fix(fetch): one status table — a persistent 4xx is permanent everywhere

status_error's catch-all called every unlisted status Transient, so a 400/405/418/451 got three retries per link before the same answer (twitter syndication's broken-token 400 being the live example), and download_status_error plus misskey's and bilibili's local fallbacks each carried their own copy of the table — bilibili and misskey classifying a 404 as Transient while the center classified it NotFound. The center now makes any client error except 408/429 a refusal (permanent), the media path delegates to it as status_error("media", ...) and its duplicated fn is deleted, and misskey/bilibili fall through to the center after their own special statuses (misskey's 400 body, bilibili's 412). A table test pins every class.

crates/x-media/src/site/bilibili/interface.rs

@@ -213,10 +213,11 @@ pub async fn fetch(dynamic_id: &str) -> Result<model::Item, FetchError> {
     if !status.is_success() {
         return Err(match status.as_u16() {
             412 => risk_control("412"),
-            // A refusal or an auth demand is not a bad moment (412 above is
+            // Everything else shares the central classes (refusals and gone
-            // bilibili's risk control, which does clear on its own).
+            // posts permanent, 429/5xx retried). The local fallback used to
-            401 | 403 => FetchError::Blocked,
+            // disagree: a bilibili 404 came back Transient here. 412 above is
-            _ => FetchError::Transient(format!("bilibili status {status}")),
+            // bilibili's risk control, which does clear on its own.
+            _ => crate::site::status_error("bilibili", status),
         });
     }
     let detail: model::Detail = response.json().await.map_err(|e| FetchError::Site {

crates/x-media/src/site/download.rs

@@ -105,7 +105,11 @@ fn download_stalled() -> FetchError {
 }
 
 /// Sends a media-download request: the response head must arrive within the
-/// idle window, and a non-2xx status is classified by [`download_status_error`].
+/// idle window, and a non-2xx status is classified by
+/// [`super::status_error`] with `"media"` as the name — the same table the
+/// site adapters use, so a dead URL and a bad moment read the same everywhere.
+/// A transport error never reaches that table — it fails in `send()` and
+/// stays [`FetchError::Http`].
 async fn send_download(request: reqwest::RequestBuilder) -> Result<reqwest::Response, FetchError> {
     let response = match tokio::time::timeout(DOWNLOAD_IDLE_TIMEOUT, request.send()).await {
         Ok(Ok(response)) => response,
@@ -115,7 +119,7 @@ async fn send_download(request: reqwest::RequestBuilder) -> Result<reqwest::Resp
     if response.status().is_success() {
         Ok(response)
     } else {
-        Err(download_status_error(response.status()))
+        Err(super::status_error("media", response.status()))
     }
 }
 
@@ -221,19 +225,6 @@ fn apply_media_headers(mut request: reqwest::RequestBuilder, url: &str) -> reqwe
     request
 }
 
-/// Maps a media download's HTTP status onto the same classes the site
-/// adapters use, so callers can tell "try again" from "this URL is dead":
-/// 4xx is a property of the media (gone, refused by the host), while 429/5xx
-/// is a property of the moment. A transport error never reaches this — it
-/// fails in `send()` and stays [`FetchError::Http`].
-fn download_status_error(status: reqwest::StatusCode) -> FetchError {
-    match status.as_u16() {
-        401 | 403 => FetchError::Blocked,
-        404 | 410 => FetchError::NotFound,
-        _ => FetchError::Transient(format!("media status {status}")),
-    }
-}
-
 /// Downloads a media file with a hard size cap: the body is streamed and the
 /// download aborts with [`FetchError::TooLarge`] the moment the cap is
 /// crossed (or when a declared Content-Length already exceeds it). Keeps the

crates/x-media/src/site/misskey/interface.rs

@@ -51,7 +51,8 @@ pub fn cache_key(url: &str) -> Option<String> {
 
 /// Fetches a note from misskey.io by id. The API answers client failures
 /// with HTTP 400 + `{"error":{"code":...}}` (NO_SUCH_NOTE → NotFound);
-/// everything else non-success is transient and retried by [`crate::site::fetch`].
+/// every other non-success status falls through to the shared classes in
+/// [`crate::site::status_error`] — persistent 4xx permanent, 429/5xx retried.
 pub async fn fetch(note_id: &str) -> Result<model::Note, FetchError> {
     let response = crate::site::CLIENT
         .post(API_URL)
@@ -62,9 +63,10 @@ pub async fn fetch(note_id: &str) -> Result<model::Note, FetchError> {
     if !status.is_success() {
         return Err(match status.as_u16() {
             400 => not_found_or_invalid(response).await,
-            // A refusal or an auth demand is not a bad moment.
+            // The local fallback used to disagree with the center: a misskey
-            401 | 403 => FetchError::Blocked,
+            // 404 came back Transient here and was fetched three more times
-            _ => FetchError::Transient(format!("misskey status {status}")),
+            // for a note that is simply gone.
+            _ => crate::site::status_error("misskey", status),
         });
     }
     response.json().await.map_err(|e| FetchError::Site {

crates/x-media/src/site/mod.rs

@@ -296,17 +296,18 @@ pub enum FetchError {
     Io(std::io::Error),
 }
 
-/// The error class for a non-success HTTP status, as the site adapters that
+/// The error class for a non-success HTTP status, shared by the site
-/// share this mapping classify it: 404/410 mean the post is gone and 401/403 a
+/// adapters, the media downloads and twitter's auth fallback: 404/410 mean
-/// refusal or an auth demand — both permanent, since retrying cannot change
+/// the post is gone (permanent), any other client error the source answers
-/// either — while everything else (429, 5xx) is transient and retried by
+/// on sight is a refusal (permanent too — three retries only delay the same
-/// [`fetch`]. `site` only names the adapter in the transient message; a site
+/// answer), and only 408/429/5xx are a bad moment, retried by [`fetch`].
-/// whose statuses mean something else (bilibili's 412 risk control, misskey's
+/// `site` only names the adapter in the message (`"media"` for downloads);
-/// 400 with `NO_SUCH_NOTE`) maps those before falling back here.
+/// a site whose statuses mean something else (bilibili's 412 risk control,
+/// misskey's 400 with `NO_SUCH_NOTE`) maps those before falling back here.
 pub fn status_error(site: &'static str, status: reqwest::StatusCode) -> FetchError {
     match status.as_u16() {
         404 | 410 => FetchError::NotFound,
-        401 | 403 => FetchError::Blocked,
+        code if status.is_client_error() && !matches!(code, 408 | 429) => FetchError::Blocked,
         _ => FetchError::Transient(format!("{site} status {status}")),
     }
 }
@@ -733,6 +734,44 @@ mod tests {
         }
     }
 
+    #[test]
+    fn persistent_client_statuses_are_permanent() {
+        use reqwest::StatusCode;
+        // The one table every caller shares now: only 408, 429 and 5xx can
+        // answer differently on a retry. A 400 used to be Transient here and
+        // in two local fallbacks — twitter syndication's broken-token 400, for
+        // one, burned three retries per link before saying the same thing.
+        assert!(matches!(
+            status_error("x", StatusCode::NOT_FOUND),
+            FetchError::NotFound
+        ));
+        assert!(matches!(
+            status_error("x", StatusCode::BAD_REQUEST),
+            FetchError::Blocked
+        ));
+        assert!(matches!(
+            status_error("x", StatusCode::PAYLOAD_TOO_LARGE),
+            FetchError::Blocked
+        ));
+        assert!(matches!(
+            status_error("x", StatusCode::REQUEST_TIMEOUT),
+            FetchError::Transient(_)
+        ));
+        assert!(matches!(
+            status_error("x", StatusCode::TOO_MANY_REQUESTS),
+            FetchError::Transient(_)
+        ));
+        assert!(matches!(
+            status_error("x", StatusCode::INTERNAL_SERVER_ERROR),
+            FetchError::Transient(_)
+        ));
+        // The download path delegates under its own name, same classes.
+        assert!(matches!(
+            status_error("media", StatusCode::BAD_REQUEST),
+            FetchError::Blocked
+        ));
+    }
+
     #[tokio::test]
     async fn disabled_site_is_reported_not_ignored() {
         // pixiv is the only token-gated site; with PIXIV_REFRESH_TOKEN set it