Default Parser error (Repro included)

[sjorn3]

Describe the bug A clear and concise description of what the bug is.

With default parser settings in flexmark 0.64.8 I'm getting an internal accumulator bug with the following minimal input. It's not valid markdown but my understanding is that it should be supported or at least give a more useful error.

x\n \n`\n x\n`

FAIL: java.lang.IllegalStateException: Accumulated range [5, 6) overlaps Transformed Range[2]: [5, 6)
java.lang.IllegalStateException: Accumulated range [5, 6) overlaps Transformed Range[2]: [5, 6)
        at com.vladsch.flexmark.util.sequence.builder.SegmentBuilderBase.processParts(SegmentBuilderBase.java:431)
        at com.vladsch.flexmark.util.sequence.builder.SegmentBuilderBase.append(SegmentBuilderBase.java:538)
        at com.vladsch.flexmark.util.sequence.builder.SegmentBuilderBase.append(SegmentBuilderBase.java:16)
        at com.vladsch.flexmark.util.sequence.builder.tree.SegmentTree.addSegments(SegmentTree.java:325)

Expected behavior A clear and concise description of what you expected to happen with sample output that would be desired, where it makes sense.

Ideally I would get a best effort at the text without markdown when calling TextCollectingVisitor().collectAndGetText, but largely just reporting the internal error.

Code to reproduce

import com.vladsch.flexmark.parser.Parser;
import com.vladsch.flexmark.util.ast.Node;
import com.vladsch.flexmark.util.ast.TextCollectingVisitor;

public class FlexmarkDefaultCheck {
  public static void main(String[] args) {
    String s = "x\n \n`\n x\n`";
    try {
      Parser p = Parser.builder().build();
      Node doc = p.parse(s);
      String out = new TextCollectingVisitor().collectAndGetText(doc);
      System.out.println("OK");
      System.out.println(out.replace("\n", "\\n"));
    } catch (Throwable t) {
      t.printStackTrace();
      System.exit(1);
    }
  }
}

Thank you for looking into it!