jsoup mistakes a token as an HTML tag

Question

I've got an html fragment as follows:

<span class=#article-title#>About《About<SomeChineseChars》Blabla</span>

sorry here I use latin chars since the editor does not allow to type Chinese chars

when I try to extract text out of this element using

doc.select(".article-title").text();

I will finally have the below as the result:

About《About》Blabla 

after debugging the programming, finding that

<SomeChineseChars> 

was treated as an HTML tag and JSoup close the tag automatically as follows

<SomeChineseChars></SomeChineseChars> 

So, if there is anyway to avoid this from happening, or if this is a BUG?

-=-=-= UPDATE =-=-=-

after dom is built and then check the parsed html, the output is

I cannot post img, so plz click me to view it

Thanks a lot, Ben

Answer 1

I made up a solution by hacking into the JSoup as following:

1. create a new package named org.jsoup.parser;

2. customize a HtmlTreeBuilder

   public class TroilaHtmlTreeBuilder extends HtmlTreeBuilder {
   
   private String zh = "[\\u4e00-\\u9fa5]+";
   
   public TroilaHtmlTreeBuilder() {
   }
   
   @Override
   Element insert(Token.StartTag startTag) {
       if (startTag.tagName.matches(zh)) {
           Token.Character ch = new Token.Character();
           ch.data(startTag.toString());
           insert(ch);
           return null;
       }
       return super.insert(startTag);
   }
   
   public Document parse(Reader input, String baseUri) {
       return super.parse(input, baseUri, ParseErrorList.noTracking(), this.defaultSettings());
   }
   
   }
   

I don't think this is a good way to solve the problem, so let me know if you have any better idea.

BTW: many thanks to @Abhilash for your help!