oss/miniflux-v2
oss/miniflux-v2
1
0
Fork 0
mirror of https://github.com/miniflux/v2.git synced 2025-06-27 16:36:00 +00:00
Code Wiki Activity

refactor(rewriter): avoid the use of regex in addDynamicImage

Browse source 
See https://dustri.org/b/parsing-noscript-tags-with-goquery.html for the whole
story.
This commit is contained in:
Julien Voisin 2024-12-16 01:56:39 +00:00 • committed by GitHub
parent 14a6e8ed3a
commit cfda948c3a
No known key found for this signature in database
GPG key ID: B5690EEEBB952194
2 changed files with 6 additions and 9 deletions
Download patch file Download diff file Expand all files Collapse all files

11
internal/reader/rewrite/rewrite_functions.go
View file

@ -23,7 +23,6 @@ var (
youtubeRegex = regexp.MustCompile(`youtube\.com/watch\?v=(.*)$`) youtubeRegex = regexp.MustCompile(`youtube\.com/watch\?v=(.*)$`)
youtubeIdRegex = regexp.MustCompile(`youtube_id"?\s*[:=]\s*"([a-zA-Z0-9_-]{11})"`) youtubeIdRegex = regexp.MustCompile(`youtube_id"?\s*[:=]\s*"([a-zA-Z0-9_-]{11})"`)
invidioRegex = regexp.MustCompile(`https?://(.*)/watch\?v=(.*)`) invidioRegex = regexp.MustCompile(`https?://(.*)/watch\?v=(.*)`)
imgRegex = regexp.MustCompile(`<img [^>]+>`)
textLinkRegex = regexp.MustCompile(`(?mi)(\bhttps?:\/\/[-A-Z0-9+&@#\/%?=~_|!:,.;]*[-A-Z0-9+&@#\/%=~_|])`) textLinkRegex = regexp.MustCompile(`(?mi)(\bhttps?:\/\/[-A-Z0-9+&@#\/%?=~_|!:,.;]*[-A-Z0-9+&@#\/%=~_|])`)
) )
@ -84,10 +83,11 @@ func addMailtoSubject(entryContent string) string {
} }
func addDynamicImage(entryContent string) string { func addDynamicImage(entryContent string) string {
doc, err := goquery.NewDocumentFromReader(strings.NewReader(entryContent)) parserHtml, err := nethtml.ParseWithOptions(strings.NewReader(entryContent), nethtml.ParseOptionEnableScripting(false))
if err != nil { if err != nil {
return entryContent return entryContent
} }
doc := goquery.NewDocumentFromNode(parserHtml)
// Ordered most preferred to least preferred. // Ordered most preferred to least preferred.
candidateAttrs := []string{ candidateAttrs := []string{
@ -149,12 +149,9 @@ func addDynamicImage(entryContent string) string {
if !changed { if !changed {
doc.Find("noscript").Each(func(i int, noscript *goquery.Selection) { doc.Find("noscript").Each(func(i int, noscript *goquery.Selection) {
matches := imgRegex.FindAllString(noscript.Text(), 2) if img := noscript.Find("img"); img.Length() == 1 {
img.Unwrap()
if len(matches) == 1 {
changed = true changed = true
noscript.ReplaceWithHtml(matches[0])
} }
}) })
} }

4
internal/reader/rewrite/rewriter_test.go
View file

@ -256,7 +256,7 @@ func TestRewriteWithNoLazyImage(t *testing.T) {
func TestRewriteWithLazyImage(t *testing.T) { func TestRewriteWithLazyImage(t *testing.T) {
controlEntry := &model.Entry{ controlEntry := &model.Entry{
Title: `A title`, Title: `A title`,
Content: `<img src="https://example.org/image.jpg" data-url="https://example.org/image.jpg" alt="Image"/><noscript><img src="https://example.org/fallback.jpg" alt="Fallback"></noscript>`, Content: `<img src="https://example.org/image.jpg" data-url="https://example.org/image.jpg" alt="Image"/><noscript><img src="https://example.org/fallback.jpg" alt="Fallback"/></noscript>`,
} }
testEntry := &model.Entry{ testEntry := &model.Entry{
Title: `A title`, Title: `A title`,
@ -272,7 +272,7 @@ func TestRewriteWithLazyImage(t *testing.T) {
func TestRewriteWithLazyDivImage(t *testing.T) { func TestRewriteWithLazyDivImage(t *testing.T) {
controlEntry := &model.Entry{ controlEntry := &model.Entry{
Title: `A title`, Title: `A title`,
Content: `<img src="https://example.org/image.jpg" alt="Image"/><noscript><img src="https://example.org/fallback.jpg" alt="Fallback"></noscript>`, Content: `<img src="https://example.org/image.jpg" alt="Image"/><noscript><img src="https://example.org/fallback.jpg" alt="Fallback"/></noscript>`,
} }
testEntry := &model.Entry{ testEntry := &model.Entry{
Title: `A title`, Title: `A title`,