrentTime(),
		ModifyTime:           utils.GetCurrentTime(),
		BaseFromLyClassifyId: classifyId,
		IndexCode:            indexCode,
		IndexName:            indexName,
		Frequency:            frequency,
		Unit:                 unit,
		EdbExist:             0,
	}
	postEdbLib, err := httpRequestFill(index, utils.ADD_LY_INDEX)
	if err != nil {
		// 有错误就不继续执行
		log.Printf("postEdbLib err: %v", err)
		return 0, err
	}
	var requestResponse models.RequestResponse[int64]
	err = json.Unmarshal(postEdbLib, &requestResponse)
	indexId := requestResponse.Data
	return int(indexId), nil
}

func httpRequestFill(data interface{}, urlMethod string) (postEdbLib []byte, err error) {
	// 转换成json
	marshal, err := json.Marshal(data)
	if err != nil {
		return nil, err
	}
	// json 转 interface
	var result map[string]interface{}
	err = json.Unmarshal(marshal, &result)
	if err != nil {
		return nil, err
	}

	postEdbLib, err = utils.PostEdbLibRequest(result, urlMethod)
	if err != nil {
		// 有错误就不继续执行
		log.Printf("postEdbLib err: %v", err)
		return nil, err
	}
	return postEdbLib, nil
}