Files
hottub/src/proxies/heavyfetish.rs
2026-04-05 21:27:47 +00:00

172 lines
5.3 KiB
Rust

use std::collections::HashMap;
use ntex::web;
use regex::Regex;
use scraper::{Html, Selector};
use url::Url;
use crate::util::requester::Requester;
#[derive(Debug, Clone)]
pub struct HeavyfetishProxy {}
impl HeavyfetishProxy {
pub fn new() -> Self {
Self {}
}
fn normalize_detail_url(endpoint: &str) -> Option<String> {
let endpoint = endpoint.trim().trim_start_matches('/');
if endpoint.is_empty() {
return None;
}
let detail_url = if endpoint.starts_with("http://") || endpoint.starts_with("https://") {
endpoint.to_string()
} else {
format!("https://{}", endpoint.trim_start_matches('/'))
};
Self::is_allowed_detail_url(&detail_url).then_some(detail_url)
}
fn is_allowed_detail_url(url: &str) -> bool {
let Some(parsed) = Url::parse(url).ok() else {
return false;
};
if parsed.scheme() != "https" {
return false;
}
let Some(host) = parsed.host_str() else {
return false;
};
(host == "heavyfetish.com" || host == "www.heavyfetish.com")
&& parsed.path().starts_with("/videos/")
}
fn normalize_url(raw: &str) -> String {
let value = raw.trim().replace("\\/", "/");
if value.is_empty() {
return String::new();
}
if value.starts_with("//") {
return format!("https:{value}");
}
if value.starts_with('/') {
return format!("https://heavyfetish.com{value}");
}
if value.starts_with("http://") {
return value.replacen("http://", "https://", 1);
}
value
}
fn quality_from_url(url: &str) -> String {
for quality in ["2160p", "1440p", "1080p", "720p", "480p", "360p", "240p"] {
if url.contains(quality) {
return quality.to_string();
}
}
"480p".to_string()
}
fn quality_score(label: &str) -> u32 {
label
.chars()
.filter(|value| value.is_ascii_digit())
.collect::<String>()
.parse::<u32>()
.unwrap_or(0)
}
fn regex(value: &str) -> Option<Regex> {
Regex::new(value).ok()
}
fn extract_js_value(block: &str, regex: &Regex) -> Option<String> {
regex
.captures(block)
.and_then(|captures| captures.get(1))
.map(|value| value.as_str().replace("\\/", "/").replace("\\'", "'"))
}
fn selector(value: &str) -> Option<Selector> {
Selector::parse(value).ok()
}
fn extract_source_url(html: &str) -> Option<String> {
let flashvars_regex = Self::regex(r#"(?s)var\s+flashvars\s*=\s*\{(.*?)\};"#)?;
let value_regex = |key: &str| Self::regex(&format!(r#"{key}:\s*'((?:\\'|[^'])*)'"#));
let mut seen = HashMap::<String, String>::new();
if let Some(flashvars) = flashvars_regex
.captures(html)
.and_then(|value| value.get(1))
.map(|value| value.as_str().to_string())
{
for key in ["video_alt_url2", "video_alt_url", "video_url"] {
let Some(url_regex) = value_regex(key) else {
continue;
};
let Some(text_regex) = value_regex(&format!("{key}_text")) else {
continue;
};
let Some(url) = Self::extract_js_value(&flashvars, &url_regex) else {
continue;
};
let normalized = Self::normalize_url(&url);
if normalized.is_empty() {
continue;
}
let quality = Self::extract_js_value(&flashvars, &text_regex)
.filter(|value| !value.is_empty())
.unwrap_or_else(|| Self::quality_from_url(&normalized));
seen.entry(quality).or_insert(normalized);
}
}
let document = Html::parse_document(html);
let Some(download_selector) = Self::selector("#download_popup a[href*='/get_file/']")
else {
return seen
.iter()
.max_by_key(|(quality, _)| Self::quality_score(quality))
.map(|(_, url)| url.clone());
};
for element in document.select(&download_selector) {
let href = element.value().attr("href").unwrap_or_default();
let normalized = Self::normalize_url(href);
if normalized.is_empty() {
continue;
}
let quality = Self::quality_from_url(&normalized);
seen.entry(quality).or_insert(normalized);
}
seen.iter()
.max_by_key(|(quality, _)| Self::quality_score(quality))
.map(|(_, url)| url.clone())
}
}
impl crate::proxies::Proxy for HeavyfetishProxy {
async fn get_video_url(&self, url: String, requester: web::types::State<Requester>) -> String {
let Some(detail_url) = Self::normalize_detail_url(&url) else {
return String::new();
};
let mut requester = requester.get_ref().clone();
let html = requester.get(&detail_url, None).await.unwrap_or_default();
if html.is_empty() {
return String::new();
}
Self::extract_source_url(&html).unwrap_or_default()
}
}