From 91eff584cbcb9aaaa71a5f2a307a2adfd2952c45 Mon Sep 17 00:00:00 2001
From: grialion <n/a>
Date: Fri, 18 Oct 2024 22:53:39 +0200
Subject: [PATCH] fix: img regex
---
src/main.rs | 7 ++-----
1 file changed, 2 insertions(+), 5 deletions(-)
diff --git a/src/main.rs b/src/main.rs
index b971a37..edd2e6d 100644
--- a/src/main.rs
+++ b/src/main.rs
@@ -21,7 +21,6 @@ async fn main() -> ExitCode {
let mut page = 0;
loop {
-
println!("now scraping page {}", page + 1);
let post_html = async || {
@@ -45,11 +44,10 @@ async fn main() -> ExitCode {
let mut wait_time = 5000;
if urls.is_empty() {
-
for reconnection_attempts in 0..4 {
println!("no urls found, retrying in {} seconds...", wait_time / 1000);
sleep(Duration::from_millis(wait_time)).await;
-
+
urls = post_html().await;
if !urls.is_empty() {
@@ -67,7 +65,6 @@ async fn main() -> ExitCode {
}
for url in urls {
-
let img_url =
extract_img_url(&client.get(url).send().await.unwrap().text().await.unwrap());
if img_url.is_empty() {
@@ -91,7 +88,7 @@ fn extract_urls(html: &str) -> Vec<String> {
fn extract_img_url(html: &str) -> String {
if let Some(img_url) =
- Regex::new(r"https://us\.rule34\.xxx//([A-Za-z0-9]+(/[A-Za-z0-9]+)+)\.[A-Za-z0-9]+")
+ Regex::new(r"https://us\.rule34\.xxx/images/([A-Za-z0-9]+(/[A-Za-z0-9]+)+)\.[A-Za-z0-9]+")
.unwrap()
.find(html)
{