jsoup
jsoup copied to clipboard
Negative source range start/end positions for first text node
Hello,
import org.jsoup.nodes.*;
import org.jsoup.parser.*;
import org.jsoup.select.*;
public class Test {
public static void main(String[] args) {
HtmlTreeBuilder treeBuilder = new HtmlTreeBuilder();
Parser parser = new Parser(treeBuilder);
parser.setTrackPosition(true);
Document document = parser.parseInput("foo<p></p>bar<p></p><div><b>baz</b></div>", "");
NodeTraversor.traverse((Node node, int depth) -> {
if (node instanceof TextNode textNode) {
Range sourceRange = textNode.sourceRange();
System.out.printf("text=%s start=%d end=%d%n",
textNode.text(),
sourceRange.start().pos(),
sourceRange.end().pos());
}
}, document);
}
}
We seeing -ve start/end positions for the source range of the first text node foo
, for example using release 1.16.1
:
java -cp ~/.m2/repository/org/jsoup/jsoup/1.16.1/jsoup-1.16.1.jar Test.java
text=foo start=-1 end=-1
text=bar start=10 end=13
text=baz start=28 end=31
Release 1.17.2
has the end position correct, but the start is still -1
java -cp ~/.m2/repository/org/jsoup/jsoup/1.17.2/jsoup-1.17.2.jar Test.java
text=foo start=-1 end=3
text=bar start=10 end=13
text=baz start=28 end=31
@KennyWongPFPT Is it possible that in Parser.java, the following might be causing the issue?
public static Document parseBodyFragment(String bodyHtml, String baseUri) { Document doc = Document.createShell(baseUri); Element body = doc.body(); List<Node> nodeList = parseFragment(bodyHtml, body, baseUri); Node[] nodes = nodeList.toArray(new Node[0]); // the node list gets modified when re-parented
for (int i = nodes.length - 1; i > 0; i--) {
nodes[i].remove();
}
for (Node node : nodes) {
body.appendChild(node);
}
return doc;}
I'm not trying to take a wild stab in the dark, but the HTML string you're passing doesn't contain an initial tag, so potentially setting the start to -1. If there's a check in place, I'm wondering if this will rectify the issue.
Thanks, fixed!