|

楼主 |
发表于 2020-5-30 23:42:27
|
显示全部楼层
- use futures::StreamExt;
- use hyper_tls::HttpsConnector;
- use std::io::Write;
- use hyper::body::HttpBody as _;
- #[tokio::main]
- async fn main() {
- let u = "https://wallhaven.cc";
- let cookie = "";
- crawl(&u, &cookie).await;
- }
- async fn crawl(base: &str, cookie: &str) {
- let mut next_pager = Some(format!("{}&page=1", base));
- loop {
- if next_pager.is_none() {
- return;
- }
- next_pager = process_one(next_pager.as_ref().unwrap(), &cookie).await;
- }
- }
- async fn process_one(u: &str, cookie: &str) -> Option<String> {
- let next_pager_url = get_next_page_url(&u, &cookie).await;
- let body = do_request(&u, &cookie).await;
- let urls = parse_image_url(&body).await;
- println!("next: {:?}", next_pager_url);
- println!("urls: {:?}", urls);
- for url in urls {
- if download(&url, &cookie).await {
- std::thread::sleep(std::time::Duration::from_millis(500));
- }
- }
- next_pager_url
- }
- async fn get_next_page_url(base: &str, cookie: &str) -> Option<String> {
- let body = do_request(base, cookie).await;
- /// class selector:
- /// grammar: .class_name .thumb-listing-page-num
- let document = scraper::Html::parse_document(&body);
- // let overlay_close_selector = scraper::Selector::parse("#overlay-close").unwrap();
- // document.select(&overlay_close_selector).next().map(|overlay| {
- // println!("overlay close title: {:?}", overlay.value().attr("title"))
- // });
- let next_pager_selector = scraper::Selector::parse(".next").unwrap();
- document.select(&next_pager_selector).next().map(|next_pager| {
- // println!("page_num_element: {:?}", next_pager.inner_html());
- String::from(next_pager.value().attr("href").unwrap())
- })
- }
- async fn do_request(u: &str, cookie: &str) -> String {
- let https = HttpsConnector::new();
- let client = hyper::Client::builder().build::<_, hyper::Body>(https);
- let request = hyper::Request::get(u).
- header("Cookie", cookie).
- body(hyper::Body::empty()).unwrap();
- let mut response = match client.request(request).await {
- Ok(response) => response,
- Err(e) => {
- println!("do request error: {:?}", e);
- panic!("");
- }
- };
- let body = hyper::body::to_bytes(response.into_body()).await.unwrap();
- let str_body: String = unsafe { String::from_utf8_unchecked(body.to_vec()) };
- str_body
- }
- async fn parse_image_url(body: &str) -> Vec<String> {
- let document = scraper::Html::parse_document(&body);
- let preview_selector = scraper::Selector::parse(".preview").unwrap();
- let urls: Vec<String> = document.select(&preview_selector).filter(|element| {
- element.value().attr("href").is_some()
- }).map(|element| {
- String::from(element.value().attr("href").unwrap())
- }).collect();
- urls
- }
- async fn download(address: &str, cookie: &str) -> bool {
- println!("address: {}", address);
- let body = do_request(&address, &cookie).await;
- let selector = r#"id="wallpaper""#;
- if let Some(index) = body.find(&selector) {
- let body = &body[index + selector.len()..];
- if let Some(src_index) = body.find("src") {
- let body = &body[src_index + 5..];
- if let Some(end_index) = body.find(r#"""#) {
- let src = &body[..end_index];
- println!("wallpaper src: {}", src);
- let slash_index = match address.rfind('/') {
- Some(index) => index,
- None => return false,
- };
- let name = &address[slash_index + 1..];
- let name = format!("./output/{}", name);
- let path = std::path::Path::new(&name);
- if path.exists() {
- return false;
- }
- let image = do_request(&src, &cookie).await;
- save(&name, image.as_bytes()).await;
- return true;
- }
- }
- return false;
- }
- return false;
- }
- async fn save(name: &str, body: &[u8]) {
- let mut file = std::fs::File::create(name).unwrap();
- file.write_all(body);
- }
复制代码 |
|