Improve whitespace handling (#869)

Author: lklein53

src/DocumentFormat.OpenXml/OpenXmlPartReader.cs

@@ -451,6 +451,12 @@ private bool MoveToNextElement()
                 case ElementState.MiscNode:
                     // cursor is end element, pop stack
                     _elementStack.Pop();
+                    if (_elementStack.Count == 0)
+                    {
+                        _elementState = ElementState.EOF;
+                        return false;
+                    }
+
                     break;
 
                 case ElementState.LeafStart:

test/DocumentFormat.OpenXml.Tests/ofapiTest/OpenXmlReaderTest.cs

@@ -399,8 +399,7 @@ public void PartReaderIgnoreWhitespaceTest(bool ignoreWhitespace)
                 "</w:body>" +
                 "</w:document>";
 
-            UTF8Encoding utf8Encoding = new UTF8Encoding();
-            using var stream = new MemoryStream(utf8Encoding.GetBytes(PartText), false);
+            using var stream = new MemoryStream(Encoding.UTF8.GetBytes(PartText), false);
 
             using var reader = OpenXmlReader.Create(stream, false, ignoreWhitespace);
             Assert.False(reader.EOF);
@@ -425,6 +424,146 @@ public void PartReaderIgnoreWhitespaceTest(bool ignoreWhitespace)
             reader.Close();
         }
 
+        /// <summary>
+        /// Test that the OpenXmlReader can handle formatted xml.
+        ///</summary>
+        [Theory]
+        [InlineData(true)]
+        [InlineData(false)]
+        public void TestIgnoreWhitespaceWithFormattedXml(bool ignoreWhitespace) {
+            const string PartText = @"
+    <w:document xmlns:v=""urn:schemas-microsoft-com:vml"" xmlns:w=""http://schemas.openxmlformats.org/wordprocessingml/2006/main"">
+      <w:body>
+        <w:p>
+          <w:r>
+            <w:t>First Text</w:t>
+          </w:r>
+        </w:p>
+      </w:body>
+    </w:document>";
+
+            using var stream = new MemoryStream(Encoding.UTF8.GetBytes(PartText), false);
+            using var reader = OpenXmlReader.Create(stream, false, ignoreWhitespace);
+            Assert.True(reader.Read());
+            Assert.False(reader.EOF);
+            Assert.Equal(typeof(Document), reader.ElementType);
+            Assert.True(reader.IsStartElement);
+            Assert.True(reader.Read());
+            Assert.Equal(typeof(Body), reader.ElementType);
+            Assert.True(reader.IsStartElement);
+            Assert.True(reader.Read());
+            Assert.Equal(typeof(Paragraph), reader.ElementType);
+            Assert.True(reader.IsStartElement);
+            Assert.True(reader.Read());
+            Assert.Equal(typeof(Run), reader.ElementType);
+            Assert.True(reader.IsStartElement);
+            Assert.True(reader.Read());
+            Assert.Equal(typeof(Text), reader.ElementType);
+            Assert.True(reader.IsStartElement);
+            Assert.True(reader.Read());
+            Assert.Equal(typeof(Text), reader.ElementType);
+            Assert.True(reader.IsEndElement);
+            Assert.True(reader.Read());
+            Assert.Equal(typeof(Run), reader.ElementType);
+            Assert.True(reader.IsEndElement);
+            Assert.True(reader.Read());
+            Assert.Equal(typeof(Paragraph), reader.ElementType);
+            Assert.True(reader.IsEndElement);
+            Assert.True(reader.Read());
+            Assert.Equal(typeof(Body), reader.ElementType);
+            Assert.True(reader.IsEndElement);
+            Assert.True(reader.Read());
+            Assert.Equal(typeof(Document), reader.ElementType);
+            Assert.True(reader.IsEndElement);
+            Assert.False(reader.Read());
+            Assert.True(reader.EOF);
+
+            reader.Close();
+    }
+
+        /// <summary>
+        /// Test that the OpenXmlReader can handle a whitespace after the last element.
+        ///</summary>
+        [Theory]
+        [InlineData(true)]
+        [InlineData(false)]
+        public void TestIgnoreWhitespaceWhitespaceAfterLastElement(bool ignoreWhitespace) {
+            const string PartText = @"
+    <w:document xmlns:v=""urn:schemas-microsoft-com:vml"" xmlns:w=""http://schemas.openxmlformats.org/wordprocessingml/2006/main"">
+      <w:body>
+        <w:p>
+        </w:p>
+      </w:body>
+    </w:document> ";
+
+            using var stream = new MemoryStream(Encoding.UTF8.GetBytes(PartText), false);
+            using var reader = OpenXmlReader.Create(stream, false, ignoreWhitespace);
+            Assert.True(reader.Read());
+            Assert.False(reader.EOF);
+            Assert.Equal(typeof(Document), reader.ElementType);
+            Assert.True(reader.IsStartElement);
+            Assert.True(reader.Read());
+            Assert.Equal(typeof(Body), reader.ElementType);
+            Assert.True(reader.IsStartElement);
+            Assert.True(reader.Read());
+            Assert.Equal(typeof(Paragraph), reader.ElementType);
+            Assert.True(reader.IsStartElement);
+            Assert.True(reader.Read());
+            Assert.Equal(typeof(Paragraph), reader.ElementType);
+            Assert.True(reader.IsEndElement);
+            Assert.True(reader.Read());
+            Assert.Equal(typeof(Body), reader.ElementType);
+            Assert.True(reader.IsEndElement);
+            Assert.True(reader.Read());
+            Assert.Equal(typeof(Document), reader.ElementType);
+            Assert.True(reader.IsEndElement);
+            Assert.False(reader.Read());
+            Assert.True(reader.EOF);
+
+            reader.Close();
+        }
+
+        /// <summary>
+        /// Test that the OpenXmlReader can handle a misc node after the last element.
+        ///</summary>
+        [Fact]
+        public void TestMiscNodeAfterDocument() {
+            const string PartText = @"
+    <w:document xmlns:v=""urn:schemas-microsoft-com:vml"" xmlns:w=""http://schemas.openxmlformats.org/wordprocessingml/2006/main"">
+      <w:body>
+        <w:p>
+        </w:p>
+      </w:body>
+    </w:document> <!--Your comment-->";
+
+            using var stream = new MemoryStream(Encoding.UTF8.GetBytes(PartText), true);
+            using var reader = OpenXmlReader.Create(stream);
+            Assert.True(reader.Read());
+            Assert.False(reader.EOF);
+            Assert.Equal(typeof(Document), reader.ElementType);
+            Assert.True(reader.IsStartElement);
+            Assert.True(reader.Read());
+            Assert.Equal(typeof(Body), reader.ElementType);
+            Assert.True(reader.IsStartElement);
+            Assert.True(reader.Read());
+            Assert.Equal(typeof(Paragraph), reader.ElementType);
+            Assert.True(reader.IsStartElement);
+            Assert.True(reader.Read());
+            Assert.Equal(typeof(Paragraph), reader.ElementType);
+            Assert.True(reader.IsEndElement);
+            Assert.True(reader.Read());
+            Assert.Equal(typeof(Body), reader.ElementType);
+            Assert.True(reader.IsEndElement);
+            Assert.True(reader.Read());
+            Assert.Equal(typeof(Document), reader.ElementType);
+            Assert.True(reader.IsEndElement);
+
+            // not reading the misc node after the end of the document
+            Assert.False(reader.Read());
+            Assert.True(reader.EOF);
+            reader.Close();
+        }
+
         /// <summary>
         ///A test for OpenXmlPartReader
         ///</summary>