385 lines
13 KiB
Rust
385 lines
13 KiB
Rust
use crate::file;
|
||
use bytes::Bytes;
|
||
use scraper::{Html, Selector};
|
||
use serde::{Deserialize, Serialize};
|
||
use std::sync::mpsc;
|
||
use std::sync::Arc;
|
||
use std::thread;
|
||
use std::time::Duration;
|
||
|
||
/// 自定义错误类型，用于区分不同类型的网络错误
|
||
#[derive(Debug)]
|
||
pub enum RequestError {
|
||
NotFound, // 404 错误
|
||
NetworkError, // 网络错误
|
||
OtherError(String), // 其他错误
|
||
}
|
||
|
||
impl From<reqwest::Error> for RequestError {
|
||
fn from(err: reqwest::Error) -> Self {
|
||
if err.is_status() {
|
||
if let Some(status) = err.status() {
|
||
if status == reqwest::StatusCode::NOT_FOUND {
|
||
return RequestError::NotFound;
|
||
}
|
||
}
|
||
RequestError::OtherError(err.to_string())
|
||
} else if err.is_timeout() || err.is_connect() {
|
||
RequestError::NetworkError
|
||
} else {
|
||
RequestError::OtherError(err.to_string())
|
||
}
|
||
}
|
||
}
|
||
|
||
/// 相册数据结构
|
||
#[derive(Debug, Serialize, Deserialize)]
|
||
struct Album {
|
||
title: String,
|
||
images: Vec<String>,
|
||
}
|
||
|
||
/// 下载结果枚举
|
||
pub(crate) enum DownloadResult {
|
||
Success,
|
||
NetworkError,
|
||
Failed,
|
||
}
|
||
|
||
/// 发送 HTTP 请求并获取响应
|
||
///
|
||
/// # 参数
|
||
/// - `url`: 请求的 URL
|
||
///
|
||
/// # 返回值
|
||
/// - `Result<Bytes, RequestError>`: 响应内容或错误
|
||
fn get_response(url: &str) -> Result<Bytes, RequestError> {
|
||
let client = reqwest::blocking::Client::builder()
|
||
.timeout(Duration::from_secs(30))
|
||
.build()
|
||
.map_err(|_| RequestError::NetworkError)?;
|
||
|
||
let response = client
|
||
.get(url)
|
||
.header("User-Agent", "Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36")
|
||
.send()
|
||
.map_err(RequestError::from)?;
|
||
|
||
// 检查 HTTP 状态码
|
||
if response.status() == reqwest::StatusCode::NOT_FOUND {
|
||
return Err(RequestError::NotFound);
|
||
}
|
||
|
||
if !response.status().is_success() {
|
||
return Err(RequestError::OtherError(format!(
|
||
"HTTP {}: {}",
|
||
response.status(),
|
||
response.status().canonical_reason().unwrap_or("Unknown")
|
||
)));
|
||
}
|
||
|
||
// 先检查内容编码，再获取 bytes
|
||
let content_encoding = response.headers()
|
||
.get("content-encoding")
|
||
.and_then(|v| v.to_str().ok())
|
||
.unwrap_or("")
|
||
.to_string();
|
||
|
||
let bytes = response.bytes().map_err(RequestError::from)?;
|
||
|
||
// 处理 gzip 压缩
|
||
if content_encoding.contains("gzip") {
|
||
|
||
use std::io::Read;
|
||
let mut decoder = flate2::read::GzDecoder::new(&bytes[..]);
|
||
let mut decompressed = Vec::new();
|
||
match decoder.read_to_end(&mut decompressed) {
|
||
Ok(_) => Ok(Bytes::from(decompressed)),
|
||
Err(e) => {
|
||
println!("⚠️ Gzip 解压失败: {}, 使用原始数据", e);
|
||
Ok(bytes)
|
||
}
|
||
}
|
||
} else {
|
||
Ok(bytes)
|
||
}
|
||
}
|
||
|
||
/// 获取相册标题
|
||
///
|
||
/// # 参数
|
||
/// - `uid`: 相册 ID
|
||
///
|
||
/// # 返回值
|
||
/// - `Result<String, DownloadResult>`: 相册标题或错误
|
||
pub(crate) fn get_title(uid: i32) -> Result<String, DownloadResult> {
|
||
let url = format!("https://www.meitu131.com/meinv/{}/", uid);
|
||
|
||
match get_response(&url) {
|
||
Ok(value) if !value.is_empty() => {
|
||
let body = String::from_utf8_lossy(&value).to_string();
|
||
Ok(get_html_title(body))
|
||
}
|
||
Ok(_) => Err(DownloadResult::NetworkError),
|
||
Err(RequestError::NotFound) => Err(DownloadResult::Failed),
|
||
Err(RequestError::NetworkError) => Err(DownloadResult::NetworkError),
|
||
Err(RequestError::OtherError(_)) => Err(DownloadResult::NetworkError),
|
||
}
|
||
}
|
||
|
||
/// 从 HTML 内容中提取标题
|
||
fn get_html_title(body: String) -> String {
|
||
let document = Html::parse_document(&body);
|
||
let title_selector = Selector::parse("h1").unwrap();
|
||
document
|
||
.select(&title_selector)
|
||
.next()
|
||
.and_then(|element| element.text().next())
|
||
.unwrap_or("未知标题")
|
||
.to_string()
|
||
}
|
||
|
||
/// 下载单张图片
|
||
///
|
||
/// # 参数
|
||
/// - `url`: 图片 URL
|
||
/// - `file_path`: 保存路径
|
||
/// - `retry_times`: 重试次数
|
||
///
|
||
/// # 返回值
|
||
/// - `DownloadResult`: 下载结果
|
||
fn download_image(url: &str, file_path: &str, retry_times: i8) -> DownloadResult {
|
||
match get_response(url) {
|
||
Ok(value) if !value.is_empty() => {
|
||
file::write_file(file_path, value).expect("❌ 写入文件错误");
|
||
DownloadResult::Success
|
||
}
|
||
Ok(_) => DownloadResult::Failed, // 空内容
|
||
Err(RequestError::NotFound) => DownloadResult::Failed, // 404 错误
|
||
Err(RequestError::NetworkError) if retry_times > 1 => {
|
||
thread::sleep(Duration::from_millis(1000));
|
||
download_image(url, file_path, retry_times - 1)
|
||
}
|
||
Err(RequestError::OtherError(_)) if retry_times > 1 => {
|
||
thread::sleep(Duration::from_millis(1000));
|
||
download_image(url, file_path, retry_times - 1)
|
||
}
|
||
Err(RequestError::NetworkError) => DownloadResult::NetworkError,
|
||
Err(RequestError::OtherError(_)) => DownloadResult::NetworkError,
|
||
}
|
||
}
|
||
|
||
/// 下载相册的所有图片
|
||
///
|
||
/// # 参数
|
||
/// - `uid`: 相册 ID
|
||
/// - `title`: 相册标题
|
||
///
|
||
/// # 返回值
|
||
/// - `DownloadResult`: 下载结果
|
||
pub(crate) fn download_albums(uid: i32, title: &str) -> DownloadResult {
|
||
let base_url = format!("https://www.meitu131.com/meinv/{}/", uid);
|
||
|
||
// 获取相册主页内容
|
||
let body = match get_response(&base_url) {
|
||
Ok(value) => String::from_utf8(value.to_vec()).expect("❌ 转换错误"),
|
||
Err(RequestError::NotFound) => {
|
||
println!("❌ 相册 {} 不存在 (404)", uid);
|
||
return DownloadResult::Failed;
|
||
}
|
||
Err(RequestError::NetworkError) => {
|
||
println!("⚠️ 相册 {} 下载时网络波动 (500)", uid);
|
||
return DownloadResult::NetworkError;
|
||
}
|
||
Err(RequestError::OtherError(_)) => {
|
||
println!("⚠️ 相册 {} 下载时网络波动 (500)", uid);
|
||
return DownloadResult::NetworkError;
|
||
}
|
||
};
|
||
|
||
let document = Html::parse_document(&body);
|
||
let img_selector = Selector::parse("p[align='center'] img").unwrap();
|
||
|
||
// 获取总页数
|
||
let total_pages = get_total_pages(&document);
|
||
|
||
// 多线程收集所有图片 URL
|
||
let images = collect_image_urls(uid, total_pages, &img_selector, &base_url);
|
||
|
||
if images.len() < total_pages as usize {
|
||
println!("⚠️ 相册 {} 下载时网络波动, 仅下载 {} 页", uid, images.len());
|
||
return DownloadResult::NetworkError;
|
||
}
|
||
|
||
// 生成 JSON 文件
|
||
generate_json_file(uid, title, &images);
|
||
|
||
// 多线程下载图片
|
||
download_images_multithreaded(uid, &images)
|
||
}
|
||
|
||
/// 获取相册总页数
|
||
fn get_total_pages(document: &Html) -> i32 {
|
||
let pages_selector = Selector::parse("#pages a").unwrap();
|
||
|
||
for page_element in document.select(&pages_selector) {
|
||
if let Some(text) = page_element.text().next() {
|
||
if text == "尾页" {
|
||
if let Some(href) = page_element.value().attr("href") {
|
||
if let Some(page_num_str) = href.split("_").last() {
|
||
if let Some(page_num) = page_num_str.split(".").next() {
|
||
if let Ok(num) = page_num.parse::<i32>() {
|
||
return num;
|
||
}
|
||
}
|
||
}
|
||
}
|
||
}
|
||
}
|
||
}
|
||
1 // 默认只有一页
|
||
}
|
||
|
||
/// 多线程收集图片 URL
|
||
fn collect_image_urls(uid: i32, total_pages: i32, img_selector: &Selector, base_url: &str) -> Vec<String> {
|
||
let (tx, rx) = mpsc::channel();
|
||
let img_selector_arc = Arc::new(img_selector.clone());
|
||
|
||
// 创建页面解析线程池
|
||
let max_page_threads = 8;
|
||
|
||
for chunk_start in (1..=total_pages).step_by(max_page_threads as usize) {
|
||
let chunk_end = std::cmp::min(chunk_start + max_page_threads as i32 - 1, total_pages);
|
||
let mut page_handles = vec![];
|
||
|
||
for page in chunk_start..=chunk_end {
|
||
let tx = tx.clone();
|
||
let img_selector = Arc::clone(&img_selector_arc);
|
||
let base_url = base_url.to_string();
|
||
|
||
let handle = thread::spawn(move || {
|
||
let page_url = if page == 1 {
|
||
base_url
|
||
} else {
|
||
format!("https://www.meitu131.com/meinv/{}/index_{}.html", uid, page)
|
||
};
|
||
|
||
let mut page_images = Vec::new();
|
||
|
||
match get_response(&page_url) {
|
||
Ok(value) if !value.is_empty() => {
|
||
let page_body = String::from_utf8(value.to_vec()).expect("❌ 转换错误");
|
||
let page_document = Html::parse_document(&page_body);
|
||
for img_element in page_document.select(&img_selector) {
|
||
if let Some(src) = img_element.value().attr("src") {
|
||
page_images.push(src.to_string());
|
||
}
|
||
}
|
||
}
|
||
Ok(_) => {
|
||
println!("⚠️ 页面 {} 内容为空, 跳过", page);
|
||
}
|
||
Err(RequestError::NotFound) => {
|
||
println!("⚠️ 页面 {} 不存在 (404), 跳过", page);
|
||
}
|
||
Err(RequestError::NetworkError) => {
|
||
println!("⚠️ 获取页面 {} 时网络错误, 跳过", page);
|
||
}
|
||
Err(RequestError::OtherError(e)) => {
|
||
println!("⚠️ 获取页面 {} 时错误: {}, 跳过", page, e);
|
||
}
|
||
}
|
||
|
||
tx.send((page, page_images)).unwrap();
|
||
});
|
||
page_handles.push(handle);
|
||
}
|
||
|
||
// 等待当前批次完成
|
||
for handle in page_handles {
|
||
handle.join().unwrap();
|
||
}
|
||
}
|
||
|
||
// 收集并排序所有页面的图片 URL
|
||
drop(tx);
|
||
let mut page_results: Vec<_> = rx.iter().collect();
|
||
page_results.sort_by_key(|(page_num, _)| *page_num);
|
||
|
||
page_results.into_iter()
|
||
.flat_map(|(_, page_images)| page_images)
|
||
.collect()
|
||
}
|
||
|
||
/// 生成 JSON 文件
|
||
fn generate_json_file(uid: i32, title: &str, images: &[String]) {
|
||
let album = Album {
|
||
title: title.to_string(),
|
||
images: images.to_vec(),
|
||
};
|
||
let json = serde_json::to_string_pretty(&album).expect("❌ 序列化相册失败");
|
||
let json_path = format!("/mnt/share/data/albums/{}/index.json", uid);
|
||
file::write_file(&json_path, Bytes::from(json)).expect("❌ 写入 JSON 文件失败");
|
||
}
|
||
|
||
/// 多线程下载图片
|
||
fn download_images_multithreaded(uid: i32, images: &[String]) -> DownloadResult {
|
||
let (tx, rx) = mpsc::channel();
|
||
let images_arc = Arc::new(images.to_vec());
|
||
|
||
// 创建线程池
|
||
let max_threads = 8;
|
||
|
||
for chunk_start in (0..images.len()).step_by(max_threads) {
|
||
let chunk_end = std::cmp::min(chunk_start + max_threads, images.len());
|
||
let mut handles = vec![];
|
||
|
||
for i in chunk_start..chunk_end {
|
||
let tx = tx.clone();
|
||
let images = Arc::clone(&images_arc);
|
||
|
||
let handle = thread::spawn(move || {
|
||
let img_url = &images[i];
|
||
let file_path = format!("/mnt/share/data/albums/{}/{}.jpg", uid, i + 1);
|
||
let result = download_image(img_url, &file_path, 3);
|
||
tx.send((i + 1, result)).unwrap();
|
||
});
|
||
handles.push(handle);
|
||
}
|
||
|
||
// 等待当前批次完成
|
||
for handle in handles {
|
||
handle.join().unwrap();
|
||
}
|
||
}
|
||
|
||
// 收集下载结果
|
||
drop(tx);
|
||
let (mut successful, mut failed, mut network_errors) = (0, 0, 0);
|
||
|
||
for (pic_num, result) in rx {
|
||
match result {
|
||
DownloadResult::Success => successful += 1,
|
||
DownloadResult::Failed => {
|
||
println!("❌ 图片 {} 下载失败（空内容或404）", pic_num);
|
||
failed += 1;
|
||
}
|
||
DownloadResult::NetworkError => {
|
||
println!("⚠️ 图片 {} 下载失败（网络错误）", pic_num);
|
||
network_errors += 1;
|
||
}
|
||
}
|
||
}
|
||
|
||
println!("📊 下载完成：成功 {} 张，失败 {} 张，网络错误 {} 张",
|
||
successful, failed, network_errors);
|
||
|
||
// 返回最终结果
|
||
if failed == 0 && network_errors == 0 {
|
||
DownloadResult::Success
|
||
} else if network_errors > 0 {
|
||
DownloadResult::NetworkError
|
||
} else {
|
||
DownloadResult::Failed
|
||
}
|
||
} |