katajisto/DuaLipaBookClubTracker
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

fix erros

Browse Source
This commit is contained in:
Tuomas Katajisto 2025-08-01 20:09:29 +03:00
parent 5d35f38292
commit 9e2fef2433
1 changed files with 3 additions and 4 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

7
main.go
View File

@ -205,7 +205,7 @@ func crawl_dua(db *sql.DB) {
bookCollector := c.Clone() bookCollector := c.Clone()
// Find links to individual book pages // Find links to individual book pages
c.OnHTML("a[href*='/books/']", func(e *colly.HTMLElement) { c.OnHTML("a", func(e *colly.HTMLElement) {
link := e.Attr("href") link := e.Attr("href")
// Make sure it's a valid book link and not something else // Make sure it's a valid book link and not something else
if strings.Contains(link, "/books/") { if strings.Contains(link, "/books/") {
@ -217,10 +217,8 @@ func crawl_dua(db *sql.DB) {
fmt.Println("Visiting", r.URL) fmt.Println("Visiting", r.URL)
}) })
bookCollector.Visit(currenturl)
// On the book page, extract the details // On the book page, extract the details
bookCollector.OnHTML(`div.book-page`, func(e *colly.HTMLElement) { bookCollector.OnHTML(`div`, func(e *colly.HTMLElement) {
title := e.ChildText("h1") title := e.ChildText("h1")
imgSrc := e.ChildAttr("img", "src") imgSrc := e.ChildAttr("img", "src")
var month string var month string
@ -242,6 +240,7 @@ func crawl_dua(db *sql.DB) {
return return
} }
if exists { if exists {
log.Println("SKipping book: ", title)
return // Silently skip if already exists return // Silently skip if already exists
} }