qq_bot/util/web_page_shot.go
2024-10-20 01:27:15 +08:00

143 lines
4.7 KiB
Go

package util
import (
"context"
"fmt"
"log"
"os"
"time"
"github.com/chromedp/cdproto/emulation"
"github.com/chromedp/cdproto/network"
"github.com/chromedp/cdproto/page"
"github.com/chromedp/chromedp"
)
// ScreenshotURL 截图函数:传入网址、输出路径、宽高、四个边距和等待的元素 ID
func ScreenshotURL(url, output string, width, height int, marginTop, marginBottom, marginRight, marginLeft int, waitClass string) error {
// 创建一个上下文,连接到 Docker 中运行的 headless-shell 实例
remoteAllocatorCtx, cancel := chromedp.NewRemoteAllocator(
context.Background(), "ws://100.124.180.117:9222/json/ws",
)
defer cancel()
ctx, cancel := chromedp.NewContext(remoteAllocatorCtx)
defer cancel()
// 设置超时时间,避免长时间无响应
ctx, cancel = context.WithTimeout(ctx, 60*time.Second)
defer cancel()
// 设置页面的宽高和缩放
if err := chromedp.Run(ctx, setViewport(width, height)); err != nil {
return fmt.Errorf("设置页面大小失败: %w", err)
}
// 启用网络请求拦截
if err := chromedp.Run(ctx, enableRequestInterception()); err != nil {
return fmt.Errorf("启用请求拦截失败: %w", err)
}
// 用于存储截图的变量
var screenshot []byte
queryAction := chromedp.WaitVisible(fmt.Sprintf(".%s", waitClass), chromedp.ByQuery)
if waitClass == "" {
queryAction = chromedp.WaitVisible(`body`, chromedp.ByQuery)
}
// 设置自定义请求头
headers := map[string]interface{}{
"accept": "text/html,application/xhtml+xml,application/xml;q=0.9,image/avif,image/webp,image/apng,*/*;q=0.8,application/signed-exchange;v=b3;q=0.7",
"accept-language": "zh-CN,zh;q=0.9,en;q=0.8,en-GB;q=0.7,en-US;q=0.6",
"cache-control": "max-age=0",
"priority": "u=0, i",
"sec-ch-ua": `"Microsoft Edge";v="129", "Not=A?Brand";v="8", "Chromium";v="129"`,
"sec-ch-ua-mobile": "?0",
"sec-ch-ua-platform": `"Windows"`,
"sec-fetch-dest": "document",
"sec-fetch-mode": "navigate",
"sec-fetch-site": "same-origin",
"sec-fetch-user": "?1",
"upgrade-insecure-requests": "1",
"user-agent": "Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/129.0.0.0 Safari/537.36 Edg/129.0.0.0",
}
// 启用网络功能并设置请求头
if err := chromedp.Run(ctx, network.Enable()); err != nil {
return fmt.Errorf("启用网络失败: %w", err)
}
if err := chromedp.Run(ctx, network.SetExtraHTTPHeaders(network.Headers(headers))); err != nil {
return fmt.Errorf("设置请求头失败: %w", err)
}
// 执行任务:打开网页并截图
err := chromedp.Run(ctx,
chromedp.Navigate(url), // 打开网页
ignoreErrors(queryAction), // 等待指定元素
chromedp.Sleep(10*time.Second),
chromedp.ActionFunc(func(ctx context.Context) error { // 自定义截图逻辑
_, _, _, _, cssLayoutViewport, _, _ := page.GetLayoutMetrics().Do(ctx)
if width == 0 || height == 0 {
width = int(cssLayoutViewport.ClientWidth)
height = int(cssLayoutViewport.ClientHeight)
}
// 计算调整后的截图区域
clip := &page.Viewport{
X: float64(marginLeft),
Y: float64(marginTop),
Width: float64(width - marginLeft - marginRight),
Height: float64(height - marginTop - marginBottom),
Scale: 1.0,
}
var err error
screenshot, err = page.CaptureScreenshot().WithClip(clip).Do(ctx)
return err
}),
)
if err != nil {
return fmt.Errorf("截图失败: %w", err)
}
// 保存截图到本地
if err := os.WriteFile(output, screenshot, 0644); err != nil {
return fmt.Errorf("保存图片失败: %w", err)
}
return nil
}
func setViewport(width, height int) chromedp.Tasks {
return chromedp.Tasks{
emulation.SetDeviceMetricsOverride(int64(width), int64(height), 1.0, false).
WithScreenOrientation(&emulation.ScreenOrientation{
Type: emulation.OrientationTypePortraitPrimary,
Angle: 0,
}),
}
}
// enableRequestInterception 启用网络请求拦截,忽略不必要的资源加载(如广告、图片等)
func enableRequestInterception() chromedp.Tasks {
return chromedp.Tasks{
chromedp.ActionFunc(func(ctx context.Context) error {
return network.Enable().Do(ctx)
}),
network.SetBlockedURLS([]string{
"pagead2.googlesyndication.com",
"optimizationguide-pa.googleapis.com",
"static.cloudflareinsights.com",
}),
}
}
// ignoreErrors 包裹 chromedp 任务,忽略执行过程中出现的错误
func ignoreErrors(task chromedp.Action) chromedp.ActionFunc {
return chromedp.ActionFunc(func(ctx context.Context) error {
err := task.Do(ctx)
if err != nil {
log.Printf("忽略错误: %v", err)
}
return nil
})
}