Complete adoption agency algorithm (#1517)

Follow adoption agency algorithm

Author: suarez12138

src/main/java/org/jsoup/parser/HtmlTreeBuilder.java

@@ -613,6 +613,14 @@ Element lastFormattingElement() {
         return formattingElements.size() > 0 ? formattingElements.get(formattingElements.size()-1) : null;
     }
 
+    int positionOfElement(Element el){
+        for (int i = 0; i < formattingElements.size(); i++){
+            if (el == formattingElements.get(i))
+                return i;
+        }
+        return -1;
+    }
+
     Element removeLastFormattingElement() {
         int size = formattingElements.size();
         if (size > 0)
@@ -623,6 +631,16 @@ Element removeLastFormattingElement() {
 
     // active formatting elements
     void pushActiveFormattingElements(Element in) {
+        this.checkActiveFormattingElements(in);
+        formattingElements.add(in);
+    }
+
+    void pushWithBookmark(Element in,int bookmark){
+        this.checkActiveFormattingElements(in);
+        formattingElements.add(bookmark, in);
+    }
+
+    void checkActiveFormattingElements(Element in){
         int numSeen = 0;
         for (int pos = formattingElements.size() -1; pos >= 0; pos--) {
             Element el = formattingElements.get(pos);
@@ -637,7 +655,6 @@ void pushActiveFormattingElements(Element in) {
                 break;
             }
         }
-        formattingElements.add(in);
     }
 
     private boolean isSameFormattingElement(Element a, Element b) {

src/main/java/org/jsoup/parser/HtmlTreeBuilderState.java

@@ -806,11 +806,14 @@ else if (!tb.onStack(formatEl)) {
                 // the spec doesn't limit to < 64, but in degenerate cases (9000+ stack depth) this prevents
                 // run-aways
                 final int stackSize = stack.size();
+                int bookmark = -1;
                 for (int si = 0; si < stackSize && si < 64; si++) {
                     el = stack.get(si);
                     if (el == formatEl) {
                         commonAncestor = stack.get(si - 1);
                         seenFormattingElement = true;
+                        // Let a bookmark note the position of the formatting element in the list of active formatting elements relative to the elements on either side of it in the list.
+                        bookmark = tb.positionOfElement(el);
                     } else if (seenFormattingElement && tb.isSpecial(el)) {
                         furthestBlock = el;
                         break;
@@ -822,8 +825,6 @@ else if (!tb.onStack(formatEl)) {
                     return true;
                 }
 
-                // todo: Let a bookmark note the position of the formatting element in the list of active formatting elements relative to the elements on either side of it in the list.
-                // does that mean: int pos of format el in list?
                 Element node = furthestBlock;
                 Element lastNode = furthestBlock;
                 for (int j = 0; j < 3; j++) {
@@ -843,8 +844,9 @@ else if (!tb.onStack(formatEl)) {
 
                     //noinspection StatementWithEmptyBody
                     if (lastNode == furthestBlock) {
-                        // todo: move the aforementioned bookmark to be immediately after the new node in the list of active formatting elements.
+                        // move the aforementioned bookmark to be immediately after the new node in the list of active formatting elements.
                         // not getting how this bookmark both straddles the element above, but is inbetween here...
+                        bookmark = tb.positionOfElement(node) + 1;
                     }
                     if (lastNode.parent() != null)
                         lastNode.remove();
@@ -871,7 +873,8 @@ else if (!tb.onStack(formatEl)) {
                 }
                 furthestBlock.appendChild(adopter);
                 tb.removeFromActiveFormattingElements(formatEl);
-                // todo: insert the new element into the list of active formatting elements at the position of the aforementioned bookmark.
+                // insert the new element into the list of active formatting elements at the position of the aforementioned bookmark.
+                tb.pushWithBookmark(adopter, bookmark);
                 tb.removeFromStack(formatEl);
                 tb.insertOnStackAfter(furthestBlock, adopter);
             }

src/test/java/org/jsoup/parser/HtmlTreeBuilderStateTest.java

@@ -1,5 +1,6 @@
 package org.jsoup.parser;
 
+import org.jsoup.Jsoup;
 import org.jsoup.parser.HtmlTreeBuilderState.Constants;
 import org.junit.jupiter.api.Test;
 
@@ -46,4 +47,54 @@ public void ensureArraysAreSorted() {
         assertEquals(38, constants.size());
     }
 
+
+    @Test
+    public void nestedAnchorElements01() {
+        String html = "<html>\n" +
+                "  <body>\n" +
+                "    <a href='#1'>\n" +
+                "        <div>\n" +
+                "          <a href='#2'>child</a>\n" +
+                "        </div>\n" +
+                "    </a>\n" +
+                "  </body>\n" +
+                "</html>";
+        String s = Jsoup.parse(html).toString();
+        assertEquals("<html> \n" +
+                " <head></head>\n" +
+                " <body> <a href=\"#1\"> </a>\n" +
+                "  <div>\n" +
+                "   <a href=\"#1\"> </a><a href=\"#2\">child</a> \n" +
+                "  </div>   \n" +
+                " </body>\n" +
+                "</html>", s);
+    }
+
+    @Test
+    public void nestedAnchorElements02() {
+        String html = "<html>\n" +
+                "  <body>\n" +
+                "    <a href='#1'>\n" +
+                "      <div>\n" +
+                "        <div>\n" +
+                "          <a href='#2'>child</a>\n" +
+                "        </div>\n" +
+                "      </div>\n" +
+                "    </a>\n" +
+                "  </body>\n" +
+                "</html>";
+        String s = Jsoup.parse(html).toString();
+        assertEquals("<html> \n" +
+                " <head></head>\n" +
+                " <body> <a href=\"#1\"> </a>\n" +
+                "  <div>\n" +
+                "   <a href=\"#1\"> </a>\n" +
+                "   <div>\n" +
+                "    <a href=\"#1\"> </a><a href=\"#2\">child</a> \n" +
+                "   </div> \n" +
+                "  </div>   \n" +
+                " </body>\n" +
+                "</html>", s);
+    }
+
 }