hongze_clpt/services/article.go

package services

import (
	"errors"
	"fmt"
	"github.com/PuerkitoBio/goquery"
	"hongze/hongze_clpt/models"
	"hongze/hongze_clpt/utils"
	"html"
	"regexp"
	"sort"
	"strconv"
	"strings"
	"time"
	"unicode/utf8"
)

func FixArticleImgUrl(body string) (contentSub string, err error) {
	r := strings.NewReader(string(body))
	doc, err := goquery.NewDocumentFromReader(r)
	if err != nil {
		fmt.Println(err)
	}
	doc.Find("img").Each(func(i int, s *goquery.Selection) {

		src, _ := s.Attr("src")
		if i == 0 && src != "" {
			contentSub = src
		}
	})
	return
}

//GetReportContentTextSubByarticle 解析文章内容
func GetReportContentTextSubByarticle(content, abstract string, articleId int) (contentSub string, err error) {
	var lenabstract int
	//如果不是研选就这么展示
	if articleId < utils.SummaryArticleId {
		abstract = html.UnescapeString(abstract)
		doc, errdoc := goquery.NewDocumentFromReader(strings.NewReader(abstract))
		if errdoc != nil {
			err = errdoc
			return
		}
		docabstract := doc.Text()
		lenabstract = utf8.RuneCountInString(docabstract)
		if lenabstract >= 20 {
			contentSub = docabstract
			return
		} else {
			contentSub, err = GetReportContentTextSub(content)
		}
	} else {
		contentSub, err = GetReportContentTextSub(content)
	}
	return
}

func GetReportContentTextSub(content string) (contentSub string, err error) {
	content = html.UnescapeString(content)
	doc, errdoc := goquery.NewDocumentFromReader(strings.NewReader(content))
	if errdoc != nil {
		err = errdoc
		return
	}
	docText := doc.Text()
	bodyRune := []rune(docText)
	bodyRuneLen := len(bodyRune)
	body := string(bodyRune[:bodyRuneLen])
	contentSub = body
	contentSub = strings.Replace(body, "Powered by Froala Editor", "", -1)
	contentSub = strings.Replace(body, "PoweredbyFroalaEditor", "", -1)
	contentSub = strings.Replace(body, " ", "", -1)
	return
}

func GetReportContentTextArticleBody(content string) (contentSub string) {
	contentSub = html.UnescapeString(content)
	contentSub = strings.Replace(contentSub, "<p data-f-id=\"pbf\" style=\"text-align: center; font-size: 14px; margin-top: 30px; opacity: 0.65; font-family: sans-serif;\">Powered by <a href=\"https://www.froala.com/wysiwyg-editor?pb=1\" title=\"Froala Editor\">Froala Editor</a></p>", "", -1)
	contentSub = strings.Replace(contentSub, "pre", "div", -1)
	return
}

//HandleArticleCategoryImg 预处理文章的封面图片
func HandleArticleCategoryImg(list []*models.ArticleListResp) (items []*models.ArticleListResp, err error) {
	//研选的五张图片
	detailResearch, e := models.GetConfigByCode("category_research_img_url")
	if e != nil {
		err = errors.New("获取研选的五张图片失败" + e.Error())
		return
	}
	researchList := strings.Split(detailResearch.ConfigValue, "{|}")
	//对应分类的所图片
	detailCategoryUrl, err := models.GetConfigByCode("category_map_img_url")
	if err != nil {
		err = errors.New("获取对应分类的所图片失败" + err.Error())
		return
	}
	categoryUrlList := strings.Split(detailCategoryUrl.ConfigValue, "{|}")
	mapCategoryUrl := make(map[string]string)
	var categoryId string
	var imgUrlChart string
	for _, v := range categoryUrlList {
		vslice := strings.Split(v, "_")
		categoryId = vslice[0]
		imgUrlChart = vslice[len(vslice)-1]
		mapCategoryUrl[categoryId] = imgUrlChart
	}

	mapChartPerssion := make(map[string]string)
	reportMappingList, err := models.GetReportMappingStrategyAll()
	if err != nil {
		err = errors.New("GetReportMappingStrategyAll err" + err.Error())
		return
	}

	for _, v := range reportMappingList {
		mapChartPerssion[strconv.Itoa(v.CategoryId)] = v.ChartPermissionName
	}
	for k, v := range list {
		if list[k].Annotation == "" {
			imgurl, _ := FixArticleImgUrl(html.UnescapeString(list[k].Body))
			if imgurl != "" {
				list[k].BodyImg = imgurl
			}
		}
		item := list[k]
		//如果文章一开始的内容是图片，优先展示第一张图片
		//newBody, _ := GetReportContentTextSubByarticle(item.Body, item.Annotation, item.ArticleId)
		list[k].Annotation = ArticleAnnotation(item)
		list[k].Body = ""
		list[k].Abstract, _ = GetReportContentTextSub(v.Abstract)
		list[k].PublishDate = utils.StrTimeToTime(item.PublishDate).Format(utils.FormatDate) //时间字符串格式转时间格式

		list[k].ChartPermissionName = mapChartPerssion[v.CategoryId]
		//如果是研选系列的任意取五张图片的中的一张
		if v.CategoryId == "0" || v.ArticleId > utils.SummaryArticleId {
			knum := v.ArticleId % 5
			list[k].ImgUrlPc = researchList[knum]
		} else {
			list[k].ImgUrlPc = mapCategoryUrl[v.CategoryId]
		}
		if list[k].ArticleId < utils.SummaryArticleId {
			list[k].HttpUrl = utils.StrategyPlatform + strconv.Itoa(v.ArticleId)
			list[k].IsNeedJump = true
		}
		list[k].Source = 1
		//添加行业默认图片
		if v.ImgUrlPc == "" {
			if v.ChartPermissionName == utils.YI_YAO_NAME {
				list[k].ImgUrlPc = utils.YI_YAO_OTHER_IMG
			} else if v.ChartPermissionName == utils.XIAO_FEI_NAME {
				list[k].ImgUrlPc = utils.XIAO_FEI_OTHER_IMG
			} else if v.ChartPermissionName == utils.KE_JI_NAME {
				list[k].ImgUrlPc = utils.KE_JI_OTHER_IMG
			} else if v.ChartPermissionName == utils.ZHI_ZAO_NAME {
				list[k].ImgUrlPc = utils.ZHI_ZAO_OTHER_IMG
			}
		}
		list[k].IsResearch = utils.InArrayByInt(utils.YAN_XUAN_ARTICLE_TYPE_ID, v.ArticleTypeId)

	}

	articleIds := make([]int, 0)
	for i := range list {
		articleIds = append(articleIds, list[i].ArticleId)
	}
	// 报告关联产业信息
	industryMap := make(map[int][]*models.IndustrialManagementIdInt, 0)
	if len(articleIds) > 0 {
		var industryCond string
		var industryPars []interface{}
		industryCond += ` AND mg.article_id IN (` + utils.GetOrmInReplace(len(articleIds)) + `)`
		industryPars = append(industryPars, articleIds)
		industryList, e := models.GetIndustrialListByarticleId(industryPars, industryCond)
		if e != nil {
			err = errors.New("GetIndustrialListByarticleId" + e.Error())
			return
		}
		for i := range industryList {
			v := industryList[i]
			industryMap[v.ArticleId] = append(industryMap[v.ArticleId], &models.IndustrialManagementIdInt{
				ArticleId:              v.ArticleId,
				IndustrialManagementId: v.IndustrialManagementId,
				IndustryName:           v.IndustryName,
				ChartPermissionId:      v.ChartPermissionId,
			})
		}
	}
	for k, v := range list {
		if len(industryMap[v.ArticleId]) > 0 {
			list[k].List = industryMap[v.ArticleId]
		} else {
			list[k].List = make([]*models.IndustrialManagementIdInt, 0)
		}
	}

	if len(list) == 0 {
		list = make([]*models.ArticleListResp, 0)
	}
	items = list
	return
}

//HandleArticleStock 处理报告关联的个股标签
func HandleArticleStock(stock string) (items []*models.ComapnyNameResp) {
	sliceSubjects := strings.Split(stock, "/")
	if len(sliceSubjects) > 0 {
		for _, vSubject := range sliceSubjects {
			sliceKuohao := strings.Split(vSubject, "(")           //过滤括号
			sliceXiahuaxian := strings.Split(sliceKuohao[0], "-") //过滤下划线
			subject := sliceXiahuaxian[0]
			items = append(items, &models.ComapnyNameResp{ComapnyName: subject})
		}
	}
	return
}

//弘则报告发布日期在三个月以内的
func GetArticNewLabelWhithActivity3Month() (labelMap map[int]bool, err error) {
	var condition string
	var pars []interface{}
	condition += ` AND publish_date  <= ?  AND article_id < ? `
	pars = append(pars, time.Now().AddDate(0, -3, 0), utils.SummaryArticleId)
	articleList, e := models.GetArticleList(condition, pars)
	if e != nil {
		err = errors.New("GetArticleList, Err: " + e.Error())
		return
	}
	var articleIds []int
	for _, v := range articleList {
		articleIds = append(articleIds, v.ArticleId)
	}
	if len(articleIds) == 0 {
		return
	}

	pars = make([]interface{}, 0)
	condition = ` AND article_id IN (` + utils.GetOrmInReplace(len(articleIds)) + `)`
	pars = append(pars, articleIds)
	industrialList, e := models.GetIndustrialArticleGroupManagementList(condition, pars)
	if e != nil {
		err = errors.New("GetIndustrialArticleGroupManagementList, Err: " + e.Error())
		return
	}
	labelMap = make(map[int]bool, 0)
	var industrialIds []int
	for _, v := range industrialList {
		industrialIds = append(industrialIds, v.IndustrialManagementId)
	}

	// 获取活动关联的产业
	var groupCond string
	var groupPars []interface{}
	groupCond += ` AND b.industrial_management_id IN (` + utils.GetOrmInReplace(len(industrialIds)) + `)  AND b.source = 1  `
	groupPars = append(groupPars, industrialIds)
	groups, e := models.GetActivityIndustryRelationList(groupCond, groupPars)
	if e != nil {
		err = errors.New("获取活动产业关联列表失败, Err: " + e.Error())
		return
	}
	for _, v := range groups {
		labelMap[v.ActivityId] = true
	}
	return
}

//GetSpecialArticleDetailUserPower 处理用户查看专项调研文章详情的权限
func GetSpecialArticleDetailUserPower(user *models.WxUserItem, articleInfo *models.ArticleDetail) (havePower bool, err error) {
	permissionStr, e := GetCompanyPermissionUpgrade(user.CompanyId)
	if e != nil {
		err = errors.New("GetCompanyPermissionUpgrade, Err: " + e.Error())
		return
	}
	reportMapDetail, e := models.GetdetailByCategoryIdPush(articleInfo.CategoryId)
	if e != nil {
		err = errors.New("GetdetailByCategoryIdPush, Err: " + e.Error())
		return
	}
	if reportMapDetail == nil {
		err = errors.New("GetdetailByCategoryIdP,获取详情失败, Err: ")
		return
	}
	//如果没有对应的升级权限，则返回
	if !strings.Contains(permissionStr, reportMapDetail.ChartPermissionName) {
		return
	} else {
		havePower = true
	}
	return
}

//处理核心观点的展示规则
//func ArticleAnnotation(item *models.ArticleListResp) (annotation string) {
//	if item.Annotation != "" {
//		annotation = strings.Replace(item.Annotation, "<br>", "", -1)
//	}
//	return
//}

func GetReportContentTextSubNew(content string) (contentSub string, err error) {
	content = html.UnescapeString(content)
	doc, errdoc := goquery.NewDocumentFromReader(strings.NewReader(content))
	if errdoc != nil {
		err = errdoc
		return
	}
	docText := doc.Text()
	bodyRune := []rune(docText)
	bodyRuneLen := len(bodyRune)
	body := string(bodyRune[:bodyRuneLen])
	contentSub = body
	contentSub = strings.Replace(contentSub, "Powered by Froala Editor", "", -1)
	contentSub = strings.Replace(contentSub, " ", "", -1)
	contentSub = strings.Replace(contentSub, "<p data-f-id=\"pbf\" style=\"text-align: center; font-size: 14px; margin-top: 30px; opacity: 0.65; font-family: sanered by <a href=\"https://www.froala.com/wysiwyg-editor?pb=1\" title=\"Froala Editor\">Froala Editor</a></p>", "", -1)
	return
}

//处理核心观点的展示规则
func ArticleAnnotation(item *models.ArticleListResp) (annotation string) {
	if item.ArticleId >= utils.SummaryArticleId {
		item.Annotation = YxArticleAnnotation(item)
	}
	if item.Annotation != "" {
		annotation = strings.Replace(item.Annotation, "<br>", "", -1)
	} else {
		return
	}
	bodyText, _ := GetReportContentTextSubNew(annotation)
	if bodyText == "" {
		return
	}
	if annotation != "" {
		annotation = html.UnescapeString(annotation)
		doc, _ := goquery.NewDocumentFromReader(strings.NewReader(annotation))
		docText := doc.Text()
		mapDoc := make(map[int]string)
		var mapSort []int
		p := doc.Find("p")
		p.Each(func(tk int, pd *goquery.Selection) {
			pdText := pd.Text()
			pdText = strings.Replace(pdText, " ", "", -1)
			if pdText != "" {
				textLen := strings.Index(docText, pdText)
				mapDoc[(strings.Index(docText, pdText))] = pdText
				mapSort = append(mapSort, textLen)
			}
		})
		li := doc.Find("li")
		li.Each(func(tk int, li *goquery.Selection) {
			liText := li.Text()
			liText = strings.Replace(liText, " ", "", -1)
			if liText != "" {
				textLen := strings.Index(docText, liText)
				mapDoc[(strings.Index(docText, liText))] = strconv.Itoa(tk+1) + "." + liText
				mapSort = append(mapSort, textLen)
			}
		})

		ul := doc.Find("ul")
		ul.Each(func(tk int, ul *goquery.Selection) {
			ulText := ul.Text()
			ulText = strings.Replace(ulText, " ", "", -1)
			if ulText != "" {
				textLen := strings.Index(docText, ulText)
				mapDoc[(strings.Index(docText, ulText))] = ulText
				mapSort = append(mapSort, textLen)
			}
		})
		if len(mapSort) == 0 {
			return
		} else {
			//排序
			sort.Ints(mapSort)
			var annotationHtml string
			for _, vSort := range mapSort {
				for k, v := range mapDoc {
					if k == vSort && v != "" {
						annotationHtml += v + "<br>"
					}
				}
			}
			annotationHtml = strings.TrimRight(annotationHtml, "<br>")
			annotationHtml = "<p>" + annotationHtml + "</p>"
			annotation = annotationHtml
		}
	}
	return
}

//解析研选内容中的核心观点
func YxArticleAnnotation(article *models.ArticleListResp) (annotation string) {
	//如果不规范，就获取内容主体
	if strings.Count(article.Body, "<hr") == 0 {
		//如果内容不规范而且，还有图片，就把核心观点置空
		if article.BodyImg != "" {
			return
		}
		annotation, _ = GetReportContentTextSub(article.Body)
		return
	}
	body := strings.ReplaceAll(article.Body, "<strong>", "")
	body = strings.ReplaceAll(body, "</strong>", "")
	body = strings.ReplaceAll(body, "</ol>", "</div>")
	body = strings.ReplaceAll(body, "<ol>", "<div>")
	body = strings.ReplaceAll(body, "</li>", "</p>")
	body = strings.ReplaceAll(body, "<li>", "<p>")
	re, _ := regexp.Compile("<strong.*?>")
	body = re.ReplaceAllString(body, "")
	reLi, _ := regexp.Compile("<li.*?>")
	body = reLi.ReplaceAllString(body, "")
	var plus int
	coreIndex := strings.Index(body, "核心观点：")
	plus = 15
	if coreIndex == -1 {
		coreIndex = strings.Index(body, "核心观点:")
		plus = 13
	}
	if coreIndex == -1 {
		coreIndex = strings.Index(body, "核心观点")
		plus = 12
	}
	if coreIndex == -1 {
		coreIndex = strings.Index(body, "核心结论：")
		plus = 15
	}
	if coreIndex == -1 {
		coreIndex = strings.Index(body, "核心结论:")
		plus = 13
	}
	if coreIndex == -1 {
		coreIndex = strings.Index(body, "核心结论")
		plus = 12
	}
	endIndex := strings.Index(body, "<hr")
	if coreIndex != -1 && endIndex != -1 {
		body = body[coreIndex+plus : endIndex]
	}
	annotation = body
	return
}