POCO C++库学习和分析 -- 流 (二)

POCO C++库学习和分析 -- 流 (二)


2. 创建自己的流

2.1 自定义流的结构

          在Poco中提供了一个框架用于创建自己的流,并且创建的流都符合C++标准。想一下标准库中流的层次和结构。每一个流都必须有对应的流缓冲,并且在流初始化时提供此流缓冲的指针。Poco中提供了3种流缓冲供选择,分别是BasicBufferedStreamBuf、BasicUnbufferedStreamBuf、BasicBufferedBidirectionalStreamBuf。首先来看一个例子:

#include "Poco/UnbufferedStreamBuf.h"
#include <ostream>
#include <cctype>
class UpperStreamBuf: public UnbufferedStreamBuf
{
public:
	UpperStreamBuf(std::ostream& ostr): _ostr(ostr)
	{}
protected:
	int writeToDevice(char c)
	{
		_ostr.put(toupper(c));
		return charToInt(c);
	}
private:
	std::ostream& _ostr;
};

class UpperIOS: public virtual std::ios
{
public:
	UpperIOS(std::ostream& ostr): _buf(ostr)
	{
		poco_ios_init(&_buf);
	}
protected:
	UpperStreamBuf _buf;
};

class UpperOutputStream: public UpperIOS, public std::ostream
{
public:
	UpperOutputStream(std::ostream& ostr):
	  UpperIOS(ostr),
		  std::ostream(&_buf)
	  {}
};

int main(int argc, char** argv)
{
	UpperOutputStream upper(std::cout);
	upper << "Hello, world!" << std::endl;
	return 0;
}

          从上面的例子中,我们看到用户自定流缓冲类型UpperStreamBuf,并自定义流类型UpperIOS和UpperOutputStream。其中UpperIOS从std::ios虚拟继承。注意这里是虚拟继承,这是为了保证UpperOutputStream多重继承时的菱形结构(std::ostream也是从std::ios虚拟继承的)。再回忆一下,std::ios的作用,其中最重要的就是定义了缓冲区指针。那么很自然的,定义UpperIOS,并让UpperOutputStream从UpperIOS继承的原因,就是为了让UpperOutputStream类拥有自己的缓冲区。再来看从流std::ostream继承的原因,很简单,为了继承"<<"操作符。那么清楚了,如果是输入流,从std::istream继承;是输入输出流从std::iostream继承。

          再从上面的例子看:为什么说,Poco中流框架提供的是一个中介流的框架。当创建UpperOutputStream对象时,我们看它做了什么.
          在调用UpperOutputStream upper(std::cout)时,传入参数是一个std::cout,这是一个已定义的流对象。这个流对象被赋予给UpperIOS类,用来初始化UpperIOS类中的UpperStreamBuf对象;而UpperStreamBuf又并不负责直接输出,它的输出依赖于内部的std::ostream对象,在这里就是std::cout。另一方面由于UpperOutputStream对象同时也继承自std::ostream,而std::ostream必须要求用一个流缓冲对象来初始化,所以在UpperIOS类中的UpperStreamBuf对象初始化后,又被赋给std::ostream对象用作初始化。

          再来看 "upper << "Hello, world!" << std::endl "语句的执行过程:
          1. "<<"操作符是std::ostream的成员函数。也就是说 "Hello, world!"被首先调用了std::ostream函数"<<"。
          2.  我们知道,流的输入输出是委托给其内部的缓冲区对象的。也就是说,在这里"<<"操作符,会委托到std::ostream类内部关联的流缓冲UpperIOS::UpperStreamBuf对象_buf上。
          2. UpperStreamBuf类型的_buf对象并不输出,它把输出任务继续委托给内部的流对象_ostr。在程序里,_ost对象为std::cout。
          3. std::cout对象在收到数据后,委托给std::cout内部关联的流缓冲对象,最终输出。至于std::cout对象关联的流对象是什么,那就不用去管了。这是系统内部实现的。

          转了半天,输出最终还是依赖于系统的默认输出。如果仅是如此,就没有意思了。Poco流在类BasicBufferedStreamBuf、BasicUnbufferedStreamBuf、BasicBufferedBidirectionalStreamBuf上开了一个小口,供用户自定义中间操作。这两个小口就是:
virtual int readFromDevice(char_type* buffer, std::streamsize length);
virtual int writeToDevice(const char_type* buffer, std::streamsize length);


          通过子类继承,用户可以自定义转换行为。任何从Poco流自定义框架出来的类都必须重新定义这两个虚函数。readFromDevice和writeToDevice会在std::ostream中的虚函数overflow和underflow中被调用。

          在了解了Poco自定义流类的流转次序后,再去看Poco中的自定义流Base64Decoder,Base64Encoder等的数据流转,会发现过程都是类似的。下面就不再对这一方面进行叙述。只是讲一下Poco中已定义的3个流缓冲类BasicBufferedStreamBuf、BasicUnbufferedStreamBuf、BasicBufferedBidirectionalStreamBuf。


2.2 BasicUnbufferedStreamBuf

          Poco::BasicUnbufferedStreamBuf是一个使用char作为模板参数类型的偏特化模板类。其完整定义为:

template <typename ch, typename tr> 
class BasicUnbufferedStreamBuf: public std::basic_streambuf<ch, tr>
	/// This is an implementation of an unbuffered streambuf
	/// that greatly simplifies the implementation of
	/// custom streambufs of various kinds.
	/// Derived classes only have to override the methods
	/// readFromDevice() or writeToDevice().
{
	// .....
	int_type _pb;
	bool     _ispb;
}

typedef BasicUnbufferedStreamBuf<char, std::char_traits<char> > UnbufferedStreamBuf;

          Poco::BasicUnbufferedStreamBuf可以说是最简单的创建用户自定流的方法,在其内部没有任何缓冲对象。从Poco::BasicUnbufferedStreamBuf继承的类需要重载readFromDevice和writeToDevice接口。

          对应重载的要求如下:
     int readFromDevice()
          读取并返回单字节无符号字符。如果没有更多的合适数据时,返回char_traits::eof()(-1)。需要注意的是,不要直接返回一个char类型值,因为char类型值是一个有符号数。调用int charToInt(char c)返回一个char转换到int的值。
     int writeToDevice(char c)
          写入一个单字节字符,如果成功返回写入的字符(已整形方式),否则返回char_traits::eof() (-1).

          在Poco中从UnbufferedStreamBuf继承的类见下图:



2.3 BasicBufferedStreamBuf

          Poco::BasicBufferedStreamBuf也同样是使用char作为模板参数类型的偏特化模板类。其完整定义为:

template <typename ch, typename tr, typename ba = BufferAllocator<ch> > 
class BasicBufferedStreamBuf: public std::basic_streambuf<ch, tr>
	/// This is an implementation of a buffered streambuf
	/// that greatly simplifies the implementation of
	/// custom streambufs of various kinds.
	/// Derived classes only have to override the methods
	/// readFromDevice() or writeToDevice().
	///
	/// This streambuf only supports unidirectional streams.
	/// In other words, the BasicBufferedStreamBuf can be
	/// used for the implementation of an istream or an
	/// ostream, but not for an iostream.
{
	// ...
	std::streamsize _bufsize;
	char_type*      _pBuffer;
	openmode        _mode;
}

typedef BasicBufferedStreamBuf<char, std::char_traits<char> > BufferedStreamBuf;

          在BasicBufferedStreamBuf内部存在一个_pBuffer,用作缓冲,供输出使用。Poco::BufferedStreamBuf的实例支持读写,但不支持同时读写。从Poco::BasicBufferedStreamBuf继承的类需要重载readFromDevice和writeToDevice接口。

          对应重载的要求如下:  
     int readFromDevice(char* buffer, std::streamsize length)
          读取指定长度的字符,并把它们放置在buffer中。返回读到的字节数,或者-1(当错误发生时)
     int writeToDevice(const char* buffer, std::streamsize length)
          写入buffer中指定数目的字节。返回读到的字节数,或者-1(当错误发生时)


          在Poco中从BasicBufferedStreamBuf继承的类见下图:




2.4 BasicBufferedBidirectionalStreamBuf

          Poco::BasicBufferedBidirectionalStreamBuf一样是使用char作为模板参数类型的偏特化模板类。其完整定义为:

template <typename ch, typename tr, typename ba = BufferAllocator<ch> > 
class BasicBufferedBidirectionalStreamBuf: public std::basic_streambuf<ch, tr>
	/// This is an implementation of a buffered bidirectional 
	/// streambuf that greatly simplifies the implementation of
	/// custom streambufs of various kinds.
	/// Derived classes only have to override the methods
	/// readFromDevice() or writeToDevice().
	///
	/// In contrast to BasicBufferedStreambuf, this class supports
	/// simultaneous read and write access, so in addition to
	/// istream and ostream this streambuf can also be used
	/// for implementing an iostream.
{
	// ....
	std::streamsize _bufsize;
	char_type*      _pReadBuffer;
	char_type*      _pWriteBuffer;
	openmode        _mode;
}

typedef BasicBufferedBidirectionalStreamBuf<char, std::char_traits<char> > BufferedBidirectionalStreamBuf;

          在BasicBufferedBidirectionalStreamBuf内部存在两个缓冲,_pReadBuffer和_pWriteBuffer,作为缓冲,分别供输入或输出使用。
          Poco::BasicBufferedBidirectionalStreamBuf的实例支持同时读写。从Poco::BasicBufferedBidirectionalStreamBuf继承的类需要重载readFromDevice和writeToDevice接口。

          对应重载的要求如下:
     int readFromDevice(char* buffer, std::streamsize length)
          读取指定长度的字符,并把它们放置在buffer中。返回读到的字节数,或者-1(当错误发生时)
     int writeToDevice(const char* buffer, std::streamsize length)
          写入buffer中指定数目的字节。返回读到的字节数,或者-1(当错误发生时)

          在Poco中从BasicBufferedBidirectionalStreamBuf继承的类见下图:



3. Poco::Base64Encoder和Poco::Base64Decoder

3.1 Base64编码的概念


          于Base64编码的概念的概念出自wiki百科。想要得到更加详细的资料,可以去看RFC 4648
          Base64是一种基于64个可打印字符来表示二进制数据的表示方法。由于2的6次方等于64,所以每6个位元为一个单元,对应某个可打印字符。三个字节有24个位元,对应于4个Base64单元,即3个字节需要用4个可打印字符来表示。它可用来作为电子邮件的传输编码。在Base64中的可打印字符包括字母A-Z、a-z、数字0-9 ,这样共有62个字符,此外两个可打印符号在不同的系统中而不同。
          Base64常用于在通常处理文本数据的场合,表示、传输、存储一些二进制数据。包括MIME的email,email via MIME, 在XML中存储复杂数据.

                                Base64索引表:
Value Char   Value Char   Value Char   Value Char
0 A 16 Q 32 g 48 w
1 B 17 R 33 h 49 x
2 C 18 S 34 i 50 y
3 D 19 T 35 j 51 z
4 E 20 U 36 k 52 0
5 F 21 V 37 l 53 1
6 G 22 W 38 m 54 2
7 H 23 X 39 n 55 3
8 I 24 Y 40 o 56 4
9 J 25 Z 41 p 57 5
10 K 26 a 42 q 58 6
11 L 27 b 43 r 59 7
12 M 28 c 44 s 60 8
13 N 29 d 45 t 61 9
14 O 30 e 46 u 62 +
15 P 31 f 47 v 63 /

    Base64是一种基于64个可打印字符来表示二进制数据的表示方法。由于2的6次方等于64,所以每6个位元为一个单元,对应某个可打印字符。三个字节有24个位元,对应于4个Base64单元,即3个字节需要用4个可打印字符来表示。它可用来作为电子邮件的传输编码。在Base64中的可打印字符包括字母A-Z、a-z、数字0-9 ,这样共有62个字符,此外两个可打印符号在不同的系统中而不同。

文本(1 Byte) A    
二进制位 0 1 0 0 0 0 0 1                                
二进制位(补0) 0 1 0 0 0 0 0 1 0 0 0 0                        
Base64编码 Q Q    
文本(2 Byte) B C  
二进制位 0 1 0 0 0 0 1 0 0 1 0 0 0 0 1 1     x x x x x x
二进制位(补0) 0 1 0 0 0 0 1 0 0 1 0 0 0 0 1 1 0 0 x x x x x x
Base64编码 Q k M  
          例子:
          举例来说,一段引用自托马斯·霍布斯的利维坦的文句:
Man is distinguished, not only by his reason, but by this singular passion from other animals, which is a lust of the mind, that by a perseverance of delight in the continued and indefatigable generation of knowledge, exceeds the short vehemence of any carnal pleasure.

          经过base64编码之后变成:
TWFuIGlzIGRpc3Rpbmd1aXNoZWQsIG5vdCBvbmx5IGJ5IGhpcyByZWFzb24sIGJ1dCBieSB0aGlz
IHNpbmd1bGFyIHBhc3Npb24gZnJvbSBvdGhlciBhbmltYWxzLCB3aGljaCBpcyBhIGx1c3Qgb2Yg
dGhlIG1pbmQsIHRoYXQgYnkgYSBwZXJzZXZlcmFuY2Ugb2YgZGVsaWdodCBpbiB0aGUgY29udGlu
dWVkIGFuZCBpbmRlZmF0aWdhYmxlIGdlbmVyYXRpb24gb2Yga25vd2xlZGdlLCBleGNlZWRzIHRo
ZSBzaG9ydCB2ZWhlbWVuY2Ugb2YgYW55IGNhcm5hbCBwbGVhc3VyZS4=

          编码“Man”
文本 M a n
ASCII编码 77 97 110
二进制位 0 1 0 0 1 1 0 1 0 1 1 0 0 0 0 1 0 1 1 0 1 1 1 0
索引 19 22 5 46
Base64编码 T W F u

          在此例中,Base64算法将三个字符编码为4个字符

          基本上说,如果那天你看到一段以“=”或者"=="结束的字符,你就可以判断这段字符用了Base64编码。就应用而言,我只在邮件内容发送和MD5 或者SHA密码保存时见过Base64编码。

         实现上Base64算法很简单,把流变成二进制,然后按6位转换为字符保存。

3.2 Poco::Base64Encode

          Poco::Base64Encode的类图如下:


3.3 Poco::Base64Decoder

          Poco::Base64Decoder的类图如下:


3.4 例子

          下面是一个例子
#include "Poco/Base64Encoder.h"
#include <iostream>
using Poco::Base64Encoder;
int main(int argc, char** argv)
{
       Base64Encoder encoder(std::cout);
       encoder << "Hello, world!";
       return 0;
}

4. Poco::HexBinaryEncoder和Poco::HexBinaryDecoder

4.1 HexBinary编码编码概念


          HexBinary编码又叫Base 16编码,Base16是一种基于16个可打印字符来表示二进制数据的表示方法。由于2的4次方等于16,所以每4个位元为一个单元,对应某个可打印字符。三个字节有24个位元,对应于6个Base64单元,即3个字节需要用6个可打印字符来表示。

          Base16索引表:
                               The Base 16 Alphabet

         Value Encoding  Value Encoding  Value Encoding  Value Encoding
             0 0             4 4             8 8            12 C
             1 1             5 5             9 9            13 D
             2 2             6 6            10 A            14 E
             3 3             7 7            11 B            15 F

           还是以编码“Man”举例:

文本                     M                          a                            n
ASCII编码                     77                         97                          110
二进制位 0 1 0 0 1 1 0 1 0 1 1 0 0 0 0 1 0 1 1 0 1 1 1 0
索引           4          13              6            1               6           14
base 16编码           4           D              6            1              6             E

         实现上Base16算法很简单,把流变成二进制,然后按4位转换为字符保存。

4.2 Poco::HexBinaryEncoder和Poco::HexBinaryDecoder

     Poco::HexBinaryEncoder 和Poco::HexBinaryDecoder的类图同Poco::Base64Encoder和Poco::Base64Decoder是类似。

       HexBinaryDecoderBuf定义如下:

class Foundation_API HexBinaryDecoderBuf: public UnbufferedStreamBuf
	/// This streambuf decodes all hexBinary-encoded data read
	/// from the istream connected to it.
	/// In hexBinary encoding, each binary octet is encoded as a character tuple,  
	/// consisting of two hexadecimal digits ([0-9a-fA-F]) representing the octet code.
	/// See also: XML Schema Part 2: Datatypes (http://www.w3.org/TR/xmlschema-2/),
	/// section 3.2.15.
	///
	/// Note: For performance reasons, the characters 
	/// are read directly from the given istream's 
	/// underlying streambuf, so the state
	/// of the istream will not reflect that of
	/// its streambuf.
{
public:
	HexBinaryDecoderBuf(std::istream& istr);
	~HexBinaryDecoderBuf();
	
private:
	int readFromDevice();
	int readOne();

	std::streambuf& _buf;
};

       构造文件为:

HexBinaryDecoderBuf::HexBinaryDecoderBuf(std::istream& istr): 
	_buf(*istr.rdbuf())
{
}

       我们看到 HexBinaryDecoderBuf直接使用了istream对象内部的缓冲作为实现。同2.1节中的例子相比,可以减少函数调用的流程.





(版权所有,转载时请注明作者和出处  http://blog.csdn.net/arau_sh/article/details/8851662













posted @ 2013-05-02 19:58  在天与地之间  阅读(1174)  评论(0编辑  收藏  举报