Question

如何读取Unicode字符？就像“ä”

public static string Read(int length, string absolutePath)
{
    StringBuilder resultAsString = new StringBuilder();

    using (MemoryMappedFile memoryMappedFile = MemoryMappedFile.CreateFromFile(absolutePath))
    using (MemoryMappedViewStream memoryMappedViewStream = memoryMappedFile.CreateViewStream(0, length))
    {
        for (int i = 0; i < length; i++)
        {                  
            int result = memoryMappedViewStream.ReadByte();

            if (result == -1)
            {
                break;
            }
            char letter = (char)result;

            resultAsString.Append(letter);
        }
    }

    return resultAsString.ToString();
}

读取的int（结果）为195，而char强制转换给我的结果不理想。

Answer 1

不确定是不是您要的，但是可以使用StreamReader

StreamReader sr = new StreamReader(stream, Encoding.Unicode);

Answer 2

如果您只想将UTF-8文件加载并读取到字符串变量中，则代码可以很简单

var text = File.ReadAllText(filePath, Encoding.UTF8);

但是，如果您坚持要逐字节处理UTF-8数据，则解析会更加复杂。
这是一个粗略的（但可行的）草图，以配合您的原始代码：

StringBuilder resultAsString = new StringBuilder();

using (MemoryMappedFile memoryMappedFile = MemoryMappedFile.CreateFromFile(filePath))
using (MemoryMappedViewStream viewStream = memoryMappedFile.CreateViewStream(0, new FileInfo(filePath).Length))
{
    int b;
    while((b = viewStream.ReadByte()) != -1)
    {
        int acc = b;

        bool readUtfDataBytes(int bytesToRead)
        {
            while (bytesToRead-- > 0)
            {
                var nextB = viewStream.ReadByte();
                if (nextB == -1) return false; // EOS reached
                if ((nextB & 0xC0) != 0x80) return false; // invalid UTF-8 data byte
                acc <<= 6;
                acc |= nextB & 0x3F;
            }
            return true;
        }

        if (b >= 0xF0) // 1111 0000
        {
            acc &= 0x07;
            if (!readUtfDataBytes(3)) break; // break on malformed UTF-8
        }
        else if (b >= 0xE0) // 1110 0000
        {
            acc &= 0x0F;
            if (!readUtfDataBytes(2)) break; // break on malformed UTF-8
        }
        else if (b >= 0xC0) // 1100 0000
        {
            acc &= 0x1F;
            if (!readUtfDataBytes(1)) break; // break on malformed UTF-8
        }
        else if (b >= 0x80)
        {
            break; // break on malformed UTF-8
        }

        if (acc == 0xFEFF)
        {
            // ignore UTF-8 BOM
        }
        else
        {
            resultAsString.Append(Char.ConvertFromUtf32(acc));
        }
    }
}

从大文本文件中读取unicode char

2 个答案: