Fixed an issue where unexpected elements in a badly nested table could be moved to the wrong location in the document.

jhy · jhy · commit 3a7f39c507ed · 2015-07-05T14:58:53.000-07:00
Fixes jhy#552 Closes jhy#591
diff --git a/CHANGES b/CHANGES
@@ -1,10 +1,14 @@
 jsoup changelog
 
 *** Release 1.8.3 [PENDING]
- * Fix an issue in Element.getElementSiblingIndex (and related methods) where sibling elements with the same content
+ * Fixed an issue in Element.getElementSiblingIndex (and related methods) where sibling elements with the same content
    would incorrectly have the same sibling index.
    <https://github.com/jhy/jsoup/issues/554>
 
+ * Fixed an issue where unexpected elements in a badly nested table could be moved to the wrong location in the
+   document.
+   <https://github.com/jhy/jsoup/issues/552>
+
 *** Release 1.8.2 [2015-Apr-13]
  * Performance improvements for parsing HTML on Android, of 1.5x to 1.9x, with larger parses getting a bigger
    speed increase. For non-Android JREs, around 1.1x to 1.2x.
diff --git a/src/main/java/org/jsoup/parser/TreeBuilder.java b/src/main/java/org/jsoup/parser/TreeBuilder.java
@@ -58,16 +58,25 @@ protected void runParser() {
     protected abstract boolean process(Token token);
 
     protected boolean processStartTag(String name) {
+        if (currentToken == start) { // don't recycle an in-use token
+            return process(new Token.StartTag().name(name));
+        }
         return process(start.reset().name(name));
     }
 
     public boolean processStartTag(String name, Attributes attrs) {
+        if (currentToken == start) { // don't recycle an in-use token
+            return process(new Token.StartTag().nameAttr(name, attrs));
+        }
         start.reset();
         start.nameAttr(name, attrs);
         return process(start);
     }
 
     protected boolean processEndTag(String name) {
+        if (currentToken == end) { // don't recycle an in-use token
+            return process(new Token.EndTag().name(name));
+        }
         return process(end.reset().name(name));
     }
 
diff --git a/src/test/java/org/jsoup/parser/HtmlParserTest.java b/src/test/java/org/jsoup/parser/HtmlParserTest.java
@@ -3,10 +3,13 @@
 import org.jsoup.Jsoup;
 import org.jsoup.TextUtil;
 import org.jsoup.helper.StringUtil;
+import org.jsoup.integration.ParseTest;
 import org.jsoup.nodes.*;
 import org.jsoup.select.Elements;
 import org.junit.Test;
 
+import java.io.File;
+import java.io.IOException;
 import java.util.List;
 
 import static org.junit.Assert.assertEquals;
@@ -847,4 +850,24 @@ public class HtmlParserTest {
         assertEquals(50000, doc.body().childNodeSize());
         assertTrue(System.currentTimeMillis() - start < 1000);
     }
+
+    @Test
+    public void testInvalidTableContents() throws IOException {
+        File in = ParseTest.getFile("/htmltests/table-invalid-elements.html");
+        Document doc = Jsoup.parse(in, "UTF-8");
+        doc.outputSettings().prettyPrint(true);
+        String rendered = doc.toString();
+        int endOfEmail = rendered.indexOf("Comment");
+        int guarantee = rendered.indexOf("Why am I here?");
+        assertTrue("Comment not found", endOfEmail > -1);
+        assertTrue("Search text not found", guarantee > -1);
+        assertTrue("Search text did not come after comment", guarantee > endOfEmail);
+    }
+
+    @Test public void testNormalisesIsIndex() {
+        Document doc = Jsoup.parse("<body><isindex action='/submit'></body>");
+        String html = doc.outerHtml();
+        assertEquals("<form action=\"/submit\"> <hr> <label>This is a searchable index. Enter search keywords: <input name=\"isindex\"></label> <hr> </form>",
+                StringUtil.normaliseWhitespace(doc.body().html()));
+    }
 }
diff --git a/src/test/resources/htmltests/table-invalid-elements.html b/src/test/resources/htmltests/table-invalid-elements.html
@@ -0,0 +1,17 @@
+<html>
+<body>
+<table>
+  <tr>
+    <td>
+      <table>
+        <tr>
+          <!--Comment-->
+          <table>
+            <p>Why am I here?</p>
+            </tr>
+          </table>
+          </td>
+        </tr>
+      </table>
+</body>
+</html>
