package utils

import (
	"os"
	"strings"

	"golang.org/x/net/html"
)

// ExtractHrefs 从HTML文件中提取所有rel="prefetch"的link标签的href属性
func ExtractHrefs(filename string) ([]string, error) {
	file, err := os.Open(filename)
	if err != nil {
		return nil, err
	}
	defer file.Close()

	doc, err := html.Parse(file)
	if err != nil {
		return nil, err
	}

	var hrefs []string
	var f func(*html.Node)
	f = func(n *html.Node) {
		if n.Type == html.ElementNode && n.Data == "link" {
			var href, rel string
			// 获取href和rel属性
			for _, a := range n.Attr {
				switch a.Key {
				case "href":
					href = a.Val
				case "rel":
					rel = a.Val
				}
			}
			// 如果是prefetch链接且href包含.html，添加到结果中
			if rel == "prefetch" && href != "" && strings.Contains(href, ".html") {
				hrefs = append(hrefs, href)
			}
		}
		for c := n.FirstChild; c != nil; c = c.NextSibling {
			f(c)
		}
	}
	f(doc)

	// 去重
	return removeDuplicates(hrefs), nil
}

// removeDuplicates 移除重复的URL
func removeDuplicates(urls []string) []string {
	seen := make(map[string]bool)
	result := []string{}

	for _, url := range urls {
		if !seen[url] {
			seen[url] = true
			result = append(result, url)
		}
	}

	return result
}