1
0
Fork 0
mirror of https://github.com/miniflux/v2.git synced 2025-08-31 18:31:01 +00:00

Add scraper rules

This commit is contained in:
Frédéric Guillot 2017-12-10 20:51:04 -08:00
parent 7a35c58f53
commit 87ccad5c7f
16 changed files with 140 additions and 34 deletions

16
reader/scraper/rules.go Normal file
View file

@ -0,0 +1,16 @@
// Copyright 2017 Frédéric Guillot. All rights reserved.
// Use of this source code is governed by the Apache 2.0
// license that can be found in the LICENSE file.
package scraper
// List of predefined scraper rules (alphabetically sorted)
// domain => CSS selectors
var predefinedRules = map[string]string{
"lemonde.fr": "div#articleBody",
"lesjoiesducode.fr": ".blog-post-content img",
"linux.com": "div.content, div[property]",
"opensource.com": "div[property]",
"phoronix.com": "div.content",
"techcrunch.com": "div.article-entry",
}

View file

@ -6,14 +6,19 @@ package scraper
import (
"errors"
"io"
"log"
"strings"
"github.com/PuerkitoBio/goquery"
"github.com/miniflux/miniflux2/http"
"github.com/miniflux/miniflux2/reader/readability"
"github.com/miniflux/miniflux2/reader/sanitizer"
"github.com/miniflux/miniflux2/url"
)
// Fetch download a web page a returns relevant contents.
func Fetch(websiteURL string) (string, error) {
func Fetch(websiteURL, rules string) (string, error) {
client := http.NewClient(websiteURL)
response, err := client.Get()
if err != nil {
@ -29,10 +34,57 @@ func Fetch(websiteURL string) (string, error) {
return "", err
}
content, err := readability.ExtractContent(page)
var content string
if rules == "" {
rules = getPredefinedScraperRules(websiteURL)
}
if rules != "" {
log.Printf(`[Scraper] Using rules "%s" for "%s"`, rules, websiteURL)
content, err = scrapContent(page, rules)
} else {
log.Printf(`[Scraper] Using readability for "%s"`, websiteURL)
content, err = readability.ExtractContent(page)
}
if err != nil {
return "", err
}
return sanitizer.Sanitize(websiteURL, content), nil
}
func scrapContent(page io.Reader, rules string) (string, error) {
document, err := goquery.NewDocumentFromReader(page)
if err != nil {
return "", err
}
contents := ""
document.Find(rules).Each(func(i int, s *goquery.Selection) {
var content string
// For some inline elements, we get the parent.
if s.Is("img") {
content, _ = s.Parent().Html()
} else {
content, _ = s.Html()
}
contents += content
})
return contents, nil
}
func getPredefinedScraperRules(websiteURL string) string {
urlDomain := url.Domain(websiteURL)
for domain, rules := range predefinedRules {
if strings.Contains(urlDomain, domain) {
return rules
}
}
return ""
}

View file

@ -0,0 +1,21 @@
// Copyright 2017 Frédéric Guillot. All rights reserved.
// Use of this source code is governed by the Apache 2.0
// license that can be found in the LICENSE file.
package scraper
import "testing"
func TestGetPredefinedRules(t *testing.T) {
if getPredefinedScraperRules("http://www.phoronix.com/") == "" {
t.Error("Unable to find rule for phoronix.com")
}
if getPredefinedScraperRules("https://www.linux.com/") == "" {
t.Error("Unable to find rule for linux.com")
}
if getPredefinedScraperRules("https://example.org/") != "" {
t.Error("A rule not defined should not return anything")
}
}