feature : 删除知乎平台特有的的zhida超链接和跳转中转链接

packages/core/src/lib/turndown.ts

@@ -14,6 +14,18 @@ import { createLogger } from './logger'
 
 const logger = createLogger('Turndown')
 
+// ============ 源平台链接清理规则 ============
+
+/** 需要去除的站内链接域名（去掉 <a> 只保留文本） */
+export const SOURCE_LINK_REMOVE_DOMAINS = [
+  'zhida.zhihu.com',
+]
+
+/** 跳转中转规则：domain → 真实 URL 所在的 query 参数名 */
+export const SOURCE_LINK_REDIRECT_RULES: Array<{ domain: string; param: string }> = [
+  { domain: 'link.zhihu.com', param: 'target' },
+]
+
 // ============ HTML 实体解码工具 ============
 
 /**
@@ -443,6 +455,31 @@ function addExtensionRules(turndownService: TurndownService): void {
     }
   })
 
+  // 源平台链接清理（站内链接去除、跳转中转还原）
+  turndownService.addRule('sourcePlatformLinks', {
+    filter: function(node) {
+      if (node.nodeName !== 'A') return false
+      const href = (node as Element).getAttribute('href') || ''
+      return SOURCE_LINK_REMOVE_DOMAINS.some(d => href.includes(d))
+        || SOURCE_LINK_REDIRECT_RULES.some(r => href.includes(r.domain))
+    },
+    replacement: function(content, node) {
+      const href = (node as Element).getAttribute('href') || ''
+      if (SOURCE_LINK_REMOVE_DOMAINS.some(d => href.includes(d))) {
+        return content
+      }
+      const rule = SOURCE_LINK_REDIRECT_RULES.find(r => href.includes(r.domain))
+      if (rule) {
+        try {
+          const url = new URL(href)
+          const real = url.searchParams.get(rule.param)
+          if (real) return '[' + content + '](' + real + ')'
+        } catch {}
+      }
+      return '[' + content + '](' + href + ')'
+    }
+  })
+
   // 保留没有表头的表格（作为 HTML）
   turndownService.keep(function(node) {
     try {

packages/extension/src/lib/content-processor.ts

@@ -9,7 +9,7 @@
  * 5. Service Worker 只做图片上传 + 调用 API
  */
 
-import { htmlToMarkdownNative, type PreprocessConfig } from '@wechatsync/core'
+import { htmlToMarkdownNative, SOURCE_LINK_REMOVE_DOMAINS, SOURCE_LINK_REDIRECT_RULES, type PreprocessConfig } from '@wechatsync/core'
 import { createLogger } from './logger'
 
 const logger = createLogger('ContentProcessor')
@@ -77,6 +77,9 @@ export function preprocessForPlatform(rawHtml: string, config: PreprocessConfig)
   // 移除 script 和 noscript（总是执行），style 根据配置决定
   removeElements(container, config.keepStyles ? ['script', 'noscript'] : ['script', 'style', 'noscript'])
 
+  // 清理源平台链接（固定步骤，在 removeLinks 之前执行）
+  cleanSourcePlatformLinks(container)
+
   if (config.removeLinks) {
     processLinks(container, config.keepLinkDomains)
   }
@@ -235,6 +238,34 @@ function processSvgImages(container: HTMLElement): void {
   })
 }
 
+/**
+ * 清理源平台链接（站内链接去除、跳转中转还原）
+ * 规则定义在 @wechatsync/core SOURCE_LINK_*_DOMAINS，新增平台只需加域名
+ */
+function cleanSourcePlatformLinks(container: HTMLElement): void {
+  const links = Array.from(container.querySelectorAll('a'))
+  for (const link of links) {
+    const href = link.getAttribute('href') || ''
+    if (SOURCE_LINK_REMOVE_DOMAINS.some(d => href.includes(d))) {
+      const parent = link.parentNode
+      if (!parent) continue
+      while (link.firstChild) {
+        parent.insertBefore(link.firstChild, link)
+      }
+      parent.removeChild(link)
+      continue
+    }
+    const rule = SOURCE_LINK_REDIRECT_RULES.find(r => href.includes(r.domain))
+    if (rule) {
+      try {
+        const url = new URL(href)
+        const real = url.searchParams.get(rule.param)
+        if (real) link.setAttribute('href', real)
+      } catch {}
+    }
+  }
+}
+
 /**
  * 处理链接
  */