用casperjs刮文本节点的最快方法

Question

I have this structure and I need get text from plain text node like this

<strong><font color="#666666">Phones:</font></strong>
<br>
<br>
<img src="/image/fgh.jpg" title="Velcom" alt="Velcom" style="margin: 2 5 -3 5;">
"+375 29"              //get this
<b>611 77 83</b>      //and this

I try to use XPath copied from chrome console

casper.thenOpen('url', function() {
    result = this.getElementInfo(x('//*[@id="main_content"]/table[2]/tbody/tr[17]/td/table/tbody/tr/td[1]/p[1]/text()[3]'));
});

casper.then(function() {
    this.echo(result.text);
});

but it is not working. Also when I try result.data

console.log(this.getElementInfo(x('//*[@id="main_content"]/table[2]/tbody/tr[17]/td/table/tbody/tr/td[1]/p[1]/text()[3]')));

returns null , but this element exists in the page, I checked it out

Answer 1

Make sure you have included:

var x = require('casper').selectXPath;

If that is still not working the following will retrieve all text from page then you can parse. This is not recommended for performance but does work if you have anchor text to parse on. You will need to slightly modify.

var casper = require("casper").create ({
    waitTimeout: 15000,
    stepTimeout: 15000,
    verbose: true,
    viewportSize: {
        width: 1400,
        height: 768
    },
    onWaitTimeout: function() {
        logConsole('Wait TimeOut Occured');
        this.capture('xWait_timeout.png');
        this.exit();
    },
    onStepTimeout: function() {
        logConsole('Step TimeOut Occured');
        this.capture('xStepTimeout.png');
        this.exit();
    }
});

casper.on('remote.message', function(msg) {
    logConsole('***remote message caught***: ' + msg);
});

casper.userAgent('Mozilla/5.0 (Macintosh; Intel Mac OS X 10_7_5) AppleWebKit/537.4 (KHTML, like Gecko) Chrome/22.0.1229.94 Safari/537.4');

// vars
var gUrl           = 'WebAddy'; //+++ Update URL

casper.start(gUrl, function() {
  var tPlainText = this.evaluate(function() {

    var bodyText        = document.body;
    var textContent     = bodyText.textContent || bodyText.innerText;
    var tCheck          = textContent.indexOf("Phones:");

    if (tCheck === -1) {
      tPlainText = 'Phone Text Not Found';
        return tPlainText;
    } else {
      // parse text
      var tSplit              = textContent.split('Phones:');
      var tStr                = (tSplit[1]) ? tSplit[1] : '';
      var tPos1               = tStr.indexOf(''); //+++ insert text to stop parse 
      var tDesiredText         = (tPos1 !== -1) ? tStr.substring(0, tPos1) : null;

        return tDesiredText;
    }
  });
  console.log(tPlainText);
});

casper.run();

Answer 2

An old question but I had the same issue. I need to get the following text, so here is how I did it.

__utils__.getElementByXPath("//bla...bla/following-sibling::node()").textContent;