rule34video: drop per-video enrichment; add on-demand page resolver proxy (off)

Listings no longer fetch every video page for formats/tags/artist, which
ran into the site's 429 limit. /proxy/rule34video/{id}/{slug} resolves the
page at play time and 302s to the signed mp4; the resolver lives in
util/rule34video_resolve.rs. Handing out proxy URLs is gated behind
USE_MEDIA_PROXY, currently false.

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01AEp724zDpCWNaSeCrMSvjx

src/util/rule34video_resolve.rs

@@ -0,0 +1,117 @@
+//! rule34video.com video-page resolution for the `/proxy/rule34video/` redirect.
+//!
+//! Lives here (always compiled) because `HOT_TUB_PROVIDER=<id>` single-provider
+//! builds only compile the selected provider, while the proxy routes are
+//! always registered.
+//!
+//! yt-dlp's Rule34Video extractor no longer resolves the page URL ("No video
+//! formats found"), so the signed mp4 URLs (360p/480p/720p, valid ~12h, no
+//! cookie/Referer needed) come from the page's `var flashvars`.
+
+use crate::util::requester::Requester;
+
+const BASE_URL: &str = "https://rule34video.com";
+
+/// Reads a single-quoted JS string value (`key: '...'`) from the
+/// `flashvars` object, honouring `\'` escapes.
+fn flashvar(block: &str, key: &str) -> Option<String> {
+    let marker = format!("{}: '", key);
+    let start = block.find(&marker)? + marker.len();
+    let mut out = String::new();
+    let mut chars = block[start..].chars();
+    while let Some(c) = chars.next() {
+        match c {
+            '\\' => out.push(chars.next()?),
+            '\'' => return Some(out),
+            other => out.push(other),
+        }
+    }
+    None
+}
+
+/// `(height, url)` for every quality on the page, best first.
+pub fn parse_formats(html: &str) -> Vec<(u32, String)> {
+    let Some(start) = html.find("var flashvars") else {
+        return vec![];
+    };
+    let block = &html[start..];
+    let block = &block[..block.len().min(6000)];
+
+    let mut formats: Vec<(u32, String)> = Vec::new();
+    for (url_key, text_key) in [
+        ("video_url", "video_url_text"),
+        ("video_alt_url", "video_alt_url_text"),
+        ("video_alt_url2", "video_alt_url2_text"),
+        ("video_alt_url3", "video_alt_url3_text"),
+    ] {
+        let Some(url) = flashvar(block, url_key).filter(|url| url.starts_with("http")) else {
+            continue;
+        };
+        let height = flashvar(block, text_key)
+            .and_then(|quality| quality.trim_end_matches('p').parse::<u32>().ok())
+            .unwrap_or(0);
+        formats.push((height, url));
+    }
+    formats.sort_by(|a, b| b.0.cmp(&a.0));
+    formats
+}
+
+/// `{id}/{slug}[/]` -> the video page URL. Only digits, letters, `-` and `_`
+/// are accepted so the endpoint cannot steer the request elsewhere.
+pub fn page_url(endpoint: &str) -> Option<String> {
+    let mut parts = endpoint.trim_matches('/').split('/');
+    let id = parts.next().filter(|id| !id.is_empty() && id.chars().all(|c| c.is_ascii_digit()))?;
+    let slug = parts.next().filter(|slug| {
+        !slug.is_empty() && slug.chars().all(|c| c.is_ascii_alphanumeric() || c == '-' || c == '_')
+    })?;
+    match parts.next() {
+        None => Some(format!("{BASE_URL}/video/{id}/{slug}/")),
+        Some(_) => None,
+    }
+}
+
+/// The best signed mp4 URL (or the one of `height`) for a video page.
+pub async fn resolve_media_url(
+    requester: &mut Requester,
+    page_url: &str,
+    height: Option<u32>,
+) -> Option<String> {
+    let headers = vec![("Referer".to_string(), format!("{BASE_URL}/"))];
+    let html = requester.get_probe(page_url, headers).await.ok().flatten()?;
+    let formats = parse_formats(&html);
+    height
+        .and_then(|h| formats.iter().find(|(height, _)| *height == h))
+        .or_else(|| formats.first())
+        .map(|(_, url)| url.clone())
+}
+
+#[cfg(test)]
+mod tests {
+    use super::*;
+
+    #[test]
+    fn parses_flashvars_formats_best_first() {
+        let html = r#"var flashvars = { video_id: '1', video_url: 'https://rule34video.com/get_file/a_360.mp4/?v-acctoken=1', video_url_text: '360p', video_alt_url: 'https://rule34video.com/get_file/a_480p.mp4/?v-acctoken=2', video_alt_url_text: '480p', video_alt_url2: 'https://rule34video.com/get_file/a_720p.mp4/?v-acctoken=3', video_alt_url2_text: '720p', video_title: 'It\'s' };"#;
+        let formats = parse_formats(html);
+        assert_eq!(formats.len(), 3);
+        assert_eq!(formats[0].0, 720);
+        assert!(formats[0].1.contains("a_720p"));
+        assert!(formats[2].1.contains("a_360"));
+    }
+
+    #[test]
+    fn page_without_flashvars_yields_nothing() {
+        assert!(parse_formats("<html></html>").is_empty());
+    }
+
+    #[test]
+    fn builds_page_url_from_safe_endpoints_only() {
+        assert_eq!(
+            page_url("123/some-slug/").as_deref(),
+            Some("https://rule34video.com/video/123/some-slug/")
+        );
+        assert_eq!(page_url("123/../x"), None);
+        assert_eq!(page_url("abc/slug"), None);
+        assert_eq!(page_url("123/slug/extra"), None);
+    }
+}