rdelaage
/
miniflux-v2
mirror of https://github.com/miniflux/v2.git
1
0
Fork 0
Code Issues Packages Projects Releases Wiki Activity

Revert "scraper follow the only link" 

This reverts commit 10207967c4.
This commit is contained in:
Frédéric Guillot 2022-11-14 17:02:45 -08:00
parent de1a06e3e8
commit 93715b542c
1 changed files with 1 additions and 51 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

52
reader/scraper/scraper.go
View File

@ -21,14 +21,6 @@ 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)
@ -54,8 +46,8 @@ func fetchURL(websiteURL, rules, userAgent string, cookie string, allowSelfSigne
return "", err
}
sameSite := url.Domain(websiteURL) == url.Domain(response.EffectiveURL)
// The entry URL could redirect somewhere else.
sameSite := url.Domain(websiteURL) == url.Domain(response.EffectiveURL)
websiteURL = response.EffectiveURL
if rules == "" {
@ -112,45 +104,3 @@ 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
}
if len(document.Find("body").Nodes) == 0 {
return content, nil
}
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)
}