Question

考虑以下课程：

public class SampleXmlGenerator
{
    public byte[] GenerateDocumentBytes()
    {
        byte[] fileBytes;
        using (var xmlStream = new MemoryStream())
        {
            using (var myWriter = new XmlTextWriter(xmlStream, Encoding.GetEncoding("UTF-8")))
            {
                myWriter.Formatting = Formatting.Indented;
                myWriter.Indentation = 4;
                myWriter.IndentChar = ' ';
                myWriter.WriteStartDocument();
                myWriter.WriteStartElement("foo");
                myWriter.WriteString("bar");
                myWriter.WriteEndElement(); // end foo

                myWriter.Flush();

                fileBytes = xmlStream.ToArray();
            }
        }

        return fileBytes;
    }
}

使用以下单元测试：

[TestClass]
public class TestSampleXmlGenerator
{
    [TestMethod]
    public void TextEmptyDocument()
    {
        var actualBytes = new SampleXmlGenerator().GenerateDocumentBytes();
        var actualUtf8String = Encoding.UTF8.GetString(actualBytes);
        Console.Out.WriteLine("// actualUtf8String");
        Console.Out.WriteLine(actualUtf8String);

        var actualDefaultString = Encoding.Default.GetString(actualBytes);
        Console.Out.WriteLine("// actualDefaultString");
        Console.Out.WriteLine(actualDefaultString);


        var expectedString = @"<?xml version=""1.0"" encoding=""utf-8""?>
<foo>bar</foo>";
        var expectedBytes = Encoding.UTF8.GetBytes(expectedString);

//      var expectedBytes = Encoding.Convert(Encoding.Default, Encoding.UTF8, Encoding.Default.GetBytes(@"<?xml version=""1.0"" encoding=""utf-8""?>
//<foo>bar</foo>"));
//      var expectedString = Encoding.UTF8.GetString(expectedBytes);

        Console.Out.WriteLine("// expectedString");
        Console.Out.WriteLine(expectedString);

        Assert.AreEqual(expectedBytes.Length, actualBytes.Length);
        //Assert.AreEqual(expectedString, actualUtf8String);
    }
}

最后输出：

Assert.AreEqual failed. Expected:<54>. Actual:<57>. 

// actualUtf8String
<?xml version="1.0" encoding="utf-8"?>
<foo>bar</foo>

// actualDefaultString
ï»¿<?xml version="1.0" encoding="utf-8"?>
<foo>bar</foo>

// expectedString
<?xml version="1.0" encoding="utf-8"?>
<foo>bar</foo>

expectedString和actualUtf8String看起来一样，但不是。

actualDefaultString在开头显示了3个额外的字符。

那是什么给出的？如何测试/比较生成的XML？我该怎么办？

Answer 1

您所看到的是BOM序言

https://msdn.microsoft.com/en-us/library/system.text.encoding.getpreamble(v=vs.110).aspx

您可以测试它，也可以不对其进行序列化。

Answer 2

根据Martin和Keith的建议，以及一些额外的研究，我最终以下列方式（基于以下SO article）从单元测试中生成的XML字节中删除BOM：

        var xmlBytes = new SampleXmlGenerator().GenerateDocumentBytes();
        var newXmlDoc = new XmlDocument {PreserveWhitespace = true};
        newXmlDoc.Load(new MemoryStream(xmlBytes));
        var actualBytes = Encoding.UTF8.GetBytes(newXmlDoc.OuterXml);

现在单元测试通过了！

单元测试Xml编码问题

2 个答案: