使用iconv进行内码转换（Big5->GB2312）

阅读更多

i conv 是一个通过unicode 作为中间码实现各种内码间相互转换的库，它基本上囊括了世界上所有编码方式，例如，ASCII 、 GB2312 、 GBK 、 GB18030 、BIG5 、UTF-8 、UCS-2 、UCS-2BE 、UCS-2LE 、UCS-4 、UCS-4BE 、UCS- 4LE 、UTF-16 、 UTF-16BE 、UTF-16LE 、UTF-32 、UTF-32BE 、UTF-32LE 、UTF-7 等等等，除此之外，还包括 泰语、日语、韩语、西欧等国家语言的编码。下面我们演示如何使用iconv 实现Big5 到GB2312 的转换，当然只要简单修改一下便可实现iconv 支 持任何编码间的转换。 

下载 
libiconv 是linux 版本的iconv ，可在 http://www.gnu.org/software/libiconv/  下载 
iconv 的win32 版本可以在 http://gnuwin32.sourceforge.net/packages/libiconv.htm  下载 

SVN 源码 
另外，还有一些演示代码，需要的可以到我的SVN 下载 
http://xcyber.googlecode.com/svn/trunk/Convert/

标签： libiconv , iconv , Big5 , GB2312 ,  大五码 ,  内码

代码片段(1)

[ 代码] [C/C++/Objective-C] 代码

001	/****************************************************************************

002	*  Big5ToGB2312 - Convert Big5 encoding file to GB2312 encoding file

003

*  File:

004	*    Big5ToGb2312.c

005	*  Description:

006	*    Convert Big5 encoding file to GB2312 encoding file using iconv library

007	*  Author:

008	*    XCyber   email:[email protected]

009

*  Date:

010	*    August 7, 2008

011

*  Other:

012	*    visit http://www.gnu.org/software/libiconv/ for more help of iconv

013	***************************************************************************/

014

015

016

#include

017

#include

018

#include

019

#include

020	#include "../iconv-1.9.2.win32/include/iconv.h"

021

022	//#pragma comment(lib, "../iconv-1.9.2.win32/lib/iconv.lib")  // using iconv dynamic-link lib, iconv.dll

023	#pragma comment(lib, "../iconv-1.9.2.win32/lib/iconv_a.lib")  // using iconv static lib

024

025	#define BUFFER_SIZE 1024   //BUFFER_SIZE must >= 2

026

027

028	void usage()

029

{

030	printf("\nBig5ToGB2312 - Convert Big5 encoding file to GB2312 encoding file\n");

031	printf("[email protected] on August 7, 2008\n");

032	printf("  Usage:\n");

033	printf("      Big5ToGB2312 [Big5 file(in)]  [GB2312 file(out)]\n\n");

034

}

035

036

037	int main(int argc, char* argv[])

038

{

039	FILE * pSrcFile = NULL;

040	FILE * pDstFile = NULL;

041

042	char szSrcBuf[BUFFER_SIZE];

043	char szDstBuf[BUFFER_SIZE];

044

045	size_t nSrc  = 0;

046	size_t nDst  = 0;

047	size_t nRead = 0;

048	size_t nRet  = 0;

049

050	char *pSrcBuf = szSrcBuf;

051	char *pDstBuf = szDstBuf;

052

053	iconv_t icv;

054	int argument = 1;

055

056	//check input arguments

057	if(argc != 3)

058

{

059

usage();

060	return -1;

061

}

062

063

064	pSrcFile = fopen(argv[1],"r");

065	if(pSrcFile == NULL)

066

{

067	printf("can't open source file!\n");

068	return -1;

069

}

070

071	pDstFile = fopen(argv[2],"w");

072

&nbs

073	p;if(pSrcFile == NULL)

074

{

075	printf("can't open destination file!\n");

076	return -1;

077

}

078

079	//initialize iconv routine, perform conversion from BIG5 to GB2312

080	//TODO: if you want to perfom other type of coversion, e.g. GB2312->BIG5, GB2312->UTF-8 ...

081	//just change following two paremeters of iconv_open()

082	icv = iconv_open("GB2312","BIG5");

083	if(icv == 0)

084

{

085	printf("can't initalize iconv routine!\n");

086	return -1;

087

}

088

089	//enable "illegal sequence discard and continue" feature, so that if met illeagal sequence,

090	//conversion will continue instead of being terminated

091	if(iconvctl (icv ,ICONV_SET_DISCARD_ILSEQ,&argument) != 0)

092

{

093	printf("can't enable \"illegal sequence discard and continue\" feature!\n");

094	return -1;

095

}

096

097	while(!feof(pSrcFile))

098

{

099	pSrcBuf = szSrcBuf;

100	pDstBuf = szDstBuf;

101	nDst = BUFFER_SIZE;

102

103	// read data from source file

104	nRead = fread(szSrcBuf + nSrc,sizeof(char),BUFFER_SIZE - nSrc,pSrcFile);

105	if(nRead == 0)

106

break;

107

108	// the amount of data to be converted should include previous left data and current read data

109	nSrc = nSrc + nRead;

110

111	//perform conversion

112	nRet = iconv(icv,(const char**)&pSrcBuf,&nSrc,&pDstBuf,&nDst);

113

114	if(nRet == -1)

115

{

116	// include all case of errno: E2BIG, EILSEQ, EINVAL

117	//     E2BIG: There is not sufficient room at *outbuf.

118	//     EILSEQ: An invalid multibyte sequence has been encountered in the input.

119	//     EINVAL: An incomplete multibyte sequence has been encountered in the input

120	// move the left data to the head of szSrcBuf in other to link it with the next data block

121	memmove(szSrcBuf,pSrcBuf,nSrc);

122

}

123

124	//wirte data to destination file

125	fwrite(szDstBuf,sizeof(char),BUFFER_SIZE - nDst,pDstFile);

126

127

}

128	iconv_close(icv);

129	fclose(pSrcFile);

130	fclose(pDstFile);

131

132	printf("conversion complete.\n");

133

134

return;

135

}