// 生成真实的url
// 传来的url可能是http://a.com, 也可能是a.com
// getRelativeUrl传来的可以是http://a.com
// url = a.com/a/?id=12&id=1221, 那么genUrl=a.com/a/index.html?id=121
func (this *LeaSpider) genUrl(url string) string {
// 去掉?后面的
queryParam, fragment := "", "" // 包含?,#
pos := strings.Index(url, "?")
if pos != -1 {
queryParam = util.Substring(url, pos)
url = util.Substr(url, 0, pos)
} else {
pos = strings.Index(url, "#")
if pos != -1 {
fragment = util.Substring(url, pos)
url = util.Substr(url, 0, pos)
}
}
// 如果url == host
if url == this.host || url == this.schemeAndHost {
return url + "/" + this.defaultFilename + queryParam + fragment
}
genFilename, needApend := this.genFilename(url)
if genFilename != "" {
if needApend {
url += "/" + genFilename + queryParam + fragment
} else {
// 是a.php => a.html
urlArr := strings.Split(url, "/")
urlArr = urlArr[:len(urlArr)-1]
url = strings.Join(urlArr, "/") + "/" + genFilename
}
}
return url
}
// 入口
func (this *LeaSpider) Fetch(url, targetPath string) {
url = strings.TrimSpace(url)
this.parseUrl(url)
// 保存路径
this.doTargetPath(targetPath)
// 去掉scheme
// a.com, a.com/index.html
url = util.Substring(url, len(this.scheme))
// url2, ok := this.getRalativeUrl("a.com/b/c/d/kk/eee.html", "http://a.com/e/c/d/kk")
// println(url2)
// println(ok)
// return
this.goDo(url, false)
this.w.Wait()
// 处理异常
this.doExceptionUrl()
}
func main() {
regular := "http:\\/|https:\\/|javascript:|mailto:|" class=|@.*?\\..+"
reg := regexp.MustCompile(regular)
url := "javascript:"
println(reg.MatchString(url))
filename := filepath.Dir("a/b/c/aaabceeeeeeee.php") // a.php
println(filename)
filename = util.Substr(filename, 0, len(filename)-len(".php")) // a
println(filename)
// _, err2 := os.Open("D:\\a.jpg")
_, err2 := os.Stat("D:/a.jpg")
if err2 != nil && os.IsNotExist(err2) {
println("file not exist!\n")
} else {
println("file exists")
}
println(trimQueryParams("a.jgp33333##"))
regular = "(?i)(src=|href=)[\"']([^#].*?)[\"']"
reg = regexp.MustCompile(regular)
println(reg.MatchString("url(a)"))
println()
re := reg.FindAllStringSubmatch("src='xaaxx3333333333331' href=\"aaaxx\"", -1)
for _, each := range re {
fmt.Println(each[2])
}
url = "http://www.a.comddd/b/c"
url = strings.Replace(url, "http://", "", 1)
index := strings.Index(url, "/")
if index == -1 {
println(url)
} else {
println(util.Substr(url, 0, index))
}
println("--------------")
arr := []string{"life", "ax", "cdj"}
sort.Strings(arr)
fmt.Println(arr)
cUrl := "//lifedddddddddddddddddddd"
println(util.Substring(cUrl, 2))
queryParam, fragment := "", ""
url = "http://a.com?id=12#ddd"
pos := strings.Index(url, "?")
if pos != -1 {
queryParam = util.Substring(url, pos)
url = util.Substr(url, 0, pos)
} else {
pos = strings.Index(url, "#")
if pos != -1 {
fragment = util.Substring(url, pos)
url = util.Substr(url, 0, pos)
}
}
println(queryParam, fragment)
urlArr := strings.Split("a/b/c/d.html", "/")
urlArr = urlArr[:len(urlArr)-1]
println(strings.Join(urlArr, "/"))
}
// url : a.com/a/b/d.html
// a.com/a/b/c genFilename: c_leaui_index.html
// 生成子的相对目录有用
func (this *LeaSpider) doHTML(pUrl, realPUrl, content string) (children []string) {
regular := "(?i)(src=|href=)[\"']([^#].*?)[\"']"
reg := regexp.MustCompile(regular)
re := reg.FindAllStringSubmatch(content, -1)
log.Println(pUrl + " => " + realPUrl)
log.Println(pUrl + " 含有: ")
//log.Println(re);
baseDir := filepath.Dir(realPUrl)
for _, each := range re {
// 为了完整替换
// 只替换src=""里的会有子串的问题, 一个url是另一个url子串
rawFullUrl := each[0] // src='http://www.uiueux.com/wp/webzine/wp-content/themes/webzine/js/googlefont.js.php?ver=1.6.4'
rawFullUrlPrefix := each[1] // src=
// http://a.com/, /a/b/c/d.html, /a/b.jgp
// 如果是/a/b.jpg, 那么是相对host的, 而不是本文件的路径
rawCUrl := each[2]
cUrl := rawCUrl // strings.TrimRight(rawCUrl, "/") // 万一就是/呢?
// 如果一个链接以//开头, 那么省略了http:, 如果以/开头, 则相对于host
prefixNotHttp := false
if strings.HasPrefix(cUrl, "//") {
cUrl = this.scheme + util.Substring(cUrl, 2)
prefixNotHttp = true
} else if strings.HasPrefix(cUrl, "/") {
cUrl = this.schemeAndHost + cUrl
}
// 如果这个url是一个目录, 新建一个文件
// 如果这个url是以http://a.com开头的, host是一样的,
// 那么content的url是相对于该url
// 生成的url, 如果是目录, 会生成一个文件
cRealUrl, ok := this.getRalativeUrl(realPUrl, cUrl)
// 错误, 不是本页面, 本host的页面
if ok == -1 {
// 如果之前//替换成了http://
if prefixNotHttp {
content = strings.Replace(content, rawFullUrl, rawFullUrlPrefix+"\""+cRealUrl+"\"", -1)
}
continue
}
// 表示已处理过, 是相对目录了, 必须把内容的替换掉
// 但要处理的还是之前的链接http://
if ok == 1 {
cRealUrl = strings.Trim(cRealUrl, "/")
// 把//变成/
for strings.Index(cRealUrl, "//") != -1 {
cRealUrl = strings.Replace(cRealUrl, "//", "/", -1)
}
log.Println(rawCUrl + " >>>>>> " + cRealUrl)
content = strings.Replace(content, rawFullUrl, rawFullUrlPrefix+"\""+cRealUrl+"\"", -1)
cUrl = strings.Replace(cUrl, this.scheme, "", 1) // 把sheme去掉, do
children = append(children, cUrl) // 不需要clean
} else {
children = append(children, this.cleanUrl(baseDir+"/"+cRealUrl))
}
}
// 把content保存起来
if !this.writeFile(realPUrl, content) {
return
}
// this.t++
// return
return
}