From ebf772c85a934a5c54bf6624adaeecdb641a8bb3 Mon Sep 17 00:00:00 2001
From: Laurent Jouanneau <laurent@xulfr.org>
Date: Tue, 23 Mar 2010 21:55:39 +0100
Subject: [PATCH] Bug 541937 - XMLSerializer should serialize content of link
 tags. r=Olli.Pettay sr=bzbarsky

---
 content/base/src/nsXHTMLContentSerializer.cpp |  69 ++++++++---
 content/base/src/nsXHTMLContentSerializer.h   |   1 +
 content/base/test/Makefile.in                 |   3 +
 content/base/test/file_bug541937.html         |   8 ++
 content/base/test/file_bug541937.xhtml        |  12 ++
 content/base/test/test_bug541937.html         | 115 ++++++++++++++++++
 6 files changed, 188 insertions(+), 20 deletions(-)
 create mode 100644 content/base/test/file_bug541937.html
 create mode 100644 content/base/test/file_bug541937.xhtml
 create mode 100644 content/base/test/test_bug541937.html

diff --git a/content/base/src/nsXHTMLContentSerializer.cpp b/content/base/src/nsXHTMLContentSerializer.cpp
index 7656d04483d..83001e129e1 100644
--- a/content/base/src/nsXHTMLContentSerializer.cpp
+++ b/content/base/src/nsXHTMLContentSerializer.cpp
@@ -461,18 +461,26 @@ nsXHTMLContentSerializer::AppendEndOfElementStart(nsIDOMElement *aOriginalElemen
     return;
   }
 
-  nsIParserService* parserService = nsContentUtils::GetParserService();
+  nsCOMPtr<nsIContent> content = do_QueryInterface(aOriginalElement);
 
-  if (parserService) {
-    PRBool isContainer;
-    parserService->IsContainer(parserService->HTMLAtomTagToId(aName),
-                             isContainer);
-    if (!isContainer) {
-      // for backward compatibility with HTML 4 user agents
-      // only non-container HTML elements can be closed immediatly,
-      // and a space is added before />
-      AppendToString(NS_LITERAL_STRING(" />"), aStr);
-      return;
+  // for non empty elements, even if they are not a container, we always
+  // serialize their content, because the XHTML element could contain non XHTML
+  // nodes useful in some context, like in an XSLT stylesheet
+  if (HasNoChildren(content)) {
+
+    nsIParserService* parserService = nsContentUtils::GetParserService();
+  
+    if (parserService) {
+      PRBool isContainer;
+      parserService->IsContainer(parserService->HTMLAtomTagToId(aName),
+                                 isContainer);
+      if (!isContainer) {
+        // for backward compatibility with HTML 4 user agents
+        // only non-container HTML elements can be closed immediatly,
+        // and a space is added before />
+        AppendToString(NS_LITERAL_STRING(" />"), aStr);
+        return;
+      }
     }
   }
   AppendToString(kGreaterThan, aStr);
@@ -605,17 +613,19 @@ nsXHTMLContentSerializer::CheckElementEnd(nsIContent * aContent,
       }
     }
 
-    nsIParserService* parserService = nsContentUtils::GetParserService();
+    if (HasNoChildren(aContent)) {
+      nsIParserService* parserService = nsContentUtils::GetParserService();
 
-    if (parserService) {
-      PRBool isContainer;
+      if (parserService) {
+        PRBool isContainer;
 
-      parserService->IsContainer(parserService->HTMLAtomTagToId(name),
-                                 isContainer);
-      if (!isContainer) {
-        // non-container HTML elements are already closed,
-        // see AppendEndOfElementStart
-        return PR_FALSE;
+        parserService->IsContainer(parserService->HTMLAtomTagToId(name),
+                                   isContainer);
+        if (!isContainer) {
+          // non-container HTML elements are already closed,
+          // see AppendEndOfElementStart
+          return PR_FALSE;
+        }
       }
     }
     // for backward compatibility with old HTML user agents,
@@ -1066,3 +1076,22 @@ nsXHTMLContentSerializer::IsFirstChildOfOL(nsIDOMElement* aElement)
   else
     return PR_FALSE;
 }
+
+PRBool
+nsXHTMLContentSerializer::HasNoChildren(nsIContent * aContent) {
+
+  PRUint32 i, childCount = aContent->GetChildCount();
+
+  for (i = 0; i < childCount; ++i) {
+
+    nsIContent* child = aContent->GetChildAt(i);
+
+    if (!child->IsNodeOfType(nsINode::eTEXT))
+      return PR_FALSE;
+
+    if (child->TextLength())
+      return PR_FALSE;
+  }
+
+  return PR_TRUE;
+}
diff --git a/content/base/src/nsXHTMLContentSerializer.h b/content/base/src/nsXHTMLContentSerializer.h
index 8088f384804..8e745b2e857 100644
--- a/content/base/src/nsXHTMLContentSerializer.h
+++ b/content/base/src/nsXHTMLContentSerializer.h
@@ -200,6 +200,7 @@ class nsXHTMLContentSerializer : public nsXMLContentSerializer {
   // Stack to store one olState struct per <OL>.
   nsAutoTArray<olState, 8> mOLStateStack;
 
+  PRBool HasNoChildren(nsIContent* aContent);
 };
 
 nsresult
diff --git a/content/base/test/Makefile.in b/content/base/test/Makefile.in
index 7fe61e1e0cc..22076cc1611 100644
--- a/content/base/test/Makefile.in
+++ b/content/base/test/Makefile.in
@@ -362,6 +362,9 @@ _TEST_FILES = 	test_bug5141.html \
 		test_bug545644.html \
 		test_bug545644.xhtml \
 		test_bug553896.xhtml \
+		test_bug541937.html \
+		file_bug541937.html \
+		file_bug541937.xhtml \
 		$(NULL)
 
 # This test fails on the Mac for some reason
diff --git a/content/base/test/file_bug541937.html b/content/base/test/file_bug541937.html
new file mode 100644
index 00000000000..af45fb65f13
--- /dev/null
+++ b/content/base/test/file_bug541937.html
@@ -0,0 +1,8 @@
+<!DOCTYPE html><html><head><meta http-equiv="content-type" content="text/html; charset=UTF-8">
+  <title>Test</title>
+   <link rel="Top" href="">  foo </link>
+
+</head><body>
+  <p>Hello world</p>
+</body>
+</html>
\ No newline at end of file
diff --git a/content/base/test/file_bug541937.xhtml b/content/base/test/file_bug541937.xhtml
new file mode 100644
index 00000000000..f4e5f3d320b
--- /dev/null
+++ b/content/base/test/file_bug541937.xhtml
@@ -0,0 +1,12 @@
+<!DOCTYPE html PUBLIC "-//W3C//DTD XHTML 1.0 Strict//EN" "http://www.w3.org/TR/xhtml1/DTD/xhtml1-strict.dtd">
+<html xmlns="http://www.w3.org/1999/xhtml">
+<head>
+   <meta http-equiv="content-type" content="text/html; charset=UTF-8" />
+  <title>Test</title>
+   <link rel="Top" href="">  foo </link>
+
+</head>
+<body>
+  <p>Hello world</p>
+</body>
+</html>
\ No newline at end of file
diff --git a/content/base/test/test_bug541937.html b/content/base/test/test_bug541937.html
new file mode 100644
index 00000000000..b1efac3eed8
--- /dev/null
+++ b/content/base/test/test_bug541937.html
@@ -0,0 +1,115 @@
+<!DOCTYPE HTML>
+<html>
+<!--
+-->
+<head>
+  <title>Test for XHTML serializer, bug 541937</title>
+  <script type="text/javascript" src="/MochiKit/packed.js"></script>
+  <script type="text/javascript" src="/tests/SimpleTest/SimpleTest.js"></script>
+  <link rel="stylesheet" type="text/css" href="/tests/SimpleTest/test.css" />
+</head>
+<body>
+<a target="_blank" href="https://bugzilla.mozilla.org/show_bug.cgi?id=541937">Mozilla Bug </a>
+<p id="display"></p>
+<div id="content" style="display: none">
+  <iframe id="testframe" src="file_bug541937.html">
+  </iframe>
+  <iframe id="testframe2" src="file_bug541937.xhtml">
+  </iframe>
+</div>
+
+<pre id="test">
+<script class="testbody" type="text/javascript">
+
+function testSerializer () {
+  netscape.security.PrivilegeManager.enablePrivilege("UniversalXPConnect");
+  const de = Components.interfaces.nsIDocumentEncoder
+  var encoder = Components.classes["@mozilla.org/layout/documentEncoder;1?type=text/html"]
+                   .createInstance(Components.interfaces.nsIDocumentEncoder);
+
+  var parser = new DOMParser();
+  var serializer = new XMLSerializer();
+
+  // with content
+  var str = '<?xml version="1.0"?><doc>\n<link xmlns="http://www.w3.org/1999/xhtml"><!-- child nodes -->  \n<content xmlns=""/></link>\n</doc>';
+  var expected = '<?xml version="1.0" encoding="UTF-8"?>\n<doc>\n<link xmlns="http://www.w3.org/1999/xhtml"><!-- child nodes -->  \n<content xmlns=""/></link>\n</doc>';
+
+  var doc = parser.parseFromString(str,"application/xml");
+  var result = serializer.serializeToString(doc);
+  is(result, expected, "serialization of a link element inside an xml document with children");
+
+  // with only whitespaces
+  str = '<?xml version="1.0"?><doc>\n<link xmlns="http://www.w3.org/1999/xhtml">       \n    </link>\n</doc>';
+  expected = '<?xml version="1.0" encoding="UTF-8"?>\n<doc>\n<link xmlns="http://www.w3.org/1999/xhtml">       \n    </link>\n</doc>';  
+
+  doc = parser.parseFromString(str,"application/xml");
+  result = serializer.serializeToString(doc);
+  is(result, expected, "serialization of a link element with only whitespaces as content, inside an xml document");
+  
+  // with only one space as content
+  str = '<?xml version="1.0"?><doc>\n<link xmlns="http://www.w3.org/1999/xhtml"> </link>\n</doc>';
+  expected = '<?xml version="1.0" encoding="UTF-8"?>\n<doc>\n<link xmlns="http://www.w3.org/1999/xhtml"> </link>\n</doc>';  
+
+  doc = parser.parseFromString(str,"application/xml");
+  result = serializer.serializeToString(doc);
+  is(result, expected, "serialization of a link element with only one space as content, inside an xml document");
+  
+  // let's remove the content
+  str = '<?xml version="1.0"?><doc>\n<link xmlns="http://www.w3.org/1999/xhtml"> <!-- child nodes -->  \ndeleted content<content xmlns=""/> </link>\n</doc>';
+  expected = '<?xml version="1.0" encoding="UTF-8"?>\n<doc>\n<link xmlns="http://www.w3.org/1999/xhtml" />\n</doc>';  
+
+  doc = parser.parseFromString(str,"application/xml");
+  doc.documentElement.firstElementChild.textContent = '';
+  result = serializer.serializeToString(doc);
+  is(result, expected, "serialization of a link element on which we removed dynamically the content, inside an xml document");
+  
+  // with no content but an ended tag
+  str = '<?xml version="1.0"?><doc>\n<link xmlns="http://www.w3.org/1999/xhtml"></link>\n</doc>';
+  expected = '<?xml version="1.0" encoding="UTF-8"?>\n<doc>\n<link xmlns="http://www.w3.org/1999/xhtml" />\n</doc>';  
+
+  doc = parser.parseFromString(str,"application/xml");
+  result = serializer.serializeToString(doc);
+  is(result, expected, "serialization of a link element with no content but with an ended tag, inside an xml document");
+  
+  // with no content
+  str = '<?xml version="1.0"?><doc>\n<link xmlns="http://www.w3.org/1999/xhtml"/>\n</doc>';
+  expected = '<?xml version="1.0" encoding="UTF-8"?>\n<doc>\n<link xmlns="http://www.w3.org/1999/xhtml" />\n</doc>';  
+
+  doc = parser.parseFromString(str,"application/xml");
+  result = serializer.serializeToString(doc);
+  is(result, expected, "serialization of a link element with no content, inside an xml document");
+  
+  
+  doc = $("testframe").contentDocument;
+  encoder.init(doc, "text/html", de.OutputLFLineBreak);
+  encoder.setCharset("UTF-8");
+  result = encoder.encodeToString();
+  expected = '<!DOCTYPE HTML>\n<html><head><meta http-equiv=\"content-type\" content=\"text/html; \ncharset=UTF-8\">\n  <title>Test</title>\n';
+  expected += '   <link rel=\"Top\" href=\"\"> ';
+  expected += ' </head><body>foo \n\n\n  <p>Hello world</p>\n</body></html>';
+  is(result, expected, "serialization of a link element with content, inside an html document");
+
+  doc = $("testframe2").contentDocument;
+  encoder.init(doc, "application/xhtml+xml", de.OutputLFLineBreak);
+  encoder.setCharset("UTF-8");
+  result = encoder.encodeToString();
+  expected = '<!DOCTYPE html PUBLIC "-//W3C//DTD XHTML 1.0 Strict//EN" "http://www.w3.org/TR/xhtml1/DTD/xhtml1-strict.dtd">\n';
+  expected += '<html xmlns="http://www.w3.org/1999/xhtml">\n<head>\n   <meta http-equiv="content-type" content="text/html; charset=UTF-8" />\n  <title>Test</title>\n';
+  expected += '   <link rel="Top" href="">  foo </link>\n';
+  expected += '\n</head>\n<body>\n  <p>Hello world</p>\n</body>\n</html>';
+  is(result, expected, "serialization of a link element with content, inside an xhtml document");
+
+  SimpleTest.finish();
+}
+
+
+SimpleTest.waitForExplicitFinish();
+
+addLoadEvent(testSerializer);
+
+</script>
+</pre>
+</body>
+</html>
+
+