Files
hottub/src/providers/noodlemagazine.rs
2026-03-20 21:05:18 +00:00

457 lines
14 KiB
Rust

use crate::DbPool;
use crate::api::ClientVersion;
use crate::providers::Provider;
use crate::status::*;
use crate::util::cache::VideoCache;
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 htmlentity::entity::{ICodedDataTrait, decode};
use std::net::IpAddr;
use url::Url;
use std::vec;
use titlecase::Titlecase;
use wreq::Version;
pub const CHANNEL_METADATA: crate::providers::ProviderChannelMetadata =
crate::providers::ProviderChannelMetadata {
group_id: "mainstream-tube",
tags: &["search", "mixed", "user-upload"],
};
error_chain! {
foreign_links {
Io(std::io::Error);
HttpRequest(wreq::Error);
}
errors {
Parse(msg: String)
}
}
#[derive(Debug, Clone)]
pub struct NoodlemagazineProvider {
url: String,
}
impl NoodlemagazineProvider {
pub fn new() -> Self {
Self {
url: "https://noodlemagazine.com".to_string(),
}
}
fn build_channel(&self, _clientversion: ClientVersion) -> Channel {
Channel {
id: "noodlemagazine".into(),
name: "Noodlemagazine".into(),
description: "The Best Search Engine of HD Videos".into(),
premium: false,
favicon: "https://www.google.com/s2/favicons?sz=64&domain=noodlemagazine.com".into(),
status: "active".into(),
categories: vec![],
options: vec![],
nsfw: true,
cacheDuration: Some(1800),
}
}
async fn get(
&self,
cache: VideoCache,
page: u8,
_sort: &str,
options: ServerOptions,
) -> Result<Vec<VideoItem>> {
let video_url = format!(
"{}/popular/recent?sort_by=views&sort_order=desc&p={}",
self.url,
page.saturating_sub(1)
);
let old_items = match cache.get(&video_url) {
Some((t, i)) if t.elapsed().unwrap_or_default().as_secs() < 300 => return Ok(i.clone()),
Some((_, i)) => i.clone(),
None => vec![],
};
let mut requester = match options.requester.clone() {
Some(r) => r,
None => return Ok(old_items),
};
let text = requester
.get(&video_url, Some(Version::HTTP_2))
.await
.unwrap_or_default();
let items = self.get_video_items_from_html(text, &options);
if items.is_empty() {
Ok(old_items)
} else {
cache.remove(&video_url);
cache.insert(video_url, items.clone());
Ok(items)
}
}
async fn query(
&self,
cache: VideoCache,
page: u8,
query: &str,
options: ServerOptions,
) -> Result<Vec<VideoItem>> {
let q = query.trim().replace(' ', "%20");
let video_url = format!("{}/video/{}?p={}", self.url, q, page.saturating_sub(1));
let old_items = match cache.get(&video_url) {
Some((t, i)) if t.elapsed().unwrap_or_default().as_secs() < 300 => return Ok(i.clone()),
Some((_, i)) => i.clone(),
None => vec![],
};
let mut requester = match options.requester.clone() {
Some(r) => r,
None => return Ok(old_items),
};
let text = requester
.get(&video_url, Some(Version::HTTP_2))
.await
.unwrap_or_default();
let items = self.get_video_items_from_html(text, &options);
if items.is_empty() {
Ok(old_items)
} else {
cache.remove(&video_url);
cache.insert(video_url, items.clone());
Ok(items)
}
}
fn get_video_items_from_html(&self, html: String, options: &ServerOptions) -> Vec<VideoItem> {
if html.is_empty() || html.contains("404 Not Found") {
return vec![];
}
let section = match html.split(">Show more</div>").next() {
Some(s) => s,
None => return vec![],
};
let list = match section
.split("<div class=\"list_videos\" id=\"list_videos\">")
.nth(1)
{
Some(l) => l,
None => return vec![],
};
list.split("<div class=\"item\">")
.skip(1)
.filter_map(|segment| {
self.get_video_item(segment.to_string(), options).ok()
})
.collect()
}
fn proxy_url(&self, options: &ServerOptions, video_url: &str) -> String {
crate::providers::build_proxy_url(
options,
"noodlemagazine",
&crate::providers::strip_url_scheme(video_url),
)
}
fn normalize_thumb_url(&self, thumb: &str) -> String {
let thumb = thumb.trim();
if thumb.is_empty() {
return String::new();
}
if thumb.starts_with("http://") || thumb.starts_with("https://") {
return thumb.to_string();
}
if thumb.starts_with("//") {
return format!("https:{thumb}");
}
if thumb.starts_with('/') {
return format!("{}{}", self.url, thumb);
}
format!("{}/{}", self.url.trim_end_matches('/'), thumb.trim_start_matches('/'))
}
fn has_allowed_image_extension(path: &str) -> bool {
let path = path.to_ascii_lowercase();
[".jpg", ".jpeg", ".png", ".webp", ".avif", ".gif"]
.iter()
.any(|ext| path.ends_with(ext))
}
fn is_disallowed_thumb_host(host: &str) -> bool {
if host.eq_ignore_ascii_case("localhost") {
return true;
}
match host.parse::<IpAddr>() {
Ok(IpAddr::V4(ip)) => {
ip.is_private()
|| ip.is_loopback()
|| ip.is_link_local()
|| ip.is_broadcast()
|| ip.is_documentation()
|| ip.is_unspecified()
}
Ok(IpAddr::V6(ip)) => {
ip.is_loopback()
|| ip.is_unspecified()
|| ip.is_multicast()
|| ip.is_unique_local()
|| ip.is_unicast_link_local()
}
Err(_) => false,
}
}
fn is_allowed_thumb_url(&self, url: &str) -> bool {
let Some(url) = Url::parse(url).ok() else {
return false;
};
if url.scheme() != "https" {
return false;
}
let Some(host) = url.host_str() else {
return false;
};
!Self::is_disallowed_thumb_host(host) && Self::has_allowed_image_extension(url.path())
}
fn proxied_thumb(&self, options: &ServerOptions, thumb: &str) -> String {
let normalized = self.normalize_thumb_url(thumb);
if normalized.is_empty() || !self.is_allowed_thumb_url(&normalized) {
return String::new();
}
crate::providers::build_proxy_url(
options,
"noodlemagazine-thumb",
&crate::providers::strip_url_scheme(&normalized),
)
}
fn get_video_item(&self, video_segment: String, options: &ServerOptions) -> Result<VideoItem> {
let href = video_segment
.split("<a href=\"")
.nth(1)
.and_then(|s| s.split('"').next())
.ok_or_else(|| Error::from("missing href"))?;
let video_url = format!("{}{}", self.url, href);
let mut title = video_segment
.split("<div class=\"title\">")
.nth(1)
.and_then(|s| s.split('<').next())
.unwrap_or("")
.trim()
.to_string();
title = decode(title.as_bytes())
.to_string()
.unwrap_or(title)
.titlecase();
let id = video_url
.split('/')
.nth(4)
.and_then(|s| s.split('.').next())
.ok_or_else(|| Error::from("missing id"))?
.to_string();
let thumb = video_segment
.split("data-src=\"")
.nth(1)
.and_then(|s| s.split('"').next())
.unwrap_or("")
.to_string();
let raw_duration = video_segment
.split("#clock-o\"></use></svg>")
.nth(1)
.and_then(|s| s.split('<').next())
.unwrap_or("0:00");
let duration = parse_time_to_seconds(raw_duration).unwrap_or(0) as u32;
let views = video_segment
.split("#eye\"></use></svg>")
.nth(1)
.and_then(|s| s.split('<').next())
.and_then(|v| parse_abbreviated_number(v.trim()))
.unwrap_or(0);
let proxy_url = self.proxy_url(options, &video_url);
let proxied_thumb = self.proxied_thumb(options, &thumb);
Ok(VideoItem::new(
id,
title,
proxy_url.clone(),
"noodlemagazine".into(),
proxied_thumb,
duration,
)
.views(views)
.formats(vec![
VideoFormat::new(proxy_url, "auto".into(), "video/mp4".into())
.format_id("auto".into())
.format_note("proxied".into())
.http_header("Referer".into(), video_url),
]))
}
}
#[async_trait]
impl Provider for NoodlemagazineProvider {
async fn get_videos(
&self,
cache: VideoCache,
pool: DbPool,
sort: String,
query: Option<String>,
page: String,
per_page: String,
options: ServerOptions,
) -> Vec<VideoItem> {
let _ = pool;
let _ = per_page;
let page = page.parse::<u8>().unwrap_or(1);
let res = match query {
Some(q) => self.query(cache, page, &q, options).await,
None => self.get(cache, page, &sort, options).await,
};
res.unwrap_or_else(|e| {
eprintln!("Noodlemagazine error: {e}");
vec![]
})
}
fn get_channel(&self, clientversion: ClientVersion) -> Option<Channel> {
Some(self.build_channel(clientversion))
}
}
#[cfg(test)]
mod tests {
use super::NoodlemagazineProvider;
use crate::videos::ServerOptions;
fn options() -> ServerOptions {
ServerOptions {
featured: None,
category: None,
sites: None,
filter: None,
language: None,
public_url_base: Some("https://example.com".to_string()),
requester: None,
network: None,
stars: None,
categories: None,
duration: None,
sort: None,
sexuality: None,
}
}
#[test]
fn rewrites_video_pages_to_hottub_proxy() {
let provider = NoodlemagazineProvider::new();
let options = options();
assert_eq!(
provider.proxy_url(&options, "https://noodlemagazine.com/watch/-123_456"),
"https://example.com/proxy/noodlemagazine/noodlemagazine.com/watch/-123_456"
);
}
#[test]
fn parses_listing_without_detail_page_requests() {
let provider = NoodlemagazineProvider::new();
let options = options();
let html = r#"
<div class="list_videos" id="list_videos">
<div class="item">
<a href="/watch/-123_456">
<img data-src="https://noodlemagazine.com/thumbs/test.jpg" />
</a>
<div class="title">sample &amp; title</div>
<svg><use></use></svg>#clock-o"></use></svg>12:34<
<svg><use></use></svg>#eye"></use></svg>1.2K<
</div>
>Show more</div>
"#;
let items = provider.get_video_items_from_html(html.to_string(), &options);
assert_eq!(items.len(), 1);
assert_eq!(
items[0].url,
"https://example.com/proxy/noodlemagazine/noodlemagazine.com/watch/-123_456"
);
assert_eq!(
items[0].thumb,
"https://example.com/proxy/noodlemagazine-thumb/noodlemagazine.com/thumbs/test.jpg"
);
assert_eq!(items[0].formats.as_ref().map(|f| f.len()), Some(1));
}
#[test]
fn keeps_https_cdn_thumbs_but_drops_non_images() {
let provider = NoodlemagazineProvider::new();
let options = options();
let html = r#"
<div class="list_videos" id="list_videos">
<div class="item">
<a href="/watch/-123_456">
<img data-src="https://cdn.example/thumb.jpg" />
</a>
<div class="title">sample</div>
<svg><use></use></svg>#clock-o"></use></svg>12:34<
<svg><use></use></svg>#eye"></use></svg>1.2K<
</div>
<div class="item">
<a href="/watch/-555_666">
<img data-src="https://noodlemagazine.com/watch/not-an-image" />
</a>
<div class="title">sample 2</div>
<svg><use></use></svg>#clock-o"></use></svg>00:42<
<svg><use></use></svg>#eye"></use></svg>123<
</div>
>Show more</div>
"#;
let items = provider.get_video_items_from_html(html.to_string(), &options);
assert_eq!(items.len(), 2);
assert_eq!(
items[0].thumb,
"https://example.com/proxy/noodlemagazine-thumb/cdn.example/thumb.jpg"
);
assert!(items[1].thumb.is_empty());
}
}