package WebCrawler

import (
	"gitee.com/fierce_wolf/go-fox-edge-common/commUtil/Slice"
	"github.com/PuerkitoBio/goquery"
	"regexp"
	"strings"
)

type webSite struct {
}

// ReadPage 读取页面的元素信息
func (e *webSite) ReadPage(host, url string) (*WebPage, error) {
	doc, err := Http.HttpGet(host+url, 0)
	if err != nil {
		return nil, err
	}

	page := &WebPage{}
	page.Host = host
	page.Url = url

	// 获得页面的头部信息
	page.Head = e.GetHead(doc.Selection)

	// 获得下一层的链接信息
	page.Hrefs = e.GetHrefs(doc.Selection)
	delete(page.Hrefs, host)

	// 获得文本相关的元素
	page.Texts = e.GetTexts(doc.Selection)

	// 获得文本相关的元素
	page.Images = e.GetImages(doc.Selection)
	return page, nil
}

// FindTexts 遍历每一个文本的元素，主要是用于分析页面的信息
// 参数：finders 遍历时的查找器
func (e *webSite) FindTexts(page *WebPage, finders []IFindElement) (*WebPage, error) {
	for _, finder := range finders {
		finder.Find(page.Texts)
	}

	return page, nil
}

// FindImages 遍历每一个文本的元素，主要是用于分析页面的信息
// 参数：finders 遍历时的查找器
func (e *webSite) FindImages(page *WebPage, finders []IFindElement) (*WebPage, error) {
	for _, finder := range finders {
		finder.Find(page.Images)
	}

	return page, nil
}

// GetHead 获得页面的常见头部信息
func (e *webSite) GetHead(selection *goquery.Selection) *WebHead {
	head := &WebHead{}

	head.Title = selection.Find("title").Text()
	head.Description = selection.Find("meta[name=description]").AttrOr("content", "")

	keywords := selection.Find("meta[name=keywords]").AttrOr("content", "")
	head.Keywords = strings.Split(keywords, ",")

	return head
}

// GetHrefs 获得下一层的链接
func (e *webSite) GetHrefs(selection *goquery.Selection) map[string]*WebHref {
	hrefs := make(map[string]*WebHref, 0)
	selection.Find("a").Each(func(i int, s *goquery.Selection) {
		// 获取<a>标签的href属性
		href, exists := s.Attr("href")
		if !exists {
			return
		}

		// 忽略指向自身的url
		if href == "" || href == "#" {
			return
		}

		hrefs[href] = &WebHref{
			Href:  href,
			Title: s.Text(),
		}
	})

	return hrefs
}

// GetTexts 获得文本节点
func (e *webSite) GetTexts(selection *goquery.Selection) []*WebElement {
	keys := make([]string, 0)
	keys = append(keys, "h")
	keys = append(keys, "h1")
	keys = append(keys, "h2")
	keys = append(keys, "h3")
	keys = append(keys, "h4")
	keys = append(keys, "h5")
	keys = append(keys, "p")
	keys = append(keys, "a")
	keys = append(keys, "span")

	result := make([]*WebElement, 0)

	for _, key := range keys {
		list := e.getTexts(selection, key)
		result = Slice.Append(result, list)
	}

	return result
}

// getTexts 获得文本节点
// 参数说明：selector 是html的节点类型，比如<h2>、<a>、<p>
func (e *webSite) getTexts(selection *goquery.Selection, selector string) []*WebElement {
	list := make([]*WebElement, 0)
	selection.Find(selector).Each(func(i int, s *goquery.Selection) {
		text := strings.TrimSpace(s.Text())
		if text == "" {
			return
		}

		list = append(list, &WebElement{
			Text:      text,
			Tag:       selector,
			Selection: s,
		})
	})

	return list
}

// GetImages 获得文本节点
// 参数说明：selector 是html的节点类型，比如<h2>、<a>、<p>
func (e *webSite) GetImages(selection *goquery.Selection) []*WebElement {
	list := make([]*WebElement, 0)
	selection.Find("img").Each(func(i int, s *goquery.Selection) {
		// 获取<a>标签的href属性
		href, exists := s.Attr("src")
		if !exists {
			return
		}

		// 忽略指向自身的url
		if len(href) <= 1 {
			return
		}

		list = append(list, &WebElement{
			Text:      href,
			Tag:       "img",
			Selection: s,
		})
	})

	return list
}

// FindElements 使用过滤器查找元素
func (e *webSite) FindElements(els []*WebElement, finder Filter) []*WebElement {
	data := make([]*WebElement, 0)

	for _, node := range els {
		if finder.Mode == ModeContain {
			if strings.Contains(strings.ToLower(node.Text), strings.ToLower(finder.Param.(string))) {
				data = append(data, node)
			}
			continue
		}
		if finder.Mode == ModeEqual {
			if node.Text == finder.Param {
				data = append(data, node)
			}
			continue
		}
		if finder.Mode == ModeRegexp {
			matches := finder.Param.(*regexp.Regexp).FindAllString(node.Text, -1)
			if len(matches) > 0 {
				data = append(data, node)
			}
			continue
		}
		if finder.Mode == ModeElTag {
			if node.Tag == finder.Param {
				data = append(data, node)
			}
			continue
		}
	}

	return data
}