mirror of
https://github.com/fish2018/pansou.git
synced 2025-11-25 03:14:59 +08:00
Merge pull request #56 from woleigedouer/main
玩偶系增加封面获取 失效源从json重构为html
This commit is contained in:
Nobody
@@ -273,7 +273,14 @@ func (p *DuoduoAsyncPlugin) parseSearchItem(s *goquery.Selection, keyword string
|
||||
return strings.Contains(title, "剧情")
|
||||
})
|
||||
plot := strings.TrimSpace(plotElement.Find(".video-info-item").Text())
|
||||
|
||||
|
||||
// 提取封面图片 (参考 Pan_mogg.js 的选择器)
|
||||
var images []string
|
||||
if picURL, exists := s.Find(".module-item-pic > img").Attr("data-src"); exists && picURL != "" {
|
||||
images = append(images, picURL)
|
||||
}
|
||||
result.Images = images
|
||||
|
||||
// 构建内容描述
|
||||
var contentParts []string
|
||||
if quality != "" {
|
||||
@@ -328,7 +335,7 @@ func (p *DuoduoAsyncPlugin) enhanceWithDetails(client *http.Client, results []mo
|
||||
}
|
||||
|
||||
itemID := parts[1]
|
||||
|
||||
|
||||
// 检查缓存
|
||||
if cached, ok := detailCache.Load(itemID); ok {
|
||||
if cachedResult, ok := cached.(model.SearchResult); ok {
|
||||
@@ -340,14 +347,19 @@ func (p *DuoduoAsyncPlugin) enhanceWithDetails(client *http.Client, results []mo
|
||||
}
|
||||
}
|
||||
atomic.AddInt64(&cacheMisses, 1)
|
||||
|
||||
// 获取详情页链接
|
||||
detailLinks := p.fetchDetailLinks(client, itemID)
|
||||
|
||||
// 获取详情页链接和图片
|
||||
detailLinks, detailImages := p.fetchDetailLinksAndImages(client, itemID)
|
||||
r.Links = detailLinks
|
||||
|
||||
|
||||
// 合并图片：优先使用详情页的海报，如果没有则使用搜索结果的图片
|
||||
if len(detailImages) > 0 {
|
||||
r.Images = detailImages
|
||||
}
|
||||
|
||||
// 缓存结果
|
||||
detailCache.Store(itemID, r)
|
||||
|
||||
|
||||
mu.Lock()
|
||||
enhancedResults = append(enhancedResults, r)
|
||||
mu.Unlock()
|
||||
@@ -387,8 +399,8 @@ func (p *DuoduoAsyncPlugin) doRequestWithRetry(req *http.Request, client *http.C
|
||||
return nil, fmt.Errorf("重试 %d 次后仍然失败: %w", maxRetries, lastErr)
|
||||
}
|
||||
|
||||
// fetchDetailLinks 获取详情页的下载链接
|
||||
func (p *DuoduoAsyncPlugin) fetchDetailLinks(client *http.Client, itemID string) []model.Link {
|
||||
// fetchDetailLinksAndImages 获取详情页的下载链接和图片
|
||||
func (p *DuoduoAsyncPlugin) fetchDetailLinksAndImages(client *http.Client, itemID string) ([]model.Link, []string) {
|
||||
// 性能统计
|
||||
start := time.Now()
|
||||
atomic.AddInt64(&detailPageRequests, 1)
|
||||
@@ -398,42 +410,48 @@ func (p *DuoduoAsyncPlugin) fetchDetailLinks(client *http.Client, itemID string)
|
||||
}()
|
||||
|
||||
detailURL := fmt.Sprintf("https://tv.yydsys.top/index.php/vod/detail/id/%s.html", itemID)
|
||||
|
||||
|
||||
// 创建带超时的上下文
|
||||
ctx, cancel := context.WithTimeout(context.Background(), DetailTimeout)
|
||||
defer cancel()
|
||||
|
||||
|
||||
// 创建请求
|
||||
req, err := http.NewRequestWithContext(ctx, "GET", detailURL, nil)
|
||||
if err != nil {
|
||||
return nil
|
||||
return nil, nil
|
||||
}
|
||||
|
||||
|
||||
// 设置请求头
|
||||
req.Header.Set("User-Agent", "Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/91.0.4472.124 Safari/537.36")
|
||||
req.Header.Set("Accept", "text/html,application/xhtml+xml,application/xml;q=0.9,image/webp,*/*;q=0.8")
|
||||
req.Header.Set("Accept-Language", "zh-CN,zh;q=0.9,en;q=0.8")
|
||||
req.Header.Set("Connection", "keep-alive")
|
||||
req.Header.Set("Referer", "https://tv.yydsys.top/")
|
||||
|
||||
|
||||
// 发送请求（带重试）
|
||||
resp, err := p.doRequestWithRetry(req, client)
|
||||
if err != nil {
|
||||
return nil
|
||||
return nil, nil
|
||||
}
|
||||
defer resp.Body.Close()
|
||||
|
||||
|
||||
if resp.StatusCode != 200 {
|
||||
return nil
|
||||
return nil, nil
|
||||
}
|
||||
|
||||
|
||||
doc, err := goquery.NewDocumentFromReader(resp.Body)
|
||||
if err != nil {
|
||||
return nil
|
||||
return nil, nil
|
||||
}
|
||||
|
||||
|
||||
var links []model.Link
|
||||
|
||||
var images []string
|
||||
|
||||
// 提取详情页的海报图片 (参考 Pan_mogg.js 的选择器)
|
||||
if posterURL, exists := doc.Find(".mobile-play .lazyload").Attr("data-src"); exists && posterURL != "" {
|
||||
images = append(images, posterURL)
|
||||
}
|
||||
|
||||
// 查找下载链接区域
|
||||
doc.Find("#download-list .module-row-one").Each(func(i int, s *goquery.Selection) {
|
||||
// 从data-clipboard-text属性提取链接
|
||||
@@ -450,7 +468,7 @@ func (p *DuoduoAsyncPlugin) fetchDetailLinks(client *http.Client, itemID string)
|
||||
}
|
||||
}
|
||||
}
|
||||
|
||||
|
||||
// 也检查直接的href属性
|
||||
s.Find("a[href]").Each(func(j int, a *goquery.Selection) {
|
||||
if linkURL, exists := a.Attr("href"); exists {
|
||||
@@ -465,7 +483,7 @@ func (p *DuoduoAsyncPlugin) fetchDetailLinks(client *http.Client, itemID string)
|
||||
break
|
||||
}
|
||||
}
|
||||
|
||||
|
||||
if !isDuplicate {
|
||||
link := model.Link{
|
||||
Type: linkType,
|
||||
@@ -479,7 +497,13 @@ func (p *DuoduoAsyncPlugin) fetchDetailLinks(client *http.Client, itemID string)
|
||||
}
|
||||
})
|
||||
})
|
||||
|
||||
|
||||
return links, images
|
||||
}
|
||||
|
||||
// fetchDetailLinks 获取详情页的下载链接（兼容性方法，仅返回链接）
|
||||
func (p *DuoduoAsyncPlugin) fetchDetailLinks(client *http.Client, itemID string) []model.Link {
|
||||
links, _ := p.fetchDetailLinksAndImages(client, itemID)
|
||||
return links
|
||||
}
|
||||
|
||||
|
||||
@@ -2,35 +2,52 @@ package erxiao
|
||||
|
||||
import (
|
||||
"fmt"
|
||||
"io"
|
||||
"net/http"
|
||||
"net/url"
|
||||
"regexp"
|
||||
"strings"
|
||||
"time"
|
||||
"context"
|
||||
"sync"
|
||||
"sync/atomic"
|
||||
|
||||
"github.com/PuerkitoBio/goquery"
|
||||
"pansou/model"
|
||||
"pansou/plugin"
|
||||
"pansou/util/json"
|
||||
)
|
||||
|
||||
const (
|
||||
// 默认超时时间 - 优化为更短时间
|
||||
// 默认超时时间
|
||||
DefaultTimeout = 8 * time.Second
|
||||
DetailTimeout = 6 * time.Second
|
||||
|
||||
// HTTP连接池配置
|
||||
MaxIdleConns = 200
|
||||
MaxIdleConnsPerHost = 50
|
||||
MaxConnsPerHost = 100
|
||||
IdleConnTimeout = 90 * time.Second
|
||||
|
||||
// 并发控制
|
||||
MaxConcurrency = 20
|
||||
|
||||
// 缓存TTL
|
||||
cacheTTL = 1 * time.Hour
|
||||
)
|
||||
|
||||
// 性能统计（原子操作）
|
||||
var (
|
||||
searchRequests int64 = 0
|
||||
totalSearchTime int64 = 0 // 纳秒
|
||||
searchRequests int64 = 0
|
||||
totalSearchTime int64 = 0 // 纳秒
|
||||
detailPageRequests int64 = 0
|
||||
totalDetailTime int64 = 0 // 纳秒
|
||||
cacheHits int64 = 0
|
||||
cacheMisses int64 = 0
|
||||
)
|
||||
|
||||
// Detail page缓存
|
||||
var (
|
||||
detailCache sync.Map
|
||||
cacheMutex sync.RWMutex
|
||||
)
|
||||
|
||||
func init() {
|
||||
@@ -41,7 +58,10 @@ func init() {
|
||||
var (
|
||||
// 密码提取正则表达式
|
||||
passwordRegex = regexp.MustCompile(`\?pwd=([0-9a-zA-Z]+)`)
|
||||
|
||||
|
||||
// 详情页ID提取正则表达式
|
||||
detailIDRegex = regexp.MustCompile(`/id/(\d+)`)
|
||||
|
||||
// 常见网盘链接的正则表达式（支持16种类型）
|
||||
quarkLinkRegex = regexp.MustCompile(`https?://pan\.quark\.cn/s/[0-9a-zA-Z]+`)
|
||||
ucLinkRegex = regexp.MustCompile(`https?://drive\.uc\.cn/s/[0-9a-zA-Z]+(\?[^"'\s]*)?`)
|
||||
@@ -100,7 +120,7 @@ func (p *ErxiaoAsyncPlugin) SearchWithResult(keyword string, ext map[string]inte
|
||||
return p.AsyncSearchWithResult(keyword, p.searchImpl, p.MainCacheKey, ext)
|
||||
}
|
||||
|
||||
// searchImpl 搜索实现
|
||||
// searchImpl 搜索实现 - HTML解析版本
|
||||
func (p *ErxiaoAsyncPlugin) searchImpl(client *http.Client, keyword string, ext map[string]interface{}) ([]model.SearchResult, error) {
|
||||
// 性能统计
|
||||
start := time.Now()
|
||||
@@ -115,271 +135,307 @@ func (p *ErxiaoAsyncPlugin) searchImpl(client *http.Client, keyword string, ext
|
||||
client = p.optimizedClient
|
||||
}
|
||||
|
||||
// 构建API搜索URL
|
||||
searchURL := fmt.Sprintf("https://erxiaofn.click/api.php/provide/vod?ac=detail&wd=%s", url.QueryEscape(keyword))
|
||||
|
||||
// 创建HTTP请求
|
||||
// 1. 构建搜索URL
|
||||
searchURL := fmt.Sprintf("https://erxiaofn.click/index.php/vod/search/wd/%s.html", url.QueryEscape(keyword))
|
||||
|
||||
// 2. 创建带超时的上下文
|
||||
ctx, cancel := context.WithTimeout(context.Background(), DefaultTimeout)
|
||||
defer cancel()
|
||||
|
||||
|
||||
// 3. 创建请求
|
||||
req, err := http.NewRequestWithContext(ctx, "GET", searchURL, nil)
|
||||
if err != nil {
|
||||
return nil, fmt.Errorf("[%s] 创建搜索请求失败: %w", p.Name(), err)
|
||||
return nil, fmt.Errorf("[%s] 创建请求失败: %w", p.Name(), err)
|
||||
}
|
||||
|
||||
// 设置请求头
|
||||
|
||||
// 4. 设置请求头
|
||||
req.Header.Set("User-Agent", "Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/91.0.4472.124 Safari/537.36")
|
||||
req.Header.Set("Accept", "application/json, text/plain, */*")
|
||||
req.Header.Set("Accept", "text/html,application/xhtml+xml,application/xml;q=0.9,image/webp,*/*;q=0.8")
|
||||
req.Header.Set("Accept-Language", "zh-CN,zh;q=0.9,en;q=0.8")
|
||||
req.Header.Set("Connection", "keep-alive")
|
||||
req.Header.Set("Referer", "https://erxiaofn.click/")
|
||||
req.Header.Set("Cache-Control", "no-cache")
|
||||
|
||||
// 发送请求
|
||||
|
||||
// 5. 发送请求
|
||||
resp, err := p.doRequestWithRetry(req, client)
|
||||
if err != nil {
|
||||
return nil, fmt.Errorf("[%s] 搜索请求失败: %w", p.Name(), err)
|
||||
}
|
||||
defer resp.Body.Close()
|
||||
|
||||
// 解析JSON响应
|
||||
body, err := io.ReadAll(resp.Body)
|
||||
|
||||
if resp.StatusCode != 200 {
|
||||
return nil, fmt.Errorf("[%s] 搜索请求返回状态码: %d", p.Name(), resp.StatusCode)
|
||||
}
|
||||
|
||||
// 6. 解析搜索结果页面
|
||||
doc, err := goquery.NewDocumentFromReader(resp.Body)
|
||||
if err != nil {
|
||||
return nil, fmt.Errorf("[%s] 读取响应失败: %w", p.Name(), err)
|
||||
return nil, fmt.Errorf("[%s] 解析搜索页面失败: %w", p.Name(), err)
|
||||
}
|
||||
|
||||
var apiResponse ErxiaoAPIResponse
|
||||
if err := json.Unmarshal(body, &apiResponse); err != nil {
|
||||
return nil, fmt.Errorf("[%s] 解析JSON响应失败: %w", p.Name(), err)
|
||||
}
|
||||
|
||||
// 检查API响应状态
|
||||
if apiResponse.Code != 1 {
|
||||
return nil, fmt.Errorf("[%s] API返回错误: %s", p.Name(), apiResponse.Msg)
|
||||
}
|
||||
|
||||
// 解析搜索结果
|
||||
|
||||
// 7. 提取搜索结果
|
||||
var results []model.SearchResult
|
||||
for _, item := range apiResponse.List {
|
||||
if result := p.parseAPIItem(item); result.Title != "" {
|
||||
|
||||
doc.Find(".module-search-item").Each(func(i int, s *goquery.Selection) {
|
||||
result := p.parseSearchItem(s, keyword)
|
||||
if result.UniqueID != "" {
|
||||
results = append(results, result)
|
||||
}
|
||||
})
|
||||
|
||||
// 8. 异步获取详情页信息
|
||||
enhancedResults := p.enhanceWithDetails(client, results)
|
||||
|
||||
// 9. 关键词过滤
|
||||
return plugin.FilterResultsByKeyword(enhancedResults, keyword), nil
|
||||
}
|
||||
|
||||
// parseSearchItem 解析单个搜索结果项
|
||||
func (p *ErxiaoAsyncPlugin) parseSearchItem(s *goquery.Selection, keyword string) model.SearchResult {
|
||||
result := model.SearchResult{}
|
||||
|
||||
// 提取详情页链接和ID
|
||||
detailLink, exists := s.Find(".video-info-header h3 a").First().Attr("href")
|
||||
if !exists {
|
||||
return result
|
||||
}
|
||||
|
||||
return results, nil
|
||||
}
|
||||
|
||||
type ErxiaoAPIResponse struct {
|
||||
Code int `json:"code"`
|
||||
Msg string `json:"msg"`
|
||||
Page int `json:"page"`
|
||||
PageCount int `json:"pagecount"`
|
||||
Limit int `json:"limit"`
|
||||
Total int `json:"total"`
|
||||
List []ErxiaoAPIItem `json:"list"`
|
||||
}
|
||||
// 提取ID
|
||||
matches := detailIDRegex.FindStringSubmatch(detailLink)
|
||||
if len(matches) < 2 {
|
||||
return result
|
||||
}
|
||||
itemID := matches[1]
|
||||
|
||||
type ErxiaoAPIItem struct {
|
||||
VodID int `json:"vod_id"`
|
||||
VodName string `json:"vod_name"`
|
||||
VodActor string `json:"vod_actor"`
|
||||
VodDirector string `json:"vod_director"`
|
||||
VodDownFrom string `json:"vod_down_from"`
|
||||
VodDownURL string `json:"vod_down_url"`
|
||||
VodRemarks string `json:"vod_remarks"`
|
||||
VodPubdate string `json:"vod_pubdate"`
|
||||
VodArea string `json:"vod_area"`
|
||||
VodYear string `json:"vod_year"`
|
||||
VodContent string `json:"vod_content"`
|
||||
VodPic string `json:"vod_pic"`
|
||||
}
|
||||
|
||||
// parseAPIItem 解析API数据项
|
||||
func (p *ErxiaoAsyncPlugin) parseAPIItem(item ErxiaoAPIItem) model.SearchResult {
|
||||
// 构建唯一ID
|
||||
uniqueID := fmt.Sprintf("%s-%d", p.Name(), item.VodID)
|
||||
|
||||
// 构建标题
|
||||
title := strings.TrimSpace(item.VodName)
|
||||
uniqueID := fmt.Sprintf("%s-%s", p.Name(), itemID)
|
||||
|
||||
// 提取标题
|
||||
title := strings.TrimSpace(s.Find(".video-info-header h3 a").First().Text())
|
||||
if title == "" {
|
||||
return model.SearchResult{}
|
||||
return result
|
||||
}
|
||||
|
||||
// 构建描述
|
||||
|
||||
// 提取分类
|
||||
category := strings.TrimSpace(s.Find(".video-info-items").First().Find(".video-info-item").First().Text())
|
||||
|
||||
// 提取导演
|
||||
directorElement := s.Find(".video-info-items").FilterFunction(func(i int, item *goquery.Selection) bool {
|
||||
title := strings.TrimSpace(item.Find(".video-info-itemtitle").Text())
|
||||
return strings.Contains(title, "导演")
|
||||
})
|
||||
director := strings.TrimSpace(directorElement.Find(".video-info-item").Text())
|
||||
|
||||
// 提取主演
|
||||
actorElement := s.Find(".video-info-items").FilterFunction(func(i int, item *goquery.Selection) bool {
|
||||
title := strings.TrimSpace(item.Find(".video-info-itemtitle").Text())
|
||||
return strings.Contains(title, "主演")
|
||||
})
|
||||
actor := strings.TrimSpace(actorElement.Find(".video-info-item").Text())
|
||||
|
||||
// 提取年份
|
||||
year := strings.TrimSpace(s.Find(".video-info-items").Last().Find(".video-info-item").First().Text())
|
||||
|
||||
// 提取质量/状态
|
||||
quality := strings.TrimSpace(s.Find(".video-info-header .video-info-remarks").Text())
|
||||
|
||||
// 提取剧情简介
|
||||
plotElement := s.Find(".video-info-items").FilterFunction(func(i int, item *goquery.Selection) bool {
|
||||
title := strings.TrimSpace(item.Find(".video-info-itemtitle").Text())
|
||||
return strings.Contains(title, "剧情")
|
||||
})
|
||||
plot := strings.TrimSpace(plotElement.Find(".video-info-item").Text())
|
||||
|
||||
// 提取封面图片
|
||||
var images []string
|
||||
if picURL, exists := s.Find(".module-item-pic > img").Attr("data-src"); exists && picURL != "" {
|
||||
images = append(images, picURL)
|
||||
}
|
||||
result.Images = images
|
||||
|
||||
// 构建内容描述
|
||||
var contentParts []string
|
||||
if item.VodActor != "" {
|
||||
contentParts = append(contentParts, fmt.Sprintf("主演: %s", item.VodActor))
|
||||
if quality != "" {
|
||||
contentParts = append(contentParts, "【"+quality+"】")
|
||||
}
|
||||
if item.VodDirector != "" {
|
||||
contentParts = append(contentParts, fmt.Sprintf("导演: %s", item.VodDirector))
|
||||
if director != "" {
|
||||
contentParts = append(contentParts, "导演："+director)
|
||||
}
|
||||
if item.VodArea != "" {
|
||||
contentParts = append(contentParts, fmt.Sprintf("地区: %s", item.VodArea))
|
||||
if actor != "" {
|
||||
contentParts = append(contentParts, "主演："+actor)
|
||||
}
|
||||
if item.VodYear != "" {
|
||||
contentParts = append(contentParts, fmt.Sprintf("年份: %s", item.VodYear))
|
||||
if year != "" {
|
||||
contentParts = append(contentParts, "年份："+year)
|
||||
}
|
||||
if item.VodRemarks != "" {
|
||||
contentParts = append(contentParts, fmt.Sprintf("状态: %s", item.VodRemarks))
|
||||
if plot != "" {
|
||||
contentParts = append(contentParts, "剧情："+plot)
|
||||
}
|
||||
content := strings.Join(contentParts, " | ")
|
||||
|
||||
// 解析下载链接
|
||||
links := p.parseDownloadLinks(item.VodDownFrom, item.VodDownURL)
|
||||
|
||||
content := strings.Join(contentParts, "\n")
|
||||
|
||||
// 构建标签
|
||||
var tags []string
|
||||
if item.VodYear != "" {
|
||||
tags = append(tags, item.VodYear)
|
||||
if year != "" {
|
||||
tags = append(tags, year)
|
||||
}
|
||||
if item.VodArea != "" {
|
||||
tags = append(tags, item.VodArea)
|
||||
}
|
||||
|
||||
return model.SearchResult{
|
||||
UniqueID: uniqueID,
|
||||
Title: title,
|
||||
Content: content,
|
||||
Links: links,
|
||||
Tags: tags,
|
||||
Channel: "", // 插件搜索结果Channel为空
|
||||
Datetime: time.Time{}, // 使用零值而不是nil，参考jikepan插件标准
|
||||
if category != "" {
|
||||
tags = append(tags, category)
|
||||
}
|
||||
|
||||
result.UniqueID = uniqueID
|
||||
result.Title = title
|
||||
result.Content = content
|
||||
result.Tags = tags
|
||||
result.Channel = "" // 插件搜索结果Channel为空
|
||||
result.Datetime = time.Time{} // 使用零值
|
||||
|
||||
return result
|
||||
}
|
||||
|
||||
// parseDownloadLinks 解析下载链接
|
||||
func (p *ErxiaoAsyncPlugin) parseDownloadLinks(vodDownFrom, vodDownURL string) []model.Link {
|
||||
if vodDownFrom == "" || vodDownURL == "" {
|
||||
return nil
|
||||
// enhanceWithDetails 异步获取详情页信息
|
||||
func (p *ErxiaoAsyncPlugin) enhanceWithDetails(client *http.Client, results []model.SearchResult) []model.SearchResult {
|
||||
var enhancedResults []model.SearchResult
|
||||
var wg sync.WaitGroup
|
||||
var mu sync.Mutex
|
||||
|
||||
// 创建信号量限制并发数
|
||||
semaphore := make(chan struct{}, MaxConcurrency)
|
||||
|
||||
for _, result := range results {
|
||||
wg.Add(1)
|
||||
go func(result model.SearchResult) {
|
||||
defer wg.Done()
|
||||
semaphore <- struct{}{} // 获取信号量
|
||||
defer func() { <-semaphore }() // 释放信号量
|
||||
|
||||
// 从UniqueID中提取itemID
|
||||
parts := strings.Split(result.UniqueID, "-")
|
||||
if len(parts) < 2 {
|
||||
mu.Lock()
|
||||
enhancedResults = append(enhancedResults, result)
|
||||
mu.Unlock()
|
||||
return
|
||||
}
|
||||
itemID := parts[1]
|
||||
|
||||
// 检查缓存
|
||||
if cached, ok := detailCache.Load(itemID); ok {
|
||||
atomic.AddInt64(&cacheHits, 1)
|
||||
r := cached.(model.SearchResult)
|
||||
mu.Lock()
|
||||
enhancedResults = append(enhancedResults, r)
|
||||
mu.Unlock()
|
||||
return
|
||||
}
|
||||
|
||||
atomic.AddInt64(&cacheMisses, 1)
|
||||
|
||||
// 获取详情页链接和图片
|
||||
detailLinks, detailImages := p.fetchDetailLinksAndImages(client, itemID)
|
||||
result.Links = detailLinks
|
||||
|
||||
// 合并图片：优先使用详情页的海报，如果没有则使用搜索结果的图片
|
||||
if len(detailImages) > 0 {
|
||||
result.Images = detailImages
|
||||
}
|
||||
|
||||
// 缓存结果
|
||||
detailCache.Store(itemID, result)
|
||||
|
||||
mu.Lock()
|
||||
enhancedResults = append(enhancedResults, result)
|
||||
mu.Unlock()
|
||||
}(result)
|
||||
}
|
||||
|
||||
// 按$$$分隔
|
||||
fromParts := strings.Split(vodDownFrom, "$$$")
|
||||
urlParts := strings.Split(vodDownURL, "$$$")
|
||||
|
||||
// 确保数组长度一致
|
||||
minLen := len(fromParts)
|
||||
if len(urlParts) < minLen {
|
||||
minLen = len(urlParts)
|
||||
|
||||
wg.Wait()
|
||||
return enhancedResults
|
||||
}
|
||||
|
||||
// fetchDetailLinksAndImages 获取详情页的下载链接和图片
|
||||
func (p *ErxiaoAsyncPlugin) fetchDetailLinksAndImages(client *http.Client, itemID string) ([]model.Link, []string) {
|
||||
// 性能统计
|
||||
start := time.Now()
|
||||
atomic.AddInt64(&detailPageRequests, 1)
|
||||
defer func() {
|
||||
duration := time.Since(start).Nanoseconds()
|
||||
atomic.AddInt64(&totalDetailTime, duration)
|
||||
}()
|
||||
|
||||
detailURL := fmt.Sprintf("https://erxiaofn.click/index.php/vod/detail/id/%s.html", itemID)
|
||||
|
||||
// 创建带超时的上下文
|
||||
ctx, cancel := context.WithTimeout(context.Background(), DetailTimeout)
|
||||
defer cancel()
|
||||
|
||||
// 创建请求
|
||||
req, err := http.NewRequestWithContext(ctx, "GET", detailURL, nil)
|
||||
if err != nil {
|
||||
return nil, nil
|
||||
}
|
||||
|
||||
|
||||
// 设置请求头
|
||||
req.Header.Set("User-Agent", "Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/91.0.4472.124 Safari/537.36")
|
||||
req.Header.Set("Accept", "text/html,application/xhtml+xml,application/xml;q=0.9,image/webp,*/*;q=0.8")
|
||||
req.Header.Set("Accept-Language", "zh-CN,zh;q=0.9,en;q=0.8")
|
||||
req.Header.Set("Connection", "keep-alive")
|
||||
req.Header.Set("Referer", "https://erxiaofn.click/")
|
||||
|
||||
// 发送请求（带重试）
|
||||
resp, err := p.doRequestWithRetry(req, client)
|
||||
if err != nil {
|
||||
return nil, nil
|
||||
}
|
||||
defer resp.Body.Close()
|
||||
|
||||
if resp.StatusCode != 200 {
|
||||
return nil, nil
|
||||
}
|
||||
|
||||
doc, err := goquery.NewDocumentFromReader(resp.Body)
|
||||
if err != nil {
|
||||
return nil, nil
|
||||
}
|
||||
|
||||
var links []model.Link
|
||||
for i := 0; i < minLen; i++ {
|
||||
fromType := strings.TrimSpace(fromParts[i])
|
||||
urlStr := strings.TrimSpace(urlParts[i])
|
||||
|
||||
if urlStr == "" {
|
||||
continue
|
||||
}
|
||||
|
||||
// 直接确定链接类型（合并验证和类型判断，避免重复正则匹配）
|
||||
linkType := p.determineLinkTypeOptimized(fromType, urlStr)
|
||||
if linkType == "" {
|
||||
continue
|
||||
}
|
||||
|
||||
// 提取密码
|
||||
password := p.extractPassword(urlStr)
|
||||
|
||||
links = append(links, model.Link{
|
||||
Type: linkType,
|
||||
URL: urlStr,
|
||||
Password: password,
|
||||
})
|
||||
var images []string
|
||||
|
||||
// 提取详情页的海报图片
|
||||
if posterURL, exists := doc.Find(".mobile-play .lazyload").Attr("data-src"); exists && posterURL != "" {
|
||||
images = append(images, posterURL)
|
||||
}
|
||||
|
||||
return links
|
||||
|
||||
// 查找下载链接区域
|
||||
doc.Find("#download-list .module-row-one").Each(func(i int, s *goquery.Selection) {
|
||||
// 从data-clipboard-text属性提取链接
|
||||
if linkURL, exists := s.Find("[data-clipboard-text]").Attr("data-clipboard-text"); exists {
|
||||
// 过滤掉无效链接
|
||||
if p.isValidNetworkDriveURL(linkURL) {
|
||||
if linkType := p.determineLinkType(linkURL); linkType != "" {
|
||||
link := model.Link{
|
||||
Type: linkType,
|
||||
URL: linkURL,
|
||||
Password: "", // 大部分网盘不需要密码
|
||||
}
|
||||
links = append(links, link)
|
||||
}
|
||||
}
|
||||
}
|
||||
})
|
||||
|
||||
return links, images
|
||||
}
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
// determineLinkTypeOptimized 优化的链接类型判断（避免重复正则匹配）
|
||||
func (p *ErxiaoAsyncPlugin) determineLinkTypeOptimized(apiType, url string) string {
|
||||
// 基本验证（包含原 isValidNetworkDriveURL 的逻辑）
|
||||
if strings.Contains(url, "javascript:") ||
|
||||
// isValidNetworkDriveURL 验证是否为有效的网盘URL
|
||||
func (p *ErxiaoAsyncPlugin) isValidNetworkDriveURL(url string) bool {
|
||||
if strings.Contains(url, "javascript:") ||
|
||||
strings.Contains(url, "#") ||
|
||||
url == "" ||
|
||||
(!strings.HasPrefix(url, "http") && !strings.HasPrefix(url, "magnet:") && !strings.HasPrefix(url, "ed2k:")) {
|
||||
return ""
|
||||
}
|
||||
|
||||
// 优先根据API标识快速映射（避免正则匹配）
|
||||
switch strings.ToUpper(apiType) {
|
||||
case "BD":
|
||||
if baiduLinkRegex.MatchString(url) {
|
||||
return "baidu"
|
||||
}
|
||||
case "KG":
|
||||
if quarkLinkRegex.MatchString(url) {
|
||||
return "quark"
|
||||
}
|
||||
case "UC":
|
||||
if ucLinkRegex.MatchString(url) {
|
||||
return "uc"
|
||||
}
|
||||
case "ALY":
|
||||
if aliyunLinkRegex.MatchString(url) {
|
||||
return "aliyun"
|
||||
}
|
||||
case "XL":
|
||||
if xunleiLinkRegex.MatchString(url) {
|
||||
return "xunlei"
|
||||
}
|
||||
case "TY":
|
||||
if tianyiLinkRegex.MatchString(url) {
|
||||
return "tianyi"
|
||||
}
|
||||
case "115":
|
||||
if link115Regex.MatchString(url) {
|
||||
return "115"
|
||||
}
|
||||
case "MB":
|
||||
if mobileLinkRegex.MatchString(url) {
|
||||
return "mobile"
|
||||
}
|
||||
case "123":
|
||||
if link123Regex.MatchString(url) {
|
||||
return "123"
|
||||
}
|
||||
case "PIKPAK":
|
||||
if pikpakLinkRegex.MatchString(url) {
|
||||
return "pikpak"
|
||||
}
|
||||
}
|
||||
|
||||
// 如果API标识匹配失败，回退到URL正则匹配（一次性匹配）
|
||||
switch {
|
||||
case baiduLinkRegex.MatchString(url):
|
||||
return "baidu"
|
||||
case ucLinkRegex.MatchString(url):
|
||||
return "uc"
|
||||
case aliyunLinkRegex.MatchString(url):
|
||||
return "aliyun"
|
||||
case xunleiLinkRegex.MatchString(url):
|
||||
return "xunlei"
|
||||
case tianyiLinkRegex.MatchString(url):
|
||||
return "tianyi"
|
||||
case link115Regex.MatchString(url):
|
||||
return "115"
|
||||
case mobileLinkRegex.MatchString(url):
|
||||
return "mobile"
|
||||
case link123Regex.MatchString(url):
|
||||
return "123"
|
||||
case pikpakLinkRegex.MatchString(url):
|
||||
return "pikpak"
|
||||
case magnetLinkRegex.MatchString(url):
|
||||
return "magnet"
|
||||
case ed2kLinkRegex.MatchString(url):
|
||||
return "ed2k"
|
||||
case quarkLinkRegex.MatchString(url):
|
||||
return "quark"
|
||||
default:
|
||||
return "" // 不支持的类型
|
||||
return false
|
||||
}
|
||||
return true
|
||||
}
|
||||
|
||||
|
||||
// determineLinkType 根据URL确定链接类型
|
||||
func (p *ErxiaoAsyncPlugin) determineLinkType(url string) string {
|
||||
switch {
|
||||
@@ -421,11 +477,11 @@ func (p *ErxiaoAsyncPlugin) extractPassword(url string) string {
|
||||
return ""
|
||||
}
|
||||
|
||||
// doRequestWithRetry 带重试的HTTP请求（优化JSON API的重试策略）
|
||||
// doRequestWithRetry 带重试的HTTP请求
|
||||
func (p *ErxiaoAsyncPlugin) doRequestWithRetry(req *http.Request, client *http.Client) (*http.Response, error) {
|
||||
maxRetries := 2 // 对于JSON API减少重试次数
|
||||
maxRetries := 2
|
||||
var lastErr error
|
||||
|
||||
|
||||
for i := 0; i < maxRetries; i++ {
|
||||
resp, err := client.Do(req)
|
||||
if err == nil {
|
||||
@@ -437,13 +493,13 @@ func (p *ErxiaoAsyncPlugin) doRequestWithRetry(req *http.Request, client *http.C
|
||||
} else {
|
||||
lastErr = err
|
||||
}
|
||||
|
||||
// JSON API快速重试：只等待很短时间
|
||||
|
||||
// 快速重试：只等待很短时间
|
||||
if i < maxRetries-1 {
|
||||
time.Sleep(100 * time.Millisecond) // 从秒级改为100毫秒
|
||||
time.Sleep(100 * time.Millisecond)
|
||||
}
|
||||
}
|
||||
|
||||
|
||||
return nil, fmt.Errorf("[%s] 请求失败，重试%d次后仍失败: %w", p.Name(), maxRetries, lastErr)
|
||||
}
|
||||
|
||||
@@ -451,15 +507,29 @@ func (p *ErxiaoAsyncPlugin) doRequestWithRetry(req *http.Request, client *http.C
|
||||
func (p *ErxiaoAsyncPlugin) GetPerformanceStats() map[string]interface{} {
|
||||
totalRequests := atomic.LoadInt64(&searchRequests)
|
||||
totalTime := atomic.LoadInt64(&totalSearchTime)
|
||||
|
||||
detailRequests := atomic.LoadInt64(&detailPageRequests)
|
||||
detailTime := atomic.LoadInt64(&totalDetailTime)
|
||||
hits := atomic.LoadInt64(&cacheHits)
|
||||
misses := atomic.LoadInt64(&cacheMisses)
|
||||
|
||||
var avgTime float64
|
||||
if totalRequests > 0 {
|
||||
avgTime = float64(totalTime) / float64(totalRequests) / 1e6 // 转换为毫秒
|
||||
}
|
||||
|
||||
|
||||
var avgDetailTime float64
|
||||
if detailRequests > 0 {
|
||||
avgDetailTime = float64(detailTime) / float64(detailRequests) / 1e6 // 转换为毫秒
|
||||
}
|
||||
|
||||
return map[string]interface{}{
|
||||
"search_requests": totalRequests,
|
||||
"avg_search_time_ms": avgTime,
|
||||
"search_requests": totalRequests,
|
||||
"avg_search_time_ms": avgTime,
|
||||
"total_search_time_ns": totalTime,
|
||||
"detail_page_requests": detailRequests,
|
||||
"avg_detail_time_ms": avgDetailTime,
|
||||
"total_detail_time_ns": detailTime,
|
||||
"cache_hits": hits,
|
||||
"cache_misses": misses,
|
||||
}
|
||||
}
|
||||
138
plugin/erxiao/html结构分析.md
Normal file
138
plugin/erxiao/html结构分析.md
Normal file
@@ -0,0 +1,138 @@
|
||||
# Erxiao HTML 数据结构分析
|
||||
|
||||
## 基本信息
|
||||
- **数据源类型**: HTML 网页
|
||||
- **搜索URL格式**: `https://erxiaofn.click/index.php/vod/search/wd/{关键词}.html`
|
||||
- **详情URL格式**: `https://erxiaofn.click/index.php/vod/detail/id/{资源ID}.html`
|
||||
- **数据特点**: 视频点播(VOD)系统网页，提供HTML格式的影视资源数据
|
||||
- **特殊说明**: 使用HTML解析替代JSON API，与zhizhen/muou插件使用相同的HTML结构
|
||||
|
||||
## HTML 页面结构
|
||||
|
||||
### 搜索结果页面 (`.module-search-item`)
|
||||
搜索结果页面包含多个搜索项，每个搜索项的HTML结构如下：
|
||||
|
||||
```html
|
||||
<div class="module-search-item">
|
||||
<div class="module-item-pic">
|
||||
<img data-src="https://..." />
|
||||
</div>
|
||||
<div class="module-item-text">
|
||||
<div class="video-info-header">
|
||||
<h3><a href="/index.php/vod/detail/id/12345.html">电影标题</a></h3>
|
||||
<span class="video-info-remarks">HD</span>
|
||||
</div>
|
||||
<div class="video-info-items">
|
||||
<div class="video-info-item">
|
||||
<span class="video-info-itemtitle">分类：</span>
|
||||
<span class="video-info-item">动作</span>
|
||||
</div>
|
||||
<div class="video-info-item">
|
||||
<span class="video-info-itemtitle">导演：</span>
|
||||
<span class="video-info-item">导演名字</span>
|
||||
</div>
|
||||
<div class="video-info-item">
|
||||
<span class="video-info-itemtitle">主演：</span>
|
||||
<span class="video-info-item">演员1,演员2</span>
|
||||
</div>
|
||||
<div class="video-info-item">
|
||||
<span class="video-info-itemtitle">年份：</span>
|
||||
<span class="video-info-item">2024</span>
|
||||
</div>
|
||||
<div class="video-info-item">
|
||||
<span class="video-info-itemtitle">剧情：</span>
|
||||
<span class="video-info-item">这是一部精彩的电影...</span>
|
||||
</div>
|
||||
</div>
|
||||
</div>
|
||||
</div>
|
||||
```
|
||||
|
||||
### 详情页面 (`.mobile-play` 和 `#download-list`)
|
||||
详情页面包含海报图片和下载链接：
|
||||
|
||||
```html
|
||||
<div class="mobile-play">
|
||||
<img class="lazyload" data-src="https://poster-url.jpg" />
|
||||
</div>
|
||||
|
||||
<div id="download-list">
|
||||
<div class="module-row-one">
|
||||
<div class="module-row-text">
|
||||
<span data-clipboard-text="https://pan.quark.cn/s/xxxxx">夸克网盘</span>
|
||||
</div>
|
||||
</div>
|
||||
<div class="module-row-one">
|
||||
<div class="module-row-text">
|
||||
<span data-clipboard-text="https://pan.baidu.com/s/xxxxx?pwd=xxxx">百度网盘</span>
|
||||
</div>
|
||||
</div>
|
||||
</div>
|
||||
```
|
||||
|
||||
## CSS 选择器参考
|
||||
|
||||
### 搜索结果提取
|
||||
- **搜索结果容器**: `.module-search-item`
|
||||
- **标题**: `.video-info-header h3 a` (文本内容)
|
||||
- **详情页链接**: `.video-info-header h3 a` (href属性)
|
||||
- **封面图片**: `.module-item-pic > img` (data-src属性)
|
||||
- **质量/状态**: `.video-info-header .video-info-remarks` (文本内容)
|
||||
|
||||
### 详情页下载链接提取
|
||||
- **海报图片**: `.mobile-play .lazyload` (data-src属性)
|
||||
- **下载链接容器**: `#download-list .module-row-one`
|
||||
- **下载链接**: `[data-clipboard-text]` (data-clipboard-text属性)
|
||||
|
||||
## 支持的网盘类型
|
||||
- **Quark网盘**: `https://pan.quark.cn/s/{分享码}`
|
||||
- **百度网盘**: `https://pan.baidu.com/s/{分享码}?pwd={密码}`
|
||||
- **阿里云盘**: `https://www.aliyundrive.com/s/{分享码}`
|
||||
- **迅雷网盘**: `https://pan.xunlei.com/s/{分享码}`
|
||||
- **天翼云盘**: `https://cloud.189.cn/t/{分享码}`
|
||||
- **UC网盘**: `https://drive.uc.cn/s/{分享码}`
|
||||
- **115网盘**: `https://115.com/s/{分享码}`
|
||||
- **123网盘**: `https://123pan.com/s/{分享码}`
|
||||
- **PikPak**: `https://mypikpak.com/s/{分享码}`
|
||||
- **移动云盘**: `https://caiyun.feixin.10086.cn/{分享码}`
|
||||
- **磁力链接**: `magnet:?xt=urn:btih:{hash}`
|
||||
- **ED2K链接**: `ed2k://|file|...`
|
||||
|
||||
## 数据流程
|
||||
|
||||
### 搜索流程
|
||||
1. **构建搜索URL**: `https://erxiaofn.click/index.php/vod/search/wd/{keyword}.html`
|
||||
2. **发送HTTP请求**: 获取搜索结果页面
|
||||
3. **解析HTML**: 使用goquery解析页面
|
||||
4. **提取搜索项**: 遍历`.module-search-item`元素
|
||||
5. **异步获取详情**: 并发请求详情页面获取下载链接
|
||||
6. **缓存管理**: 使用sync.Map缓存详情页结果，TTL为1小时
|
||||
7. **关键词过滤**: 过滤不相关的结果
|
||||
|
||||
### 详情页请求示例
|
||||
```go
|
||||
detailURL := fmt.Sprintf("https://erxiaofn.click/index.php/vod/detail/id/%s.html", itemID)
|
||||
```
|
||||
|
||||
## 并发控制
|
||||
- **最大并发数**: 20 (MaxConcurrency)
|
||||
- **搜索超时**: 8秒 (DefaultTimeout)
|
||||
- **详情页超时**: 6秒 (DetailTimeout)
|
||||
- **缓存TTL**: 1小时 (cacheTTL)
|
||||
|
||||
## 性能统计
|
||||
- **搜索请求数**: 总搜索请求数
|
||||
- **平均搜索时间**: 单次搜索平均耗时(毫秒)
|
||||
- **详情页请求数**: 总详情页请求数
|
||||
- **平均详情页时间**: 单次详情页请求平均耗时(毫秒)
|
||||
- **缓存命中数**: 详情页缓存命中次数
|
||||
- **缓存未命中数**: 详情页缓存未命中次数
|
||||
|
||||
## 注意事项
|
||||
1. **HTML解析**: 使用goquery库进行HTML解析
|
||||
2. **异步获取详情**: 搜索结果只包含基本信息，需要异步请求详情页获取下载链接
|
||||
3. **并发控制**: 使用信号量限制并发数为20
|
||||
4. **缓存管理**: 使用sync.Map缓存详情页结果，避免重复请求
|
||||
5. **链接验证**: 过滤掉无效链接（如包含`javascript:`、`#`等）
|
||||
6. **密码提取**: 从URL中提取`?pwd=`参数作为密码
|
||||
|
||||
133
plugin/huban/html结构分析.md
Normal file
133
plugin/huban/html结构分析.md
Normal file
@@ -0,0 +1,133 @@
|
||||
# Huban HTML 数据结构分析
|
||||
|
||||
## 基本信息
|
||||
- **数据源类型**: HTML 网页
|
||||
- **搜索URL格式**: `http://xsayang.fun:12512/index.php/vod/search/wd/{关键词}.html`
|
||||
- **详情URL格式**: `http://xsayang.fun:12512/index.php/vod/detail/id/{资源ID}.html`
|
||||
- **数据特点**: 视频点播(VOD)系统网页，提供HTML格式的影视资源数据
|
||||
- **特殊说明**: 使用HTML解析替代JSON API，与erxiao/zhizhen/muou插件使用相同的HTML结构
|
||||
|
||||
## HTML 页面结构
|
||||
|
||||
### 搜索结果页面 (`.module-search-item`)
|
||||
搜索结果页面包含多个搜索项，每个搜索项的HTML结构如下：
|
||||
|
||||
```html
|
||||
<div class="module-search-item">
|
||||
<div class="module-item-pic">
|
||||
<img data-src="https://..." />
|
||||
</div>
|
||||
<div class="module-item-text">
|
||||
<div class="video-info-header">
|
||||
<h3><a href="/index.php/vod/detail/id/12345.html">电影标题</a></h3>
|
||||
<span class="video-info-remarks">HD</span>
|
||||
</div>
|
||||
<div class="video-info-items">
|
||||
<div class="video-info-item">
|
||||
<span class="video-info-itemtitle">分类：</span>
|
||||
<span class="video-info-item">动作</span>
|
||||
</div>
|
||||
<div class="video-info-item">
|
||||
<span class="video-info-itemtitle">导演：</span>
|
||||
<span class="video-info-item">导演名字</span>
|
||||
</div>
|
||||
<div class="video-info-item">
|
||||
<span class="video-info-itemtitle">主演：</span>
|
||||
<span class="video-info-item">演员1,演员2</span>
|
||||
</div>
|
||||
<div class="video-info-item">
|
||||
<span class="video-info-itemtitle">年份：</span>
|
||||
<span class="video-info-item">2024</span>
|
||||
</div>
|
||||
<div class="video-info-item">
|
||||
<span class="video-info-itemtitle">剧情：</span>
|
||||
<span class="video-info-item">这是一部精彩的电影...</span>
|
||||
</div>
|
||||
</div>
|
||||
</div>
|
||||
</div>
|
||||
```
|
||||
|
||||
### 详情页面 (`.mobile-play` 和 `#download-list`)
|
||||
详情页面包含海报图片和下载链接：
|
||||
|
||||
```html
|
||||
<div class="mobile-play">
|
||||
<img class="lazyload" data-src="https://poster-url.jpg" />
|
||||
</div>
|
||||
|
||||
<div id="download-list">
|
||||
<div class="module-row-one">
|
||||
<div class="module-row-text">
|
||||
<span data-clipboard-text="https://pan.quark.cn/s/xxxxx">夸克网盘</span>
|
||||
</div>
|
||||
</div>
|
||||
<div class="module-row-one">
|
||||
<div class="module-row-text">
|
||||
<span data-clipboard-text="https://pan.baidu.com/s/xxxxx?pwd=xxxx">百度网盘</span>
|
||||
</div>
|
||||
</div>
|
||||
</div>
|
||||
```
|
||||
|
||||
## CSS 选择器参考
|
||||
|
||||
### 搜索结果提取
|
||||
- **搜索结果容器**: `.module-search-item`
|
||||
- **标题**: `.video-info-header h3 a` (文本内容)
|
||||
- **详情页链接**: `.video-info-header h3 a` (href属性)
|
||||
- **封面图片**: `.module-item-pic > img` (data-src属性)
|
||||
- **质量/状态**: `.video-info-header .video-info-remarks` (文本内容)
|
||||
|
||||
### 详情页下载链接提取
|
||||
- **海报图片**: `.mobile-play .lazyload` (data-src属性)
|
||||
- **下载链接容器**: `#download-list .module-row-one`
|
||||
- **下载链接**: `[data-clipboard-text]` (data-clipboard-text属性)
|
||||
|
||||
## 支持的网盘类型
|
||||
- **Quark网盘**: `https://pan.quark.cn/s/{分享码}`
|
||||
- **百度网盘**: `https://pan.baidu.com/s/{分享码}?pwd={密码}`
|
||||
- **阿里云盘**: `https://www.aliyundrive.com/s/{分享码}`
|
||||
- **迅雷网盘**: `https://pan.xunlei.com/s/{分享码}`
|
||||
- **天翼云盘**: `https://cloud.189.cn/t/{分享码}`
|
||||
- **UC网盘**: `https://drive.uc.cn/s/{分享码}`
|
||||
- **115网盘**: `https://115.com/s/{分享码}`
|
||||
- **123网盘**: `https://123pan.com/s/{分享码}`
|
||||
- **PikPak**: `https://mypikpak.com/s/{分享码}`
|
||||
- **移动云盘**: `https://caiyun.feixin.10086.cn/{分享码}`
|
||||
- **磁力链接**: `magnet:?xt=urn:btih:{hash}`
|
||||
- **ED2K链接**: `ed2k://|file|...`
|
||||
|
||||
## 数据流程
|
||||
|
||||
### 搜索流程
|
||||
1. **构建搜索URL**: `http://xsayang.fun:12512/index.php/vod/search/wd/{keyword}.html`
|
||||
2. **发送HTTP请求**: 获取搜索结果页面
|
||||
3. **解析HTML**: 使用goquery解析页面
|
||||
4. **提取搜索项**: 遍历`.module-search-item`元素
|
||||
5. **异步获取详情**: 并发请求详情页面获取下载链接
|
||||
6. **缓存管理**: 使用sync.Map缓存详情页结果，TTL为1小时
|
||||
7. **关键词过滤**: 过滤不相关的结果
|
||||
|
||||
## 并发控制
|
||||
- **最大并发数**: 20 (MaxConcurrency)
|
||||
- **搜索超时**: 8秒 (DefaultTimeout)
|
||||
- **详情页超时**: 6秒 (DetailTimeout)
|
||||
- **缓存TTL**: 1小时 (cacheTTL)
|
||||
|
||||
## 性能统计
|
||||
- **搜索请求数**: 总搜索请求数
|
||||
- **平均搜索时间**: 单次搜索平均耗时(毫秒)
|
||||
- **详情页请求数**: 总详情页请求数
|
||||
- **平均详情页时间**: 单次详情页请求平均耗时(毫秒)
|
||||
- **缓存命中数**: 详情页缓存命中次数
|
||||
- **缓存未命中数**: 详情页缓存未命中次数
|
||||
|
||||
## 注意事项
|
||||
1. **HTML解析**: 使用goquery库进行HTML解析
|
||||
2. **异步获取详情**: 搜索结果只包含基本信息，需要异步请求详情页获取下载链接
|
||||
3. **并发控制**: 使用信号量限制并发数为20
|
||||
4. **缓存管理**: 使用sync.Map缓存详情页结果，避免重复请求
|
||||
5. **链接验证**: 过滤掉无效链接（如包含`javascript:`、`#`等）
|
||||
6. **密码提取**: 从URL中提取`?pwd=`参数作为密码
|
||||
|
||||
@@ -2,41 +2,58 @@ package huban
|
||||
|
||||
import (
|
||||
"fmt"
|
||||
"io"
|
||||
"net/http"
|
||||
"net/url"
|
||||
"regexp"
|
||||
"strings"
|
||||
"time"
|
||||
"context"
|
||||
"sync"
|
||||
"sync/atomic"
|
||||
|
||||
"github.com/PuerkitoBio/goquery"
|
||||
"pansou/model"
|
||||
"pansou/plugin"
|
||||
"pansou/util/json"
|
||||
)
|
||||
|
||||
const (
|
||||
// 默认超时时间 - 优化为更短时间
|
||||
// 默认超时时间
|
||||
DefaultTimeout = 8 * time.Second
|
||||
DetailTimeout = 6 * time.Second
|
||||
|
||||
// HTTP连接池配置
|
||||
MaxIdleConns = 200
|
||||
MaxIdleConnsPerHost = 50
|
||||
MaxConnsPerHost = 100
|
||||
IdleConnTimeout = 90 * time.Second
|
||||
|
||||
|
||||
// 并发控制
|
||||
MaxConcurrency = 20
|
||||
|
||||
// 缓存TTL
|
||||
cacheTTL = 1 * time.Hour
|
||||
|
||||
// 请求来源控制 - 默认开启，提高安全性
|
||||
EnableRefererCheck = false
|
||||
|
||||
|
||||
// 调试日志开关
|
||||
DebugLog = false
|
||||
)
|
||||
|
||||
// 性能统计（原子操作）
|
||||
var (
|
||||
searchRequests int64 = 0
|
||||
totalSearchTime int64 = 0 // 纳秒
|
||||
searchRequests int64 = 0
|
||||
totalSearchTime int64 = 0 // 纳秒
|
||||
detailPageRequests int64 = 0
|
||||
totalDetailTime int64 = 0 // 纳秒
|
||||
cacheHits int64 = 0
|
||||
cacheMisses int64 = 0
|
||||
)
|
||||
|
||||
// Detail page缓存
|
||||
var (
|
||||
detailCache sync.Map
|
||||
cacheMutex sync.RWMutex
|
||||
)
|
||||
|
||||
// 请求来源控制配置
|
||||
@@ -59,6 +76,9 @@ var (
|
||||
// 密码提取正则表达式
|
||||
passwordRegex = regexp.MustCompile(`\?pwd=([0-9a-zA-Z]+)`)
|
||||
password115Regex = regexp.MustCompile(`password=([0-9a-zA-Z]+)`)
|
||||
|
||||
// 详情页ID提取正则表达式
|
||||
detailIDRegex = regexp.MustCompile(`/id/(\d+)`)
|
||||
|
||||
// 常见网盘链接的正则表达式（支持16种类型）
|
||||
quarkLinkRegex = regexp.MustCompile(`https?://pan\.quark\.cn/s/[0-9a-zA-Z]+`)
|
||||
@@ -149,7 +169,7 @@ func (p *HubanAsyncPlugin) SearchWithResult(keyword string, ext map[string]inter
|
||||
return p.AsyncSearchWithResult(keyword, p.searchImpl, p.MainCacheKey, ext)
|
||||
}
|
||||
|
||||
// searchImpl 搜索实现（双域名支持）
|
||||
// searchImpl 搜索实现 - HTML解析版本
|
||||
func (p *HubanAsyncPlugin) searchImpl(client *http.Client, keyword string, ext map[string]interface{}) ([]model.SearchResult, error) {
|
||||
// 性能统计
|
||||
start := time.Now()
|
||||
@@ -164,259 +184,296 @@ func (p *HubanAsyncPlugin) searchImpl(client *http.Client, keyword string, ext m
|
||||
client = p.optimizedClient
|
||||
}
|
||||
|
||||
// 定义双域名 - 主备模式
|
||||
urls := []string{
|
||||
fmt.Sprintf("http://xsayang.fun:12512/api.php/provide/vod?ac=detail&wd=%s", url.QueryEscape(keyword)),
|
||||
fmt.Sprintf("http://103.45.162.207:20720/api.php/provide/vod?ac=detail&wd=%s", url.QueryEscape(keyword)),
|
||||
}
|
||||
|
||||
// 主备模式：优先使用第一个域名，失败时切换到第二个
|
||||
for i, searchURL := range urls {
|
||||
if results, err := p.tryRequest(searchURL, client); err == nil {
|
||||
return results, nil
|
||||
} else if i == 0 {
|
||||
// 第一个域名失败，记录日志但继续尝试第二个
|
||||
// fmt.Printf("[%s] 域名1失败，尝试域名2: %v\n", p.Name(), err)
|
||||
}
|
||||
}
|
||||
|
||||
return nil, fmt.Errorf("[%s] 所有域名都请求失败", p.Name())
|
||||
}
|
||||
// 1. 构建搜索URL
|
||||
searchURL := fmt.Sprintf("http://103.45.162.207:20720/index.php/vod/search/wd/%s.html", url.QueryEscape(keyword))
|
||||
|
||||
// tryRequest 尝试单个域名请求
|
||||
func (p *HubanAsyncPlugin) tryRequest(searchURL string, client *http.Client) ([]model.SearchResult, error) {
|
||||
// 创建HTTP请求
|
||||
// 2. 创建带超时的上下文
|
||||
ctx, cancel := context.WithTimeout(context.Background(), DefaultTimeout)
|
||||
defer cancel()
|
||||
|
||||
|
||||
// 3. 创建请求
|
||||
req, err := http.NewRequestWithContext(ctx, "GET", searchURL, nil)
|
||||
if err != nil {
|
||||
return nil, fmt.Errorf("创建搜索请求失败: %w", err)
|
||||
return nil, fmt.Errorf("[%s] 创建请求失败: %w", p.Name(), err)
|
||||
}
|
||||
|
||||
// 设置请求头
|
||||
|
||||
// 4. 设置请求头
|
||||
req.Header.Set("User-Agent", "Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/91.0.4472.124 Safari/537.36")
|
||||
req.Header.Set("Accept", "application/json, text/plain, */*")
|
||||
req.Header.Set("Accept", "text/html,application/xhtml+xml,application/xml;q=0.9,image/webp,*/*;q=0.8")
|
||||
req.Header.Set("Accept-Language", "zh-CN,zh;q=0.9,en;q=0.8")
|
||||
req.Header.Set("Connection", "keep-alive")
|
||||
req.Header.Set("Cache-Control", "no-cache")
|
||||
|
||||
// 发送请求
|
||||
req.Header.Set("Referer", "http://103.45.162.207:20720/")
|
||||
|
||||
// 5. 发送请求
|
||||
resp, err := p.doRequestWithRetry(req, client)
|
||||
if err != nil {
|
||||
return nil, fmt.Errorf("搜索请求失败: %w", err)
|
||||
return nil, fmt.Errorf("[%s] 搜索请求失败: %w", p.Name(), err)
|
||||
}
|
||||
defer resp.Body.Close()
|
||||
|
||||
// 解析JSON响应
|
||||
body, _ := io.ReadAll(resp.Body)
|
||||
|
||||
var apiResponse HubanAPIResponse
|
||||
if err := json.Unmarshal(body, &apiResponse); err != nil {
|
||||
return nil, fmt.Errorf("解析JSON响应失败: %w", err)
|
||||
|
||||
if resp.StatusCode != 200 {
|
||||
return nil, fmt.Errorf("[%s] 搜索请求返回状态码: %d", p.Name(), resp.StatusCode)
|
||||
}
|
||||
|
||||
// 检查API响应状态
|
||||
if apiResponse.Code != 1 {
|
||||
return nil, fmt.Errorf("API返回错误: %s", apiResponse.Msg)
|
||||
|
||||
// 6. 解析搜索结果页面
|
||||
doc, err := goquery.NewDocumentFromReader(resp.Body)
|
||||
if err != nil {
|
||||
return nil, fmt.Errorf("[%s] 解析搜索页面失败: %w", p.Name(), err)
|
||||
}
|
||||
|
||||
// 解析搜索结果
|
||||
|
||||
// 7. 提取搜索结果
|
||||
var results []model.SearchResult
|
||||
for _, item := range apiResponse.List {
|
||||
if result := p.parseAPIItem(item); result.Title != "" {
|
||||
|
||||
doc.Find(".module-search-item").Each(func(i int, s *goquery.Selection) {
|
||||
result := p.parseSearchItem(s, keyword)
|
||||
if result.UniqueID != "" {
|
||||
results = append(results, result)
|
||||
}
|
||||
})
|
||||
|
||||
// 8. 异步获取详情页信息
|
||||
enhancedResults := p.enhanceWithDetails(client, results)
|
||||
|
||||
// 9. 关键词过滤
|
||||
return plugin.FilterResultsByKeyword(enhancedResults, keyword), nil
|
||||
}
|
||||
|
||||
// parseSearchItem 解析单个搜索结果项
|
||||
func (p *HubanAsyncPlugin) parseSearchItem(s *goquery.Selection, keyword string) model.SearchResult {
|
||||
result := model.SearchResult{}
|
||||
|
||||
// 提取详情页链接和ID
|
||||
detailLink, exists := s.Find(".video-info-header h3 a").First().Attr("href")
|
||||
if !exists {
|
||||
return result
|
||||
}
|
||||
|
||||
return results, nil
|
||||
}
|
||||
|
||||
// HubanAPIResponse API响应结构
|
||||
type HubanAPIResponse struct {
|
||||
Code int `json:"code"`
|
||||
Msg string `json:"msg"`
|
||||
Page int `json:"page"`
|
||||
PageCount int `json:"pagecount"`
|
||||
Limit interface{} `json:"limit"` // 可能是字符串或数字
|
||||
Total int `json:"total"`
|
||||
List []HubanAPIItem `json:"list"`
|
||||
}
|
||||
// 提取ID
|
||||
matches := detailIDRegex.FindStringSubmatch(detailLink)
|
||||
if len(matches) < 2 {
|
||||
return result
|
||||
}
|
||||
itemID := matches[1]
|
||||
|
||||
// HubanAPIItem API数据项
|
||||
type HubanAPIItem struct {
|
||||
VodID int `json:"vod_id"`
|
||||
VodName string `json:"vod_name"`
|
||||
VodActor string `json:"vod_actor"`
|
||||
VodDirector string `json:"vod_director"`
|
||||
VodDownFrom string `json:"vod_down_from"`
|
||||
VodDownURL string `json:"vod_down_url"`
|
||||
VodRemarks string `json:"vod_remarks"`
|
||||
VodPubdate string `json:"vod_pubdate"`
|
||||
VodArea string `json:"vod_area"`
|
||||
VodLang string `json:"vod_lang"`
|
||||
VodYear string `json:"vod_year"`
|
||||
VodContent string `json:"vod_content"`
|
||||
VodBlurb string `json:"vod_blurb"`
|
||||
VodPic string `json:"vod_pic"`
|
||||
}
|
||||
|
||||
// parseAPIItem 解析API数据项
|
||||
func (p *HubanAsyncPlugin) parseAPIItem(item HubanAPIItem) model.SearchResult {
|
||||
// 构建唯一ID
|
||||
uniqueID := fmt.Sprintf("%s-%d", p.Name(), item.VodID)
|
||||
|
||||
// 构建标题
|
||||
title := strings.TrimSpace(item.VodName)
|
||||
uniqueID := fmt.Sprintf("%s-%s", p.Name(), itemID)
|
||||
|
||||
// 提取标题
|
||||
title := strings.TrimSpace(s.Find(".video-info-header h3 a").First().Text())
|
||||
if title == "" {
|
||||
return model.SearchResult{}
|
||||
return result
|
||||
}
|
||||
|
||||
// 构建描述（需要清理数据）
|
||||
content := p.buildContent(item)
|
||||
|
||||
// 解析下载链接（huban特殊格式）
|
||||
links := p.parseHubanLinks(item.VodDownFrom, item.VodDownURL)
|
||||
|
||||
|
||||
// 提取分类
|
||||
category := strings.TrimSpace(s.Find(".video-info-items").First().Find(".video-info-item").First().Text())
|
||||
|
||||
// 提取导演
|
||||
directorElement := s.Find(".video-info-items").FilterFunction(func(i int, item *goquery.Selection) bool {
|
||||
title := strings.TrimSpace(item.Find(".video-info-itemtitle").Text())
|
||||
return strings.Contains(title, "导演")
|
||||
})
|
||||
director := strings.TrimSpace(directorElement.Find(".video-info-item").Text())
|
||||
|
||||
// 提取主演
|
||||
actorElement := s.Find(".video-info-items").FilterFunction(func(i int, item *goquery.Selection) bool {
|
||||
title := strings.TrimSpace(item.Find(".video-info-itemtitle").Text())
|
||||
return strings.Contains(title, "主演")
|
||||
})
|
||||
actor := strings.TrimSpace(actorElement.Find(".video-info-item").Text())
|
||||
|
||||
// 提取年份
|
||||
year := strings.TrimSpace(s.Find(".video-info-items").Last().Find(".video-info-item").First().Text())
|
||||
|
||||
// 提取质量/状态
|
||||
quality := strings.TrimSpace(s.Find(".video-info-header .video-info-remarks").Text())
|
||||
|
||||
// 提取剧情简介
|
||||
plotElement := s.Find(".video-info-items").FilterFunction(func(i int, item *goquery.Selection) bool {
|
||||
title := strings.TrimSpace(item.Find(".video-info-itemtitle").Text())
|
||||
return strings.Contains(title, "剧情")
|
||||
})
|
||||
plot := strings.TrimSpace(plotElement.Find(".video-info-item").Text())
|
||||
|
||||
// 提取封面图片
|
||||
coverImage, _ := s.Find(".module-item-pic > img").Attr("data-src")
|
||||
|
||||
// 构建内容描述
|
||||
var contentParts []string
|
||||
if category != "" {
|
||||
contentParts = append(contentParts, fmt.Sprintf("分类: %s", category))
|
||||
}
|
||||
if director != "" {
|
||||
contentParts = append(contentParts, fmt.Sprintf("导演: %s", director))
|
||||
}
|
||||
if actor != "" {
|
||||
contentParts = append(contentParts, fmt.Sprintf("主演: %s", actor))
|
||||
}
|
||||
if quality != "" {
|
||||
contentParts = append(contentParts, fmt.Sprintf("质量: %s", quality))
|
||||
}
|
||||
if plot != "" {
|
||||
contentParts = append(contentParts, fmt.Sprintf("剧情: %s", plot))
|
||||
}
|
||||
|
||||
// 构建标签
|
||||
var tags []string
|
||||
if item.VodYear != "" {
|
||||
tags = append(tags, item.VodYear)
|
||||
if year != "" {
|
||||
tags = append(tags, year)
|
||||
}
|
||||
// area通常为空，不添加
|
||||
|
||||
|
||||
// 构建图片数组
|
||||
var images []string
|
||||
if coverImage != "" {
|
||||
images = append(images, coverImage)
|
||||
}
|
||||
|
||||
return model.SearchResult{
|
||||
UniqueID: uniqueID,
|
||||
Title: title,
|
||||
Content: content,
|
||||
Links: links,
|
||||
Content: strings.Join(contentParts, " | "),
|
||||
Images: images,
|
||||
Tags: tags,
|
||||
Channel: "", // 插件搜索结果Channel为空
|
||||
Datetime: time.Time{}, // 使用零值而不是nil，参考jikepan插件标准
|
||||
Channel: "",
|
||||
Datetime: time.Time{},
|
||||
}
|
||||
}
|
||||
|
||||
// buildContent 构建内容描述（清理特殊字符）
|
||||
func (p *HubanAsyncPlugin) buildContent(item HubanAPIItem) string {
|
||||
var contentParts []string
|
||||
|
||||
// 清理演员字段（移除前后逗号）
|
||||
if item.VodActor != "" {
|
||||
actor := strings.Trim(item.VodActor, ",")
|
||||
actor = strings.TrimSpace(actor)
|
||||
if actor != "" {
|
||||
contentParts = append(contentParts, fmt.Sprintf("主演: %s", actor))
|
||||
}
|
||||
}
|
||||
|
||||
// 清理导演字段（移除前后逗号）
|
||||
if item.VodDirector != "" {
|
||||
director := strings.Trim(item.VodDirector, ",")
|
||||
director = strings.TrimSpace(director)
|
||||
if director != "" {
|
||||
contentParts = append(contentParts, fmt.Sprintf("导演: %s", director))
|
||||
}
|
||||
}
|
||||
|
||||
if item.VodYear != "" {
|
||||
contentParts = append(contentParts, fmt.Sprintf("年份: %s", item.VodYear))
|
||||
}
|
||||
|
||||
if item.VodRemarks != "" {
|
||||
contentParts = append(contentParts, fmt.Sprintf("状态: %s", item.VodRemarks))
|
||||
}
|
||||
|
||||
return strings.Join(contentParts, " | ")
|
||||
}
|
||||
// enhanceWithDetails 异步获取详情页信息
|
||||
func (p *HubanAsyncPlugin) enhanceWithDetails(client *http.Client, results []model.SearchResult) []model.SearchResult {
|
||||
var enhancedResults []model.SearchResult
|
||||
var wg sync.WaitGroup
|
||||
var mu sync.Mutex
|
||||
|
||||
// parseHubanLinks 解析huban特殊格式的链接
|
||||
func (p *HubanAsyncPlugin) parseHubanLinks(vodDownFrom, vodDownURL string) []model.Link {
|
||||
if vodDownFrom == "" || vodDownURL == "" {
|
||||
return nil
|
||||
}
|
||||
|
||||
// 按$$$分隔网盘类型
|
||||
fromParts := strings.Split(vodDownFrom, "$$$")
|
||||
urlParts := strings.Split(vodDownURL, "$$$")
|
||||
|
||||
var links []model.Link
|
||||
minLen := len(fromParts)
|
||||
if len(urlParts) < minLen {
|
||||
minLen = len(urlParts)
|
||||
}
|
||||
|
||||
for i := 0; i < minLen; i++ {
|
||||
linkType := p.mapHubanCloudType(fromParts[i])
|
||||
if linkType == "" {
|
||||
continue
|
||||
}
|
||||
|
||||
// 解析单个网盘类型的多个链接
|
||||
// 格式: "来源$链接1#标题1$链接2#标题2#"
|
||||
urlSection := urlParts[i]
|
||||
|
||||
// 移除来源前缀（如"小虎斑$"）
|
||||
if strings.Contains(urlSection, "$") {
|
||||
urlSection = urlSection[strings.Index(urlSection, "$")+1:]
|
||||
}
|
||||
|
||||
// 按#分隔多个链接
|
||||
linkParts := strings.Split(urlSection, "#")
|
||||
for j := 0; j < len(linkParts); j++ {
|
||||
linkURL := strings.TrimSpace(linkParts[j])
|
||||
|
||||
// 跳过空链接和标题（标题通常不是链接格式）
|
||||
if linkURL == "" || !p.isValidNetworkDriveURL(linkURL) {
|
||||
continue
|
||||
// 创建信号量限制并发数
|
||||
semaphore := make(chan struct{}, MaxConcurrency)
|
||||
|
||||
for _, result := range results {
|
||||
wg.Add(1)
|
||||
go func(result model.SearchResult) {
|
||||
defer wg.Done()
|
||||
semaphore <- struct{}{} // 获取信号量
|
||||
defer func() { <-semaphore }() // 释放信号量
|
||||
|
||||
// 从UniqueID中提取itemID
|
||||
parts := strings.Split(result.UniqueID, "-")
|
||||
if len(parts) < 2 {
|
||||
mu.Lock()
|
||||
enhancedResults = append(enhancedResults, result)
|
||||
mu.Unlock()
|
||||
return
|
||||
}
|
||||
|
||||
// 提取密码
|
||||
password := p.extractPassword(linkURL)
|
||||
|
||||
links = append(links, model.Link{
|
||||
Type: linkType,
|
||||
URL: linkURL,
|
||||
Password: password,
|
||||
})
|
||||
}
|
||||
itemID := parts[1]
|
||||
|
||||
// 检查缓存
|
||||
if cached, ok := detailCache.Load(itemID); ok {
|
||||
atomic.AddInt64(&cacheHits, 1)
|
||||
r := cached.(model.SearchResult)
|
||||
mu.Lock()
|
||||
enhancedResults = append(enhancedResults, r)
|
||||
mu.Unlock()
|
||||
return
|
||||
}
|
||||
|
||||
atomic.AddInt64(&cacheMisses, 1)
|
||||
|
||||
// 获取详情页链接和图片
|
||||
detailLinks, detailImages := p.fetchDetailLinksAndImages(client, itemID)
|
||||
result.Links = detailLinks
|
||||
|
||||
// 合并图片：优先使用详情页的海报，如果没有则使用搜索结果的图片
|
||||
if len(detailImages) > 0 {
|
||||
result.Images = detailImages
|
||||
}
|
||||
|
||||
// 缓存结果
|
||||
detailCache.Store(itemID, result)
|
||||
|
||||
mu.Lock()
|
||||
enhancedResults = append(enhancedResults, result)
|
||||
mu.Unlock()
|
||||
}(result)
|
||||
}
|
||||
|
||||
// 去重（可能存在重复链接）
|
||||
return p.deduplicateLinks(links)
|
||||
|
||||
wg.Wait()
|
||||
return enhancedResults
|
||||
}
|
||||
|
||||
// mapHubanCloudType 映射huban特有的网盘标识符
|
||||
func (p *HubanAsyncPlugin) mapHubanCloudType(apiType string) string {
|
||||
switch strings.ToUpper(apiType) {
|
||||
case "UCWP":
|
||||
return "uc"
|
||||
case "KKWP":
|
||||
return "quark"
|
||||
case "ALWP":
|
||||
return "aliyun"
|
||||
case "BDWP":
|
||||
return "baidu"
|
||||
case "123WP":
|
||||
return "123"
|
||||
case "115WP":
|
||||
return "115"
|
||||
case "TYWP":
|
||||
return "tianyi"
|
||||
case "XYWP":
|
||||
return "xunlei"
|
||||
case "WYWP":
|
||||
return "weiyun"
|
||||
case "LZWP":
|
||||
return "lanzou"
|
||||
case "JGYWP":
|
||||
return "jianguoyun"
|
||||
case "PKWP":
|
||||
return "pikpak"
|
||||
default:
|
||||
return ""
|
||||
// fetchDetailLinksAndImages 获取详情页的下载链接和图片
|
||||
func (p *HubanAsyncPlugin) fetchDetailLinksAndImages(client *http.Client, itemID string) ([]model.Link, []string) {
|
||||
// 性能统计
|
||||
start := time.Now()
|
||||
atomic.AddInt64(&detailPageRequests, 1)
|
||||
defer func() {
|
||||
duration := time.Since(start).Nanoseconds()
|
||||
atomic.AddInt64(&totalDetailTime, duration)
|
||||
}()
|
||||
|
||||
detailURL := fmt.Sprintf("http://103.45.162.207:20720/index.php/vod/detail/id/%s.html", itemID)
|
||||
|
||||
// 创建带超时的上下文
|
||||
ctx, cancel := context.WithTimeout(context.Background(), DetailTimeout)
|
||||
defer cancel()
|
||||
|
||||
// 创建请求
|
||||
req, err := http.NewRequestWithContext(ctx, "GET", detailURL, nil)
|
||||
if err != nil {
|
||||
return nil, nil
|
||||
}
|
||||
|
||||
// 设置请求头
|
||||
req.Header.Set("User-Agent", "Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/91.0.4472.124 Safari/537.36")
|
||||
req.Header.Set("Accept", "text/html,application/xhtml+xml,application/xml;q=0.9,image/webp,*/*;q=0.8")
|
||||
req.Header.Set("Accept-Language", "zh-CN,zh;q=0.9,en;q=0.8")
|
||||
req.Header.Set("Connection", "keep-alive")
|
||||
req.Header.Set("Referer", "http://103.45.162.207:20720/")
|
||||
|
||||
// 发送请求（带重试）
|
||||
resp, err := p.doRequestWithRetry(req, client)
|
||||
if err != nil {
|
||||
return nil, nil
|
||||
}
|
||||
defer resp.Body.Close()
|
||||
|
||||
if resp.StatusCode != 200 {
|
||||
return nil, nil
|
||||
}
|
||||
|
||||
doc, err := goquery.NewDocumentFromReader(resp.Body)
|
||||
if err != nil {
|
||||
return nil, nil
|
||||
}
|
||||
|
||||
var links []model.Link
|
||||
var images []string
|
||||
|
||||
// 提取详情页的海报图片
|
||||
if posterURL, exists := doc.Find(".mobile-play .lazyload").Attr("data-src"); exists && posterURL != "" {
|
||||
images = append(images, posterURL)
|
||||
}
|
||||
|
||||
// 查找下载链接区域
|
||||
doc.Find("#download-list .module-row-one").Each(func(i int, s *goquery.Selection) {
|
||||
// 从data-clipboard-text属性提取链接
|
||||
if linkURL, exists := s.Find("[data-clipboard-text]").Attr("data-clipboard-text"); exists {
|
||||
// 过滤掉无效链接
|
||||
if p.isValidNetworkDriveURL(linkURL) {
|
||||
if linkType := p.determineLinkType(linkURL); linkType != "" {
|
||||
link := model.Link{
|
||||
Type: linkType,
|
||||
URL: linkURL,
|
||||
Password: "", // 大部分网盘不需要密码
|
||||
}
|
||||
links = append(links, link)
|
||||
}
|
||||
}
|
||||
}
|
||||
})
|
||||
|
||||
return links, images
|
||||
}
|
||||
|
||||
|
||||
|
||||
// isValidNetworkDriveURL 检查URL是否为有效的网盘链接
|
||||
func (p *HubanAsyncPlugin) isValidNetworkDriveURL(url string) bool {
|
||||
// 过滤掉明显无效的链接
|
||||
@@ -497,27 +554,11 @@ func (p *HubanAsyncPlugin) extractPassword(url string) string {
|
||||
return ""
|
||||
}
|
||||
|
||||
// deduplicateLinks 去重链接
|
||||
func (p *HubanAsyncPlugin) deduplicateLinks(links []model.Link) []model.Link {
|
||||
seen := make(map[string]bool)
|
||||
var result []model.Link
|
||||
|
||||
for _, link := range links {
|
||||
key := fmt.Sprintf("%s-%s", link.Type, link.URL)
|
||||
if !seen[key] {
|
||||
seen[key] = true
|
||||
result = append(result, link)
|
||||
}
|
||||
}
|
||||
|
||||
return result
|
||||
}
|
||||
|
||||
// doRequestWithRetry 带重试的HTTP请求（优化JSON API的重试策略）
|
||||
// doRequestWithRetry 带重试的HTTP请求
|
||||
func (p *HubanAsyncPlugin) doRequestWithRetry(req *http.Request, client *http.Client) (*http.Response, error) {
|
||||
maxRetries := 2 // 对于JSON API减少重试次数
|
||||
maxRetries := 2
|
||||
var lastErr error
|
||||
|
||||
|
||||
for i := 0; i < maxRetries; i++ {
|
||||
resp, err := client.Do(req)
|
||||
if err == nil {
|
||||
@@ -529,13 +570,13 @@ func (p *HubanAsyncPlugin) doRequestWithRetry(req *http.Request, client *http.Cl
|
||||
} else {
|
||||
lastErr = err
|
||||
}
|
||||
|
||||
// JSON API快速重试：只等待很短时间
|
||||
|
||||
// 快速重试：只等待很短时间
|
||||
if i < maxRetries-1 {
|
||||
time.Sleep(100 * time.Millisecond) // 从秒级改为100毫秒
|
||||
time.Sleep(100 * time.Millisecond)
|
||||
}
|
||||
}
|
||||
|
||||
|
||||
return nil, fmt.Errorf("[%s] 请求失败，重试%d次后仍失败: %w", p.Name(), maxRetries, lastErr)
|
||||
}
|
||||
|
||||
@@ -543,16 +584,30 @@ func (p *HubanAsyncPlugin) doRequestWithRetry(req *http.Request, client *http.Cl
|
||||
func (p *HubanAsyncPlugin) GetPerformanceStats() map[string]interface{} {
|
||||
totalRequests := atomic.LoadInt64(&searchRequests)
|
||||
totalTime := atomic.LoadInt64(&totalSearchTime)
|
||||
|
||||
detailRequests := atomic.LoadInt64(&detailPageRequests)
|
||||
detailTime := atomic.LoadInt64(&totalDetailTime)
|
||||
hits := atomic.LoadInt64(&cacheHits)
|
||||
misses := atomic.LoadInt64(&cacheMisses)
|
||||
|
||||
var avgTime float64
|
||||
if totalRequests > 0 {
|
||||
avgTime = float64(totalTime) / float64(totalRequests) / 1e6 // 转换为毫秒
|
||||
}
|
||||
|
||||
|
||||
var avgDetailTime float64
|
||||
if detailRequests > 0 {
|
||||
avgDetailTime = float64(detailTime) / float64(detailRequests) / 1e6 // 转换为毫秒
|
||||
}
|
||||
|
||||
return map[string]interface{}{
|
||||
"search_requests": totalRequests,
|
||||
"avg_search_time_ms": avgTime,
|
||||
"search_requests": totalRequests,
|
||||
"avg_search_time_ms": avgTime,
|
||||
"total_search_time_ns": totalTime,
|
||||
"detail_page_requests": detailRequests,
|
||||
"avg_detail_time_ms": avgDetailTime,
|
||||
"total_detail_time_ns": detailTime,
|
||||
"cache_hits": hits,
|
||||
"cache_misses": misses,
|
||||
}
|
||||
}
|
||||
|
||||
|
||||
@@ -239,7 +239,14 @@ func (p *LabiAsyncPlugin) parseSearchItem(s *goquery.Selection, keyword string)
|
||||
return strings.Contains(title, "剧情")
|
||||
})
|
||||
plot := strings.TrimSpace(plotElement.Find(".video-info-item").Text())
|
||||
|
||||
|
||||
// 提取封面图片 (参考 Pan_wogg.js 的选择器)
|
||||
var images []string
|
||||
if picURL, exists := s.Find(".module-item-pic > img").Attr("data-src"); exists && picURL != "" {
|
||||
images = append(images, picURL)
|
||||
}
|
||||
result.Images = images
|
||||
|
||||
// 构建内容描述
|
||||
var contentParts []string
|
||||
if quality != "" {
|
||||
@@ -258,11 +265,11 @@ func (p *LabiAsyncPlugin) parseSearchItem(s *goquery.Selection, keyword string)
|
||||
if plot != "" {
|
||||
contentParts = append(contentParts, plot)
|
||||
}
|
||||
|
||||
|
||||
result.Content = strings.Join(contentParts, "\n")
|
||||
result.Channel = "" // 插件搜索结果不设置频道名，只有Telegram频道结果才设置
|
||||
result.Datetime = time.Time{} // 使用零值而不是nil，参考jikepan插件标准
|
||||
|
||||
|
||||
return result
|
||||
}
|
||||
|
||||
@@ -305,10 +312,15 @@ func (p *LabiAsyncPlugin) enhanceWithDetails(client *http.Client, results []mode
|
||||
}
|
||||
}
|
||||
|
||||
// 获取详情页链接
|
||||
detailLinks := p.fetchDetailLinks(client, itemID)
|
||||
// 获取详情页链接和图片
|
||||
detailLinks, detailImages := p.fetchDetailLinksAndImages(client, itemID)
|
||||
r.Links = detailLinks
|
||||
|
||||
|
||||
// 合并图片：优先使用详情页的海报，如果没有则使用搜索结果的图片
|
||||
if len(detailImages) > 0 {
|
||||
r.Images = detailImages
|
||||
}
|
||||
|
||||
// 缓存结果
|
||||
detailCache.Store(itemID, r)
|
||||
|
||||
@@ -351,45 +363,51 @@ func (p *LabiAsyncPlugin) doRequestWithRetry(req *http.Request, client *http.Cli
|
||||
return nil, fmt.Errorf("重试 %d 次后仍然失败: %w", maxRetries, lastErr)
|
||||
}
|
||||
|
||||
// fetchDetailLinks 获取详情页的下载链接
|
||||
func (p *LabiAsyncPlugin) fetchDetailLinks(client *http.Client, itemID string) []model.Link {
|
||||
// fetchDetailLinksAndImages 获取详情页的下载链接和图片
|
||||
func (p *LabiAsyncPlugin) fetchDetailLinksAndImages(client *http.Client, itemID string) ([]model.Link, []string) {
|
||||
detailURL := fmt.Sprintf("http://xiaocge.fun/index.php/vod/detail/id/%s.html", itemID)
|
||||
|
||||
|
||||
// 创建带超时的上下文
|
||||
ctx, cancel := context.WithTimeout(context.Background(), DetailTimeout)
|
||||
defer cancel()
|
||||
|
||||
|
||||
// 创建请求
|
||||
req, err := http.NewRequestWithContext(ctx, "GET", detailURL, nil)
|
||||
if err != nil {
|
||||
return nil
|
||||
return nil, nil
|
||||
}
|
||||
|
||||
|
||||
// 设置请求头
|
||||
req.Header.Set("User-Agent", "Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/91.0.4472.124 Safari/537.36")
|
||||
req.Header.Set("Accept", "text/html,application/xhtml+xml,application/xml;q=0.9,image/webp,*/*;q=0.8")
|
||||
req.Header.Set("Accept-Language", "zh-CN,zh;q=0.9,en;q=0.8")
|
||||
req.Header.Set("Connection", "keep-alive")
|
||||
req.Header.Set("Referer", "http://xiaocge.fun/")
|
||||
|
||||
|
||||
// 发送请求（带重试）
|
||||
resp, err := p.doRequestWithRetry(req, client)
|
||||
if err != nil {
|
||||
return nil
|
||||
return nil, nil
|
||||
}
|
||||
defer resp.Body.Close()
|
||||
|
||||
|
||||
if resp.StatusCode != 200 {
|
||||
return nil
|
||||
return nil, nil
|
||||
}
|
||||
|
||||
|
||||
doc, err := goquery.NewDocumentFromReader(resp.Body)
|
||||
if err != nil {
|
||||
return nil
|
||||
return nil, nil
|
||||
}
|
||||
|
||||
|
||||
var links []model.Link
|
||||
|
||||
var images []string
|
||||
|
||||
// 提取详情页的海报图片 (参考 Pan_wogg.js 的选择器)
|
||||
if posterURL, exists := doc.Find(".module-item-pic > img").Attr("data-src"); exists && posterURL != "" {
|
||||
images = append(images, posterURL)
|
||||
}
|
||||
|
||||
// 查找下载链接区域
|
||||
doc.Find("#download-list .module-row-one").Each(func(i int, s *goquery.Selection) {
|
||||
// 从data-clipboard-text属性提取链接
|
||||
@@ -404,7 +422,7 @@ func (p *LabiAsyncPlugin) fetchDetailLinks(client *http.Client, itemID string) [
|
||||
links = append(links, link)
|
||||
}
|
||||
}
|
||||
|
||||
|
||||
// 也检查直接的href属性
|
||||
s.Find("a[href]").Each(func(j int, a *goquery.Selection) {
|
||||
if linkURL, exists := a.Attr("href"); exists {
|
||||
@@ -418,7 +436,7 @@ func (p *LabiAsyncPlugin) fetchDetailLinks(client *http.Client, itemID string) [
|
||||
break
|
||||
}
|
||||
}
|
||||
|
||||
|
||||
if !isDuplicate {
|
||||
link := model.Link{
|
||||
Type: "quark",
|
||||
@@ -431,7 +449,13 @@ func (p *LabiAsyncPlugin) fetchDetailLinks(client *http.Client, itemID string) [
|
||||
}
|
||||
})
|
||||
})
|
||||
|
||||
|
||||
return links, images
|
||||
}
|
||||
|
||||
// fetchDetailLinks 获取详情页的下载链接（兼容性方法，仅返回链接）
|
||||
func (p *LabiAsyncPlugin) fetchDetailLinks(client *http.Client, itemID string) []model.Link {
|
||||
links, _ := p.fetchDetailLinksAndImages(client, itemID)
|
||||
return links
|
||||
}
|
||||
|
||||
|
||||
@@ -135,7 +135,7 @@ func (p *MuouAsyncPlugin) searchImpl(client *http.Client, keyword string, ext ma
|
||||
}
|
||||
|
||||
// 1. 构建搜索URL
|
||||
searchURL := fmt.Sprintf("http://123.666291.xyz/index.php/vod/search/wd/%s.html", url.QueryEscape(keyword))
|
||||
searchURL := fmt.Sprintf("https://666.666291.xyz/index.php/vod/search/wd/%s.html", url.QueryEscape(keyword))
|
||||
|
||||
// 2. 创建带超时的上下文
|
||||
ctx, cancel := context.WithTimeout(context.Background(), DefaultTimeout)
|
||||
@@ -154,7 +154,7 @@ func (p *MuouAsyncPlugin) searchImpl(client *http.Client, keyword string, ext ma
|
||||
req.Header.Set("Connection", "keep-alive")
|
||||
req.Header.Set("Upgrade-Insecure-Requests", "1")
|
||||
req.Header.Set("Cache-Control", "max-age=0")
|
||||
req.Header.Set("Referer", "http://123.666291.xyz/")
|
||||
req.Header.Set("Referer", "https://666.666291.xyz/")
|
||||
|
||||
// 5. 发送请求（带重试机制）
|
||||
resp, err := p.doRequestWithRetry(req, client)
|
||||
@@ -256,7 +256,14 @@ func (p *MuouAsyncPlugin) parseSearchItem(s *goquery.Selection, keyword string)
|
||||
return strings.Contains(title, "剧情")
|
||||
})
|
||||
plot := strings.TrimSpace(plotElement.Find(".video-info-item").Text())
|
||||
|
||||
|
||||
// 提取封面图片 (参考 Pan_mogg.js 的选择器)
|
||||
var images []string
|
||||
if picURL, exists := s.Find(".module-item-pic > img").Attr("data-src"); exists && picURL != "" {
|
||||
images = append(images, picURL)
|
||||
}
|
||||
result.Images = images
|
||||
|
||||
// 构建内容描述
|
||||
var contentParts []string
|
||||
if quality != "" {
|
||||
@@ -275,11 +282,11 @@ func (p *MuouAsyncPlugin) parseSearchItem(s *goquery.Selection, keyword string)
|
||||
if plot != "" {
|
||||
contentParts = append(contentParts, plot)
|
||||
}
|
||||
|
||||
|
||||
result.Content = strings.Join(contentParts, "\n")
|
||||
result.Channel = "" // 插件搜索结果不设置频道名，只有Telegram频道结果才设置
|
||||
result.Datetime = time.Time{} // 使用零值而不是nil，参考jikepan插件标准
|
||||
|
||||
|
||||
return result
|
||||
}
|
||||
|
||||
@@ -311,7 +318,7 @@ func (p *MuouAsyncPlugin) enhanceWithDetails(client *http.Client, results []mode
|
||||
}
|
||||
|
||||
itemID := parts[1]
|
||||
|
||||
|
||||
// 检查缓存
|
||||
if cached, ok := detailCache.Load(itemID); ok {
|
||||
if cachedResult, ok := cached.(model.SearchResult); ok {
|
||||
@@ -323,14 +330,19 @@ func (p *MuouAsyncPlugin) enhanceWithDetails(client *http.Client, results []mode
|
||||
}
|
||||
}
|
||||
atomic.AddInt64(&cacheMisses, 1)
|
||||
|
||||
// 获取详情页链接
|
||||
detailLinks := p.fetchDetailLinks(client, itemID)
|
||||
|
||||
// 获取详情页链接和图片
|
||||
detailLinks, detailImages := p.fetchDetailLinksAndImages(client, itemID)
|
||||
r.Links = detailLinks
|
||||
|
||||
|
||||
// 合并图片：优先使用详情页的海报，如果没有则使用搜索结果的图片
|
||||
if len(detailImages) > 0 {
|
||||
r.Images = detailImages
|
||||
}
|
||||
|
||||
// 缓存结果
|
||||
detailCache.Store(itemID, r)
|
||||
|
||||
|
||||
mu.Lock()
|
||||
enhancedResults = append(enhancedResults, r)
|
||||
mu.Unlock()
|
||||
@@ -370,8 +382,8 @@ func (p *MuouAsyncPlugin) doRequestWithRetry(req *http.Request, client *http.Cli
|
||||
return nil, fmt.Errorf("重试 %d 次后仍然失败: %w", maxRetries, lastErr)
|
||||
}
|
||||
|
||||
// fetchDetailLinks 获取详情页的下载链接
|
||||
func (p *MuouAsyncPlugin) fetchDetailLinks(client *http.Client, itemID string) []model.Link {
|
||||
// fetchDetailLinksAndImages 获取详情页的下载链接和图片
|
||||
func (p *MuouAsyncPlugin) fetchDetailLinksAndImages(client *http.Client, itemID string) ([]model.Link, []string) {
|
||||
// 性能统计
|
||||
start := time.Now()
|
||||
atomic.AddInt64(&detailPageRequests, 1)
|
||||
@@ -380,43 +392,49 @@ func (p *MuouAsyncPlugin) fetchDetailLinks(client *http.Client, itemID string) [
|
||||
atomic.AddInt64(&totalDetailTime, duration)
|
||||
}()
|
||||
|
||||
detailURL := fmt.Sprintf("http://123.666291.xyz/index.php/vod/detail/id/%s.html", itemID)
|
||||
|
||||
detailURL := fmt.Sprintf("https://666.666291.xyz/index.php/vod/detail/id/%s.html", itemID)
|
||||
|
||||
// 创建带超时的上下文
|
||||
ctx, cancel := context.WithTimeout(context.Background(), DetailTimeout)
|
||||
defer cancel()
|
||||
|
||||
|
||||
// 创建请求
|
||||
req, err := http.NewRequestWithContext(ctx, "GET", detailURL, nil)
|
||||
if err != nil {
|
||||
return nil
|
||||
return nil, nil
|
||||
}
|
||||
|
||||
|
||||
// 设置请求头
|
||||
req.Header.Set("User-Agent", "Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/91.0.4472.124 Safari/537.36")
|
||||
req.Header.Set("Accept", "text/html,application/xhtml+xml,application/xml;q=0.9,image/webp,*/*;q=0.8")
|
||||
req.Header.Set("Accept-Language", "zh-CN,zh;q=0.9,en;q=0.8")
|
||||
req.Header.Set("Connection", "keep-alive")
|
||||
req.Header.Set("Referer", "http://123.666291.xyz/")
|
||||
|
||||
req.Header.Set("Referer", "https://666.666291.xyz/")
|
||||
|
||||
// 发送请求（带重试）
|
||||
resp, err := p.doRequestWithRetry(req, client)
|
||||
if err != nil {
|
||||
return nil
|
||||
return nil, nil
|
||||
}
|
||||
defer resp.Body.Close()
|
||||
|
||||
|
||||
if resp.StatusCode != 200 {
|
||||
return nil
|
||||
return nil, nil
|
||||
}
|
||||
|
||||
|
||||
doc, err := goquery.NewDocumentFromReader(resp.Body)
|
||||
if err != nil {
|
||||
return nil
|
||||
return nil, nil
|
||||
}
|
||||
|
||||
|
||||
var links []model.Link
|
||||
|
||||
var images []string
|
||||
|
||||
// 提取详情页的海报图片 (参考 Pan_mogg.js 的选择器)
|
||||
if posterURL, exists := doc.Find(".mobile-play .lazyload").Attr("data-src"); exists && posterURL != "" {
|
||||
images = append(images, posterURL)
|
||||
}
|
||||
|
||||
// 查找下载链接区域
|
||||
doc.Find("#download-list .module-row-one").Each(func(i int, s *goquery.Selection) {
|
||||
// 从data-clipboard-text属性提取链接
|
||||
@@ -433,7 +451,7 @@ func (p *MuouAsyncPlugin) fetchDetailLinks(client *http.Client, itemID string) [
|
||||
}
|
||||
}
|
||||
}
|
||||
|
||||
|
||||
// 也检查直接的href属性
|
||||
s.Find("a[href]").Each(func(j int, a *goquery.Selection) {
|
||||
if linkURL, exists := a.Attr("href"); exists {
|
||||
@@ -448,7 +466,7 @@ func (p *MuouAsyncPlugin) fetchDetailLinks(client *http.Client, itemID string) [
|
||||
break
|
||||
}
|
||||
}
|
||||
|
||||
|
||||
if !isDuplicate {
|
||||
link := model.Link{
|
||||
Type: linkType,
|
||||
@@ -462,7 +480,13 @@ func (p *MuouAsyncPlugin) fetchDetailLinks(client *http.Client, itemID string) [
|
||||
}
|
||||
})
|
||||
})
|
||||
|
||||
|
||||
return links, images
|
||||
}
|
||||
|
||||
// fetchDetailLinks 获取详情页的下载链接（兼容性方法，仅返回链接）
|
||||
func (p *MuouAsyncPlugin) fetchDetailLinks(client *http.Client, itemID string) []model.Link {
|
||||
links, _ := p.fetchDetailLinksAndImages(client, itemID)
|
||||
return links
|
||||
}
|
||||
|
||||
|
||||
@@ -229,7 +229,13 @@ func (p *OugeAsyncPlugin) parseAPIItem(item OugeAPIItem) model.SearchResult {
|
||||
|
||||
// 解析下载链接
|
||||
links := p.parseDownloadLinks(item.VodDownFrom, item.VodDownURL)
|
||||
|
||||
|
||||
// 提取封面图片
|
||||
var images []string
|
||||
if item.VodPic != "" {
|
||||
images = append(images, item.VodPic)
|
||||
}
|
||||
|
||||
// 构建标签
|
||||
var tags []string
|
||||
if item.VodYear != "" {
|
||||
@@ -238,13 +244,14 @@ func (p *OugeAsyncPlugin) parseAPIItem(item OugeAPIItem) model.SearchResult {
|
||||
if item.VodArea != "" {
|
||||
tags = append(tags, item.VodArea)
|
||||
}
|
||||
|
||||
|
||||
return model.SearchResult{
|
||||
UniqueID: uniqueID,
|
||||
Title: title,
|
||||
Content: content,
|
||||
Links: links,
|
||||
Tags: tags,
|
||||
Images: images,
|
||||
Channel: "", // 插件搜索结果Channel为空
|
||||
Datetime: time.Time{}, // 使用零值而不是nil，参考jikepan插件标准
|
||||
}
|
||||
|
||||
@@ -229,7 +229,13 @@ func (p *WanouAsyncPlugin) parseAPIItem(item WanouAPIItem) model.SearchResult {
|
||||
|
||||
// 解析下载链接
|
||||
links := p.parseDownloadLinks(item.VodDownFrom, item.VodDownURL)
|
||||
|
||||
|
||||
// 提取封面图片
|
||||
var images []string
|
||||
if item.VodPic != "" {
|
||||
images = append(images, item.VodPic)
|
||||
}
|
||||
|
||||
// 构建标签
|
||||
var tags []string
|
||||
if item.VodYear != "" {
|
||||
@@ -238,13 +244,14 @@ func (p *WanouAsyncPlugin) parseAPIItem(item WanouAPIItem) model.SearchResult {
|
||||
if item.VodArea != "" {
|
||||
tags = append(tags, item.VodArea)
|
||||
}
|
||||
|
||||
|
||||
return model.SearchResult{
|
||||
UniqueID: uniqueID,
|
||||
Title: title,
|
||||
Content: content,
|
||||
Links: links,
|
||||
Tags: tags,
|
||||
Images: images,
|
||||
Channel: "", // 插件搜索结果Channel为空
|
||||
Datetime: time.Time{}, // 使用零值而不是nil，参考jikepan插件标准
|
||||
}
|
||||
|
||||
@@ -1,84 +1,117 @@
|
||||
# Zhizhen API 数据结构分析
|
||||
# Zhizhen HTML 数据结构分析
|
||||
|
||||
## 基本信息
|
||||
- **数据源类型**: JSON API
|
||||
- **API URL格式**: `https://xiaomi666.fun/api.php/provide/vod?ac=detail&wd={关键词}`
|
||||
- **数据特点**: 视频点播(VOD)系统API，提供结构化影视资源数据
|
||||
- **特殊说明**: 使用独立域名，网盘标识符与wanou/ouge略有不同
|
||||
- **数据源类型**: HTML 网页
|
||||
- **搜索URL格式**: `https://xiaomi666.fun/index.php/vod/search/wd/{关键词}.html`
|
||||
- **详情URL格式**: `https://xiaomi666.fun/index.php/vod/detail/id/{资源ID}.html`
|
||||
- **数据特点**: 视频点播(VOD)系统网页，提供HTML格式的影视资源数据
|
||||
- **特殊说明**: 使用独立域名，HTML结构与muou插件相同
|
||||
|
||||
## API响应结构
|
||||
## HTML 页面结构
|
||||
|
||||
### 顶层结构
|
||||
```json
|
||||
{
|
||||
"code": 1, // 状态码：1表示成功
|
||||
"msg": "数据列表", // 响应消息
|
||||
"page": 1, // 当前页码
|
||||
"pagecount": 1, // 总页数
|
||||
"limit": 20, // 每页限制条数
|
||||
"total": 6, // 总记录数
|
||||
"list": [] // 数据列表数组
|
||||
}
|
||||
### 搜索结果页面 (`.module-search-item`)
|
||||
搜索结果页面包含多个搜索项，每个搜索项的HTML结构如下：
|
||||
|
||||
```html
|
||||
<div class="module-search-item">
|
||||
<div class="module-item-pic">
|
||||
<img data-src="https://..." />
|
||||
</div>
|
||||
<div class="video-info-header">
|
||||
<h3>
|
||||
<a href="/index.php/vod/detail/id/12345.html">资源标题</a>
|
||||
</h3>
|
||||
</div>
|
||||
<div class="video-serial">更新至11集</div>
|
||||
<div class="video-info-aux">
|
||||
<span class="tag-link">
|
||||
<a>分类1</a>
|
||||
<a>分类2</a>
|
||||
</span>
|
||||
</div>
|
||||
<div class="video-info-items">
|
||||
<div>
|
||||
<span class="video-info-itemtitle">导演：</span>
|
||||
<a class="video-info-actor">导演名</a>
|
||||
</div>
|
||||
<div>
|
||||
<span class="video-info-itemtitle">主演：</span>
|
||||
<a class="video-info-actor">演员1</a>
|
||||
<a class="video-info-actor">演员2</a>
|
||||
</div>
|
||||
<div>
|
||||
<span class="video-info-itemtitle">剧情：</span>
|
||||
<span class="video-info-item">剧情简介内容</span>
|
||||
</div>
|
||||
</div>
|
||||
</div>
|
||||
```
|
||||
|
||||
### `list`数组中的数据项结构
|
||||
```json
|
||||
{
|
||||
"vod_id": 11455, // 资源唯一ID
|
||||
"vod_name": "凡人修仙传真人版", // 资源标题
|
||||
"vod_actor": "杨洋,金晨,汪铎...", // 主演（逗号分隔）
|
||||
"vod_director": "杨阳", // 导演
|
||||
"vod_area": "大陆", // 地区
|
||||
"vod_lang": "国语", // 语言
|
||||
"vod_year": "2025", // 年份
|
||||
"vod_remarks": "更新至11集", // 更新状态/备注
|
||||
"vod_pubdate": "", // 发布日期（可能为空）
|
||||
"vod_blurb": "该剧改编自忘语...", // 简介
|
||||
"vod_content": "该剧改编自忘语...", // 内容描述
|
||||
"vod_pic": "https://...", // 封面图片URL
|
||||
|
||||
// 关键字段：下载链接相关
|
||||
"vod_down_from": "kuake$$$BAIDUI$$$kuake",
|
||||
"vod_down_url": "https://pan.quark.cn/s/d228bf3a6e44$$$https://pan.baidu.com/s/1kOWHnazfGFe6wJ-tin2pNQ?pwd=b2s4$$$https://pan.quark.cn/s/12e29bdacec4"
|
||||
}
|
||||
### 详情页面 (`.module-row-one`)
|
||||
详情页面包含下载链接区域，每个链接的HTML结构如下：
|
||||
|
||||
```html
|
||||
<div id="download-list">
|
||||
<div class="module-row-one">
|
||||
<button data-clipboard-text="https://pan.quark.cn/s/xxx">复制链接</button>
|
||||
<a href="https://pan.quark.cn/s/xxx">打开链接</a>
|
||||
</div>
|
||||
</div>
|
||||
```
|
||||
|
||||
## 插件所需字段映射
|
||||
|
||||
| 源字段 | 目标字段 | 说明 |
|
||||
|--------|----------|------|
|
||||
| `vod_id` | `UniqueID` | 格式: `zhizhen-{vod_id}` |
|
||||
| `vod_name` | `Title` | 资源标题 |
|
||||
| `vod_actor`, `vod_director`, `vod_area`, `vod_year`, `vod_remarks` | `Content` | 组合描述信息 |
|
||||
| `vod_year`, `vod_area` | `Tags` | 标签数组 |
|
||||
| `vod_down_from` + `vod_down_url` | `Links` | 解析为Link数组 |
|
||||
| 详情页URL中的ID | `UniqueID` | 格式: `zhizhen-{id}` |
|
||||
| `.video-info-header h3 a` 文本 | `Title` | 资源标题 |
|
||||
| 质量、导演、主演、剧情 | `Content` | 组合描述信息 |
|
||||
| `.video-info-aux .tag-link a` | `Tags` | 标签数组 |
|
||||
| 详情页 `#download-list` 中的链接 | `Links` | 解析为Link数组 |
|
||||
| `.module-item-pic > img` 的 `data-src` | `Images` | 封面图片 |
|
||||
| `""` | `Channel` | 插件搜索结果Channel为空 |
|
||||
| `time.Now()` | `Datetime` | 当前时间 |
|
||||
| `time.Time{}` | `Datetime` | 使用零值 |
|
||||
|
||||
## 下载链接解析
|
||||
|
||||
### 分隔符规则
|
||||
- **多个下载源**: 使用 `$$$` 分隔
|
||||
- **对应关系**: `vod_down_from`、`vod_down_url` 按相同位置对应
|
||||
### 链接提取方式
|
||||
- **从 `data-clipboard-text` 属性**: 优先从按钮的 `data-clipboard-text` 属性提取链接
|
||||
- **从 `href` 属性**: 如果没有 `data-clipboard-text`，则从 `<a>` 标签的 `href` 属性提取
|
||||
- **去重处理**: 避免重复添加相同的链接
|
||||
|
||||
### 下载源标识映射（zhizhen特有）
|
||||
| API标识 | 网盘类型 | 域名示例 | 备注 |
|
||||
|---------|----------|----------|------|
|
||||
| `kuake` | quark (夸克网盘) | `pan.quark.cn` | ⚠️ 使用`kuake`而非`KG` |
|
||||
| `BAIDUI` | baidu (百度网盘) | `pan.baidu.com` | ⚠️ 使用`BAIDUI`而非`bd` |
|
||||
| `UC` | uc (UC网盘) | `drive.uc.cn` | 与标准一致 |
|
||||
### 链接类型识别
|
||||
通过正则表达式匹配URL来自动识别网盘类型，支持16种网盘类型：
|
||||
|
||||
### 多源示例数据
|
||||
```
|
||||
vod_down_from: "kuake$$$BAIDUI$$$UC$$$BAIDUI"
|
||||
vod_down_url: "https://pan.quark.cn/s/24afb59cd9ae$$$https://pan.baidu.com/s/1d8bHaARjn60rlY_5mN3phA?pwd=ceda$$$https://drive.uc.cn/s/40f6a8d5c9804?public=1$$$https://pan.baidu.com/s/19CVP2d8_ka901b9myBh68w?pwd=begh"
|
||||
```go
|
||||
// 主流网盘
|
||||
quark: https://pan.quark.cn/s/...
|
||||
baidu: https://pan.baidu.com/s/...?pwd=...
|
||||
aliyun: https://aliyundrive.com/s/... 或 https://www.alipan.com/s/...
|
||||
uc: https://drive.uc.cn/s/...
|
||||
xunlei: https://pan.xunlei.com/s/...
|
||||
|
||||
// 运营商网盘
|
||||
tianyi: https://cloud.189.cn/t/...
|
||||
mobile: https://caiyun.feixin.10086.cn/...
|
||||
|
||||
// 专业网盘
|
||||
115: https://115.com/s/...
|
||||
weiyun: https://share.weiyun.com/...
|
||||
lanzou: https://lanzou.com/... 或其他变体
|
||||
jianguoyun: https://jianguoyun.com/p/...
|
||||
123: https://123pan.com/s/...
|
||||
pikpak: https://mypikpak.com/s/...
|
||||
|
||||
// 其他协议
|
||||
magnet: magnet:?xt=urn:btih:...
|
||||
ed2k: ed2k://|file|...|
|
||||
```
|
||||
|
||||
### 链接格式示例
|
||||
### 密码提取
|
||||
从URL中提取 `?pwd=` 参数作为密码，例如：
|
||||
```
|
||||
夸克网盘: https://pan.quark.cn/s/d228bf3a6e44
|
||||
百度网盘: https://pan.baidu.com/s/1kOWHnazfGFe6wJ-tin2pNQ?pwd=b2s4
|
||||
UC网盘: https://drive.uc.cn/s/40f6a8d5c9804?public=1
|
||||
https://pan.baidu.com/s/1kOWHnazfGFe6wJ-tin2pNQ?pwd=b2s4
|
||||
提取密码: b2s4
|
||||
```
|
||||
|
||||
## 支持的网盘类型（16种）
|
||||
@@ -109,76 +142,64 @@ UC网盘: https://drive.uc.cn/s/40f6a8d5c9804?public=1
|
||||
|
||||
## 插件开发指导
|
||||
|
||||
### 请求示例
|
||||
### 搜索请求示例
|
||||
```go
|
||||
searchURL := fmt.Sprintf("https://xiaomi666.fun/api.php/provide/vod?ac=detail&wd=%s", url.QueryEscape(keyword))
|
||||
searchURL := fmt.Sprintf("https://xiaomi666.fun/index.php/vod/search/wd/%s.html", url.QueryEscape(keyword))
|
||||
```
|
||||
|
||||
### 详情页请求示例
|
||||
```go
|
||||
detailURL := fmt.Sprintf("https://xiaomi666.fun/index.php/vod/detail/id/%s.html", itemID)
|
||||
```
|
||||
|
||||
### HTML解析流程
|
||||
1. **搜索页面解析**: 使用 goquery 解析搜索结果页面
|
||||
2. **提取搜索项**: 遍历 `.module-search-item` 元素
|
||||
3. **提取基本信息**: 从搜索项中提取标题、分类、导演、主演等
|
||||
4. **异步获取详情**: 并发请求详情页面获取下载链接
|
||||
5. **缓存管理**: 使用 sync.Map 缓存详情页结果，TTL为1小时
|
||||
|
||||
### SearchResult构建示例
|
||||
```go
|
||||
result := model.SearchResult{
|
||||
UniqueID: fmt.Sprintf("zhizhen-%d", item.VodID),
|
||||
Title: item.VodName,
|
||||
Content: buildContent(item),
|
||||
Links: parseDownloadLinks(item.VodDownFrom, item.VodDownURL),
|
||||
Tags: []string{item.VodYear, item.VodArea},
|
||||
UniqueID: fmt.Sprintf("zhizhen-%s", itemID),
|
||||
Title: title,
|
||||
Content: strings.Join(contentParts, "\n"),
|
||||
Links: detailLinks,
|
||||
Tags: tags,
|
||||
Images: images,
|
||||
Channel: "", // 插件搜索结果Channel为空
|
||||
Datetime: time.Now(),
|
||||
Datetime: time.Time{}, // 使用零值
|
||||
}
|
||||
```
|
||||
|
||||
### 特殊映射函数
|
||||
```go
|
||||
func (p *ZhizhenAsyncPlugin) mapCloudType(apiType, url string) string {
|
||||
// 优先根据API标识映射（zhizhen特有）
|
||||
switch strings.ToUpper(apiType) {
|
||||
case "KUAKE":
|
||||
return "quark"
|
||||
case "BAIDUI":
|
||||
return "baidu"
|
||||
case "UC":
|
||||
return "uc"
|
||||
}
|
||||
|
||||
// 如果API标识无法识别，则通过URL模式匹配
|
||||
return p.determineLinkType(url)
|
||||
}
|
||||
```
|
||||
|
||||
### 链接解析逻辑
|
||||
```go
|
||||
// 按$$$分隔
|
||||
fromParts := strings.Split(item.VodDownFrom, "$$$")
|
||||
urlParts := strings.Split(item.VodDownURL, "$$$")
|
||||
|
||||
// 遍历对应位置
|
||||
for i := 0; i < min(len(fromParts), len(urlParts)); i++ {
|
||||
linkType := p.mapCloudType(fromParts[i], urlParts[i])
|
||||
password := extractPassword(urlParts[i])
|
||||
// ...
|
||||
}
|
||||
```
|
||||
### 并发控制
|
||||
- **最大并发数**: 20 (MaxConcurrency)
|
||||
- **搜索超时**: 8秒 (DefaultTimeout)
|
||||
- **详情页超时**: 6秒 (DetailTimeout)
|
||||
- **缓存TTL**: 1小时 (cacheTTL)
|
||||
|
||||
## 与其他插件的差异
|
||||
|
||||
| 特性 | zhizhen | wanou/ouge | 说明 |
|
||||
|------|---------|------------|------|
|
||||
| **API域名** | `xiaomi666.fun` | `woog.nxog.eu.org` | 不同域名 |
|
||||
| **夸克标识** | `kuake` | `KG` | 标识符不同 |
|
||||
| **百度标识** | `BAIDUI` | `bd` | 标识符不同 |
|
||||
| **UC标识** | `UC` | `UC` | 一致 |
|
||||
| **数据结构** | 相同 | 相同 | JSON结构完全一致 |
|
||||
| 特性 | zhizhen | muou | 说明 |
|
||||
|------|---------|------|------|
|
||||
| **域名** | `xiaomi666.fun` | `666.666291.xyz` | 不同域名 |
|
||||
| **数据格式** | HTML | HTML | 都是HTML格式 |
|
||||
| **HTML结构** | 相同 | 相同 | 使用相同的CSS选择器 |
|
||||
| **并发数** | 20 | 20 | 相同 |
|
||||
| **缓存TTL** | 1小时 | 1小时 | 相同 |
|
||||
|
||||
## 注意事项
|
||||
1. **标识符差异**: 需要专门处理`kuake`和`BAIDUI`标识符
|
||||
2. **数据格式**: 纯JSON API，无需HTML解析
|
||||
3. **分隔符处理**: 多个值使用`$$$`分隔，需要split处理
|
||||
4. **密码提取**: 部分百度网盘链接包含`?pwd=`参数
|
||||
5. **错误处理**: API可能返回`code != 1`的错误状态
|
||||
6. **链接验证**: 应过滤无效链接（如`javascript:;`等）
|
||||
1. **HTML解析**: 使用 goquery 库进行HTML解析
|
||||
2. **异步获取详情**: 搜索结果只包含基本信息，需要异步请求详情页获取下载链接
|
||||
3. **并发控制**: 使用信号量限制并发数为20
|
||||
4. **缓存管理**: 使用 sync.Map 缓存详情页结果，避免重复请求
|
||||
5. **链接验证**: 过滤掉无效链接（如包含`javascript:`、`#`等）
|
||||
6. **密码提取**: 从URL中提取 `?pwd=` 参数作为密码
|
||||
7. **去重处理**: 避免在详情页中重复添加相同的链接
|
||||
|
||||
## 开发建议
|
||||
- **基于wanou改造**: 可以复制wanou插件实现，修改域名和标识符映射
|
||||
- **映射函数重点**: 关键是正确处理`kuake`→`quark`和`BAIDUI`→`baidu`的映射
|
||||
- **测试覆盖**: 重点测试多种网盘类型的混合链接解析
|
||||
- **缓存策略**: 建议使用相同的缓存机制和TTL设置
|
||||
- **参考muou插件**: zhizhen的HTML结构与muou完全相同，可以直接参考muou的实现
|
||||
- **关键差异**: 仅需修改域名和插件名称
|
||||
- **测试覆盖**: 重点测试多种网盘类型的链接解析和缓存功能
|
||||
- **性能优化**: 使用并发请求详情页，提高搜索速度
|
||||
@@ -1,36 +1,47 @@
|
||||
package zhizhen
|
||||
|
||||
import (
|
||||
"context"
|
||||
"fmt"
|
||||
"io"
|
||||
"net/http"
|
||||
"net/url"
|
||||
"regexp"
|
||||
"strings"
|
||||
"time"
|
||||
"context"
|
||||
"sync/atomic"
|
||||
|
||||
"pansou/model"
|
||||
"pansou/plugin"
|
||||
"pansou/util/json"
|
||||
"regexp"
|
||||
"strings"
|
||||
"sync"
|
||||
"sync/atomic"
|
||||
"time"
|
||||
|
||||
"github.com/PuerkitoBio/goquery"
|
||||
)
|
||||
|
||||
const (
|
||||
// 默认超时时间 - 优化为更短时间
|
||||
DefaultTimeout = 8 * time.Second
|
||||
DetailTimeout = 6 * time.Second
|
||||
|
||||
// 并发数限制 - 大幅提高并发数
|
||||
MaxConcurrency = 20
|
||||
|
||||
// HTTP连接池配置
|
||||
MaxIdleConns = 200
|
||||
MaxIdleConnsPerHost = 50
|
||||
MaxConnsPerHost = 100
|
||||
IdleConnTimeout = 90 * time.Second
|
||||
|
||||
// 缓存TTL - 更短的缓存时间
|
||||
cacheTTL = 1 * time.Hour
|
||||
)
|
||||
|
||||
// 性能统计（原子操作）
|
||||
var (
|
||||
searchRequests int64 = 0
|
||||
totalSearchTime int64 = 0 // 纳秒
|
||||
searchRequests int64 = 0
|
||||
detailPageRequests int64 = 0
|
||||
cacheHits int64 = 0
|
||||
cacheMisses int64 = 0
|
||||
totalSearchTime int64 = 0 // 纳秒
|
||||
totalDetailTime int64 = 0 // 纳秒
|
||||
)
|
||||
|
||||
func init() {
|
||||
@@ -39,9 +50,12 @@ func init() {
|
||||
|
||||
// 预编译的正则表达式
|
||||
var (
|
||||
// 从详情页URL中提取ID的正则表达式
|
||||
detailIDRegex = regexp.MustCompile(`/vod/detail/id/(\d+)\.html`)
|
||||
|
||||
// 密码提取正则表达式
|
||||
passwordRegex = regexp.MustCompile(`\?pwd=([0-9a-zA-Z]+)`)
|
||||
|
||||
|
||||
// 常见网盘链接的正则表达式（支持16种类型）
|
||||
quarkLinkRegex = regexp.MustCompile(`https?://pan\.quark\.cn/s/[0-9a-zA-Z]+`)
|
||||
ucLinkRegex = regexp.MustCompile(`https?://drive\.uc\.cn/s/[0-9a-zA-Z]+(\?[^"'\s]*)?`)
|
||||
@@ -58,6 +72,9 @@ var (
|
||||
pikpakLinkRegex = regexp.MustCompile(`https?://mypikpak\.com/s/[0-9a-zA-Z]+`)
|
||||
magnetLinkRegex = regexp.MustCompile(`magnet:\?xt=urn:btih:[0-9a-fA-F]{40}`)
|
||||
ed2kLinkRegex = regexp.MustCompile(`ed2k://\|file\|.+\|\d+\|[0-9a-fA-F]{32}\|/`)
|
||||
|
||||
// 缓存相关
|
||||
detailCache = sync.Map{} // 缓存详情页解析结果
|
||||
)
|
||||
|
||||
// ZhizhenAsyncPlugin Zhizhen异步插件
|
||||
@@ -104,7 +121,7 @@ func (p *ZhizhenAsyncPlugin) SearchWithResult(keyword string, ext map[string]int
|
||||
return p.AsyncSearchWithResult(keyword, p.searchImpl, p.MainCacheKey, ext)
|
||||
}
|
||||
|
||||
// searchImpl 搜索实现
|
||||
// searchImpl 实现具体的搜索逻辑
|
||||
func (p *ZhizhenAsyncPlugin) searchImpl(client *http.Client, keyword string, ext map[string]interface{}) ([]model.SearchResult, error) {
|
||||
// 性能统计
|
||||
start := time.Now()
|
||||
@@ -119,222 +136,160 @@ func (p *ZhizhenAsyncPlugin) searchImpl(client *http.Client, keyword string, ext
|
||||
client = p.optimizedClient
|
||||
}
|
||||
|
||||
// 构建API搜索URL - 使用zhizhen专用域名
|
||||
searchURL := fmt.Sprintf("https://xiaomi666.fun/api.php/provide/vod?ac=detail&wd=%s", url.QueryEscape(keyword))
|
||||
|
||||
// 创建HTTP请求
|
||||
// 1. 构建搜索URL
|
||||
searchURL := fmt.Sprintf("https://xiaomi666.fun/index.php/vod/search/wd/%s.html", url.QueryEscape(keyword))
|
||||
|
||||
// 2. 创建带超时的上下文
|
||||
ctx, cancel := context.WithTimeout(context.Background(), DefaultTimeout)
|
||||
defer cancel()
|
||||
|
||||
|
||||
// 3. 创建请求
|
||||
req, err := http.NewRequestWithContext(ctx, "GET", searchURL, nil)
|
||||
if err != nil {
|
||||
return nil, fmt.Errorf("[%s] 创建搜索请求失败: %w", p.Name(), err)
|
||||
return nil, fmt.Errorf("[%s] 创建请求失败: %w", p.Name(), err)
|
||||
}
|
||||
|
||||
// 设置请求头
|
||||
|
||||
// 4. 设置完整的请求头（避免反爬虫）
|
||||
req.Header.Set("User-Agent", "Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/91.0.4472.124 Safari/537.36")
|
||||
req.Header.Set("Accept", "application/json, text/plain, */*")
|
||||
req.Header.Set("Accept", "text/html,application/xhtml+xml,application/xml;q=0.9,image/webp,*/*;q=0.8")
|
||||
req.Header.Set("Accept-Language", "zh-CN,zh;q=0.9,en;q=0.8")
|
||||
req.Header.Set("Connection", "keep-alive")
|
||||
req.Header.Set("Upgrade-Insecure-Requests", "1")
|
||||
req.Header.Set("Cache-Control", "max-age=0")
|
||||
req.Header.Set("Referer", "https://xiaomi666.fun/")
|
||||
req.Header.Set("Cache-Control", "no-cache")
|
||||
|
||||
// 发送请求
|
||||
|
||||
// 5. 发送请求（带重试机制）
|
||||
resp, err := p.doRequestWithRetry(req, client)
|
||||
if err != nil {
|
||||
return nil, fmt.Errorf("[%s] 搜索请求失败: %w", p.Name(), err)
|
||||
}
|
||||
defer resp.Body.Close()
|
||||
|
||||
// 解析JSON响应
|
||||
body, err := io.ReadAll(resp.Body)
|
||||
|
||||
if resp.StatusCode != 200 {
|
||||
return nil, fmt.Errorf("[%s] 搜索请求返回状态码: %d", p.Name(), resp.StatusCode)
|
||||
}
|
||||
|
||||
// 6. 解析搜索结果页面
|
||||
doc, err := goquery.NewDocumentFromReader(resp.Body)
|
||||
if err != nil {
|
||||
return nil, fmt.Errorf("[%s] 读取响应失败: %w", p.Name(), err)
|
||||
return nil, fmt.Errorf("[%s] 解析搜索页面失败: %w", p.Name(), err)
|
||||
}
|
||||
|
||||
var apiResponse ZhizhenAPIResponse
|
||||
if err := json.Unmarshal(body, &apiResponse); err != nil {
|
||||
return nil, fmt.Errorf("[%s] 解析JSON响应失败: %w", p.Name(), err)
|
||||
}
|
||||
|
||||
// 检查API响应状态
|
||||
if apiResponse.Code != 1 {
|
||||
return nil, fmt.Errorf("[%s] API返回错误: %s", p.Name(), apiResponse.Msg)
|
||||
}
|
||||
|
||||
// 解析搜索结果
|
||||
|
||||
// 7. 提取搜索结果
|
||||
var results []model.SearchResult
|
||||
for _, item := range apiResponse.List {
|
||||
if result := p.parseAPIItem(item); result.Title != "" {
|
||||
|
||||
doc.Find(".module-search-item").Each(func(i int, s *goquery.Selection) {
|
||||
result := p.parseSearchItem(s, keyword)
|
||||
if result.UniqueID != "" {
|
||||
results = append(results, result)
|
||||
}
|
||||
}
|
||||
|
||||
return results, nil
|
||||
})
|
||||
|
||||
// 8. 异步获取详情页信息
|
||||
enhancedResults := p.enhanceWithDetails(client, results)
|
||||
|
||||
// 9. 关键词过滤
|
||||
return plugin.FilterResultsByKeyword(enhancedResults, keyword), nil
|
||||
}
|
||||
|
||||
// ZhizhenAPIResponse API响应结构
|
||||
type ZhizhenAPIResponse struct {
|
||||
Code int `json:"code"`
|
||||
Msg string `json:"msg"`
|
||||
Page int `json:"page"`
|
||||
PageCount int `json:"pagecount"`
|
||||
Limit int `json:"limit"`
|
||||
Total int `json:"total"`
|
||||
List []ZhizhenAPIItem `json:"list"`
|
||||
}
|
||||
// parseSearchItem 解析单个搜索结果项
|
||||
func (p *ZhizhenAsyncPlugin) parseSearchItem(s *goquery.Selection, keyword string) model.SearchResult {
|
||||
result := model.SearchResult{}
|
||||
|
||||
// ZhizhenAPIItem API数据项
|
||||
type ZhizhenAPIItem struct {
|
||||
VodID int `json:"vod_id"`
|
||||
VodName string `json:"vod_name"`
|
||||
VodActor string `json:"vod_actor"`
|
||||
VodDirector string `json:"vod_director"`
|
||||
VodDownFrom string `json:"vod_down_from"`
|
||||
VodDownURL string `json:"vod_down_url"`
|
||||
VodRemarks string `json:"vod_remarks"`
|
||||
VodPubdate string `json:"vod_pubdate"`
|
||||
VodArea string `json:"vod_area"`
|
||||
VodLang string `json:"vod_lang"`
|
||||
VodYear string `json:"vod_year"`
|
||||
VodContent string `json:"vod_content"`
|
||||
VodPic string `json:"vod_pic"`
|
||||
}
|
||||
// 提取详情页链接和ID (修正：使用正确的选择器)
|
||||
detailLink, exists := s.Find(".video-info-header h3 a").First().Attr("href")
|
||||
if !exists {
|
||||
return result
|
||||
}
|
||||
|
||||
// parseAPIItem 解析API数据项
|
||||
func (p *ZhizhenAsyncPlugin) parseAPIItem(item ZhizhenAPIItem) model.SearchResult {
|
||||
// 构建唯一ID
|
||||
uniqueID := fmt.Sprintf("%s-%d", p.Name(), item.VodID)
|
||||
|
||||
// 构建标题
|
||||
title := strings.TrimSpace(item.VodName)
|
||||
if title == "" {
|
||||
return model.SearchResult{}
|
||||
// 提取ID
|
||||
matches := detailIDRegex.FindStringSubmatch(detailLink)
|
||||
if len(matches) < 2 {
|
||||
return result
|
||||
}
|
||||
|
||||
// 构建描述
|
||||
var contentParts []string
|
||||
if item.VodActor != "" {
|
||||
contentParts = append(contentParts, fmt.Sprintf("主演: %s", item.VodActor))
|
||||
}
|
||||
if item.VodDirector != "" {
|
||||
contentParts = append(contentParts, fmt.Sprintf("导演: %s", item.VodDirector))
|
||||
}
|
||||
if item.VodArea != "" {
|
||||
contentParts = append(contentParts, fmt.Sprintf("地区: %s", item.VodArea))
|
||||
}
|
||||
if item.VodLang != "" {
|
||||
contentParts = append(contentParts, fmt.Sprintf("语言: %s", item.VodLang))
|
||||
}
|
||||
if item.VodYear != "" {
|
||||
contentParts = append(contentParts, fmt.Sprintf("年份: %s", item.VodYear))
|
||||
}
|
||||
if item.VodRemarks != "" {
|
||||
contentParts = append(contentParts, fmt.Sprintf("状态: %s", item.VodRemarks))
|
||||
}
|
||||
content := strings.Join(contentParts, " | ")
|
||||
|
||||
// 解析下载链接
|
||||
links := p.parseDownloadLinks(item.VodDownFrom, item.VodDownURL)
|
||||
|
||||
// 构建标签
|
||||
|
||||
itemID := matches[1]
|
||||
result.UniqueID = fmt.Sprintf("%s-%s", p.Name(), itemID)
|
||||
|
||||
// 提取标题
|
||||
titleElement := s.Find(".video-info-header h3 a")
|
||||
result.Title = strings.TrimSpace(titleElement.Text())
|
||||
|
||||
// 提取资源类型/质量
|
||||
qualityElement := s.Find(".video-serial")
|
||||
quality := strings.TrimSpace(qualityElement.Text())
|
||||
|
||||
// 提取分类信息
|
||||
var tags []string
|
||||
if item.VodYear != "" {
|
||||
tags = append(tags, item.VodYear)
|
||||
}
|
||||
if item.VodArea != "" {
|
||||
tags = append(tags, item.VodArea)
|
||||
}
|
||||
|
||||
return model.SearchResult{
|
||||
UniqueID: uniqueID,
|
||||
Title: title,
|
||||
Content: content,
|
||||
Links: links,
|
||||
Tags: tags,
|
||||
Channel: "", // 插件搜索结果Channel为空
|
||||
Datetime: time.Time{}, // 使用零值而不是nil，参考jikepan插件标准
|
||||
}
|
||||
}
|
||||
|
||||
// parseDownloadLinks 解析下载链接
|
||||
func (p *ZhizhenAsyncPlugin) parseDownloadLinks(vodDownFrom, vodDownURL string) []model.Link {
|
||||
if vodDownFrom == "" || vodDownURL == "" {
|
||||
return nil
|
||||
}
|
||||
|
||||
// 按$$$分隔
|
||||
fromParts := strings.Split(vodDownFrom, "$$$")
|
||||
urlParts := strings.Split(vodDownURL, "$$$")
|
||||
|
||||
// 确保数组长度一致
|
||||
minLen := len(fromParts)
|
||||
if len(urlParts) < minLen {
|
||||
minLen = len(urlParts)
|
||||
}
|
||||
|
||||
var links []model.Link
|
||||
for i := 0; i < minLen; i++ {
|
||||
fromType := strings.TrimSpace(fromParts[i])
|
||||
urlStr := strings.TrimSpace(urlParts[i])
|
||||
|
||||
if urlStr == "" || !p.isValidNetworkDriveURL(urlStr) {
|
||||
continue
|
||||
s.Find(".video-info-aux .tag-link a").Each(func(i int, tag *goquery.Selection) {
|
||||
tagText := strings.TrimSpace(tag.Text())
|
||||
if tagText != "" {
|
||||
tags = append(tags, tagText)
|
||||
}
|
||||
|
||||
// 映射网盘类型
|
||||
linkType := p.mapCloudType(fromType, urlStr)
|
||||
if linkType == "" {
|
||||
continue
|
||||
}
|
||||
|
||||
// 提取密码
|
||||
password := p.extractPassword(urlStr)
|
||||
|
||||
links = append(links, model.Link{
|
||||
Type: linkType,
|
||||
URL: urlStr,
|
||||
Password: password,
|
||||
})
|
||||
}
|
||||
|
||||
return links
|
||||
}
|
||||
})
|
||||
result.Tags = tags
|
||||
|
||||
// mapCloudType 映射网盘类型（zhizhen特有标识符）
|
||||
func (p *ZhizhenAsyncPlugin) mapCloudType(apiType, url string) string {
|
||||
// 优先根据API标识映射（zhizhen特有）
|
||||
switch strings.ToUpper(apiType) {
|
||||
case "KUAKE": // ⚠️ zhizhen特有：kuake -> quark
|
||||
return "quark"
|
||||
case "BAIDUI": // ⚠️ zhizhen特有：BAIDUI -> baidu
|
||||
return "baidu"
|
||||
case "UC": // ✅ 标准：UC -> uc
|
||||
return "uc"
|
||||
case "ALY":
|
||||
return "aliyun"
|
||||
case "XL":
|
||||
return "xunlei"
|
||||
case "TY":
|
||||
return "tianyi"
|
||||
case "115":
|
||||
return "115"
|
||||
case "MB":
|
||||
return "mobile"
|
||||
case "WY":
|
||||
return "weiyun"
|
||||
case "LZ":
|
||||
return "lanzou"
|
||||
case "JGY":
|
||||
return "jianguoyun"
|
||||
case "123":
|
||||
return "123"
|
||||
case "PK":
|
||||
return "pikpak"
|
||||
// 提取导演信息
|
||||
director := ""
|
||||
s.Find(".video-info-items").Each(func(i int, item *goquery.Selection) {
|
||||
title := strings.TrimSpace(item.Find(".video-info-itemtitle").Text())
|
||||
if strings.Contains(title, "导演") {
|
||||
director = strings.TrimSpace(item.Find(".video-info-actor a").Text())
|
||||
}
|
||||
})
|
||||
|
||||
// 提取主演信息
|
||||
var actors []string
|
||||
s.Find(".video-info-items").Each(func(i int, item *goquery.Selection) {
|
||||
title := strings.TrimSpace(item.Find(".video-info-itemtitle").Text())
|
||||
if strings.Contains(title, "主演") {
|
||||
item.Find(".video-info-actor a").Each(func(j int, actor *goquery.Selection) {
|
||||
actorName := strings.TrimSpace(actor.Text())
|
||||
if actorName != "" {
|
||||
actors = append(actors, actorName)
|
||||
}
|
||||
})
|
||||
}
|
||||
})
|
||||
|
||||
// 提取剧情简介
|
||||
plotElement := s.Find(".video-info-items").FilterFunction(func(i int, item *goquery.Selection) bool {
|
||||
title := strings.TrimSpace(item.Find(".video-info-itemtitle").Text())
|
||||
return strings.Contains(title, "剧情")
|
||||
})
|
||||
plot := strings.TrimSpace(plotElement.Find(".video-info-item").Text())
|
||||
|
||||
// 提取封面图片 (参考 Pan_mogg.js 的选择器)
|
||||
var images []string
|
||||
if picURL, exists := s.Find(".module-item-pic > img").Attr("data-src"); exists && picURL != "" {
|
||||
images = append(images, picURL)
|
||||
}
|
||||
|
||||
// 如果API标识无法识别，则通过URL模式匹配
|
||||
return p.determineLinkType(url)
|
||||
result.Images = images
|
||||
|
||||
// 构建内容描述
|
||||
var contentParts []string
|
||||
if quality != "" {
|
||||
contentParts = append(contentParts, "【"+quality+"】")
|
||||
}
|
||||
if director != "" {
|
||||
contentParts = append(contentParts, "导演："+director)
|
||||
}
|
||||
if len(actors) > 0 {
|
||||
actorStr := strings.Join(actors[:min(3, len(actors))], "、") // 只显示前3个演员
|
||||
if len(actors) > 3 {
|
||||
actorStr += "等"
|
||||
}
|
||||
contentParts = append(contentParts, "主演："+actorStr)
|
||||
}
|
||||
if plot != "" {
|
||||
contentParts = append(contentParts, plot)
|
||||
}
|
||||
|
||||
result.Content = strings.Join(contentParts, "\n")
|
||||
result.Channel = "" // 插件搜索结果不设置频道名，只有Telegram频道结果才设置
|
||||
result.Datetime = time.Time{} // 使用零值而不是nil，参考jikepan插件标准
|
||||
|
||||
return result
|
||||
}
|
||||
|
||||
// isValidNetworkDriveURL 检查URL是否为有效的网盘链接
|
||||
@@ -412,45 +367,243 @@ func (p *ZhizhenAsyncPlugin) extractPassword(url string) string {
|
||||
return ""
|
||||
}
|
||||
|
||||
// doRequestWithRetry 带重试的HTTP请求（优化JSON API的重试策略）
|
||||
func (p *ZhizhenAsyncPlugin) doRequestWithRetry(req *http.Request, client *http.Client) (*http.Response, error) {
|
||||
maxRetries := 2 // 对于JSON API减少重试次数
|
||||
var lastErr error
|
||||
|
||||
for i := 0; i < maxRetries; i++ {
|
||||
resp, err := client.Do(req)
|
||||
if err == nil {
|
||||
if resp.StatusCode == http.StatusOK {
|
||||
return resp, nil
|
||||
// enhanceWithDetails 异步获取详情页信息以获取下载链接
|
||||
func (p *ZhizhenAsyncPlugin) enhanceWithDetails(client *http.Client, results []model.SearchResult) []model.SearchResult {
|
||||
var enhancedResults []model.SearchResult
|
||||
var mu sync.Mutex
|
||||
var wg sync.WaitGroup
|
||||
|
||||
// 限制并发数
|
||||
semaphore := make(chan struct{}, MaxConcurrency)
|
||||
|
||||
for _, result := range results {
|
||||
wg.Add(1)
|
||||
go func(r model.SearchResult) {
|
||||
defer wg.Done()
|
||||
|
||||
// 获取信号量
|
||||
semaphore <- struct{}{}
|
||||
defer func() { <-semaphore }()
|
||||
|
||||
// 从UniqueID提取ID
|
||||
parts := strings.Split(r.UniqueID, "-")
|
||||
if len(parts) < 2 {
|
||||
mu.Lock()
|
||||
enhancedResults = append(enhancedResults, r)
|
||||
mu.Unlock()
|
||||
return
|
||||
}
|
||||
resp.Body.Close()
|
||||
lastErr = fmt.Errorf("HTTP状态码: %d", resp.StatusCode)
|
||||
} else {
|
||||
lastErr = err
|
||||
}
|
||||
|
||||
// JSON API快速重试：只等待很短时间
|
||||
if i < maxRetries-1 {
|
||||
time.Sleep(100 * time.Millisecond) // 从秒级改为100毫秒
|
||||
}
|
||||
|
||||
itemID := parts[1]
|
||||
|
||||
// 检查缓存
|
||||
if cached, ok := detailCache.Load(itemID); ok {
|
||||
if cachedResult, ok := cached.(model.SearchResult); ok {
|
||||
atomic.AddInt64(&cacheHits, 1)
|
||||
mu.Lock()
|
||||
enhancedResults = append(enhancedResults, cachedResult)
|
||||
mu.Unlock()
|
||||
return
|
||||
}
|
||||
}
|
||||
atomic.AddInt64(&cacheMisses, 1)
|
||||
|
||||
// 获取详情页链接和图片
|
||||
detailLinks, detailImages := p.fetchDetailLinksAndImages(client, itemID)
|
||||
r.Links = detailLinks
|
||||
|
||||
// 合并图片：优先使用详情页的海报，如果没有则使用搜索结果的图片
|
||||
if len(detailImages) > 0 {
|
||||
r.Images = detailImages
|
||||
}
|
||||
|
||||
// 缓存结果
|
||||
detailCache.Store(itemID, r)
|
||||
|
||||
mu.Lock()
|
||||
enhancedResults = append(enhancedResults, r)
|
||||
mu.Unlock()
|
||||
}(result)
|
||||
}
|
||||
|
||||
return nil, fmt.Errorf("[%s] 请求失败，重试%d次后仍失败: %w", p.Name(), maxRetries, lastErr)
|
||||
|
||||
wg.Wait()
|
||||
return enhancedResults
|
||||
}
|
||||
|
||||
// doRequestWithRetry 带重试机制的HTTP请求
|
||||
func (p *ZhizhenAsyncPlugin) doRequestWithRetry(req *http.Request, client *http.Client) (*http.Response, error) {
|
||||
maxRetries := 3
|
||||
var lastErr error
|
||||
|
||||
for i := 0; i < maxRetries; i++ {
|
||||
if i > 0 {
|
||||
// 指数退避
|
||||
backoff := time.Duration(1<<uint(i-1)) * 200 * time.Millisecond
|
||||
time.Sleep(backoff)
|
||||
}
|
||||
|
||||
// 克隆请求
|
||||
reqClone := req.Clone(req.Context())
|
||||
|
||||
resp, err := client.Do(reqClone)
|
||||
if err == nil && resp.StatusCode == 200 {
|
||||
return resp, nil
|
||||
}
|
||||
|
||||
if resp != nil {
|
||||
resp.Body.Close()
|
||||
}
|
||||
lastErr = err
|
||||
}
|
||||
|
||||
return nil, fmt.Errorf("重试 %d 次后仍然失败: %w", maxRetries, lastErr)
|
||||
}
|
||||
|
||||
// fetchDetailLinksAndImages 获取详情页的下载链接和图片
|
||||
func (p *ZhizhenAsyncPlugin) fetchDetailLinksAndImages(client *http.Client, itemID string) ([]model.Link, []string) {
|
||||
// 性能统计
|
||||
start := time.Now()
|
||||
atomic.AddInt64(&detailPageRequests, 1)
|
||||
defer func() {
|
||||
duration := time.Since(start).Nanoseconds()
|
||||
atomic.AddInt64(&totalDetailTime, duration)
|
||||
}()
|
||||
|
||||
detailURL := fmt.Sprintf("https://xiaomi666.fun/index.php/vod/detail/id/%s.html", itemID)
|
||||
|
||||
// 创建带超时的上下文
|
||||
ctx, cancel := context.WithTimeout(context.Background(), DetailTimeout)
|
||||
defer cancel()
|
||||
|
||||
// 创建请求
|
||||
req, err := http.NewRequestWithContext(ctx, "GET", detailURL, nil)
|
||||
if err != nil {
|
||||
return nil, nil
|
||||
}
|
||||
|
||||
// 设置请求头
|
||||
req.Header.Set("User-Agent", "Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/91.0.4472.124 Safari/537.36")
|
||||
req.Header.Set("Accept", "text/html,application/xhtml+xml,application/xml;q=0.9,image/webp,*/*;q=0.8")
|
||||
req.Header.Set("Accept-Language", "zh-CN,zh;q=0.9,en;q=0.8")
|
||||
req.Header.Set("Connection", "keep-alive")
|
||||
req.Header.Set("Referer", "https://xiaomi666.fun/")
|
||||
|
||||
// 发送请求（带重试）
|
||||
resp, err := p.doRequestWithRetry(req, client)
|
||||
if err != nil {
|
||||
return nil, nil
|
||||
}
|
||||
defer resp.Body.Close()
|
||||
|
||||
if resp.StatusCode != 200 {
|
||||
return nil, nil
|
||||
}
|
||||
|
||||
doc, err := goquery.NewDocumentFromReader(resp.Body)
|
||||
if err != nil {
|
||||
return nil, nil
|
||||
}
|
||||
|
||||
var links []model.Link
|
||||
var images []string
|
||||
|
||||
// 提取详情页的海报图片 (参考 Pan_mogg.js 的选择器)
|
||||
if posterURL, exists := doc.Find(".mobile-play .lazyload").Attr("data-src"); exists && posterURL != "" {
|
||||
images = append(images, posterURL)
|
||||
}
|
||||
|
||||
// 查找下载链接区域
|
||||
doc.Find("#download-list .module-row-one").Each(func(i int, s *goquery.Selection) {
|
||||
// 从data-clipboard-text属性提取链接
|
||||
if linkURL, exists := s.Find("[data-clipboard-text]").Attr("data-clipboard-text"); exists {
|
||||
// 过滤掉无效链接
|
||||
if p.isValidNetworkDriveURL(linkURL) {
|
||||
if linkType := p.determineLinkType(linkURL); linkType != "" {
|
||||
link := model.Link{
|
||||
Type: linkType,
|
||||
URL: linkURL,
|
||||
Password: "", // 大部分网盘不需要密码
|
||||
}
|
||||
links = append(links, link)
|
||||
}
|
||||
}
|
||||
}
|
||||
|
||||
// 也检查直接的href属性
|
||||
s.Find("a[href]").Each(func(j int, a *goquery.Selection) {
|
||||
if linkURL, exists := a.Attr("href"); exists {
|
||||
// 过滤掉无效链接
|
||||
if p.isValidNetworkDriveURL(linkURL) {
|
||||
if linkType := p.determineLinkType(linkURL); linkType != "" {
|
||||
// 避免重复添加
|
||||
isDuplicate := false
|
||||
for _, existingLink := range links {
|
||||
if existingLink.URL == linkURL {
|
||||
isDuplicate = true
|
||||
break
|
||||
}
|
||||
}
|
||||
|
||||
if !isDuplicate {
|
||||
link := model.Link{
|
||||
Type: linkType,
|
||||
URL: linkURL,
|
||||
Password: "",
|
||||
}
|
||||
links = append(links, link)
|
||||
}
|
||||
}
|
||||
}
|
||||
}
|
||||
})
|
||||
})
|
||||
|
||||
return links, images
|
||||
}
|
||||
|
||||
// fetchDetailLinks 获取详情页的下载链接（兼容性方法，仅返回链接）
|
||||
func (p *ZhizhenAsyncPlugin) fetchDetailLinks(client *http.Client, itemID string) []model.Link {
|
||||
links, _ := p.fetchDetailLinksAndImages(client, itemID)
|
||||
return links
|
||||
}
|
||||
|
||||
// min 返回两个整数中的较小值
|
||||
func min(a, b int) int {
|
||||
if a < b {
|
||||
return a
|
||||
}
|
||||
return b
|
||||
}
|
||||
|
||||
// GetPerformanceStats 获取性能统计信息
|
||||
func (p *ZhizhenAsyncPlugin) GetPerformanceStats() map[string]interface{} {
|
||||
totalRequests := atomic.LoadInt64(&searchRequests)
|
||||
totalTime := atomic.LoadInt64(&totalSearchTime)
|
||||
|
||||
var avgTime float64
|
||||
if totalRequests > 0 {
|
||||
avgTime = float64(totalTime) / float64(totalRequests) / 1e6 // 转换为毫秒
|
||||
totalSearchRequests := atomic.LoadInt64(&searchRequests)
|
||||
totalDetailRequests := atomic.LoadInt64(&detailPageRequests)
|
||||
totalCacheHits := atomic.LoadInt64(&cacheHits)
|
||||
totalCacheMisses := atomic.LoadInt64(&cacheMisses)
|
||||
totalSearchTime := atomic.LoadInt64(&totalSearchTime)
|
||||
totalDetailTime := atomic.LoadInt64(&totalDetailTime)
|
||||
|
||||
var avgSearchTime, avgDetailTime, cacheHitRate float64
|
||||
if totalSearchRequests > 0 {
|
||||
avgSearchTime = float64(totalSearchTime) / float64(totalSearchRequests) / 1e6 // 转换为毫秒
|
||||
}
|
||||
|
||||
if totalDetailRequests > 0 {
|
||||
avgDetailTime = float64(totalDetailTime) / float64(totalDetailRequests) / 1e6 // 转换为毫秒
|
||||
}
|
||||
if totalCacheHits+totalCacheMisses > 0 {
|
||||
cacheHitRate = float64(totalCacheHits) / float64(totalCacheHits+totalCacheMisses) * 100
|
||||
}
|
||||
|
||||
return map[string]interface{}{
|
||||
"search_requests": totalRequests,
|
||||
"avg_search_time_ms": avgTime,
|
||||
"total_search_time_ns": totalTime,
|
||||
"search_requests": totalSearchRequests,
|
||||
"detail_page_requests": totalDetailRequests,
|
||||
"cache_hits": totalCacheHits,
|
||||
"cache_misses": totalCacheMisses,
|
||||
"cache_hit_rate": cacheHitRate,
|
||||
"avg_search_time_ms": avgSearchTime,
|
||||
"avg_detail_time_ms": avgDetailTime,
|
||||
"total_search_time_ns": totalSearchTime,
|
||||
"total_detail_time_ns": totalDetailTime,
|
||||
}
|
||||
}
|
||||
Reference in New Issue
Block a user