使用 JAXB 在 XML 输出中避免特殊字符

Question

I am reading tweets and forming an XML out of it, for which I am using JAXB Marshaller and UTF-8 encoding.

JAXB Marshaller setting is:

JAXBContext jaxbContext;
StringWriter writer = new StringWriter();

jaxbContext = JAXBContext.newInstance(obj.getClass());
Marshaller m = jaxbContext.createMarshaller();  
m.setProperty(Marshaller.JAXB_FORMATTED_OUTPUT, true);
m.marshal(obj, writer);

Here, obj is my class object which contains tweet text and other information.

My problem is, the generated XML contains special characters like:

> x85, x93, xAQ

Sample Output XML:

   <tweet>
        <id>500923859663872000</id>
        <createdAt>2014-08-17T14:05:29+05:30</createdAt>
        **<text>Ԁhughwizzy: 55% of all '14-'15 @PremierLeague players will wear @Nike** Boots. (@adidas 35%, @Puma 5%). http://t.co/VHit1Es7KlԠ@Yup_Yup9</text>
        <langISOCode>en</langISOCode>
        <place>NA</place>
        <favouriteCount>0</favouriteCount>
        <retweetCount>0</retweetCount>
        <isPossiblySensitive>false</isPossiblySensitive>
        <user>
            <id>39481349</id>
            <createdAt>2009-05-12T17:12:37+05:30</createdAt>
            <location>NA</location>
            <followersCount>281</followersCount>
            <listedCount>4</listedCount>
            <preferredLang>en</preferredLang>
            <isVerified>false</isVerified>
            <isTranslator>false</isTranslator>
        </user>
    </tweet>

I found that these are UTF-8 encoded characters, but it makes my XML invalid.

Is there a way to avoid these characters in the generated XML.

Answer 1

You can Base64Encode the string before you set it to your variable in your object. If you want to display the content in the browser you can extract the xml value for that text , then you can display after decoding it.

obj.setText(DatatypeConverter.printBase64Binary(byte[] your_tweet_as_a_byte_array));
// serialize it
//in the front end extract the xml value for your text using javascript. lets say xml contains like this:=>  <text>adiufgdb12bsre==<text>
// you can use the following code in the javascript to decode it.
atob(extracted_encoded _string) // will give you the decoded string.

Using this way you can avoid creating an invalid xml, by avoiding xml not supported unicode characters.