随着Windows8的发布,微软给出了一个Windows Runtime(以下简称WinRT),据说是用COM技术实现的。在结合使用.NET和WinRT时,你会发现它们对相同的概念,有不同的实现,或者说是类,比如异步操作,.NET中用Task概念,而WinRT则是用IAsyncInfo,IAsyncAction等,而在流的概念中,.NET围绕Stream类建立,而WinRT则先定义了三个主要的接口,然后逐一实现之。本文就是集中在“流”的相互转换上,因为你在编写Metro App时,会用到WinRT组件。
首先,.NET的Stream可谓是集读、写以及流定位于一身的一个类,那么在WinRT中则将它抽象成三个不同的接口,分别为:IInputStream、IOutupStream和IRandomAccessStream,其实这三个接口就是对应Stream所提供的功能。当然还有别的接口,这里暂不介绍。还有一个要介绍的是IBuffer,这个接口提供了对我们传统放置字节数组的byte[]的抽象,而且只提供了Capcity和Length两个属性(没有方法),Capcity是说这个IBuffer能够容纳多少字节,而Length则说明实际有多少字节,WinRT中也会有一个实现了该接口的类,称为Buffer,我们使用输入输出流时,都会用到。
然后,我们在转换时,最好加入System.IO命名空间,这个空间提供了我们需要的转换的扩展方法。
1. 将IBuffer转换成一个.NET Stream:
由于已经知道了一个字节块(IBuffer),那么我们可以把它放入到一个内存的随机访问流中,就是从Buffer中读取byte到内存流中(InMemoryRandomAccessStream,它实现了上诉的三大接口),再通过扩展方法转换成Stream,代码如下
InMemoryRandomAccessStream memoryStremWRT = new InMemoryRandomAccessStream(); await memoryStremWRT.ReadAsync(buffer,buffer.Length,InputStreamOptions.None); Stream stream = memoryStremWRT.AsStream();
2.将一个IOutputStream转换成为Stream:
stream = outputStream.AsStreamForWrite();
3.将一个IInputStream转换成Stream:
stream = inputStream.AsStreamForRead();
4.将IRandomAccess转换成Stream:
stream=randomAccess.AsStream();
5.Stream 转成Buffer:
由于buffer需要读取数据,所以要一个输入流,此处使用DataReader来加载,它的构造参数就是一个输入流。
IBuffer buffer=null; var inputstream=stream.AsInputStream(); using(var dataReader=new DataReader(inputstream)) { await dataReader.LoadAsync((uint)stream.Length); buffer=dataReader.DetachBuffer(); }
6.Stream 转成IIputStream
var inputstream=stream.AsInputStream();
7.Stream转成IOutputStream
var outputstream=stream.AsOutputStream();
8.Stream转成 IRandomAccess:
此处没有直接提供扩展方法,所以我们的思路还是先构造出一个输入流来获取数据:
IBuffer buffer=null; var inputstream=stream.AsInputStream(); using(var dataReader=new DataReader(inputstream)) { await dataReader.LoadAsync((uint)stream.Length); buffer=dataReader.DetachBuffer(); } var randomAccessStream =new InMemoryRandomAccessStream (); await randomAccessStream.WriteAsync(buffer);
以上用到了DataReader类,对应的还有DataWriter类,这两个类和.NET中的StreamReader和StreamWriter的用法和概念一样,使用了适配器模式,将我们平时用到的类型,比如文本啊,整形数据啊,输入到流或者从流中读出,那么底层的字符,或者整形与byte之间的转换就不需要我们操心了,最多我们要指明是用大端还是小端表示,或者使用什么字节编码。
对于DataReader的含义,就是说我们要从一个输入流中读取数据(数据源是输入流,目标是从流中组装的具体变量值),至于数据的具体含义,那么就看你自己的需求了,一般情况下你是知道流到底是应该编码成string,还是组成int,long,亦或是两者都有,只要顺序搞对就行。
对于DataWriter的含义,则与Reader相对,就是我们输入我们需要的数据,无论是byte,还是string 还是int,long,经过适度的编码以及分解,然后输出到一个流中。
以上的参考代码如下:
// Initialize the in-memory stream where data will be stored. using (var stream = new Windows.Storage.Streams.InMemoryRandomAccessStream()) { // Create the data writer object backed by the in-memory stream. using (var dataWriter = new Windows.Storage.Streams.DataWriter(stream)) { dataWriter.UnicodeEncoding = Windows.Storage.Streams.UnicodeEncoding.Utf8; dataWriter.ByteOrder = Windows.Storage.Streams.ByteOrder.LittleEndian; // Parse the input stream and write each element separately. string[] inputElements = ElementsToWrite.Text.Split(';'); foreach (string inputElement in inputElements) { uint inputElementSize = dataWriter.MeasureString(inputElement); dataWriter.WriteUInt32(inputElementSize); dataWriter.WriteString(inputElement); } // Send the contents of the writer to the backing stream. await dataWriter.StoreAsync(); // For the in-memory stream implementation we are using, the flushAsync call // is superfluous,but other types of streams may require it. await dataWriter.FlushAsync(); // In order to prolong the lifetime of the stream, detach it from the // DataWriter so that it will not be closed when Dispose() is called on // dataWriter. Were we to fail to detach the stream, the call to // dataWriter.Dispose() would close the underlying stream, preventing // its subsequent use by the DataReader below. dataWriter.DetachStream(); } // Create the input stream at position 0 so that the stream can be read // from the beginning. using (var inputStream = stream.GetInputStreamAt(0)) { using (var dataReader = new Windows.Storage.Streams.DataReader(inputStream)) { // The encoding and byte order need to match the settings of the writer // we previously used. dataReader.UnicodeEncoding = Windows.Storage.Streams.UnicodeEncoding.Utf8; dataReader.ByteOrder = Windows.Storage.Streams.ByteOrder.LittleEndian; // Once we have written the contents successfully we load the stream. await dataReader.LoadAsync((uint)stream.Size); var receivedStrings = ""; // Keep reading until we consume the complete stream. while (dataReader.UnconsumedBufferLength > 0) { // Note that the call to readString requires a length of "code units" // to read. This is the reason each string is preceded by its length // when "on the wire". uint bytesToRead = dataReader.ReadUInt32(); receivedStrings += dataReader.ReadString(bytesToRead) + "\n"; } // Populate the ElementsRead text block with the items we read // from the stream. ElementsRead.Text = receivedStrings; } }