golang 解析html时把url转绝对路径

golang 解析html时把url转绝对路径

在golang解析html时，常遇到img、a标签的资源地址为相对路径[…/，./]的情况，这时就需要一个方法把这些地址转换为绝对路径。

// 返回url的绝对地址
// currUrl 当前url
func absUrl(currUrl, baseUrl string) string {
	urlInfo, err := url.Parse(currUrl)
	if err != nil {
		return ""
	}
	if urlInfo.Scheme != "" {
		return currUrl
	}
	baseInfo, err := url.Parse(baseUrl)
	if err != nil {
		return ""
	}

	u := baseInfo.Scheme + "://" + baseInfo.Host
	var path string
	if strings.Index(urlInfo.Path, "/") == 0 {
		path = urlInfo.Path
	} else {
		path = filepath.Dir(baseInfo.Path) + "/" + urlInfo.Path
	}

	rst := make([]string, 0)
	pathArr := strings.Split(path, "/")

	// 如果path是已/开头，那在rst加入一个空元素
	if pathArr[0] == "" {
		rst = append(rst, "")
	}
	for _, p := range pathArr {
		if p == ".." {
			if rst[len(rst)-1] == ".." {
				rst = append(rst, "..")
			} else {
				rst = rst[:len(rst)-1]
			}
		} else if p != "" && p != "." {
			rst = append(rst, p)
		}
	}
	return u + strings.Join(rst, "/")
}

测试

func Test(t *testing.T) {
	str := "https://xxx.com/articles/2876/1.html"
	s := "../../2.html"
	s1 := "./../2.html"
	s2 := "/2.html"
	s3 := "./2.html"
	s4 := "https://xxx.com/2.html"
	println(absUrl(s, str)) // https://xxx.com/2.html
	println(absUrl(s1, str)) // https://xxx.com/articles/2.html
	println(absUrl(s2, str)) // https://xxx.com/2.html
	println(absUrl(s3, str)) // https://xxx.com/articles/2876/2.html
	println(absUrl(s4, str)) // https://xxx.com/2.html
}