新增插件kkmao

2026-05-06 21:51:31 +08:00 · 2025-11-26 18:49:45 +08:00
parent b01da5847e
commit 5bd5e26805
3 changed files with 557 additions and 0 deletions
--- a/plugin/kkmao/html结构分析.md
+++ b/plugin/kkmao/html结构分析.md
@@ -0,0 +1,159 @@
+# kkmao (夸克猫) HTML结构分析
+
+## 网站信息
+- **网站名称**: 夸克猫资源
+- **域名**: `www.kuakemao.com`
+- **类型**: 夸克网盘影视资源分享站（WordPress 主题站）
+- **特点**: 每篇文章提供 1~N 个夸克网盘链接，正文结构高度统一，仅包含夸克网盘
+
+## 搜索页结构
+
+### 1. 搜索入口
+```
+https://www.kuakemao.com/?s={关键词}
+
+示例:
+https://www.kuakemao.com/?s=物
+```
+- 直接使用 UTF-8 中文或 URL 编码均可
+- 页面为标准 WordPress 搜索结果页
+
+### 2. 结果容器
+- **父容器**: `section.container > div.content-wrap > div.content`
+- **结果项**: `article.excerpt`（会附带 `excerpt-1/2` 等序号类名）
+
+### 3. 单个结果结构
+
+#### 封面/详情链接
+```html
+<a class="focus" href="https://www.kuakemao.com/653.html">
+    <img data-src="https://img.kuakemao.com/.../c4ac4195bed96c7-220x150.webp" class="thumb">
+</a>
+```
+- `href` 即详情页地址，形如 `/数字.html`
+
+#### 标题
+```html
+<header>
+  <h2>
+    <a href="https://www.kuakemao.com/653.html"
+       title="某种物质 (2024) 夸克网盘 法国 恐怖 4K 豆瓣7.5 - 夸克猫资源">
+       某种物质 (2024) 夸克网盘 法国 恐怖 4K 豆瓣7.5
+    </a>
+  </h2>
+</header>
+```
+- 提取要素:
+  - **标题**: `h2 > a` 文本
+  - **详情页 URL**: `h2 > a` 的 `href`
+
+#### 简介
+```html
+<p class="note">
+  某种物质 夸克网盘资源 https://pan.quark.cn/s/631243a6189a ... 
+</p>
+```
+- 用于填充 `SearchResult.Content`
+- 文本中偶尔包含裸露的夸克链接，但仍需访问详情页获取规范链接
+
+#### 元数据
+```html
+<div class="meta">
+  <time>2025-11-26</time>
+  <a class="cat" href="https://www.kuakemao.com/dy">电影</a>
+  <span class="pv">阅读(...)</span>
+</div>
+```
+- **发布时间**: `<time>` 文本（`YYYY-MM-DD`）
+- **分类标签**: `.meta a.cat` 文本
+
+## 详情页结构
+
+### 1. URL 规则
+```
+https://www.kuakemao.com/{文章ID}.html
+示例: https://www.kuakemao.com/653.html
+```
+- 文章 ID 可由 `/{id}.html` 提取，用于唯一 ID
+
+### 2. 主要节点
+- **标题**: `.article-title`
+- **元信息**: `.article-meta .item`（日期、分类、阅读数等）
+- **正文容器**: `.article-content`
+
+### 3. 夸克链接位置
+```html
+<div class="article-content">
+  <h2>某种物质 夸克网盘资源</h2>
+  <p>
+    <a rel="nofollow" href="https://pan.quark.cn/s/631243a6189a" target="_blank">
+      https://pan.quark.cn/s/631243a6189a
+    </a>
+  </p>
+  ...
+</div>
+```
+- 所有下载链接位于 `.article-content` 中
+- 仅出现夸克域名 (`pan.quark.cn`)
+- 提取码通常在链接同一段落后续文字，需解析 `提取码/密码/pwd/code` 关键词
+
+## CSS 选择器速查表
+
+| 数据项 | 选择器 / 规则 | 备注 |
+|--------|---------------|------|
+| 结果列表 | `article.excerpt` | 遍历搜索结果 |
+| 标题 | `article.excerpt h2 a` | 文本 & `href` |
+| 简介 | `article.excerpt p.note` | 文本描述 |
+| 分类 | `article.excerpt .meta a.cat` | 可能 0/1 个 |
+| 发布时间 | `article.excerpt .meta time` | `YYYY-MM-DD` |
+| 详情正文 | `.article-content` | 包含所有下载信息 |
+| 夸克链接 | `.article-content a[href*="pan.quark.cn"]` | href 即下载地址 |
+| 提取码 | 链接文本 / 父节点文本 | 关键词：`提取码/密码/pwd/code` |
+
+## 实现要点
+
+1. **请求策略**
+   - 搜索页：`GET https://www.kuakemao.com/?s=关键词`
+   - 设置常规浏览器 UA、Referer，必要时加入重试
+2. **列表解析**
+   - 遍历 `article.excerpt`，提取标题、摘要、分类、时间
+   - 由详情 URL 提取 `articleID` 作为唯一后缀
+3. **详情页抓取**
+   - 进入 `.article-content`，收集 `a[href*="pan.quark.cn"]`
+   - 一篇可能提供多条夸克链接，需要全部返回
+   - 通过父节点/兄弟文本匹配提取码
+4. **链接过滤**
+   - 本站只提供夸克网盘，其他域名全部忽略
+5. **结果构建**
+   - `UniqueID = kkmao-{articleID}`
+   - `Channel` 置空
+   - `Datetime` 使用搜索结果页的 `<time>`（格式 `2006-01-02`）
+   - `Links` 仅包含 `Type="quark"` 的条目
+
+## 示例流程
+```
+关键词: 物
+↓
+搜索页: https://www.kuakemao.com/?s=物
+  - 解析 article.excerpt
+  - 取得标题「某种物质 (2024)...」、详情链接 https://www.kuakemao.com/653.html
+↓
+详情页: https://www.kuakemao.com/653.html
+  - 在 .article-content 中找到 <a href="https://pan.quark.cn/s/631243a6189a">
+↓
+结果:
+  UniqueID: kkmao-653
+  Title: 某种物质 (2024) 夸克网盘 法国 恐怖 4K 豆瓣7.5
+  Content: 搜索结果页的摘要
+  Links: [{Type:"quark", URL:"https://pan.quark.cn/s/631243a6189a", Password:""}]
+  Tags: ["电影"]
+  Datetime: 2025-11-26
+```
+
+## 注意事项
+1. 搜索页的 `<time>` 可能缺失，需兜底为当前时间
+2. `.note` 中的裸露链接可忽略，以详情页数据为准
+3. 页面加载较快，但仍建议设置 10~12 秒超时与 2~3 次重试
+4. 站点仅有夸克网盘，插件实现时可直接过滤其它域名
+5. 文章正文含大量 `<h2>` 与 `<pre>`，解析提取码时需遍历父节点文本，避免遗漏
+
--- a/plugin/kkmao/kkmao.go
+++ b/plugin/kkmao/kkmao.go
@@ -0,0 +1,397 @@
+package kkmao
+
+import (
+	"context"
+	"fmt"
+	"net/http"
+	"net/url"
+	"regexp"
+	"strings"
+	"sync"
+	"time"
+
+	"github.com/PuerkitoBio/goquery"
+
+	"pansou/model"
+	"pansou/plugin"
+)
+
+var (
+	articleIDRegex = regexp.MustCompile(`/(\d+)\.html`)
+	quarkRegex     = regexp.MustCompile(`https?://pan\.quark\.cn/s/[0-9A-Za-z]+`)
+	pwdPatterns    = []*regexp.Regexp{
+		regexp.MustCompile(`提取码[:：]?\s*([0-9A-Za-z]+)`),
+		regexp.MustCompile(`密码[:：]?\s*([0-9A-Za-z]+)`),
+		regexp.MustCompile(`pwd\s*[=:：]\s*([0-9A-Za-z]+)`),
+		regexp.MustCompile(`code\s*[=:：]\s*([0-9A-Za-z]+)`),
+	}
+	detailCache = sync.Map{}
+
+	cacheTTL             = 1 * time.Hour
+	cacheCleanupInterval = 30 * time.Minute
+)
+
+type detailCacheEntry struct {
+	links     []model.Link
+	expiresAt time.Time
+}
+
+const (
+	pluginName            = "kkmao"
+	defaultPriority       = 2
+	searchTimeout         = 12 * time.Second
+	detailTimeout         = 10 * time.Second
+	maxConcurrency        = 8
+	maxIdleConns          = 64
+	maxIdlePerHost        = 8
+	maxConnsPerHost       = 32
+	idleConnLifetime      = 90 * time.Second
+	tlsHandshakeTimeout   = 10 * time.Second
+	expectContinueTimeout = 1 * time.Second
+
+	searchMaxRetries = 3
+	detailMaxRetries = 2
+	retryBaseDelay   = 200 * time.Millisecond
+)
+
+// KkMaoPlugin 夸克猫插件
+type KkMaoPlugin struct {
+	*plugin.BaseAsyncPlugin
+	client *http.Client
+}
+
+func init() {
+	plugin.RegisterGlobalPlugin(NewKkMaoPlugin())
+	go startDetailCacheCleaner()
+}
+
+// NewKkMaoPlugin 构造函数
+func NewKkMaoPlugin() *KkMaoPlugin {
+	return &KkMaoPlugin{
+		BaseAsyncPlugin: plugin.NewBaseAsyncPlugin(pluginName, defaultPriority),
+		client:          newHTTPClient(),
+	}
+}
+
+// Search 兼容方法
+func (p *KkMaoPlugin) Search(keyword string, ext map[string]interface{}) ([]model.SearchResult, error) {
+	result, err := p.SearchWithResult(keyword, ext)
+	if err != nil {
+		return nil, err
+	}
+	return result.Results, nil
+}
+
+// SearchWithResult 主搜索实现
+func (p *KkMaoPlugin) SearchWithResult(keyword string, ext map[string]interface{}) (model.PluginSearchResult, error) {
+	return p.AsyncSearchWithResult(keyword, p.searchImpl, p.MainCacheKey, ext)
+}
+
+func newHTTPClient() *http.Client {
+	transport := &http.Transport{
+		MaxIdleConns:          maxIdleConns,
+		MaxIdleConnsPerHost:   maxIdlePerHost,
+		MaxConnsPerHost:       maxConnsPerHost,
+		IdleConnTimeout:       idleConnLifetime,
+		TLSHandshakeTimeout:   tlsHandshakeTimeout,
+		ExpectContinueTimeout: expectContinueTimeout,
+		ForceAttemptHTTP2:     true,
+	}
+	return &http.Client{
+		Transport: transport,
+		Timeout:   searchTimeout,
+	}
+}
+
+func (p *KkMaoPlugin) searchImpl(client *http.Client, keyword string, ext map[string]interface{}) ([]model.SearchResult, error) {
+	if p.client != nil {
+		client = p.client
+	}
+
+	searchURL := fmt.Sprintf("https://www.kuakemao.com/?s=%s", url.QueryEscape(keyword))
+	ctx, cancel := context.WithTimeout(context.Background(), searchTimeout)
+	defer cancel()
+
+	req, err := http.NewRequestWithContext(ctx, http.MethodGet, searchURL, nil)
+	if err != nil {
+		return nil, fmt.Errorf("[%s] 创建请求失败: %w", p.Name(), err)
+	}
+
+	setCommonHeaders(req, "https://www.kuakemao.com/")
+
+	resp, err := p.doRequestWithRetry(req, client, searchMaxRetries, retryBaseDelay)
+	if err != nil {
+		return nil, fmt.Errorf("[%s] 搜索请求失败: %w", p.Name(), err)
+	}
+	defer resp.Body.Close()
+
+	if resp.StatusCode != http.StatusOK {
+		return nil, fmt.Errorf("[%s] 搜索返回状态码: %d", p.Name(), resp.StatusCode)
+	}
+
+	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)
+	}
+
+	var (
+		results []model.SearchResult
+		wg      sync.WaitGroup
+		mu      sync.Mutex
+		sem     = make(chan struct{}, maxConcurrency)
+	)
+
+	doc.Find("article.excerpt").Each(func(_ int, item *goquery.Selection) {
+		titleSel := item.Find("header h2 a")
+		title := strings.TrimSpace(titleSel.Text())
+		detailURL, ok := titleSel.Attr("href")
+		if !ok || title == "" || detailURL == "" {
+			return
+		}
+
+		articleID := extractArticleID(detailURL)
+		if articleID == "" {
+			return
+		}
+
+		summary := strings.TrimSpace(item.Find("p.note").Text())
+
+		var tags []string
+		category := strings.TrimSpace(item.Find(".meta a.cat").First().Text())
+		if category != "" {
+			tags = append(tags, category)
+		}
+
+		rawTime := strings.TrimSpace(item.Find(".meta time").Text())
+		publishTime := parsePublishTime(rawTime)
+
+		wg.Add(1)
+		sem <- struct{}{}
+		go func(title, detailURL, articleID, summary string, tags []string, publishTime time.Time) {
+			defer wg.Done()
+			defer func() { <-sem }()
+
+			links := p.fetchDetailLinks(client, detailURL, articleID)
+			if len(links) == 0 {
+				return
+			}
+
+			result := model.SearchResult{
+				UniqueID: fmt.Sprintf("%s-%s", p.Name(), articleID),
+				Title:    title,
+				Content:  summary,
+				Links:    links,
+				Tags:     tags,
+				Channel:  "",
+				Datetime: publishTime,
+			}
+
+			mu.Lock()
+			results = append(results, result)
+			mu.Unlock()
+		}(title, detailURL, articleID, summary, tags, publishTime)
+	})
+
+	wg.Wait()
+
+	return plugin.FilterResultsByKeyword(results, keyword), nil
+}
+
+func extractArticleID(detailURL string) string {
+	matches := articleIDRegex.FindStringSubmatch(detailURL)
+	if len(matches) >= 2 {
+		return matches[1]
+	}
+	return ""
+}
+
+func parsePublishTime(value string) time.Time {
+	value = strings.TrimSpace(value)
+	if value == "" {
+		return time.Now()
+	}
+
+	layouts := []string{
+		"2006-01-02",
+		"2006-01-02 15:04:05",
+		time.RFC3339,
+	}
+
+	for _, layout := range layouts {
+		if t, err := time.Parse(layout, value); err == nil {
+			return t
+		}
+	}
+
+	return time.Now()
+}
+
+func (p *KkMaoPlugin) fetchDetailLinks(client *http.Client, detailURL, articleID string) []model.Link {
+	if cached, ok := detailCache.Load(articleID); ok {
+		if entry, valid := cached.(detailCacheEntry); valid {
+			if time.Now().Before(entry.expiresAt) && len(entry.links) > 0 {
+				return entry.links
+			}
+			detailCache.Delete(articleID)
+		}
+	}
+
+	ctx, cancel := context.WithTimeout(context.Background(), detailTimeout)
+	defer cancel()
+
+	req, err := http.NewRequestWithContext(ctx, http.MethodGet, detailURL, nil)
+	if err != nil {
+		return nil
+	}
+	setCommonHeaders(req, detailURL)
+
+	resp, err := p.doRequestWithRetry(req, client, detailMaxRetries, retryBaseDelay)
+	if err != nil {
+		return nil
+	}
+	defer resp.Body.Close()
+
+	if resp.StatusCode != http.StatusOK {
+		return nil
+	}
+
+	doc, err := goquery.NewDocumentFromReader(resp.Body)
+	if err != nil {
+		return nil
+	}
+
+	links := extractQuarkLinks(doc)
+	if len(links) > 0 {
+		detailCache.Store(articleID, detailCacheEntry{
+			links:     links,
+			expiresAt: time.Now().Add(cacheTTL),
+		})
+	}
+	return links
+}
+
+func extractQuarkLinks(doc *goquery.Document) []model.Link {
+	var (
+		results []model.Link
+		seen    = make(map[string]struct{})
+	)
+
+	doc.Find(".article-content a[href]").Each(func(_ int, link *goquery.Selection) {
+		href, _ := link.Attr("href")
+		href = strings.TrimSpace(href)
+		if href == "" {
+			return
+		}
+
+		loc := quarkRegex.FindString(href)
+		if loc == "" {
+			return
+		}
+
+		if _, exists := seen[loc]; exists {
+			return
+		}
+
+		password := extractPassword(link)
+
+		results = append(results, model.Link{
+			Type:     "quark",
+			URL:      loc,
+			Password: password,
+		})
+		seen[loc] = struct{}{}
+	})
+
+	return results
+}
+
+func extractPassword(link *goquery.Selection) string {
+	if pwd := matchPassword(link.Text()); pwd != "" {
+		return pwd
+	}
+
+	if title, ok := link.Attr("title"); ok {
+		if pwd := matchPassword(title); pwd != "" {
+			return pwd
+		}
+	}
+
+	if parent := link.Parent(); parent != nil && parent.Length() > 0 {
+		if pwd := matchPassword(parent.Text()); pwd != "" {
+			return pwd
+		}
+		if next := parent.Next(); next.Length() > 0 {
+			if pwd := matchPassword(next.Text()); pwd != "" {
+				return pwd
+			}
+		}
+	}
+
+	if sibling := link.Next(); sibling.Length() > 0 {
+		if pwd := matchPassword(sibling.Text()); pwd != "" {
+			return pwd
+		}
+	}
+
+	return ""
+}
+
+func matchPassword(text string) string {
+	text = strings.TrimSpace(text)
+	if text == "" {
+		return ""
+	}
+
+	for _, pattern := range pwdPatterns {
+		if matches := pattern.FindStringSubmatch(text); len(matches) >= 2 {
+			return strings.TrimSpace(matches[1])
+		}
+	}
+
+	return ""
+}
+
+func setCommonHeaders(req *http.Request, referer string) {
+	req.Header.Set("User-Agent", "Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/120.0.0.0 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", referer)
+}
+
+func (p *KkMaoPlugin) doRequestWithRetry(req *http.Request, client *http.Client, maxRetries int, baseDelay time.Duration) (*http.Response, error) {
+	var lastErr error
+
+	for attempt := 0; attempt < maxRetries; attempt++ {
+		resp, err := client.Do(req.Clone(req.Context()))
+		if err == nil && resp.StatusCode == http.StatusOK {
+			return resp, nil
+		}
+		if resp != nil {
+			resp.Body.Close()
+		}
+		lastErr = err
+		if attempt < maxRetries-1 {
+			backoff := baseDelay * time.Duration(1<<attempt)
+			time.Sleep(backoff)
+		}
+	}
+
+	return nil, fmt.Errorf("重试 %d 次后失败: %w", maxRetries, lastErr)
+}
+
+func startDetailCacheCleaner() {
+	ticker := time.NewTicker(cacheCleanupInterval)
+	defer ticker.Stop()
+
+	for range ticker.C {
+		now := time.Now()
+		detailCache.Range(func(key, value interface{}) bool {
+			entry, ok := value.(detailCacheEntry)
+			if !ok || now.After(entry.expiresAt) {
+				detailCache.Delete(key)
+			}
+			return true
+		})
+	}
+}