1
0
Fork 0
mirror of https://github.com/miniflux/v2.git synced 2025-07-02 16:38:37 +00:00

Add rewrite rule to remove dom elements

This commit is contained in:
Lukas Dietrich 2021-09-01 23:42:23 +02:00 committed by fguillot
parent 9fbcfc213b
commit 93596c1218
3 changed files with 112 additions and 38 deletions

View file

@ -229,3 +229,15 @@ func replaceCustom(entryContent string, searchTerm string, replaceTerm string) s
} }
return entryContent return entryContent
} }
func removeCustom(entryContent string, selector string) string {
doc, err := goquery.NewDocumentFromReader(strings.NewReader(entryContent))
if err != nil {
return entryContent
}
doc.Find(selector).Remove()
output, _ := doc.Find("body").First().Html()
return output
}

View file

@ -5,14 +5,18 @@
package rewrite // import "miniflux.app/reader/rewrite" package rewrite // import "miniflux.app/reader/rewrite"
import ( import (
"regexp" "strconv"
"strings" "strings"
"text/scanner"
"miniflux.app/logger" "miniflux.app/logger"
"miniflux.app/url" "miniflux.app/url"
) )
var customReplaceRuleRegex = regexp.MustCompile(`replace\("(.*)"\|"(.*)"\)`) type rule struct {
name string
args []string
}
// Rewriter modify item contents with a set of rewriting rules. // Rewriter modify item contents with a set of rewriting rules.
func Rewriter(entryURL, entryContent, customRewriteRules string) string { func Rewriter(entryURL, entryContent, customRewriteRules string) string {
@ -21,46 +25,78 @@ func Rewriter(entryURL, entryContent, customRewriteRules string) string {
rulesList = customRewriteRules rulesList = customRewriteRules
} }
rules := strings.Split(rulesList, ",") rules := parseRules(rulesList)
rules = append(rules, "add_pdf_download_link") rules = append(rules, rule{name: "add_pdf_download_link"})
logger.Debug(`[Rewrite] Applying rules %v for %q`, rules, entryURL) logger.Debug(`[Rewrite] Applying rules %v for %q`, rules, entryURL)
for _, rule := range rules { for _, rule := range rules {
rule := strings.TrimSpace(rule) entryContent = applyRule(entryURL, entryContent, rule)
switch rule { }
case "add_image_title":
entryContent = addImageTitle(entryURL, entryContent) return entryContent
case "add_mailto_subject": }
entryContent = addMailtoSubject(entryURL, entryContent)
case "add_dynamic_image": func parseRules(rulesText string) (rules []rule) {
entryContent = addDynamicImage(entryURL, entryContent) scan := scanner.Scanner{Mode: scanner.ScanIdents | scanner.ScanStrings}
case "add_youtube_video": scan.Init(strings.NewReader(rulesText))
entryContent = addYoutubeVideo(entryURL, entryContent)
case "add_invidious_video": for {
entryContent = addInvidiousVideo(entryURL, entryContent) switch scan.Scan() {
case "add_youtube_video_using_invidious_player": case scanner.Ident:
entryContent = addYoutubeVideoUsingInvidiousPlayer(entryURL, entryContent) rules = append(rules, rule{name: scan.TokenText()})
case "add_pdf_download_link":
entryContent = addPDFLink(entryURL, entryContent) case scanner.String:
case "nl2br": if l := len(rules) - 1; l >= 0 {
entryContent = replaceLineFeeds(entryContent) text := scan.TokenText()
case "convert_text_link", "convert_text_links": text, _ = strconv.Unquote(text)
entryContent = replaceTextLinks(entryContent)
case "fix_medium_images": rules[l].args = append(rules[l].args, text)
entryContent = fixMediumImages(entryURL, entryContent)
case "use_noscript_figure_images":
entryContent = useNoScriptImages(entryURL, entryContent)
default:
if strings.Contains(rule, "replace") {
// Format: replace("search-term"|"replace-term")
args := customReplaceRuleRegex.FindStringSubmatch(rule)
if len(args) >= 3 {
entryContent = replaceCustom(entryContent, args[1], args[2])
} else {
logger.Debug("[Rewrite] Cannot find search and replace terms for replace rule %s", rule)
}
} }
case scanner.EOF:
return
}
}
}
func applyRule(entryURL, entryContent string, rule rule) string {
switch rule.name {
case "add_image_title":
entryContent = addImageTitle(entryURL, entryContent)
case "add_mailto_subject":
entryContent = addMailtoSubject(entryURL, entryContent)
case "add_dynamic_image":
entryContent = addDynamicImage(entryURL, entryContent)
case "add_youtube_video":
entryContent = addYoutubeVideo(entryURL, entryContent)
case "add_invidious_video":
entryContent = addInvidiousVideo(entryURL, entryContent)
case "add_youtube_video_using_invidious_player":
entryContent = addYoutubeVideoUsingInvidiousPlayer(entryURL, entryContent)
case "add_pdf_download_link":
entryContent = addPDFLink(entryURL, entryContent)
case "nl2br":
entryContent = replaceLineFeeds(entryContent)
case "convert_text_link", "convert_text_links":
entryContent = replaceTextLinks(entryContent)
case "fix_medium_images":
entryContent = fixMediumImages(entryURL, entryContent)
case "use_noscript_figure_images":
entryContent = useNoScriptImages(entryURL, entryContent)
case "replace":
// Format: replace("search-term"|"replace-term")
if len(rule.args) >= 2 {
entryContent = replaceCustom(entryContent, rule.args[0], rule.args[1])
} else {
logger.Debug("[Rewrite] Cannot find search and replace terms for replace rule %s", rule)
}
case "remove":
// Format: remove("#selector > .element, .another")
if len(rule.args) >= 1 {
entryContent = removeCustom(entryContent, rule.args[0])
} else {
logger.Debug("[Rewrite] Cannot find selector for remove rule %s", rule)
} }
} }

View file

@ -5,10 +5,26 @@
package rewrite // import "miniflux.app/reader/rewrite" package rewrite // import "miniflux.app/reader/rewrite"
import ( import (
"reflect"
"strings" "strings"
"testing" "testing"
) )
func TestParseRules(t *testing.T) {
rulesText := `add_dynamic_image,replace("article/(.*).svg"|"article/$1.png"),remove(".spam, .ads:not(.keep)")`
expected := []rule{
{name: "add_dynamic_image"},
{name: "replace", args: []string{"article/(.*).svg", "article/$1.png"}},
{name: "remove", args: []string{".spam, .ads:not(.keep)"}},
}
actual := parseRules(rulesText)
if !reflect.DeepEqual(expected, actual) {
t.Errorf(`Parsed rules do not match expected rules: got %v instead of %v`, actual, expected)
}
}
func TestReplaceTextLinks(t *testing.T) { func TestReplaceTextLinks(t *testing.T) {
scenarios := map[string]string{ scenarios := map[string]string{
`This is a link to example.org`: `This is a link to example.org`, `This is a link to example.org`: `This is a link to example.org`,
@ -234,7 +250,17 @@ func TestRewriteNoScriptImageWithNoScriptTag(t *testing.T) {
func TestRewriteReplaceCustom(t *testing.T) { func TestRewriteReplaceCustom(t *testing.T) {
content := `<img src="http://example.org/logo.svg"><img src="https://example.org/article/picture.svg">` content := `<img src="http://example.org/logo.svg"><img src="https://example.org/article/picture.svg">`
expected := `<img src="http://example.org/logo.svg"><img src="https://example.org/article/picture.png">` expected := `<img src="http://example.org/logo.svg"><img src="https://example.org/article/picture.png">`
output := Rewriter("https://example.org/artcle", content, `replace("article/(.*).svg"|"article/$1.png")`) output := Rewriter("https://example.org/article", content, `replace("article/(.*).svg"|"article/$1.png")`)
if expected != output {
t.Errorf(`Not expected output: %s`, output)
}
}
func TestRewriteRemoveCustom(t *testing.T) {
content := `<div>Lorem Ipsum <span class="spam">I dont want to see this</span><span class="ads keep">Super important info</span></div>`
expected := `<div>Lorem Ipsum <span class="ads keep">Super important info</span></div>`
output := Rewriter("https://example.org/article", content, `remove(".spam, .ads:not(.keep)")`)
if expected != output { if expected != output {
t.Errorf(`Not expected output: %s`, output) t.Errorf(`Not expected output: %s`, output)