Merge pull request #102 from VictorLoktev/master

Added check for restricted XML characters.

DocX/HelperFunctions.cs

@@ -17,7 +17,18 @@ namespace Novacode
         public const string DOCUMENT_DOCUMENTTYPE = "application/vnd.openxmlformats-officedocument.wordprocessingml.document.main+xml";
         public const string TEMPLATE_DOCUMENTTYPE = "application/vnd.openxmlformats-officedocument.wordprocessingml.template.main+xml";
 
-            public static bool IsNullOrWhiteSpace(this string value)
+		/// <summary>
+		///  List of restricted character in xml: [#x1-#x8] | [#xB-#xC] | [#xE-#x1F] | [#x7F-#x84] | [#x86-#x9F]
+		///  See:  https://www.w3.org/TR/xml11/#sec-xml11
+		/// </summary>
+		public static readonly char[] RestrictedXmlChar = new char[] {
+			'\x1','\x2','\x3','\x4','\x5','\x6','\x7','\x8','\xb','\xc','\xe','\xf',
+			'\x10','\x11','\x12','\x13','\x14','\x15','\x16','\x17','\x18','\x19','\x1a','\x1b','\x1c','\x1e','\x1f',
+			'\x7f','\x80','\x81','\x82','\x83','\x84','\x86','\x87','\x88','\x89','\x8a','\x8b','\x8c','\x8d','\x8e','\x8f',
+			'\x90','\x91','\x92','\x93','\x94','\x95','\x96','\x97','\x98','\x99','\x9a','\x9b','\x9c','\x9d','\x9e','\x9f'
+		};
+
+		public static bool IsNullOrWhiteSpace(this string value)
             {
                 if (value == null) return true;
                 return string.IsNullOrEmpty(value.Trim());
@@ -589,7 +600,15 @@ namespace Novacode
                         break;
 
                     default:
-                        sb.Append(c);
+						// Check the character against restricted list:
+						// RestrictedChar	   ::=   	[#x1-#x8] | [#xB-#xC] | [#xE-#x1F] | [#x7F-#x84] | [#x86-#x9F]
+						// See https://www.w3.org/TR/xml11/#sec-xml11 
+						if( RestrictedXmlChar.Contains( c ) )
+						{
+							// skip the character
+						}
+						else
+							sb.Append(c);
                         break;
                 }
                 

DocX/Paragraph.cs

@@ -1097,7 +1097,7 @@ namespace Novacode
             XElement h_xml;
             if (index == 0)
             {
-                // Add this hyperlink as the last element.
+                // Add this hyperlink as the first element.
                 Xml.AddFirst(h.Xml);
 
                 // Extract the picture back out of the DOM.
@@ -1138,10 +1138,12 @@ namespace Novacode
                     h_xml = (XElement)run.Xml.NextNode;
                 }
 
-                h_xml.SetAttributeValue(DocX.r + "id", Id);
-            }
+			}
+			h_xml.SetAttributeValue( DocX.r + "id", Id );
 
-            return this;
+			this.runs = Xml.Elements().Last().Elements( XName.Get( "r", DocX.w.NamespaceName ) ).ToList();
+
+			return this;
         }
 
         /// <summary>

DocX/Table.cs

@@ -785,19 +785,29 @@ namespace Novacode
 
                 if (design == TableDesign.Custom)
                 {
-                    if (string.IsNullOrEmpty(_customTableDesignName))
-                    {
-                        design = TableDesign.None;
-                        if (style != null)
-                            style.Remove();
-
-                    }
-                    else
-                    {
-                        val.Value = _customTableDesignName;
-                    }
-                }
-                else
+					#region Code is commented out
+					// The code gives a problem while copiing a table.
+					// Look at Test_Clone_Table_Twice method in test.
+					//Example:
+					//Table tab1 = doc.Tables[ 0 ];
+					//Table tab2 = doc.InsertTable( tab1 );
+					//Table tab3 = doc.InsertTable( tab2 ); - here we have exception at "var styleElement =" line below in this method
+					// The source of the problem is loosing the "<w:tblStyle w:val="a3"/>" by the commented code
+
+					//if (string.IsNullOrEmpty(_customTableDesignName))
+					//{
+					//    design = TableDesign.None;
+					//    if (style != null)
+					//        style.Remove();
+
+					//}
+					//else
+					//{
+					//    val.Value = _customTableDesignName;
+					//}
+					#endregion
+				}
+				else
                 {
                     switch (design)
                     {
@@ -1132,9 +1142,10 @@ namespace Novacode
                         let styleId = e.Attribute(XName.Get("styleId", DocX.w.NamespaceName))
                         where (styleId != null && styleId.Value == val.Value)
                         select e
-                    ).First();
+                    ).FirstOrDefault();
 
-                    Document.styles.Element(XName.Get("styles", DocX.w.NamespaceName)).Add(styleElement);
+					if( styleElement != null )
+						Document.styles.Element(XName.Get("styles", DocX.w.NamespaceName)).Add(styleElement);
                 }
             }
         }

DocX/bin/Release/DocX.XML

@@ -475,6 +475,12 @@
             The custom property to display.
             </summary>
         </member>
+        <member name="F:Novacode.HelperFunctions.RestrictedXmlChar">
+            <summary>
+             List of restricted character in xml: [#x1-#x8] | [#xB-#xC] | [#xE-#x1F] | [#x7F-#x84] | [#x86-#x9F]
+             See:  https://www.w3.org/TR/xml11/#sec-xml11
+            </summary>
+        </member>
         <member name="M:Novacode.HelperFunctions.ContainsEveryChildOf(System.Xml.Linq.XElement,System.Xml.Linq.XElement,Novacode.MatchFormattingOptions)">
             <summary>
             Checks whether 'toCheck' has all children that 'desired' has and values of 'val' attributes are the same

UnitTests/DocXUnitTests.cs

@@ -150,6 +150,28 @@ namespace UnitTests
             }
         }
 
+		[Test]
+		public void Test_InvalidCharacter()
+		{
+			using( var output = File.Open( Path.Combine( _directoryWithFiles, "InvalidCharacters.docx" ), FileMode.Create ) )
+			{
+				using( var doc = DocX.Create( output ) )
+				{
+					doc.InsertParagraph( "\b" );
+					Exception ex = null;
+					try
+					{
+						doc.Save();
+					}
+					catch( Exception e )
+					{
+						ex = e;
+					}
+					Assert.IsTrue( ex == null );
+				}
+			}
+		}
+
 		/// <summary>
 		/// TextRemove should not remove empty paragraphs in case the paragraph is alone in the cell.
 		/// In the rest cases empty paragraph may be removed.
@@ -274,6 +296,29 @@ namespace UnitTests
 			}
 		}
 
+		[Test]
+		public void Test_Clone_Table_Twice()
+		{
+			using( var input = File.Open( Path.Combine( _directoryWithFiles, "TableSpecifiedHeights.docx" ), FileMode.Open ) )
+			{
+				using( var doc = DocX.Load( input ) )
+				{
+					// Make sure content of the file is ok for test
+					Assert.IsTrue( doc.Tables.Count == 1 );
+
+					Table tab1 = doc.Tables[ 0 ];
+					doc.InsertParagraph( "" );
+					Table tab2 = doc.InsertTable( tab1 );
+					Assert.IsTrue( doc.Tables.Count == 2 );
+					doc.InsertParagraph( "" );
+					Table tab3 = doc.InsertTable( tab2 );
+					Assert.IsTrue( doc.Tables.Count == 3 );
+
+					doc.SaveAs( Path.Combine( _directoryWithFiles, "TwoClonedTables.docx" ) );
+				}
+			}
+		}
+
 		public string ReplaceFunc(string findStr)
         {
             var testPatterns = new Dictionary<string, string>
@@ -803,7 +848,105 @@ namespace UnitTests
             }
         }
 
-        [Test]
+		/// <summary>
+		/// This test fills two tables with hyperlinks.
+		/// </summary>
+		[Test]
+		public void Test_Insert_Hyperlink_In_Tables()
+		{
+			using( var input = File.Open( Path.Combine( _directoryWithFiles, "TableSpecifiedHeights.docx" ), FileMode.Open ) )
+			{
+				using( var doc = DocX.Load( input ) )
+				{
+					// Make sure content of the file is ok for test
+					Assert.IsTrue( doc.Tables.Count > 0 );
+					Table tab1 = doc.Tables[ 0 ];
+					Assert.IsTrue( tab1.RowCount > 0 );
+					Assert.IsTrue( tab1.Rows[0].ColumnCount > 0 );
+					doc.InsertParagraph( "" );
+					Table tab2 = doc.InsertTable( tab1 );
+					Assert.IsTrue( tab2.RowCount > 0 );
+
+					Row row1 = tab1.Rows[ 0 ];
+					Row row2 = tab2.Rows[ 0 ];
+
+					// 10 times insert hyperlinks in both tables in tic-tak order
+					for( int index = 0; index < 10; index++ )
+					{
+						Row newRow1 = tab1.InsertRow( row1 );
+						Row newRow2 = tab2.InsertRow( row2 );
+
+						Hyperlink h1 = doc.AddHyperlink(
+							string.Format( "Table {0}, Row {1}. Google searches for {0} {1}", 1, index + 1 ),
+							new Uri( string.Format( "https://www.google.com/search?q=Table{0}Row{1}", 1, index + 1 ) ) );
+						newRow1.Cells[ 0 ].Paragraphs[ 0 ].InsertHyperlink( h1 );
+
+						Hyperlink h2 = doc.AddHyperlink(
+							string.Format( "Table {0}, Row {1}. Google searches for {0} {1}", 2, index + 1 ),
+							new Uri( string.Format( "https://www.google.com/search?q=Table{0}Row{1}", 2, index + 1 ) ) );
+						newRow2.Cells[ 0 ].Paragraphs[ 0 ].InsertHyperlink( h2 );
+
+					}
+					//Make sure links are ok and in right order
+					for( int index = 0; index < doc.Hyperlinks.Count; index++ )
+					{
+						Hyperlink h = doc.Hyperlinks[ index ];
+						string text = string.Format( "Table {0}, Row {1}. Google searches for {0} {1}", ( index / 10 ) + 1, ( index ) % 10 + 1 );
+						string uri = string.Format( "https://www.google.com/search?q=Table{0}Row{1}", ( index / 10 ) + 1, ( index ) % 10 + 1 );
+						Assert.IsTrue( string.Compare( h.Text, text ) == 0 );
+						Assert.IsTrue( h.Uri != null );
+						Assert.IsTrue( string.Compare( h.Uri.ToString(), uri ) == 0 );
+					}
+					doc.SaveAs( Path.Combine( _directoryDocuments, "Test_Insert_Hyperlink_In_Tables.docx" ) );
+				}
+			}
+		}
+
+		/// <summary>
+		/// This test makes 2 file. The first uses InsertHyperlink. The second uses AppendHyperlink.
+		/// The both hyperlink collections should be equal to each other.
+		/// We need be sure the bug in InsertHyperlink is fixed (id attribute in hyperlink was empty and order of inserteed hyperlinks was broken).
+		/// </summary>
+		[Test]
+		public void Test_Compare_InsertHyperlink_And_AppendHyperLinks()
+		{
+			string fileName1 = Path.Combine( _directoryDocuments, "Test_InsertHyperLinks.docx" );
+			string fileName2 = Path.Combine( _directoryDocuments, "Test_AppendHyperlinks.docx" );
+			using( DocX document1 = DocX.Create( fileName1 ) )
+			{
+				using( DocX document2 = DocX.Create( fileName2 ) )
+				{
+					for( int index = 0; index < 10; index++ )
+					{
+						Hyperlink h = document1.AddHyperlink(
+							string.Format( "Google searches for {0}", index + 1 ),
+							new Uri( string.Format( "https://www.google.com/search?q={0}", index + 1 ) ) );
+						document1.InsertParagraph( "" ).InsertHyperlink( h );
+					}
+					document1.Save();
+
+					for( int index = 0; index < 10; index++ )
+					{
+						Hyperlink h = document2.AddHyperlink(
+							string.Format( "Google searches for {0}", index + 1 ),
+							new Uri( string.Format( "https://www.google.com/search?q={0}", index + 1 ) ) );
+						document2.InsertParagraph( "" ).AppendHyperlink( h );
+					}
+					document2.Save();
+
+					Assert.IsTrue( document1.Hyperlinks.Count == document2.Hyperlinks.Count );
+					for( int index = 0; index < document1.Hyperlinks.Count; index++ )
+					{
+						Hyperlink h1 = document1.Hyperlinks[ index ];
+						Hyperlink h2 = document2.Hyperlinks[ index ];
+						Assert.IsTrue( string.Compare( h1.Text, h2.Text ) == 0 );
+						Assert.IsTrue( string.Compare( h1.Uri.ToString(), h2.Uri.ToString() ) == 0 );
+					}
+				}
+			}
+		}
+
+		[Test]
         public void Test_Insert_Hyperlink()
         {
             // Load test document.