Merge pull request #1486 from antony-liu/npoi-bug-1481

fix: avoid corrupting docx file with comments
nissl-lab · Feb 7, 2025 · 58de0e4 · 58de0e4
2 parents 735c986 + 4e59a37
commit 58de0e4
Show file tree

Hide file tree

Showing 3 changed files with 73 additions and 6 deletions.
diff --git a/OpenXmlFormats/Wordprocessing/Markup.cs b/OpenXmlFormats/Wordprocessing/Markup.cs
@@ -182,15 +182,49 @@ public class CT_Comments
 
         private List<CT_Comment> commentField;
 
+        private Dictionary<string, string> extraXmlNamespace = new Dictionary<string, string>();
+        private Dictionary<string, string> defaultXmlNamespace = new Dictionary<string, string>(){
+            { "o", "urn:schemas-microsoft-com:office:office" },
+            { "ve", "http://schemas.openxmlformats.org/markup-compatibility/2006" },
+            { "r", "http://schemas.openxmlformats.org/officeDocument/2006/relationships" },
+            { "v", "urn:schemas-microsoft-com:vml" },
+            { "m", "http://schemas.openxmlformats.org/officeDocument/2006/math" },
+            { "w", "http://schemas.openxmlformats.org/wordprocessingml/2006/main" },
+            { "wp", "http://schemas.openxmlformats.org/drawingml/2006/wordprocessingDrawing" },
+            { "w10", "urn:schemas-microsoft-com:office:word" },
+            { "w14", "http://schemas.microsoft.com/office/word/2010/wordml" },
+            { "wne", "http://schemas.microsoft.com/office/word/2006/wordml" },
+
+        };
         public CT_Comments()
         {
             this.commentField = new List<CT_Comment>();
         }
+        private void FindExtraNamespace(XmlNode node)
+        {
+            if (node.Attributes != null)
+            {
+                foreach (XmlAttribute attr in node.Attributes)
+                {
+                    if (attr.Name.StartsWith("xmlns:"))
+                    {
+                        string[] strs = attr.Name.Split(':');
+                        if (strs.Length == 2)
+                        {
+                            if (!extraXmlNamespace.ContainsKey(strs[1]) &&
+                                !defaultXmlNamespace.ContainsKey(strs[1]))
+                                extraXmlNamespace.Add(strs[1], attr.Value);
+                        }
+                    }
+                }
+            }
+        }
         public static CT_Comments Parse(XmlNode node, XmlNamespaceManager namespaceManager)
         {
             if (node == null)
                 return null;
             CT_Comments ctObj = new CT_Comments();
+            ctObj.FindExtraNamespace(node);
             ctObj.comment = new List<CT_Comment>();
             foreach (XmlNode childNode in node.ChildNodes)
             {
@@ -201,16 +235,21 @@ public static CT_Comments Parse(XmlNode node, XmlNamespaceManager namespaceManag
         }
 
 
-
+        private void WriteXmlNamespace(StreamWriter sw, Dictionary<string, string> ns)
+        {
+            if(ns == null || ns.Count == 0)
+                return;
+            foreach (var kv in ns)
+            {
+                sw.Write(string.Format($"xmlns:{kv.Key}=\"{kv.Value}\" ", kv.Key, kv.Value));
+            }
+        }
         internal void Write(StreamWriter sw)
         {
             sw.Write("<?xml version=\"1.0\" encoding=\"UTF-8\" standalone=\"yes\"?>");
             sw.Write(string.Format("<w:comments "));
-            sw.Write("xmlns:ve=\"http://schemas.openxmlformats.org/markup-compatibility/2006\" xmlns:o=\"urn:schemas-microsoft-com:office:office\" ");
-            sw.Write("xmlns:r=\"http://schemas.openxmlformats.org/officeDocument/2006/relationships\" xmlns:m=\"http://schemas.openxmlformats.org/officeDocument/2006/math\" ");
-            sw.Write("xmlns:v=\"urn:schemas-microsoft-com:vml\" xmlns:wp=\"http://schemas.openxmlformats.org/drawingml/2006/wordprocessingDrawing\" ");
-            sw.Write("xmlns:w10=\"urn:schemas-microsoft-com:office:word\" xmlns:w=\"http://schemas.openxmlformats.org/wordprocessingml/2006/main\" ");
-            sw.Write("xmlns:wne=\"http://schemas.microsoft.com/office/word/2006/wordml\">");
+            WriteXmlNamespace(sw, defaultXmlNamespace);
+            WriteXmlNamespace(sw, extraXmlNamespace);
             sw.Write(">");
             if (this.comment != null)
             {

diff --git a/testcases/ooxml/XWPF/UserModel/TestXWPFComments.cs b/testcases/ooxml/XWPF/UserModel/TestXWPFComments.cs
@@ -3,6 +3,8 @@
     using NPOI.XWPF.UserModel;
     using NUnit.Framework;
     using System.Collections.Generic;
+    using System.IO;
+    using System.Reflection.Metadata;
 
     [TestFixture]
     public class TestXWPFComments
@@ -56,5 +58,31 @@ public void TestReadComments()
                 Assert.AreEqual(1, allPictures.Count);
             }
         }
+
+        [Test]
+        public void TestNPOIBug1481()
+        {
+            using(XWPFDocument doc = XWPFTestDataSamples.OpenSampleDocument("NPOI-bug-1481.docx"))
+            {
+                XWPFComment[] comments = doc.GetComments();
+                Assert.AreEqual(1, comments.Length);
+
+                XWPFComment comment = comments[0];
+                Assert.AreEqual("Claudio Pais", comment.GetAuthor());
+                Assert.AreEqual("2025-01-23T17:18:00Z", comment.Date);
+                Assert.AreEqual("Bla bla", comment.GetText());
+
+                XWPFDocument docIn = XWPFTestDataSamples.WriteOutAndReadBack(doc);
+
+                comments = docIn.GetComments();
+                Assert.AreEqual(1, comments.Length);
+
+                comment = comments[0];
+                Assert.AreEqual("Claudio Pais", comment.GetAuthor());
+                Assert.AreEqual("2025-01-23T17:18:00Z", comment.Date);
+                Assert.AreEqual("Bla bla", comment.GetText());
+
+            }
+        }
     }
 }
diff --git a/testcases/test-data/document/NPOI-bug-1481.docx b/testcases/test-data/document/NPOI-bug-1481.docx