Jsoup：如何提取文件名中有空格的img？

Question

I am trying to extract img using Jsoup. It works fine for images without any space in filename but it extract only the first part if there is a white space.

I tried with below.

String result = Jsoup.clean(content,"https://rally1.rallydev.com/", Whitelist.relaxed().preserveRelativeLinks(true), new Document.OutputSettings().prettyPrint(false));
        Document doc = Jsoup.parse(result);
        Elements images = doc.select("img");

eg HTML content

Description:<div>some text content<br /></div> 
<div><img src=/slm/attachment/43647556403/My file with space.png /></div>
<div><img src=/slm/attachment/43648152373/my_file_without_space.png/></div>

result content is:

Description:Some text content<br> <img src="/slm/attachment/43647556403/My"><img src="/slm/attachment/43648152373/my_file_without_space.png/">

in "result" for the image with space in file name has only first part "My". It ignored the content after whitespace.

How to extract filename if that contains space?

Answer 1

The problem can't be easily solved in Jsoup, since the src attribute value of the example with spaces actually is correctly identified to be only My . The file , with and space.png parts are in this example also attributes without values. Of course you can use JSoup to concatenate the attribute keys that follow the src attribute to its value. For example like this:

String test =""
        + "<div><img src=/slm/attachment/43647556403/My file with space.png /></div>"
        + "<div><img src=/slm/attachment/43647556403/My file with space.png name=whatever/></div>"
        + "<div><img src=/slm/attachment/43647556403/This  breaks  it.png name=whatever/></div>"
        + "<div><img src=\"/slm/attachment/43647556403/This  works.png\" name=whatever/></div>"
        + "<div><img src=/slm/attachment/43648152373/my_file_without_space.png/></div>";
Document doc = Jsoup.parse(test);
Elements imgs = doc.select("img");
for (Element img : imgs){
    Attribute src = null;
    StringBuffer newSrcVal = new StringBuffer();
    List<String> toRemove = new ArrayList<>();
    for (Attribute a : img.attributes()){
        if (a.getKey().equals("src")){
            newSrcVal.append(a.getValue());
            src = a;
        }
        else if (newSrcVal.length()>0){
            //we already found the scr tag
            if (a.getValue().isEmpty()){
                newSrcVal.append(" ").append(a.getKey());
                toRemove.add(a.getKey());
            }
            else{
                //the empty attributes, i.e. file name parts are over
                break;
            }
        }               
    }
    for (String toRemAttr : toRemove){
        img.removeAttr(toRemAttr);
    }
    src.setValue(newSrcVal.toString());
}
System.out.println(doc);

This algorithm cycles over all img elements and within each img it cycles over its attributes. When it finds the src attribute it keeps it for reference and starts to fill the newSrcBuf StringBuffer. All following value-less attributes will be added to to newSrcBuf until either another attribute with value is found or there are no more attributes. Finally the scr attribute value is reset with the contents of newSrcBuf and the former empty attributes are removed from the DOM.

Note that this will not work when your filename contains two or more consecutive spaces. JSoup discards those spaces between attributes and therefore you can't get them back after parsing. If you need that, then you need to manipulate the input html before parsing.

Answer 2

You can something like this:

 Elements images = doc.select("img");

 for(Element image: images){
 String imgSrc = image.attr("src");
 imgSrc = imgSrc.subString(imgSrc.lastIndexOf("/"), imgSrc.length()); // this will give you name.png
 }