Nested inlineable elements should indent

Fixes #1926
jhy · Apr 29, 2023 · 8e2b868 · 8e2b868
1 parent d9544a4
commit 8e2b868
Show file tree

Hide file tree

Showing 4 changed files with 34 additions and 13 deletions.
diff --git a/CHANGES b/CHANGES
@@ -38,6 +38,10 @@ Release 1.16.1 [PENDING]
   * Bugfix: when pretty-printing a <pre> containing block tags, those tags were incorrectly indented.
     <https://github.com/jhy/jsoup/issues/1891>
 
+  * Bugfix: when pretty-printing nested inlineable blocks (such as a <p> in a <td>), the inner element should be
+    indented.
+    <https://github.com/jhy/jsoup/issues/1926>
+
   * Bugfix: <br> tags should be wrap-indented when in block tags (and not when in inline tags).
     <https://github.com/jhy/jsoup/issues/1911>
 

diff --git a/src/main/java/org/jsoup/nodes/Element.java b/src/main/java/org/jsoup/nodes/Element.java
@@ -1844,7 +1844,7 @@ public void onContentsChanged() {
     }
 
     private boolean isFormatAsBlock(Document.OutputSettings out) {
-        return tag.formatAsBlock() || (parent() != null && parent().tag().formatAsBlock()) || out.outline();
+        return tag.isBlock() || (parent() != null && parent().tag().formatAsBlock()) || out.outline();
     }
 
     private boolean isInlineable(Document.OutputSettings out) {

diff --git a/src/test/java/org/jsoup/nodes/ElementTest.java b/src/test/java/org/jsoup/nodes/ElementTest.java
@@ -2701,4 +2701,28 @@ void prettySerializationRoundTrips(Document.OutputSettings settings) {
         Document doc = Jsoup.parse(html);
         assertEquals(html, doc.body().html());
     }
+
+    @Test void nestedFormatAsInlinePrintsAsBlock() {
+        // https://github.com/jhy/jsoup/issues/1926
+        String h = "        <table>\n" +
+            "            <tr>\n" +
+            "                <td>\n" +
+            "                    <p style=\"display:inline;\">A</p>\n" +
+            "                    <p style=\"display:inline;\">B</p>\n" +
+            "                </td>\n" +
+            "            </tr>\n" +
+            "        </table>";
+        Document doc = Jsoup.parse(h);
+        String out = doc.body().html();
+        assertEquals("<table>\n" +
+            " <tbody>\n" +
+            "  <tr>\n" +
+            "   <td>\n" +
+            "    <p style=\"display:inline;\">A</p>\n" +
+            "    <p style=\"display:inline;\">B</p></td>\n" +
+            "  </tr>\n" +
+            " </tbody>\n" +
+            "</table>", out);
+        // todo - I would prefer the </td> to wrap down there - but need to reimplement pretty printer to simplify and track indented state
+    }
 }
diff --git a/src/test/java/org/jsoup/parser/HtmlParserTest.java b/src/test/java/org/jsoup/parser/HtmlParserTest.java
@@ -720,15 +720,8 @@ public class HtmlParserTest {
         // and the <i> inside the table and does not leak out.
         String h = "<p><b>One</p> <table><tr><td><p><i>Three<p>Four</i></td></tr></table> <p>Five</p>";
         Document doc = Jsoup.parse(h);
-        String want = "<p><b>One</b></p><b>\n" +
-            " <table>\n" +
-            "  <tbody>\n" +
-            "   <tr>\n" +
-            "    <td><p><i>Three</i></p><p><i>Four</i></p></td>\n" +
-            "   </tr>\n" +
-            "  </tbody>\n" +
-            " </table><p>Five</p></b>";
-        assertEquals(want, doc.body().html());
+        String want = "<p><b>One</b></p><b><table><tbody><tr><td><p><i>Three</i></p><p><i>Four</i></p></td></tr></tbody></table><p>Five</p></b>";
+        assertEquals(want, TextUtil.stripNewlines(doc.body().html()));
     }
 
     @Test public void commentBeforeHtml() {
@@ -777,7 +770,7 @@ public class HtmlParserTest {
 
         Document two = Jsoup.parse("<title>One<b>Two <p>Test</p>"); // no title, so <b> causes </title> breakout
         assertEquals("One", two.title());
-        assertEquals("<b>Two <p>Test</p></b>", two.body().html());
+        assertEquals("<b>Two \n <p>Test</p></b>", two.body().html());
     }
 
     @Test public void handlesUnclosedScriptAtEof() {
@@ -1470,7 +1463,7 @@ private boolean didAddElements(String input) {
         assertEquals(1, nodes.size());
         Node node = nodes.get(0);
         assertEquals("h2", node.nodeName());
-        assertEquals("<p><h2>text</h2></p>", node.parent().outerHtml());
+        assertEquals("<p>\n <h2>text</h2></p>", node.parent().outerHtml());
     }
 
     @Test public void nestedPFragments() {
@@ -1479,7 +1472,7 @@ private boolean didAddElements(String input) {
         List<Node> nodes = new Document("").parser().parseFragmentInput(bareFragment, new Element("p"), "");
         assertEquals(2, nodes.size());
         Node node = nodes.get(0);
-        assertEquals("<p><p></p><a></a></p>", node.parent().outerHtml()); // mis-nested because fragment forced into the element, OK
+        assertEquals("<p>\n <p></p><a></a></p>", node.parent().outerHtml()); // mis-nested because fragment forced into the element, OK
     }
 
     @Test public void nestedAnchorAdoption() {