1
0
Fork 0

scraper follow the only link

* in some cases, what the scraper got is only a landing page, user can use scraper rules to extract the link of the landing page and follow it
* it also fix the  wrong scrape rule apply when the server redirects it to another host
This commit is contained in:
jebbs 2021-12-08 16:46:33 +08:00 committed by Frédéric Guillot
parent 242eeaf07e
commit 10207967c4

View file

@ -21,6 +21,14 @@ import (
// Fetch downloads a web page and returns relevant contents.
func Fetch(websiteURL, rules, userAgent string, cookie string, allowSelfSignedCertificates, useProxy bool) (string, error) {
content, err := fetchURL(websiteURL, rules, userAgent, cookie, allowSelfSignedCertificates, useProxy)
if err != nil {
return "", err
}
return followTheOnlyLink(websiteURL, content, rules, userAgent, cookie, allowSelfSignedCertificates, useProxy)
}
func fetchURL(websiteURL, rules, userAgent string, cookie string, allowSelfSignedCertificates, useProxy bool) (string, error) {
clt := client.NewClientWithConfig(websiteURL, config.Opts)
clt.WithUserAgent(userAgent)
clt.WithCookie(cookie)
@ -46,6 +54,7 @@ func Fetch(websiteURL, rules, userAgent string, cookie string, allowSelfSignedCe
return "", err
}
sameSite := url.Domain(websiteURL) == url.Domain(response.EffectiveURL)
// The entry URL could redirect somewhere else.
websiteURL = response.EffectiveURL
@ -54,7 +63,7 @@ func Fetch(websiteURL, rules, userAgent string, cookie string, allowSelfSignedCe
}
var content string
if rules != "" {
if sameSite && rules != "" {
logger.Debug(`[Scraper] Using rules %q for %q`, rules, websiteURL)
content, err = scrapContent(response.Body, rules)
} else {
@ -103,3 +112,35 @@ func isAllowedContentType(contentType string) bool {
return strings.HasPrefix(contentType, "text/html") ||
strings.HasPrefix(contentType, "application/xhtml+xml")
}
func followTheOnlyLink(websiteURL, content string, rules, userAgent string, cookie string, allowSelfSignedCertificates, useProxy bool) (string, error) {
document, err := goquery.NewDocumentFromReader(strings.NewReader(content))
if err != nil {
return "", err
}
body := document.Find("body").Nodes[0]
if body.FirstChild.NextSibling != nil ||
body.FirstChild.Data != "a" {
return content, nil
}
// the body has only one child of <a>
var href string
for _, attr := range body.FirstChild.Attr {
if attr.Key == "href" {
href = attr.Val
break
}
}
if href == "" {
return content, nil
}
href, err = url.AbsoluteURL(websiteURL, href)
if err != nil {
return "", err
}
sameSite := url.Domain(websiteURL) == url.Domain(href)
if sameSite {
return fetchURL(href, rules, userAgent, cookie, allowSelfSignedCertificates, useProxy)
}
return fetchURL(href, rules, userAgent, "", false, false)
}