The text is escaped either using XML escape entities, or CDATA sections.
using System;
using System.Xml;
using System.Text.RegularExpressions;
namespace log4net.Util
{
/// <summary>
/// Utility class for transforming XML strings.
/// </summary>
public static class Transform
{
private const string CdataEnd = "]]>";
private const string CdataUnescapableToken = "]]";
/// <summary>
/// Characters illegal in XML 1.0.
/// </summary>
private static readonly Regex InvalidChars = new Regex(@"[^\x09\x0A\x0D\x20-\uD7FF\uE000-\uFFFD]", RegexOptions.Compiled);
/// <summary>
/// Replace invalid XML characters in text string.
/// </summary>
/// <param name="textData">the XML text input string</param>
/// <param name="mask">the string to use in place of invalid characters. This defaults to a ?. Set it to the empty string to simply remove offending characters.</param>
/// <returns>A string that does not contain invalid XML characters.</returns>
/// <remarks>
/// <para>
/// Certain Unicode code points are not allowed in the XML InfoSet, for
/// details see: <a href="http://www.w3.org/TR/REC-xml/#charsets">http://www.w3.org/TR/REC-xml/#charsets</a>.
/// </para>
/// <para>
/// This method replaces any illegal characters in the input string
/// with the mask string specified.
/// </para>
/// </remarks>
public static string MaskXmlInvalidCharacters(string textData, string mask = "?")
{
return InvalidChars.Replace(textData, mask);
}
/// <summary>
/// Write a string to an <see cref="XmlWriter"/>.
/// The text is escaped either using XML escape entities, or CDATA sections.
/// </summary>
/// <param name="writer">the <see cref="XmlWriter"/> to write to.</param>
/// <param name="textData">the string to write.</param>
/// <param name="invalidCharReplacement">The string that will be used to replace non-XML compliant characters. This defaults to a ?. Set it to the empty string to simply remove offending characters.</param>
public static void WriteEscapedXmlString(XmlWriter writer, string textData, string invalidCharReplacement = "?")
{
var stringData = MaskXmlInvalidCharacters(textData, invalidCharReplacement);
// Write either escaped text or CDATA sections
var weightCData = 12 * (1 + CountSubstrings(stringData, CdataEnd));
var weightStringEscapes = 3 * (CountSubstrings(stringData, "<") + CountSubstrings(stringData, ">")) + 4 * CountSubstrings(stringData, "&");
if (weightStringEscapes <= weightCData)
{
// Write string using string escapes
writer.WriteString(stringData);
}
else
{
// Write string using CDATA section
var end = stringData.IndexOf(CdataEnd, StringComparison.Ordinal);
if (end < 0)
{
writer.WriteCData(stringData);
}
else
{
var start = 0;
while (end > -1)
{
writer.WriteCData(stringData.Substring(start, end - start));
if (end == stringData.Length - 3)
{
start = stringData.Length;
writer.WriteString(CdataEnd);
break;
}
writer.WriteString(CdataUnescapableToken);
start = end + 2;
end = stringData.IndexOf(CdataEnd, start, StringComparison.Ordinal);
}
if (start < stringData.Length)
{
writer.WriteCData(stringData.Substring(start));
}
}
}
}
/// <summary>
/// Count the number of times that the substring occurs in the text.
/// </summary>
/// <param name="text">the text to search</param>
/// <param name="substring">the substring to find</param>
/// <returns>the number of times the substring occurs in the text</returns>
/// <remarks>
/// <para>
/// The substring is assumed to be non repeating within itself.
/// </para>
/// </remarks>
private static int CountSubstrings(string text, string substring)
{
var length = text.Length;
if (length == 0)
{
return 0;
}
var substringLength = substring.Length;
if (substringLength == 0)
{
return 0;
}
int count = 0, offset = 0;
while (offset < length)
{
var index = text.IndexOf(substring, offset, StringComparison.Ordinal);
if (index == -1)
{
break;
}
count++;
offset = index + substringLength;
}
return count;
}
}
}
// ---------------------------------------------------
// Tests
// ---------------------------------------------------
using log4net.Util;
using NUnit.Framework;
namespace log4net.Tests.Util
{
[TestFixture]
public class TransformTest
{
[Test]
public void MaskXmlInvalidCharactersAllowsJapaneseCharacters()
{
string kome = "\u203B";
Assert.AreEqual(kome, Transform.MaskXmlInvalidCharacters(kome, "?"));
}
[Test]
public void MaskXmlInvalidCharactersMasks0Char()
{
string c = "\u0000";
Assert.AreEqual("?", Transform.MaskXmlInvalidCharacters(c, "?"));
}
}
}