Files
hottub/src/providers/omgxxx.rs
2025-10-04 17:40:27 +00:00

466 lines
16 KiB
Rust

use crate::DbPool;
use crate::api::ClientVersion;
use crate::providers::Provider;
use crate::util::cache::VideoCache;
use crate::util::parse_abbreviated_number;
use crate::util::time::parse_time_to_seconds;
use crate::videos::{ServerOptions, VideoItem};
use crate::{status::*, util};
use async_trait::async_trait;
use error_chain::error_chain;
use htmlentity::entity::{ICodedDataTrait, decode};
use std::sync::{Arc, RwLock};
use std::vec;
use std::{env, thread};
error_chain! {
foreign_links {
Io(std::io::Error);
HttpRequest(wreq::Error);
}
}
#[derive(Debug, Clone)]
pub struct OmgxxxProvider {
url: String,
sites: Arc<RwLock<Vec<FilterOption>>>,
networks: Arc<RwLock<Vec<FilterOption>>>,
}
impl OmgxxxProvider {
pub fn new() -> Self {
let provider = OmgxxxProvider {
url: "https://www.omg.xxx".to_string(),
sites: Arc::new(RwLock::new(vec![FilterOption {
id: "all".to_string(),
title: "All".to_string(),
}])),
networks: Arc::new(RwLock::new(vec![FilterOption {
id: "all".to_string(),
title: "All".to_string(),
}])),
};
// Kick off the background load but return immediately
provider.spawn_initial_load();
provider
}
fn spawn_initial_load(&self) {
let url = self.url.clone();
let sites = Arc::clone(&self.sites);
let networks = Arc::clone(&self.networks);
thread::spawn(move || {
// Create a tiny runtime just for these async tasks
let rt = tokio::runtime::Builder::new_current_thread()
.enable_all()
.build()
.expect("build tokio runtime");
rt.block_on(async move {
// If you have a streaming sites loader, call it here too
if let Err(e) = Self::load_sites(&url, sites).await {
eprintln!("load_sites_into failed: {e}");
}
if let Err(e) = Self::load_networks(&url, networks).await {
eprintln!("load_networks failed: {e}");
}
});
});
}
async fn load_sites(base_url: &str, networks: Arc<RwLock<Vec<FilterOption>>>) -> Result<()> {
let mut requester = util::requester::Requester::new();
let mut page = 0;
loop {
page += 1;
let text = requester
.get(format!("{}/sites/{}/", &base_url, page).as_str())
.await
.unwrap();
if text.contains("404 Not Found") || text.is_empty() {
break;
}
let sites_div = text
.split("id=\"list_content_sources_sponsors_list_items\"").collect::<Vec<&str>>()[1]
.split("class=\"pagination\"").collect::<Vec<&str>>()[0];
for sites_element in
sites_div.split("class=\"headline\"").collect::<Vec<&str>>()[1..].to_vec()
{
let site_url = sites_element.split("href=\"").collect::<Vec<&str>>()[1]
.split("\"")
.collect::<Vec<&str>>()[0];
let site_id = site_url.split("/").collect::<Vec<&str>>()[4].to_string();
let site_name = sites_element.split("<h2>").collect::<Vec<&str>>()[1]
.split("<")
.collect::<Vec<&str>>()[0]
.to_string();
Self::push_unique(
&networks,
FilterOption {
id: site_id,
title: site_name,
},
);
}
}
return Ok(());
}
async fn load_networks(base_url: &str, networks: Arc<RwLock<Vec<FilterOption>>>) -> Result<()> {
let mut requester = util::requester::Requester::new();
let text = requester.get(&base_url).await.unwrap();
let networks_div = text.split("class=\"sites__list\"").collect::<Vec<&str>>()[1]
.split("</div>")
.collect::<Vec<&str>>()[0];
for network_element in
networks_div.split("sites__item").collect::<Vec<&str>>()[1..].to_vec()
{
if network_element.contains("sites__all") {
continue;
}
let network_url = network_element.split("href=\"").collect::<Vec<&str>>()[1]
.split("\"")
.collect::<Vec<&str>>()[0];
let network_id = network_url.split("/").collect::<Vec<&str>>()[4].to_string();
let network_name = network_element.split(">").collect::<Vec<&str>>()[1]
.split("<")
.collect::<Vec<&str>>()[0]
.to_string();
Self::push_unique(
&networks,
FilterOption {
id: network_id,
title: network_name,
},
);
}
return Ok(());
}
// Push one item with minimal lock time and dedup by id
fn push_unique(target: &Arc<RwLock<Vec<FilterOption>>>, item: FilterOption) {
if let Ok(mut vec) = target.write() {
if !vec.iter().any(|x| x.id == item.id) {
vec.push(item);
// Optional: keep it sorted for nicer UX
// vec.sort_by(|a,b| a.title.cmp(&b.title));
}
}
}
fn build_channel(&self, clientversion: ClientVersion) -> Channel {
let _ = clientversion;
let sites: Vec<FilterOption> = self
.sites
.read()
.map(|g| g.clone()) // or: .map(|g| g.to_vec())
.unwrap_or_default(); // or: .unwrap_or_else(|_| Vec::new())
let networks: Vec<FilterOption> = self
.networks
.read()
.map(|g| g.clone()) // or: .map(|g| g.to_vec())
.unwrap_or_default(); // or: .unwrap_or_else(|_| Vec::new())
Channel {
id: "omgxxx".to_string(),
name: "OMG XXX".to_string(),
description: "OMG look at that Collection!".to_string(),
premium: false,
favicon: "https://www.google.com/s2/favicons?sz=64&domain=www.omg.xxx".to_string(),
status: "active".to_string(),
categories: vec![],
options: vec![
ChannelOption {
id: "sort".to_string(),
title: "Sort".to_string(),
description: "Sort the Videos".to_string(),
systemImage: "list.number".to_string(),
colorName: "blue".to_string(),
options: vec![
FilterOption {
id: "latest-updates".into(),
title: "Latest".into(),
},
FilterOption {
id: "most-popular".into(),
title: "Most Viewed".into(),
},
FilterOption {
id: "top-rated".into(),
title: "Top Rated".into(),
},
],
multiSelect: false,
},
ChannelOption {
id: "sites".to_string(),
title: "Sites".to_string(),
description: "Sort the Videos".to_string(),
systemImage: "list.bullet.indent".to_string(),
colorName: "green".to_string(),
options: sites,
multiSelect: false,
},
ChannelOption {
id: "networks".to_string(),
title: "Networks".to_string(),
description: "Sort the Videos".to_string(),
systemImage: "list.dash".to_string(),
colorName: "purple".to_string(),
options: networks,
multiSelect: false,
},
],
nsfw: true,
cacheDuration: None,
}
}
async fn get(
&self,
cache: VideoCache,
page: u8,
sort: &str,
options: ServerOptions,
) -> Result<Vec<VideoItem>> {
let mut sort_string: String = match sort {
"top-rated" => "top-rated".to_string(),
"most-popular" => "most-popular".to_string(),
_ => "latest-updates".to_string(),
};
if options.network.is_some()
&& !options.network.as_ref().unwrap().is_empty()
&& options.network.as_ref().unwrap() != "all"
{
sort_string = format!("networks/{}", options.network.as_ref().unwrap());
}
let video_url = format!("{}/{}/{}/", self.url, sort_string, page);
let old_items = match cache.get(&video_url) {
Some((time, items)) => {
if time.elapsed().unwrap_or_default().as_secs() < 60 * 5 {
println!("Cache hit for URL: {}", video_url);
return Ok(items.clone());
} else {
items.clone()
}
}
None => {
vec![]
}
};
let mut requester = options.requester.clone().unwrap();
let text = requester.get(&video_url).await.unwrap();
let video_items: Vec<VideoItem> = self.get_video_items_from_html(text.clone());
if !video_items.is_empty() {
cache.remove(&video_url);
cache.insert(video_url.clone(), video_items.clone());
} else {
return Ok(old_items);
}
Ok(video_items)
}
async fn query(
&self,
cache: VideoCache,
page: u8,
query: &str,
options: ServerOptions,
) -> Result<Vec<VideoItem>> {
let mut search_type = "search";
if query.starts_with("@models:") {
search_type = "models";
}
let video_url = format!(
"{}/{}/{}/{}/",
self.url,
search_type,
query
.to_lowercase()
.trim()
.replace(" ", "-")
.replace("@models:", ""),
page
);
// Check our Video Cache. If the result is younger than 1 hour, we return it.
let old_items = match cache.get(&video_url) {
Some((time, items)) => {
if time.elapsed().unwrap_or_default().as_secs() < 60 * 5 {
return Ok(items.clone());
} else {
let _ = cache.check().await;
return Ok(items.clone());
}
}
None => {
vec![]
}
};
let mut requester = options.requester.clone().unwrap();
let text = requester.get(&video_url).await.unwrap();
let video_items: Vec<VideoItem> = self.get_video_items_from_html(text.clone());
if !video_items.is_empty() {
cache.remove(&video_url);
cache.insert(video_url.clone(), video_items.clone());
} else {
return Ok(old_items);
}
Ok(video_items)
}
fn get_video_items_from_html(&self, html: String) -> Vec<VideoItem> {
if html.is_empty() {
println!("HTML is empty");
return vec![];
}
let mut items: Vec<VideoItem> = Vec::new();
let raw_videos = html.split("videos_list_pagination").collect::<Vec<&str>>()[0]
.split(" class=\"pagination\" ")
.collect::<Vec<&str>>()[0]
.split("class=\"list-videos\"")
.collect::<Vec<&str>>()[1]
.split("class=\"item\"")
.collect::<Vec<&str>>()[1..]
.to_vec();
for video_segment in &raw_videos {
// let vid = video_segment.split("\n").collect::<Vec<&str>>();
// for (index, line) in vid.iter().enumerate() {
// println!("Line {}: {}", index, line);
// }
let video_url: String = video_segment.split("<a href=\"").collect::<Vec<&str>>()[1]
.split("\"")
.collect::<Vec<&str>>()[0]
.to_string();
let mut title = video_segment.split(" title=\"").collect::<Vec<&str>>()[1]
.split("\"")
.collect::<Vec<&str>>()[0]
.to_string();
// html decode
title = decode(title.as_bytes()).to_string().unwrap_or(title);
let id = video_url.split("/").collect::<Vec<&str>>()[4].to_string();
let thumb = match video_segment.split("img loading").collect::<Vec<&str>>()[1]
.contains("data-src=\"")
{
true => video_segment.split("img loading").collect::<Vec<&str>>()[1]
.split("data-src=\"")
.collect::<Vec<&str>>()[1]
.split("\"")
.collect::<Vec<&str>>()[0]
.to_string(),
false => video_segment.split("img loading").collect::<Vec<&str>>()[1]
.split("data-original=\"")
.collect::<Vec<&str>>()[1]
.split("\"")
.collect::<Vec<&str>>()[0]
.to_string(),
};
let raw_duration = video_segment
.split("<span class=\"duration\">")
.collect::<Vec<&str>>()[1]
.split("<")
.collect::<Vec<&str>>()[0]
.split(" ")
.collect::<Vec<&str>>()
.last()
.unwrap_or(&"")
.to_string();
let duration = parse_time_to_seconds(raw_duration.as_str()).unwrap_or(0) as u32;
let views = parse_abbreviated_number(
video_segment
.split("<div class=\"views\">")
.collect::<Vec<&str>>()[1]
.split("<")
.collect::<Vec<&str>>()[0]
.to_string()
.as_str(),
)
.unwrap_or(0) as u32;
let preview = video_segment
.split("data-preview=\"")
.collect::<Vec<&str>>()[1]
.split("\"")
.collect::<Vec<&str>>()[0]
.to_string();
let tags = match video_segment.contains("class=\"models\">") {
true => video_segment
.split("class=\"models\">")
.collect::<Vec<&str>>()[1]
.split("</div>")
.collect::<Vec<&str>>()[0]
.split("href=\"")
.collect::<Vec<&str>>()[1..]
.into_iter()
.map(|s| {
format!(
"@models:{}",
s.split("/").collect::<Vec<&str>>()[4].to_string()
)
})
.collect::<Vec<String>>()
.to_vec(),
false => vec![],
};
let video_item = VideoItem::new(
id,
title,
video_url.to_string(),
"omgxxx".to_string(),
thumb,
duration,
)
.views(views)
.preview(preview)
.tags(tags);
items.push(video_item);
}
return items;
}
}
#[async_trait]
impl Provider for OmgxxxProvider {
async fn get_videos(
&self,
cache: VideoCache,
pool: DbPool,
sort: String,
query: Option<String>,
page: String,
per_page: String,
options: ServerOptions,
) -> Vec<VideoItem> {
let _ = per_page;
let _ = pool;
let videos: std::result::Result<Vec<VideoItem>, Error> = match query {
Some(q) => {
self.query(cache, page.parse::<u8>().unwrap_or(1), &q, options)
.await
}
None => {
self.get(cache, page.parse::<u8>().unwrap_or(1), &sort, options)
.await
}
};
match videos {
Ok(v) => v,
Err(e) => {
println!("Error fetching videos: {}", e);
vec![]
}
}
}
fn get_channel(&self, clientversion: ClientVersion) -> crate::status::Channel {
self.build_channel(clientversion)
}
}