Phantomjs: Phantomjs can not show — or some other UTF-8 words.

Created on 16 Mar 2014  ·  3Comments  ·  Source: ariya/phantomjs

I write a url2src.js to convert a html file to another html file that all javascripts are processed.
I found if there is some UTF-8 words(such as —), the result is not correct.

I use Windows 8.1 Chinese version. and run this command in windows cmd:

d:\epub\components>phantomjs  --output-encoding=utf8 --script-encoding=utf8 url2src.js activities2.html activities2-processed.html
d:\epub\components>phantomjs --version
1.9.7

url2src.js's content is:

var page = require('webpage').create(),
    system = require('system'),
    t, address, output;

if (system.args.length !== 3) {
    console.log('Usage: url2src.js <some URL> <output File path>');
    phantom.exit();
}

t = Date.now();
address = system.args[1];
output = system.args[2];
page.open(address, function (status) {
    if (status !== 'success') {
        console.log('FAIL to load the address : ' + address);
    } else {
        t = Date.now() - t;
        //console.log('Loading time ' + t + ' ms');
        var js = page.evaluate(function () {
            return document;
        });
        //console.log(js.all[0].outerHTML); 
        var fs = require('fs');
        try {
            fs.write(output, js.all[0].outerHTML, 'w');
        } catch(e) {
            console.log(e);
        }
    }
    phantom.exit();
});

More detail info is in the attachment here:
https://groups.google.com/forum/#!topic/phantomjs/oqvK8mkk6aY

Any help is appreciated.

Most helpful comment

I know the reason now, because the input html document was treated as ISO-8859-1 as default.
Add meta to set charset=utf-8 can solve this problem.

<html>
<head>
 <meta http-equiv="content-type" content="text/html; charset=UTF-8">
 </head>
 <body>
    <p> The user interface for an activity is provided by a hierarchy of views—objects derived from the <code>View</code> class. </p> 
 </body>
</html>

All 3 comments

I know the reason now, because the input html document was treated as ISO-8859-1 as default.
Add meta to set charset=utf-8 can solve this problem.

<html>
<head>
 <meta http-equiv="content-type" content="text/html; charset=UTF-8">
 </head>
 <body>
    <p> The user interface for an activity is provided by a hierarchy of views—objects derived from the <code>View</code> class. </p> 
 </body>
</html>

You are right.However, I can not change the html which other people give to me.How to do?

I have some documents that don't have any header information (I'm getting them from third party apps)... a way to force the encoding would be nice.

Was this page helpful?
0 / 5 - 0 ratings

Related issues

gustavohenke picture gustavohenke  ·  4Comments

simplegroupware picture simplegroupware  ·  5Comments

dhilipsiva picture dhilipsiva  ·  4Comments

maboiteaspam picture maboiteaspam  ·  3Comments

machadolab picture machadolab  ·  5Comments