在Java上逐元素解析HTML

Question

I have an HTML file:

 <div> DIV1 <div> DIV2 <div> DIV3 </div> </div> </div> 

I want to parse that HTML. However I don't want to get whole parsed HTML as a string:

DIV1 DIV2 DIV3

I would like to get values element by element but none of them duplicated. I mean I don't want that:

When you get first div's value it is:

DIV1 DIV2 DIV3

Seconds div's value:

DIV2 DIV3

Third div's value:

DIV3

The result that I don't want is:

DIV1 DIV2 DIV3
DIV2 DIV3
DIV3

I want that result:

DIV1
DIV2
DIV2

I will apply some procedure to them and I don't want duplicated values as well. I want to use a Java parser to solve my problem. I've considered to use Jsoup but you get entire HTML parsed when you use it.

Answer 1

It sounds like you want to do a pre order depth first search for all text nodes in an HTML document. Luckily most parsing libraries including XML ones will give you all the nodes in pre order as an iterator.

I recommend you use Jericho and call getNodeIterator() and just check to see if its a text node and if it is you print it out. Nootice the link has example code but I will paste it here for your convenience:

 for (Iterator<Segment> nodeIterator=segment.getNoteIterator(); nodeIterator.hasNext();) {
   Segment nodeSegment=nodeIterator.next();
   if (nodeSegment instanceof Tag) {
     Tag tag=(Tag)nodeSegment;
     // HANDLE TAG
     // Uncomment the following line to ensure each tag is valid XML:
     // writer.write(tag.tidy()); continue;
   } else if (nodeSegment instanceof CharacterReference) {
     CharacterReference characterReference=(CharacterReference)nodeSegment;
     // HANDLE CHARACTER REFERENCE
     // Uncomment the following line to decode all character references instead of copying them verbatim:
     // characterReference.appendCharTo(writer); continue;
   } else {
     // HANDLE PLAIN TEXT
   }
   // unless specific handling has prevented getting to here, simply output the segment as is:
   //writer.write(nodeSegment.toString());
 }

In the // HANDLE CHARACTER REFERENCE and // HANDLE PLAIN TEXT are where you want to add your string appending code.