Fix table insertion into TH tags

jhy · jhy · commit 3b4f9dfa91b6 · 2015-07-05T21:21:02.000-07:00
Fixes jhy#575
diff --git a/CHANGES b/CHANGES
@@ -15,6 +15,9 @@ jsoup changelog
    document.
    <https://github.com/jhy/jsoup/issues/552>
 
+ * Fixed an issue where a table nested within a TH cell would parse to an incorrect tree.
+   <https://github.com/jhy/jsoup/issues/575>
+
 *** 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/HtmlTreeBuilder.java b/src/main/java/org/jsoup/parser/HtmlTreeBuilder.java
@@ -391,7 +391,7 @@ void resetInsertionMode() {
             if ("select".equals(name)) {
                 transition(HtmlTreeBuilderState.InSelect);
                 break; // frag
-            } else if (("td".equals(name) || "td".equals(name) && !last)) {
+            } else if (("td".equals(name) || "th".equals(name) && !last)) {
                 transition(HtmlTreeBuilderState.InCell);
                 break;
             } else if ("tr".equals(name)) {
diff --git a/src/test/java/org/jsoup/parser/HtmlParserTest.java b/src/test/java/org/jsoup/parser/HtmlParserTest.java
@@ -870,4 +870,10 @@ public void testInvalidTableContents() throws IOException {
         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()));
     }
+
+    @Test public void testReinsertionModeForThCelss() {
+        String body = "<body> <table> <tr> <th> <table><tr><td></td></tr></table> <div> <table><tr><td></td></tr></table> </div> <div></div> <div></div> <div></div> </th> </tr> </table> </body>";
+        Document doc = Jsoup.parse(body);
+        assertEquals(1, doc.body().children().size());
+    }
 }

Original file line number	Diff line number	Diff line change
`@@ -870,4 +870,10 @@ public void testInvalidTableContents() throws IOException {`
`870`	`870`	`assertEquals("<form action=\"/submit\"> <hr> <label>This is a searchable index. Enter search keywords: <input name=\"isindex\"></label> <hr> </form>",`
`871`	`871`	`StringUtil.normaliseWhitespace(doc.body().html()));`
`872`	`872`	`}`
	`873`	`+`
	`874`	`+ @Test public void testReinsertionModeForThCelss() {`
	`875`	`+ String body = "<body> <table> <tr> <th> <table><tr><td></td></tr></table> <div> <table><tr><td></td></tr></table> </div> <div></div> <div></div> <div></div> </th> </tr> </table> </body>";`
	`876`	`+ Document doc = Jsoup.parse(body);`
	`877`	`+ assertEquals(1, doc.body().children().size());`
	`878`	`+ }`
`873`	`879`	`}`