一 前言

在WebView加载网页的过程中，有时会页面加载不出来的情况，如下：

error

那么如何知道网页加载错误了呢？

二 解决方法

webview.webViewClient = object : WebViewClient() {
    override fun onPageFinished(view: WebView?, url: String?) {
        super.onPageFinished(view, url)
        Log.d(TAG, "onPageFinished: ")
    }

    override fun onReceivedError(
        view: WebView?,
        request: WebResourceRequest?,
        error: WebResourceError?
    ) {
        super.onReceivedError(view, request, error)
        Log.d(TAG, "onReceivedError: request= " + request.json())
        Log.d(TAG, "onReceivedError: error= " + error.json())
    }
}
webview.loadUrl("https://www.baidu.com/")

request输出：

{"a":{"a":"https://www.baidu.com/","b":true,"c":false,"d":false,"e":"GET","f":{"User-Agent":"Mozilla/5.0 (Linux; Android 11; sdk_gphone_x86_arm Build/RSR1.200819.001.A1; wv) AppleWebKit/537.36 (KHTML, like Gecko) Version/4.0 Chrome/83.0.4103.106 Mobile Safari/537.36","Accept":"text/html,application/xhtml+xml,application/xml;q\u003d0.9,image/webp,image/apng,/;q\u003d0.8,application/signed-exchange;v\u003db3;q\u003d0.9","Upgrade-Insecure-Requests":"1"}}}

error输出：

{"a":{"a":-1,"b":"net::ERR_CACHE_MISS"}}

/**
* 需要注意的是：与老版本不同，新版本已经针对所有的资源加载错误都会回调*(iframe, image, etc.), 不仅仅是加载的主页面
 */
public void onReceivedError(WebView view, WebResourceRequest request, WebResourceError error) {
    if (request.isForMainFrame()) {
        onReceivedError(view,
                error.getErrorCode(), error.getDescription().toString(),
                request.getUrl().toString());
    }
}

所以，通过以上监听，即可知道对应的错误内容，进而做相应的业务处理。

三 常见的错误案例分析

1.错误：ERR_UNKNOWN_URL_SCHEME

复现条件:打开百度新闻资讯网页，并点击评论

error.gif

来看一下页面加载shouldOverrideUrlLoading 中回调request输出：

{"a":{"a":"baiduboxapp://v1/easybrowse/hybrid?upgrade\u003d1\u0026type\u003dhybrid\u0026tpl_id\u003dlanding_app.html\u0026newbrowser\u003d1\u0026style\u003d%7B%22toolbaricons%22%3A%7B%22tids%22%3A%5B%224%22%2C%221%22%2C%222%22%2C%223%22%5D%2C%22menumode%22%3A%222%22%2C%22actionBarConfig%22%3A%7B%22extCase%22%3A%220%22%7D%7D%7D\u0026slog\u003d%7B%22from%22%3A%22feed%22%7D\u0026context\u003d%7B%22nid%22%3A%22news_9777178803801641010%22%7D\u0026ch_url\u003dhttps%3A%2F%2Fmbd.baidu.com%2Fnewspage%2Fdata%2Flandingreact%3FpageType%3D2%26nid%3Dnews_9777178803801641010%26uk%3D%26sourceFrom%3DlandingShare\u0026commentInfo\u003d%7B%22topic_id%22%3A1066000039731532%2C%22opentype%22%3A2%7D\u0026logargs\u003d%7B%22source%22%3A%221020283l%22%2C%22channel%22%3A%221020283m%22%7D\u0026needlog\u003d1","b":false,"c":true,"d":false,"e":"GET","f":{"Accept":"text/html,application/xhtml+xml,application/xml;q\u003d0.9,image/avif,image/webp,image/apng,/;q\u003d0.8,application/signed-exchange;v\u003db3;q\u003d0.9","Referer":"https://mbd.baidu.com/newspage/data/landingpage?s_type\u003dnews\u0026dsp\u003dwise\u0026context\u003d%7B%22nid%22%3A%22news_9777178803801641010%22%7D\u0026pageType\u003d1\u0026n_type\u003d1\u0026p_from\u003d-1\u0026innerIframe\u003d1","Upgrade-Insecure-Requests":"1"}}}

可以发现协议不是http类开头，而是百度自己定义的"baiduboxapp"。因此，不做处理，当然是识别不了的。

2.实现UC、华为浏览器处理的效果——>无视

webview.webViewClient = object : WebViewClient() {
            override fun shouldOverrideUrlLoading(
                view: WebView?,
                request: WebResourceRequest?
            ): Boolean {
                Log.d(TAG, "shouldOverrideUrlLoading: ${request.json()}")
                if (Build.VERSION.SDK_INT >= Build.VERSION_CODES.LOLLIPOP) {
                    Log.d(TAG, "shouldOverrideUrlLoading: url->${request?.url}")
                    if (view != null && request?.url?.toString() != null) {
                        if (request.url.toString().startsWith("http")) {
                            view.loadUrl(request.url.toString())
                        }
                return true
            }

3.如何实现Chrome浏览器处理的效果——>能跳转

假定这是百度适配Chrome的效果，而不是Chrome主动去适配百度的代码页面，那么说明客户端的WebView也是能做到这样效果的。

基于此，我们来urldecode一下协议内容来看看request有什么特殊：

baiduboxapp://v1/easybrowse/hybrid?upgrade=1&type=hybrid&tpl_id=landing_app.html&newbrowser=1&style={"toolbaricons":{"tids":["4","1","2","3"],"menumode":"2","actionBarConfig":{"extCase":"0"}}}&slog={"from":"feed"}&context={"nid":"news_9777178803801641010"}&ch_url=https://mbd.baidu.com/newspage/data/landingreact?pageType=2&nid=news_9777178803801641010&uk=&sourceFrom=landingShare&commentInfo={"topic_id":1066000039731532,"opentype":2}&logargs={"source":"1020283l","channel":"1020283m"}&needlog=1

整体来看，仍旧是符合URI规则的。这样的话就可以考虑使用Intent处理的方式了。

webview.webViewClient = object : WebViewClient() {
    override fun shouldOverrideUrlLoading(
        view: WebView?,
        request: WebResourceRequest?
    ): Boolean {
        Log.d(TAG, "shouldOverrideUrlLoading: ${request.json()}")
        if (Build.VERSION.SDK_INT >= Build.VERSION_CODES.LOLLIPOP) {
            Log.d(TAG, "shouldOverrideUrlLoading: url->${request?.url}")
            if (view != null && request?.url?.toString() != null) {
                if (request.url.toString().startsWith("http")) {
                    view.loadUrl(request.url.toString())
                } else {
                    try {
                        Log.d(TAG, "shouldOverrideUrlLoading: 开始跳转")
                        startActivity(Intent(Intent.ACTION_VIEW, request.url))
                    } catch (e: Exception) {
                        Log.d(TAG, "shouldOverrideUrlLoading: ${e.json()}")
                    }
                }
            }
        }

        return true
    }

运行查看效果如下：

resolved.gif

从结果来看，能跳转至百度的App，假定成立。
也就是针对非http协议地址的情况，可使用 startActivity(Intent(Intent.ACTION_VIEW, request.url)) 之类来进行协议内容处理。

四 总结

最后，根据自家业务来定协议并处理，毕竟不是谁都愿意自家用户跳出自家App的。