libiconv库是一个基于GNU协议的开源库,主要是解决多语言编码处理转换等应用问题。 怎样学习使用libiconv库?对于刚接触到人来说,这篇文章不妨去看一看,若已经用到过该库的人,在应用的过程中可能遇到一些问题,我们可以一起来探讨,我的联系方式是 。
几个函数原型: iconv_t iconv_open(const char *tocode, const char *fromcode); size_t iconv(iconv_t cd, char **inbuf, size_t *inbytesleft, char **outbuf, size_t *outbytesleft); int iconv_close(iconv_t cd);
其中: iconv_open是打开一个编码流,类似于打开一个编码管道(通道),出错则返回 -1; iconv用于具体输入的转换,如果出错,则返回 -1,否则返回 0; iconv_close是关闭该管道(通道)。 举个例子: #include #include #include #include
#define OUTLEN 255 int covert(char *, char *, char *, size_t , char *, size_t );
int main(int argc, char *argv[]) { char *input = "中国"; size_t len = strlen(input); char *output = (char *)malloc(OUTLEN); covert("UTF-8", "GBK", input, len, output, OUTLEN); printf("%s\n", output); return 0; }
int covert(char *desc, char *src, char *input, size_t ilen, char *output, size_t olen) { char **pin = &input; char **pout = &output; iconv_t cd = iconv_open(desc, src); if (cd == (iconv_t)-1) { return -1; } memset(output, 0, olen); if (iconv(cd, pin, &ilen, pout, &olen)) return -1; iconv_close(cd); return 0; }
这里面covert函数就是用于将编码进行转换,其中要注意的地方是iconv函数的传递参数: 1,iconv传递有5个参数; 2,第3个参数和第5个参数一般是input和output实际分配的大小,一般是 sizeof(type)*strlen(string); 3,第4个参数是不能直接传递指针的地址,因为iconv函数能够改变指针的值,所以需要复制一份指针变量; 如果对于大量需要转换的编码,上述函数covert不适合该方式,一是内存的限制不能一次调用,二是若分多次调用会频繁打开一个编码管道(通道),导致资源浪费,最好的办法还是拆开该函数根据情况使用。 这里补充一下代码: translateSP.h: #ifndef __TRANSLATESP_H_ #define __TRANSLATESP_H_ #include class TranslateSP { public: TranslateSP():i_cd(0){} TranslateSP(const char *from_charset,const char *to_charset) { i_cd = iconv_open(to_charset, from_charset); if ((iconv_t)-1 == i_cd) printf("iconv open error!\n"); } ~TranslateSP() { if (i_cd) iconv_close(i_cd); } public: size_t translate(char *src, size_t srcLen, char *desc, size_t descLen); size_t convert(const char *from_charset, const char *to_charset, char *src, size_t srcLen, char *desc, size_t descLen); private: iconv_t i_cd; }; #endif
translateSP.cpp: #include "translateSP.h" #define MAX_LEN 200 size_t TranslateSP::translate(char *src, size_t srcLen, char *desc, size_t descLen) { char **inbuf = &src; char **outbuf = &desc; memset(desc, 0, descLen); return iconv(i_cd, inbuf, &srcLen, outbuf, &descLen); } size_t TranslateSP::convert(const char *from_charset, const char *to_charset, char *src, size_t srcLen, char *desc, size_t descLen) { char **inbuf = &src; char **outbuf = &desc; iconv_t cd = iconv_open(to_charset, from_charset); if ((iconv_t)-1 == cd) return (size_t)-1; memset(desc, 0, descLen); size_t n = iconv(cd, inbuf, &srcLen, outbuf, &descLen); iconv_close(cd); return n; } int main(int argc, char *argv[]) { char *str = "我爱zhong国! %#@#"; char *str1 = "i大量需要转换的编码"; char *str2 = "函数就是用于将hello进行转换"; char newstr[MAX_LEN]; TranslateSP tsp; tsp.convert("utf-8", "gbk", str, strlen(str), newstr, MAX_LEN); printf("%s\n", newstr); TranslateSP newtsp("UTF-8", "GBK"); newtsp.translate(str1, strlen(str1), newstr, MAX_LEN); printf("%s\n", newstr); newtsp.translate(str2, strlen(str2), newstr, MAX_LEN); printf("%s\n", newstr); return 0; }
编译: g++ translateSP.cpp -o test ./test 我爱zhong国! %#@# i大量需要转换的编码 函数就是用于将hello进行转换
(以上输出是GBK编码) 在Linux上进行编码转换时,既可以利用iconv函数族编程实现,在虚拟机我们只需直接使用就可以了,但是要在开发板上运行就要自己交叉编译一个iconv的库,下面就大概说一下我的编译步骤和使用的例程。 1.首先下载一个libiconv-1.11.tar.gz压缩包,就放到/root/cross下,然后右击解压到此目录。 下载地址为:[url]ftp://ftp.eenet.ee/pub/gentoo/distfiles/libiconv-1.11.tar.gz[/url] ,可以直接用迅雷下载。 2.在终端中进入此目录: [root@localhost ~]# [color=red]cd /root/cross/libiconv-1.11[/color] 3.配置并生成Makefile: [root@localhost libiconv-1.11]# [color=red]./configure CC=arm-linux-gcc --build=i686-pc-linux --target=arm-linux --host=arm-linux prefix=/usr/local/arm/3.4.1/arm-linux --enable-shared --enable-shared --enable-static[/color] 4.编译并安装: [root@localhost libiconv-1.11]# [color=red]make && make install[/color] 5.将生成好的iconv库拷贝到nfs下: [root@localhost libiconv-1.11]# [color=red]cp /usr/local/arm/3.4.1/arm-linux/lib/libiconv.* /nfs/lib/[/color] 6.编写应用程序: [root@localhost libiconv-1.11]# [color=red]cd /nfs/ [/color][root@localhost nfs]# [color=red]gedit conv.c[/color] [color=darkred]#include "/usr/local/arm/3.4.1/arm-linux/include/iconv.h" #include #include int code_convert(char *from_charset,char *to_charset,char *inbuf,int inlen,char *outbuf,int outlen) { iconv_t cd; char **pin = &inbuf; char **pout = &outbuf; cd = iconv_open(to_charset,from_charset); if (cd==0) return -1; memset(outbuf,0,outlen); if (iconv(cd,pin,&inlen,pout,&outlen)==-1) return -1; iconv_close(cd); return 0; }[/color] [color=darkred]int u2g(char *inbuf,int inlen,char *outbuf,int outlen) //UNICODE码转为GB2312码 { return code_convert("utf-8","gb2312",inbuf,inlen,outbuf,outlen); }[/color] [color=darkred]int g2u(char *inbuf,size_t inlen,char *outbuf,size_t outlen) //GB2312码转为UNICODE码 { return code_convert("gb2312","utf-8",inbuf,inlen,outbuf,outlen); }
int main(void) { unsigned char in_utf8[] = {0x20,0xe8,0xbf,0x99,0xe6,0x98,0xaf,0xe4,0xb8,0x80,0xe4,0xb8,0xaa,0x69,0x63,0x6f,0x6e,0x76,0xe7,0x9a,0x84,0xe6,0xb5,0x8b,0xe8,0xaf,0x95,0xe4,0xbe,0x8b,0xe7,0xa8,0x8b,0xef,0xbc,0x9a,0xa,0xe2,0x80,0x9c,0xe7,0xba,0xb5,0xe6,0xb5,0xb7,0xe6,0x9d,0xaf,0xe2,0x80,0x9d,0xe4,0xb8,0x9c,0xe5,0x8d,0x97,0xe5,0xa4,0xa7,0xe5,0xad,0xa6,0xe7,0xac,0xac,0xe4,0xb8,0x89,0xe5,0xb1,0x8a,0xe5,0xb5,0x8c,0xe5,0x85,0xa5,0xe5,0xbc,0x8f,0xe7,0xb3,0xbb,0xe7,0xbb,0x9f,0xe8,0xae,0xbe,0xe8,0xae,0xa1,0xe9,0x82,0x80,0xe8,0xaf,0xb7,0xe8,0xb5,0x9b}; //UTF-8编码 char out[100]; int rc;[/color] [color=darkred] rc = u2g(in_utf8,strlen(in_utf8),out,100); printf("%s\n",out); } [/color]保存并退出。 7.交叉编译此程序: [root@localhost nfs]# [color=red]arm-linux-gcc -o iconv conv.c -liconv[/color] 8.在开发板上运行此程序: / # [b]./iconv[/b] [color=magenta] 这是一个iconv的测试例程: “纵海杯”东南大学第三届嵌入式系统设计邀请赛[/color]
[[i] 本帖最后由 youren4548 于 2010-5-10 22:40 编辑 [/i]]zyfo2 发表于 2010-5-10 23:53 回复 1# youren4548 的帖子学长,我试了下,utf8和gb2312的确可以转换,但unicode 却不识别,是少什么库文件吧?youren4548 发表于 2010-5-11 00:00 iconv也可以实现Unicode至GB2312的转换,下面是一个测试代码: #include "/usr/local/arm/3.4.1/arm-linux/include/iconv.h" #include #include int code_convert(char *from_charset,char *to_charset,char *inbuf,int inlen,char *outbuf,int outlen) { iconv_t cd; char **pin = &inbuf; char **pout = &outbuf; cd = iconv_open(to_charset,from_charset); if (cd==0) return -1; memset(outbuf,0,outlen); if (iconv(cd,pin,&inlen,pout,&outlen)==-1) return -1; iconv_close(cd); return 0; }
int u2g(char *inbuf,int inlen,char *outbuf,int outlen) //UNICODE码转为GB2312码 { return code_convert("UTF-16LE","gb2312",inbuf,inlen,outbuf,outlen); }
int main(void) { unsigned char utf_16le[] = {0xd9,0x8f,0x2f,0x66,0x00,0x4e,0x2a,0x4e,0x69,0x0,0x63,0x00,0x6f,0x0,0x6e,0x0,0x76,0x0,0x84,0x76,0x4b,0x6d,0xd5,0x8b,0x8b,0x4f,0xb,0x7a,0x1a,0xff,0xa,0x0,0x1c,0x20,0xb5,0x7e,0x77,0x6d,0x6f,0x67,0x1d,0x20,0x1c,0x4e,0x57,0x53,0x27,0x59,0x66,0x5b,0x2c,0x7b,0x9,0x4e,0x4a,0x5c,0x4c,0x5d,0x65,0x51,0xf,0x5f,0xfb,0x7c,0xdf,0x7e,0xbe,0x8b,0xa1,0x8b}; //Unicode编码 char out[100]; int rc; rc = u2g(utf_16le,sizeof(utf_16le),out,100); printf("%s\n",out); } |
|