tnypxl/rollup
1
0
Fork 0
mirror of https://github.com/tnypxl/rollup.git synced 2025-12-15 15:03:17 +00:00
Code Issues Packages Projects Releases Wiki Activity

fix: Remove print media emulation and improve CSS selector extraction

Browse Source
This commit is contained in:
Arik Jones (aider)
2024-09-14 15:14:53 -05:00
parent a3b23a6d34
commit 1a57be80fa
1 changed files with 6 additions and 12 deletions
Download Patch File Download Diff File Expand all files Collapse all files

18
internal/scraper/scraper.go
View File

@@ -65,14 +65,6 @@ func FetchWebpageContent(urlStr string) (string, error) {
}
defer page.Close()
err = page.EmulateMedia(playwright.PageEmulateMediaOptions{
Media: playwright.MediaPrint,
})
if err != nil {
log.Printf("Error emulating print media: %v\n", err)
return "", fmt.Errorf("could not emulate print media: %v", err)
}
time.Sleep(time.Duration(rand.Intn(2000)+1000) * time.Millisecond)
log.Printf("Navigating to URL: %s\n", urlStr)
@@ -250,13 +242,15 @@ func ExtractContentWithCSS(content, selector string) (string, error) {
return "", fmt.Errorf("error parsing HTML: %v", err)
}
selectedContent, err := doc.Find(selector).Html()
selection := doc.Find(selector)
if selection.Length() == 0 {
return "", fmt.Errorf("no content found with CSS selector: %s", selector)
}
selectedContent, err := selection.Html()
if err != nil {
return "", fmt.Errorf("error extracting content with CSS selector: %v", err)
}
if selectedContent == "" {
return "", fmt.Errorf("no content found with CSS selector: %s", selector)
}
log.Printf("Extracted content length: %d\n", len(selectedContent))
return selectedContent, nil