Update: Docs

Author: coder-hxl

README.md

@@ -104,7 +104,7 @@ const myXCrawl = xCrawl({
 */
 myXCrawl.startPolling({ d: 1 }, async (count, stopPolling) => {
   // Call crawlPage API to crawl Page
-  const { jsdom } = await myXCrawl.crawlPage('https://zh.airbnb.com/s/*/plus_homes')
+  const { jsdom, page } = await myXCrawl.crawlPage('https://zh.airbnb.com/s/*/plus_homes')
 
   // Get the cover image elements for Plus listings
   const imgEls = jsdom.window.document.querySelector('.a1stauiv')?.querySelectorAll('picture img')
@@ -115,6 +115,9 @@ myXCrawl.startPolling({ d: 1 }, async (count, stopPolling) => {
 
   // Call the crawlFile API to crawl pictures
   myXCrawl.crawlFile({ requestConfig, fileConfig: { storeDir: './upload' } })
+
+  // Close page
+  page.close()
 })
 ```
 
@@ -218,6 +221,8 @@ The browser instance is a headless browser without a UI shell. What he does is t
 
 It is an instance object of [Page](https://pptr.dev/api/puppeteer.page). The instance can also perform interactive operations such as events. For specific usage, please refer to [page](https://pptr.dev /api/puppeteer. page).
 
+The browser instance will retain a reference to the page instance. If it is no longer used in the future, the page instance needs to be closed by itself, otherwise it will cause a memory leak.
+
 **Parse the page by yourself**
 
 Take the jsdom library as an example:
@@ -323,9 +328,12 @@ myXCrawl.startPolling({ h: 2, m: 30 }, async (count, stopPolling) => {
   // will be executed every two and a half hours
   // crawlPage/crawlData/crawlFile
   const { jsdom, browser, page } = await myXCrawl.crawlPage('https://xxx.com')
+  page.close()
 })
 ```
 
+**Using crawlPage in polling Note:** Calling page.close() is to prevent the browser instance from retaining references to the page instance. If it is no longer used in the future, you need to close the page instance yourself, otherwise it will cause memory leaks.
+
 Callback function parameters:
 
 - The count attribute records the current number of polling operations.

docs/cn.md

@@ -99,9 +99,9 @@ const myXCrawl = xCrawl({
 
 // 3.设置爬取任务
 // 调用 startPolling API 开始轮询功能，每隔一天会调用回调函数
-myXCrawl.startPolling({ d: 1 }, async () => {
+myXCrawl.startPolling({ d: 1 }, async (count, stopPolling) => {
   // 调用 crawlPage API 爬取 Page
-  const { jsdom } = await myXCrawl.crawlPage('https://www.bilibili.com/guochuang/')
+  const { jsdom, page } = await myXCrawl.crawlPage('https://www.bilibili.com/guochuang/')
 
   // 获取轮播图片元素
   const imgEls = jsdom.window.document.querySelectorAll('.chief-recom-item img')
@@ -112,6 +112,9 @@ myXCrawl.startPolling({ d: 1 }, async () => {
 
   // 调用 crawlFile API 爬取图片
   myXCrawl.crawlFile({ requestConfig, fileConfig: { storeDir: './upload' } })
+
+  // 关闭页面
+  page.close()
 })
 ```
 
@@ -213,6 +216,8 @@ browser 实例他是个无头浏览器，并无 UI 外壳，他做的是将浏
 
 它是 [Page](https://pptr.dev/api/puppeteer.page) 的实例对象，实例还可以做事件之类的交互操作，具体使用可以参考 [page](https://pptr.dev/api/puppeteer.page) 。
 
+browser 实例内部会保留着对 page 实例的引用，如果后续不再使用需要自行关闭 page 实例，否则会造成内存泄露。
+
 **自行解析页面**
 
 以使用 jsdom 库为例：
@@ -317,9 +322,12 @@ myXCrawl.startPolling({ h: 2, m: 30 }, async (count, stopPolling) => {
   // 每隔两个半小时会执行一次
   // crawlPage/crawlData/crawlFile
   const { jsdom, browser, page } = await myXCrawl.crawlPage('https://xxx.com')
+  page.close()
 })
 ```
 
+**在轮询中使用 crawlPage 注意：** 调用 page.close() 是为了防止 browser 实例内部还保留着对 page 实例的引用，如果后续不再使用需要自行关闭 page 实例，否则会造成内存泄露。
+
 回调函数参数：
 
 - count 属性记录当前是第几次轮询操作。

publish/README.md

@@ -104,7 +104,7 @@ const myXCrawl = xCrawl({
 */
 myXCrawl.startPolling({ d: 1 }, async (count, stopPolling) => {
   // Call crawlPage API to crawl Page
-  const { jsdom } = await myXCrawl.crawlPage('https://zh.airbnb.com/s/*/plus_homes')
+  const { jsdom, page } = await myXCrawl.crawlPage('https://zh.airbnb.com/s/*/plus_homes')
 
   // Get the cover image elements for Plus listings
   const imgEls = jsdom.window.document.querySelector('.a1stauiv')?.querySelectorAll('picture img')
@@ -115,6 +115,9 @@ myXCrawl.startPolling({ d: 1 }, async (count, stopPolling) => {
 
   // Call the crawlFile API to crawl pictures
   myXCrawl.crawlFile({ requestConfig, fileConfig: { storeDir: './upload' } })
+
+  // Close page
+  page.close()
 })
 ```
 
@@ -218,6 +221,8 @@ The browser instance is a headless browser without a UI shell. What he does is t
 
 It is an instance object of [Page](https://pptr.dev/api/puppeteer.page). The instance can also perform interactive operations such as events. For specific usage, please refer to [page](https://pptr.dev /api/puppeteer. page).
 
+The browser instance will retain a reference to the page instance. If it is no longer used in the future, the page instance needs to be closed by itself, otherwise it will cause a memory leak.
+
 **Parse the page by yourself**
 
 Take the jsdom library as an example:
@@ -323,9 +328,12 @@ myXCrawl.startPolling({ h: 2, m: 30 }, async (count, stopPolling) => {
   // will be executed every two and a half hours
   // crawlPage/crawlData/crawlFile
   const { jsdom, browser, page } = await myXCrawl.crawlPage('https://xxx.com')
+  page.close()
 })
 ```
 
+**Using crawlPage in polling Note:** Calling page.close() is to prevent the browser instance from retaining references to the page instance. If it is no longer used in the future, you need to close the page instance yourself, otherwise it will cause memory leaks.
+
 Callback function parameters:
 
 - The count attribute records the current number of polling operations.