viralxxxporn and xfree bugfix

src/providers/mod.rs

@@ -36,6 +36,7 @@ pub mod porn00;
 pub mod pornzog;
 pub mod sxyprn;
 pub mod tnaflix;
+pub mod viralxxxporn;
 pub mod xfree;
 pub mod xxthots;
 pub mod youjizz;
@@ -135,6 +136,10 @@ pub static ALL_PROVIDERS: Lazy<HashMap<&'static str, DynProvider>> = Lazy::new(|
         "tnaflix",
         Arc::new(tnaflix::TnaflixProvider::new()) as DynProvider,
     );
+    m.insert(
+        "viralxxxporn",
+        Arc::new(viralxxxporn::ViralxxxpornProvider::new()) as DynProvider,
+    );
     // m.insert("pornxp", Arc::new(pornxp::PornxpProvider::new()) as DynProvider);
     m.insert(
         "rule34gen",

src/providers/viralxxxporn.rs

@@ -0,0 +1,722 @@
+use crate::DbPool;
+use crate::api::ClientVersion;
+use crate::providers::{Provider, report_provider_error_background, requester_or_default};
+use crate::status::*;
+use crate::util::cache::VideoCache;
+use crate::util::discord::send_discord_error_report;
+use crate::util::parse_abbreviated_number;
+use crate::util::time::parse_time_to_seconds;
+use crate::videos::{ServerOptions, VideoFormat, VideoItem};
+use async_trait::async_trait;
+use error_chain::error_chain;
+use futures::stream::{FuturesUnordered, StreamExt};
+use htmlentity::entity::{ICodedDataTrait, decode};
+use regex::Regex;
+use std::collections::HashSet;
+use std::fmt::Write;
+use std::vec;
+use url::form_urlencoded::Serializer;
+
+error_chain! {
+    foreign_links {
+        Io(std::io::Error);
+        HttpRequest(wreq::Error);
+    }
+}
+
+#[derive(Debug, Clone)]
+pub struct ViralxxxpornProvider {
+    url: String,
+}
+
+#[derive(Debug, Clone)]
+struct RawVideo {
+    id: String,
+    title: String,
+    detail_url: String,
+    thumb: String,
+    duration: u32,
+    views: Option<u32>,
+    rating: Option<f32>,
+    quality: Option<String>,
+}
+
+impl ViralxxxpornProvider {
+    pub fn new() -> Self {
+        Self {
+            url: "https://viralxxxporn.com".to_string(),
+        }
+    }
+
+    fn build_channel(&self, _clientversion: ClientVersion) -> Channel {
+        Channel {
+            id: "viralxxxporn".to_string(),
+            name: "ViralXXXPorn".to_string(),
+            description: "Trending free porn and onlyfans leaks".to_string(),
+            premium: false,
+            favicon: "https://www.google.com/s2/favicons?sz=64&domain=viralxxxporn.com".to_string(),
+            status: "active".to_string(),
+            categories: vec![],
+            options: vec![
+                ChannelOption {
+                    id: "sort".to_string(),
+                    title: "Sort".to_string(),
+                    description: "Browse different video feeds".to_string(),
+                    systemImage: "list.number".to_string(),
+                    colorName: "blue".to_string(),
+                    options: vec![
+                        FilterOption {
+                            id: "latest".to_string(),
+                            title: "Latest".to_string(),
+                        },
+                        FilterOption {
+                            id: "top-rated".to_string(),
+                            title: "Top Rated".to_string(),
+                        },
+                        FilterOption {
+                            id: "most-viewed".to_string(),
+                            title: "Most Viewed".to_string(),
+                        },
+                        FilterOption {
+                            id: "premium".to_string(),
+                            title: "Premium".to_string(),
+                        },
+                    ],
+                    multiSelect: false,
+                },
+                ChannelOption {
+                    id: "filter".to_string(),
+                    title: "Period".to_string(),
+                    description: "Time range for top-rated feed".to_string(),
+                    systemImage: "clock".to_string(),
+                    colorName: "green".to_string(),
+                    options: vec![
+                        FilterOption {
+                            id: "all-time".to_string(),
+                            title: "All Time".to_string(),
+                        },
+                        FilterOption {
+                            id: "month".to_string(),
+                            title: "This Month".to_string(),
+                        },
+                        FilterOption {
+                            id: "week".to_string(),
+                            title: "This Week".to_string(),
+                        },
+                        FilterOption {
+                            id: "today".to_string(),
+                            title: "Today".to_string(),
+                        },
+                    ],
+                    multiSelect: false,
+                },
+            ],
+            nsfw: true,
+            cacheDuration: Some(300),
+        }
+    }
+
+    fn normalize_ws(s: &str) -> String {
+        s.split_whitespace().collect::<Vec<_>>().join(" ")
+    }
+
+    fn decode_html(s: &str) -> String {
+        decode(s.as_bytes())
+            .to_string()
+            .unwrap_or_else(|_| s.to_string())
+    }
+
+    fn absolute_url(&self, raw: &str) -> String {
+        if raw.starts_with("http://") || raw.starts_with("https://") {
+            return raw.to_string();
+        }
+        if raw.starts_with("//") {
+            return format!("https:{raw}");
+        }
+        if raw.starts_with('/') {
+            return format!("{}{}", self.url, raw);
+        }
+        format!("{}/{}", self.url, raw.trim_start_matches('/'))
+    }
+
+    fn slugify_query(query: &str) -> String {
+        let mut out = String::new();
+        let mut prev_dash = false;
+        for c in query.to_ascii_lowercase().chars() {
+            if c.is_ascii_alphanumeric() {
+                out.push(c);
+                prev_dash = false;
+            } else if !prev_dash {
+                out.push('-');
+                prev_dash = true;
+            }
+        }
+        out.trim_matches('-').to_string()
+    }
+
+    fn encode_query_value(value: &str) -> String {
+        let mut serializer = Serializer::new(String::new());
+        serializer.append_pair("q", value);
+        let encoded = serializer.finish();
+        encoded.strip_prefix("q=").unwrap_or(&encoded).to_string()
+    }
+
+    fn with_page(url: &str, page: u8) -> Vec<String> {
+        if page <= 1 {
+            return vec![url.to_string()];
+        }
+        let base = url.trim_end_matches('/');
+        let mut candidates = vec![
+            format!("{base}/page/{page}/"),
+            format!("{base}/{page}/"),
+            format!("{base}/?page={page}"),
+        ];
+        if url.contains('?') {
+            candidates.push(format!("{url}&page={page}"));
+        }
+        candidates
+    }
+
+    fn unique_urls(urls: Vec<String>) -> Vec<String> {
+        let mut seen = HashSet::new();
+        let mut out = vec![];
+        for u in urls {
+            if seen.insert(u.clone()) {
+                out.push(u);
+            }
+        }
+        out
+    }
+
+    fn build_listing_urls(&self, page: u8, query: &str, options: &ServerOptions) -> Vec<String> {
+        let sort = options
+            .sort
+            .clone()
+            .unwrap_or_else(|| "latest".to_string())
+            .to_ascii_lowercase();
+        let filter = options
+            .filter
+            .clone()
+            .unwrap_or_else(|| "all-time".to_string())
+            .to_ascii_lowercase();
+        let query = query.trim();
+        let slug = Self::slugify_query(query);
+        let encoded = Self::encode_query_value(query);
+
+        let mut urls = vec![];
+
+        if !query.is_empty() {
+            urls.extend(Self::with_page(&format!("{}/tags/{slug}/", self.url), page));
+            urls.extend(Self::with_page(
+                &format!("{}/search/{}/", self.url, slug),
+                page,
+            ));
+            urls.extend(Self::with_page(
+                &format!("{}/?s={}", self.url, encoded),
+                page,
+            ));
+            urls.extend(Self::with_page(
+                &format!("{}/search?q={}", self.url, encoded),
+                page,
+            ));
+            return Self::unique_urls(urls);
+        }
+
+        match sort.as_str() {
+            "top-rated" => {
+                urls.extend(Self::with_page(&format!("{}/top-rated/", self.url), page));
+                urls.extend(Self::with_page(
+                    &format!("{}/top-rated/?period={}", self.url, filter),
+                    page,
+                ));
+                urls.extend(Self::with_page(
+                    &format!("{}/top-rated/{}/", self.url, filter),
+                    page,
+                ));
+            }
+            "most-viewed" => {
+                urls.extend(Self::with_page(
+                    &format!("{}/most-popular/", self.url),
+                    page,
+                ));
+                urls.extend(Self::with_page(&format!("{}/most-viewed/", self.url), page));
+            }
+            "premium" => {
+                urls.extend(Self::with_page(&format!("{}/premium/", self.url), page));
+            }
+            _ => {
+                urls.extend(Self::with_page(
+                    &format!("{}/latest-updates/", self.url),
+                    page,
+                ));
+                urls.extend(Self::with_page(&self.url, page));
+            }
+        }
+
+        Self::unique_urls(urls)
+    }
+
+    fn strip_tags(text: &str) -> String {
+        let Ok(tag_re) = Regex::new(r"(?is)<[^>]+>") else {
+            return text.to_string();
+        };
+        tag_re.replace_all(text, " ").to_string()
+    }
+
+    fn clean_media_url(raw: &str) -> String {
+        let mut out = raw
+            .trim_matches(|c: char| c == '"' || c == '\'' || c == '\\' || c.is_whitespace())
+            .to_string();
+        out = out
+            .replace("\\u0026", "&")
+            .replace("\\u002F", "/")
+            .replace("\\/", "/")
+            .replace("&amp;", "&");
+        out = out
+            .trim_end_matches(|c: char| matches!(c, ',' | ';' | ')' | ']' | '}'))
+            .to_string();
+        if out.starts_with("//") {
+            return format!("https:{out}");
+        }
+        out
+    }
+
+    fn is_media_url(url: &str) -> bool {
+        let lower = url.to_ascii_lowercase();
+        (lower.starts_with("http://") || lower.starts_with("https://"))
+            && (lower.contains(".mp4") || lower.contains(".m3u8"))
+    }
+
+    fn extract_duration_seconds(text: &str) -> Option<u32> {
+        let Ok(duration_re) = Regex::new(r"\b(\d{1,2}:\d{2}(?::\d{2})?)\b") else {
+            return None;
+        };
+        duration_re
+            .captures(text)
+            .and_then(|c| c.get(1))
+            .and_then(|m| parse_time_to_seconds(m.as_str()))
+            .map(|v| v as u32)
+    }
+
+    fn extract_quality(text: &str) -> Option<String> {
+        let Ok(q_re) = Regex::new(r"(?i)\b((?:\d{3,4})p|hd|4k)\b") else {
+            return None;
+        };
+        q_re.captures(text)
+            .and_then(|c| c.get(1))
+            .map(|m| m.as_str().to_ascii_lowercase())
+    }
+
+    fn extract_rating(text: &str) -> Option<f32> {
+        let Ok(r_re) = Regex::new(r"(?i)\b(\d{1,3})%") else {
+            return None;
+        };
+        r_re.captures(text)
+            .and_then(|c| c.get(1))
+            .and_then(|m| m.as_str().parse::<f32>().ok())
+    }
+
+    fn extract_views(text: &str) -> Option<u32> {
+        let Ok(v_re) = Regex::new(r"(?i)\b([0-9]+(?:\.[0-9]+)?\s*[kmb]?)\s+\d{1,3}%") else {
+            return None;
+        };
+        let raw = v_re
+            .captures(text)
+            .and_then(|c| c.get(1))
+            .map(|m| m.as_str().to_string())?;
+        parse_abbreviated_number(&raw)
+    }
+
+    fn parse_listing_items(&self, html: &str) -> Vec<RawVideo> {
+        if html.trim().is_empty() {
+            return vec![];
+        }
+
+        let Ok(link_re) = Regex::new(
+            r#"(?is)<a[^>]+href="(?P<href>/video/(?P<id>\d+)/[^"]+)"[^>]*>(?P<body>.*?)</a>"#,
+        ) else {
+            return vec![];
+        };
+        let Ok(title_attr_re) = Regex::new(r#"(?is)\btitle="([^"]+)""#) else {
+            return vec![];
+        };
+        let Ok(thumb_re) = Regex::new(r#"(?is)(?:src|data-src|data-original|poster)="([^"]+)""#)
+        else {
+            return vec![];
+        };
+
+        let mut out = vec![];
+        let mut seen = HashSet::new();
+
+        for caps in link_re.captures_iter(html) {
+            let Some(id) = caps.name("id").map(|m| m.as_str().to_string()) else {
+                continue;
+            };
+            if !seen.insert(id.clone()) {
+                continue;
+            }
+            let href = caps.name("href").map(|m| m.as_str()).unwrap_or_default();
+            let body = caps.name("body").map(|m| m.as_str()).unwrap_or_default();
+
+            let m0 = match caps.get(0) {
+                Some(v) => v,
+                None => continue,
+            };
+            let seg_start = m0.start().saturating_sub(600);
+            let seg_end = (m0.end() + 1800).min(html.len());
+            let segment = html.get(seg_start..seg_end).unwrap_or(body);
+
+            let title_from_attr = title_attr_re
+                .captures(m0.as_str())
+                .and_then(|c| c.get(1))
+                .map(|m| m.as_str().to_string())
+                .unwrap_or_default();
+            let title_from_body = Self::strip_tags(body);
+            let title = Self::normalize_ws(&Self::decode_html(if !title_from_attr.is_empty() {
+                &title_from_attr
+            } else {
+                &title_from_body
+            }));
+            if title.is_empty() {
+                continue;
+            }
+
+            let thumb = thumb_re
+                .captures(segment)
+                .and_then(|c| c.get(1))
+                .map(|m| self.absolute_url(m.as_str()))
+                .unwrap_or_default();
+
+            let duration = Self::extract_duration_seconds(segment).unwrap_or(0);
+            let rating = Self::extract_rating(segment);
+            let views = Self::extract_views(segment);
+            let quality = Self::extract_quality(segment);
+
+            out.push(RawVideo {
+                id,
+                title,
+                detail_url: self.absolute_url(href),
+                thumb,
+                duration,
+                views,
+                rating,
+                quality,
+            });
+        }
+
+        out
+    }
+
+    fn extract_media_urls(&self, html: &str) -> Vec<String> {
+        let patterns = [
+            r#"https?:\\?/\\?/[^"' <>\s]+?\.(?:mp4|m3u8)[^"' <>\s]*"#,
+            r#"https?://[^"' <>\s]+?\.(?:mp4|m3u8)[^"' <>\s]*"#,
+            r#"(?is)<source[^>]+src=["']([^"']+\.(?:mp4|m3u8)[^"']*)["']"#,
+            r#"(?is)["']file["']\s*:\s*["']([^"']+\.(?:mp4|m3u8)[^"']*)["']"#,
+            r#"(?is)["']src["']\s*:\s*["']([^"']+\.(?:mp4|m3u8)[^"']*)["']"#,
+        ];
+
+        let mut urls = vec![];
+        let mut seen = HashSet::new();
+
+        for pattern in patterns {
+            let Ok(re) = Regex::new(pattern) else {
+                continue;
+            };
+            for caps in re.captures_iter(html) {
+                let raw = caps
+                    .get(1)
+                    .map(|m| m.as_str())
+                    .unwrap_or_else(|| caps.get(0).map(|m| m.as_str()).unwrap_or(""));
+                if raw.is_empty() {
+                    continue;
+                }
+                let url = Self::clean_media_url(raw);
+                if !Self::is_media_url(&url) {
+                    continue;
+                }
+                if seen.insert(url.clone()) {
+                    urls.push(url);
+                }
+            }
+        }
+
+        urls
+    }
+
+    fn extract_tags_from_detail(html: &str) -> Vec<String> {
+        let Ok(tag_re) = Regex::new(r#"(?is)href="/tags?/([^"?#/]+)/""#) else {
+            return vec![];
+        };
+        let mut out = vec![];
+        let mut seen = HashSet::new();
+        for caps in tag_re.captures_iter(html) {
+            let Some(raw) = caps.get(1).map(|m| m.as_str()) else {
+                continue;
+            };
+            let tag = raw.replace('-', " ").to_ascii_lowercase();
+            if !tag.is_empty() && seen.insert(tag.clone()) {
+                out.push(tag);
+            }
+        }
+        out
+    }
+
+    fn quality_from_url(url: &str, fallback: Option<&String>) -> String {
+        let Ok(q_re) = Regex::new(r"(?i)(\d{3,4})p") else {
+            return fallback
+                .cloned()
+                .unwrap_or_else(|| "1080p".to_string())
+                .replace('p', "");
+        };
+        if let Some(q) = q_re.captures(url).and_then(|c| c.get(1)) {
+            return q.as_str().to_string();
+        }
+        if url.to_ascii_lowercase().contains(".m3u8") {
+            return "hls".to_string();
+        }
+        fallback
+            .cloned()
+            .unwrap_or_else(|| "1080p".to_string())
+            .replace('p', "")
+    }
+
+    async fn fetch_detailed_item(
+        &self,
+        raw: RawVideo,
+        mut requester: crate::util::requester::Requester,
+    ) -> Option<VideoItem> {
+        let detail_html = match requester.get(&raw.detail_url, None).await {
+            Ok(text) => text,
+            Err(e) => {
+                report_provider_error_background(
+                    "viralxxxporn",
+                    "detail.request",
+                    &format!("url={}; error={e}", raw.detail_url),
+                );
+                return None;
+            }
+        };
+
+        let media_urls = self.extract_media_urls(&detail_html);
+        if media_urls.is_empty() {
+            report_provider_error_background(
+                "viralxxxporn",
+                "detail.media",
+                &format!("no_media_url_found; url={}", raw.detail_url),
+            );
+            return None;
+        }
+
+        let mut formats = vec![];
+        for media in &media_urls {
+            let format_kind = if media.to_ascii_lowercase().contains(".m3u8") {
+                "m3u8".to_string()
+            } else {
+                "mp4".to_string()
+            };
+            let quality = Self::quality_from_url(media, raw.quality.as_ref());
+            formats.push(VideoFormat::new(media.clone(), quality, format_kind));
+        }
+
+        let url = media_urls
+            .iter()
+            .find(|u| u.to_ascii_lowercase().contains(".mp4"))
+            .cloned()
+            .unwrap_or_else(|| media_urls.first().cloned().unwrap_or_default());
+        if url.is_empty() {
+            return None;
+        }
+
+        let mut tags = Self::extract_tags_from_detail(&detail_html);
+        if tags.is_empty() {
+            tags = vec![];
+        }
+
+        let mut item = VideoItem::new(
+            raw.id,
+            raw.title,
+            url,
+            "viralxxxporn".to_string(),
+            raw.thumb,
+            raw.duration,
+        )
+        .formats(formats)
+        .preview(
+            media_urls
+                .first()
+                .cloned()
+                .unwrap_or_else(|| raw.detail_url.clone()),
+        );
+
+        if let Some(views) = raw.views {
+            item = item.views(views);
+        }
+        if let Some(rating) = raw.rating {
+            item = item.rating(rating);
+        }
+        if !tags.is_empty() {
+            item = item.tags(tags);
+        }
+
+        Some(item)
+    }
+
+    async fn collect_video_items(
+        &self,
+        html: String,
+        requester: crate::util::requester::Requester,
+    ) -> Vec<VideoItem> {
+        let raw_items = self.parse_listing_items(&html);
+        if raw_items.is_empty() {
+            return vec![];
+        }
+
+        let mut in_flight = FuturesUnordered::new();
+        let mut out = vec![];
+        let mut iter = raw_items.into_iter();
+        const MAX_IN_FLIGHT: usize = 6;
+
+        loop {
+            while in_flight.len() < MAX_IN_FLIGHT {
+                let Some(raw) = iter.next() else {
+                    break;
+                };
+                in_flight.push(self.fetch_detailed_item(raw, requester.clone()));
+            }
+
+            let Some(result) = in_flight.next().await else {
+                break;
+            };
+            if let Some(item) = result {
+                out.push(item);
+            }
+        }
+
+        out
+    }
+
+    async fn fetch(
+        &self,
+        cache: VideoCache,
+        page: u8,
+        query: &str,
+        options: ServerOptions,
+    ) -> Result<Vec<VideoItem>> {
+        let mut requester = requester_or_default(&options, "viralxxxporn", "fetch");
+        let urls = self.build_listing_urls(page, query, &options);
+        let mut stale_items: Vec<VideoItem> = vec![];
+
+        for url in urls {
+            if let Some((time, items)) = cache.get(&url) {
+                if time.elapsed().unwrap_or_default().as_secs() < 300 {
+                    return Ok(items.clone());
+                }
+                if stale_items.is_empty() && !items.is_empty() {
+                    stale_items = items.clone();
+                }
+            }
+
+            let html = match requester.get(&url, None).await {
+                Ok(text) => text,
+                Err(e) => {
+                    report_provider_error_background(
+                        "viralxxxporn",
+                        "listing.request",
+                        &format!("url={url}; error={e}"),
+                    );
+                    continue;
+                }
+            };
+
+            let items = self.collect_video_items(html, requester.clone()).await;
+            if !items.is_empty() {
+                cache.remove(&url);
+                cache.insert(url, items.clone());
+                return Ok(items);
+            }
+        }
+
+        Ok(stale_items)
+    }
+}
+
+#[async_trait]
+impl Provider for ViralxxxpornProvider {
+    async fn get_videos(
+        &self,
+        cache: VideoCache,
+        _pool: DbPool,
+        _sort: String,
+        query: Option<String>,
+        page: String,
+        _per_page: String,
+        options: ServerOptions,
+    ) -> Vec<VideoItem> {
+        let page = page.parse::<u8>().unwrap_or(1);
+        let query = query.unwrap_or_default();
+
+        match self.fetch(cache, page, &query, options).await {
+            Ok(v) => v,
+            Err(e) => {
+                let mut chain = String::new();
+                for (idx, cause) in e.iter().enumerate() {
+                    let _ = writeln!(chain, "{}. {}", idx + 1, cause);
+                }
+                send_discord_error_report(
+                    e.to_string(),
+                    Some(chain),
+                    Some("Viralxxxporn Provider"),
+                    Some("Failed to fetch videos"),
+                    file!(),
+                    line!(),
+                    module_path!(),
+                )
+                .await;
+                vec![]
+            }
+        }
+    }
+
+    fn get_channel(&self, clientversion: ClientVersion) -> Option<Channel> {
+        Some(self.build_channel(clientversion))
+    }
+}
+
+#[cfg(test)]
+mod tests {
+    use super::ViralxxxpornProvider;
+
+    #[test]
+    fn parses_listing_card() {
+        let provider = ViralxxxpornProvider::new();
+        let html = r#"
+            <a href="/video/12345/test-title/" title="Test Title">
+              <img src="https://cdn.example.com/thumb.jpg"/>
+              <span>1080p</span><span>12:34</span><span>4.2K 95%</span>
+            </a>
+        "#;
+        let items = provider.parse_listing_items(html);
+        assert_eq!(items.len(), 1);
+        assert_eq!(items[0].id, "12345");
+        assert_eq!(items[0].title, "Test Title");
+        assert_eq!(items[0].duration, 754);
+        assert_eq!(items[0].views, Some(4200));
+        assert_eq!(items[0].rating, Some(95.0));
+    }
+
+    #[test]
+    fn extracts_media_urls() {
+        let provider = ViralxxxpornProvider::new();
+        let html = r#"
+            <script>
+                var src = "https:\/\/cdn.site.com\/video_720p.mp4?token=1\u0026x=2";
+                var hls = "https://cdn.site.com/master.m3u8";
+            </script>
+        "#;
+        let urls = provider.extract_media_urls(html);
+        assert_eq!(urls.len(), 2);
+        assert!(urls.iter().any(|u| u.contains("video_720p.mp4")));
+        assert!(urls.iter().any(|u| u.contains("master.m3u8")));
+    }
+}

src/providers/xfree.rs

@@ -293,15 +293,35 @@ impl XfreeProvider {
     }
 
     fn extract_views(text: &str) -> Option<u32> {
-        let Ok(views_re) = Regex::new(r"(?i)\b([0-9]+(?:\.[0-9]+)?\s*[kmb]?)\s*(?:views?|view)\b")
-        else {
-            return None;
-        };
-        let raw = views_re
-            .captures(text)
-            .and_then(|c| c.get(1))
-            .map(|m| m.as_str().to_string())?;
-        parse_abbreviated_number(&raw)
+        let patterns = [
+            r#"(?is)(?:class=["'][^"']*views[^"']*["'][^>]*>|data-views=["']?)([0-9]+(?:\.[0-9]+)?\s*[kmb]?)"#,
+            r#"(?is)(?:fa-eye|icon-eye|eye[^>]*></i>)[^0-9]{0,20}([0-9]+(?:\.[0-9]+)?\s*[kmb]?)"#,
+            r#"(?i)\b([0-9]+(?:\.[0-9]+)?\s*[kmb])\s*(?:views?|view)\b"#,
+        ];
+
+        for pattern in patterns {
+            let Ok(re) = Regex::new(pattern) else {
+                continue;
+            };
+            let Some(raw) = re
+                .captures(text)
+                .and_then(|c| c.get(1))
+                .map(|m| m.as_str().to_string())
+            else {
+                continue;
+            };
+            let Some(parsed) = parse_abbreviated_number(&raw) else {
+                continue;
+            };
+            let has_suffix = raw
+                .chars()
+                .any(|c| matches!(c, 'k' | 'K' | 'm' | 'M' | 'b' | 'B'));
+            if has_suffix || parsed >= 100 {
+                return Some(parsed);
+            }
+        }
+
+        None
     }
 
     fn extract_tags(text: &str) -> Vec<String> {
@@ -390,7 +410,7 @@ impl XfreeProvider {
             }
 
             let seg_start = full.start().saturating_sub(400);
-            let seg_end = (full.end() + 1600).min(html.len());
+            let seg_end = (full.end() + 700).min(html.len());
             let segment = html.get(seg_start..seg_end).unwrap_or(full.as_str());
 
             let title_from_attr = title_attr_re
@@ -748,4 +768,19 @@ mod tests {
         assert!(urls.iter().any(|u| u.contains("clip_720p.mp4")));
         assert!(urls.iter().any(|u| u.contains("master.m3u8")));
     }
+
+    #[test]
+    fn does_not_take_views_from_plain_title_text() {
+        let provider = XfreeProvider::new();
+        let html = r#"
+            <a href="/video?id=777&title=18-view-example" title="18 View Example">
+                <img src="https://thumbs.xfree.com/x/y/test.jpg" />
+                <span>1:01</span>
+                <span>18 View Example</span>
+            </a>
+        "#;
+        let items = provider.parse_listing_items(html);
+        assert_eq!(items.len(), 1);
+        assert_eq!(items[0].views, None);
+    }
 }