Fix xxdbx site-relative thumbs; cumlouder answers 404 immediately

xxdbx assets are now /id.jpg, not //host/..; prefixing https: produced
https:/id.jpg. cumlouder's unknown-tag 404 went through the requester's
mirror/FlareSolverr fallbacks (146s); read the raw response and return
an empty list at once while still retrying 5xx.

Co-Authored-By: Claude Sonnet 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01MUYdCh2nCVsQYgJAT7gwSi

src/providers/cumlouder.rs

@@ -357,22 +357,33 @@ impl CumlouderProvider {
         let headers = vec![("Referer".to_string(), format!("{BASE_URL}/"))];
 
         // The origin sporadically answers its first uncached request with a
-        // Cloudflare 522; a single retry succeeds.
+        // Cloudflare 522; a retry succeeds. Use the raw response so a real
+        // 404 (unknown tag/category) answers immediately instead of
+        // escalating through the requester's mirror/FlareSolverr fallbacks,
+        // which cost minutes.
         let mut last_err = String::new();
         let mut html = None;
         for _ in 0..3 {
             match requester
-                .get_with_headers(&url, headers.clone(), None)
+                .get_raw_with_headers(&url, headers.clone())
                 .await
             {
-                Ok(body) if body.contains("muestra-escena") => {
-                    html = Some(body);
-                    break;
-                }
-                Ok(body) => {
+                Ok(response) => {
+                    let status = response.status().as_u16();
+                    let body = response.text().await.unwrap_or_default();
+                    if status == 404 {
+                        return Ok(vec![]);
+                    }
+                    if body.contains("muestra-escena") {
+                        html = Some(body);
+                        break;
+                    }
+                    if status >= 500 || status == 429 {
+                        last_err = format!("HTTP {status}");
+                        continue;
+                    }
+                    // A page with no cards is a real empty archive.
                     last_err = format!("no cards in {} bytes", body.len());
-                    // A page with no cards is a real empty archive (404 body
-                    // included); only keep retrying on transport errors.
                     html = Some(body);
                     break;
                 }

src/providers/xxdbx.rs

@@ -52,6 +52,23 @@ impl XxdbxProvider {
         provider
     }
 
+    /// Resolves the card's asset URLs, which the site serves protocol-relative
+    /// (`//prev.xxdbx.com/..`) or, now, site-relative (`/1234.jpg`).
+    fn absolute_url(raw: &str) -> String {
+        let raw = raw.trim();
+        if raw.is_empty() {
+            String::new()
+        } else if raw.starts_with("http://") || raw.starts_with("https://") {
+            raw.to_string()
+        } else if let Some(rest) = raw.strip_prefix("//") {
+            format!("https://{rest}")
+        } else if raw.starts_with('/') {
+            format!("https://xxdbx.com{raw}")
+        } else {
+            format!("https://xxdbx.com/{raw}")
+        }
+    }
+
     fn build_channel(&self, clientversion: ClientVersion) -> Channel {
         let _ = clientversion;
         Channel {
@@ -272,26 +289,24 @@ impl XxdbxProvider {
                 .unwrap_or_default()
                 .to_string();
 
-            let thumb = format!(
-                "https:{}",
-                video_segment
-                    .split("<img ")
-                    .collect::<Vec<&str>>()
-                    .get(1)
-                    .copied()
-                    .unwrap_or_default()
-                    .split("src=\"")
-                    .collect::<Vec<&str>>()
-                    .last()
-                    .copied()
-                    .unwrap_or_default()
-                    .split("\"")
-                    .collect::<Vec<&str>>()
-                    .get(0)
-                    .copied()
-                    .unwrap_or_default()
-                    .to_string()
-            );
+            let thumb_raw = video_segment
+                .split("<img ")
+                .collect::<Vec<&str>>()
+                .get(1)
+                .copied()
+                .unwrap_or_default()
+                .split("src=\"")
+                .collect::<Vec<&str>>()
+                .last()
+                .copied()
+                .unwrap_or_default()
+                .split("\"")
+                .collect::<Vec<&str>>()
+                .get(0)
+                .copied()
+                .unwrap_or_default()
+                .to_string();
+            let thumb = Self::absolute_url(&thumb_raw);
             let raw_duration = video_segment
                 .split("<div class=\"v_dur\">")
                 .collect::<Vec<&str>>()
@@ -306,21 +321,19 @@ impl XxdbxProvider {
                 .to_string();
             let duration = parse_time_to_seconds(raw_duration.as_str()).unwrap_or(0) as u32;
 
-            let preview = format!(
-                "https:{}",
-                video_segment
-                    .split("data-preview=\"")
-                    .collect::<Vec<&str>>()
-                    .get(1)
-                    .copied()
-                    .unwrap_or_default()
-                    .split("\"")
-                    .collect::<Vec<&str>>()
-                    .get(0)
-                    .copied()
-                    .unwrap_or_default()
-                    .to_string()
-            );
+            let preview_raw = video_segment
+                .split("data-preview=\"")
+                .collect::<Vec<&str>>()
+                .get(1)
+                .copied()
+                .unwrap_or_default()
+                .split("\"")
+                .collect::<Vec<&str>>()
+                .get(0)
+                .copied()
+                .unwrap_or_default()
+                .to_string();
+            let preview = Self::absolute_url(&preview_raw);
             let tags = video_segment
                 .split("<div class=\"v_tags\">")
                 .collect::<Vec<&str>>()
@@ -439,3 +452,25 @@ impl Provider for XxdbxProvider {
         Some(self.build_channel(clientversion))
     }
 }
+
+#[cfg(test)]
+mod tests {
+    use super::XxdbxProvider;
+
+    #[test]
+    fn resolves_site_relative_protocol_relative_and_absolute_assets() {
+        assert_eq!(
+            XxdbxProvider::absolute_url("/7004779764736.jpg"),
+            "https://xxdbx.com/7004779764736.jpg"
+        );
+        assert_eq!(
+            XxdbxProvider::absolute_url("//prev.xxdbx.com/700477973230.mp4"),
+            "https://prev.xxdbx.com/700477973230.mp4"
+        );
+        assert_eq!(
+            XxdbxProvider::absolute_url("https://cdn.example/a.jpg"),
+            "https://cdn.example/a.jpg"
+        );
+        assert_eq!(XxdbxProvider::absolute_url(""), "");
+    }
+}