No2:luacom中文内容输出BUG及修正

quasiceo 2014-01-03

展开全文

No2:luacom中文内容输出BUG及修正

2012-03-05 18:33 764人阅读评论(9) 收藏举报

lua delete null transformation byte 测试

使用cell.Value2输出中文内容时总是乱码。怀疑是utf-8的原因，转换后结果仍然是乱码。自己再写个转换的再测试，依然是乱码，莫非有BUG！？

下个LUACOM的源码，查看函数tLuaCOMTypeHandler::com2lua和tStringBuffer tUtil::bstr2string。整个过程看起来都OK，但再测试发现，结果字符串少了一个bytes。

在tLuaCOMTypeHandler::com2lua VT_BSTR分支中返回结果恰好是减去1，将其修改

lua_pushlstring(L, str, str.getSize()-1);修改为lua_pushlstring(L, str, str.getSize());

重新编译LUACOM，再看cell.Value2输出结果，终于正确了。由于没进行全面测试，不知道其此修改会不会引入错误。资源里有个已经编译好的。

--lc是从网上抄来的unicode utf-8 ansi相互转换的函数

[cpp] view plain copy print ?

package.cpath=[[C:\Program Files\Lua\5.1\clibs\?.dll;d:\loonlib\sample\lc\?.dll]]
require "luacom"
require "lc"
function print_table(t) for k,v in pairs(t) do print(k,v) end end
excel = luacom.CreateObject("Excel.Application")
excel.Visible = true
excel.Workbooks:Add();
--luacom.ViewTypeLib(excel);
sheet=excel.Sheets(1);
local r=sheet:Range("E6");
local s = "严中";
ws, s2=lc.a2w(s); --0x25 0x4e 0x2d 0x4e 0x00 0x00 6
print("unicode : " .. lc.bstr(ws, s2));
us, s2=lc.w2u(ws, s2); --0xe4 0xb8 0xa5 0xe4 0xb8 0xad 0x00 0x00 8
print("utf8 : " .. lc.bstr(us, s2));
r.Value2=us;
ws, s2=lc.u2w(r.Value2, s2);
print("unicode : " .. lc.bstr(ws, s2));
as, s2=lc.w2a(ws, s2);
print("ansi : " .. lc.bstr(as, s2));
print(as);

lc.def

[ruby] view plain copy print ?

LIBRARY "lc"
EXPORTS
luaopen_lc

lc.h

[ruby] view plain copy print ?

extern "C" {
#include "lua.h"
#include "lualib.h"
#include "lauxlib.h"
int luaopen_local(lua_State* L);
}
#include <locale.h>
#include <cstring>
#ifdef WIN32
#include <windows.h>
#include <winnls.h>
#else
#include <cstdlib>
#endif
#define LN_lc "lc"
int lua_a2w(lua_State* L);
int lua_u2w(lua_State* L);
int lua_w2a(lua_State* L);
int lua_w2u(lua_State* L);
int lua_u2a(lua_State* L);
int lua_a2u(lua_State* L);
int lua_bstr(lua_State* L);
int lua_help(lua_State* L);
wchar_t* mb2wc(const char* mbstr, int& s2, int cp);
char* wc2mb(const wchar_t* wcstr, int& s2, int cp);

lc.cpp

[ruby] view plain copy print ?

#include "lc.h"
//g++ -shared -s -o lc.dll -O3 lc.cpp lc.def -llua5.1 -DWIN32 -I%loon%/lua/src -L%loon%/lib/gcc_dll/debug -Wl,--out-implib,liblc.a
int lua_bstr(lua_State* L) {
const char* s = luaL_optstring(L, 1, "");
int len = luaL_optnumber(L, 2, 0);
if (strcmp(s, "")==0 || 0==len) {
lua_pushstring(L, s);
} else {
luaL_Buffer b;
luaL_buffinit(L, &b);
char* byte = (char*)malloc(64);
for (int i=0; i<len; ++i) {
sprintf(byte, "0x%02x ", (unsigned char)*s++);
luaL_addstring(&b, byte);
}
free(byte);
luaL_pushresult(&b);
}
return 1;
}
int lua_u2w(lua_State* L) {
int result = 0;
size_t len = 0;
const char* mbstr = lua_tolstring(L, 1, &len);
if (mbstr && len>0) {
int s2 = 0;
wchar_t* wcstr = mb2wc(mbstr, s2, CP_UTF8);
if (wcstr) {
lua_pushlstring(L, (const char*)wcstr, s2);
lua_pushnumber(L, s2);
delete[] wcstr;
result = 2;
}
}
return result;
}
int lua_a2w(lua_State* L) {
int result = 0;
size_t len = 0;
const char* mbstr = lua_tolstring(L, 1, &len);
if (mbstr && len>0) {
int s2 = 0;
wchar_t* wcstr = mb2wc(mbstr, s2, CP_ACP);
if (wcstr) {
lua_pushlstring(L, (const char*)wcstr, s2);
lua_pushnumber(L, s2);
delete[] wcstr;
result = 2;
}
}
return result;
}
int lua_w2a(lua_State* L) {
int result = 0;
size_t len = 0;
const char* wcstr = lua_tolstring(L, 1, &len);
if (wcstr && len>0) {
int s2 = 0;
char* mbstr = wc2mb((wchar_t*)wcstr, s2, CP_ACP);
if (mbstr) {
lua_pushlstring(L, mbstr, s2);
lua_pushnumber(L, s2);
delete[] mbstr;
result = 2;
}
}
return result;
}
int lua_w2u(lua_State* L) {
int result = 0;
size_t len = 0;
const char* wcstr = lua_tolstring(L, 1, &len);
if (wcstr && len>0) {
int s2 = 0;
char* mbstr = wc2mb((wchar_t*)wcstr, s2, CP_UTF8);
if (mbstr) {
lua_pushlstring(L, mbstr, s2);
lua_pushnumber(L, s2);
delete[] mbstr;
result = 2;
}
}
return result;
}
int lua_u2a(lua_State* L) {
int result = 0;
size_t len = 0;
const char* mbstr = lua_tolstring(L, 1, &len);
if (mbstr && len>0) {
int s2 = 0;
wchar_t* wcstr = mb2wc(mbstr, s2, CP_UTF8);
if (wcstr) {
char* nmbstr = wc2mb(wcstr, s2, CP_ACP);
if (nmbstr) {
lua_pushlstring(L, nmbstr, s2);
lua_pushnumber(L, s2);
result = 2;
delete[] nmbstr;
}
delete[] wcstr;
}
}
return result;
}
int lua_a2u(lua_State* L) {
int result = 0;
size_t len = 0;
const char* mbstr = lua_tolstring(L, 1, &len);
if (mbstr && len>0) {
int s2 = 0;
wchar_t* wcstr = mb2wc(mbstr, s2, CP_ACP);
if (wcstr) {
char* nmbstr = wc2mb(wcstr, s2, CP_UTF8);
if (nmbstr) {
lua_pushlstring(L, nmbstr, s2);
lua_pushnumber(L, s2);
result = 2;
delete[] nmbstr;
}
delete[] wcstr;
}
}
return result;
}
wchar_t* mb2wc(const char* mbstr, int& s2, int cp) {
wchar_t* wcstr = NULL;
#ifdef WIN32
int size = MultiByteToWideChar(cp, 0, mbstr, -1, NULL, 0);
#else
size_t size = mbstowcs(NULL, mbstr, 0);
#endif
wcstr = new wchar_t[size];
if (wcstr) {
memset(wcstr, 0, size * sizeof(wchar_t));
#ifdef WIN32
int ret = MultiByteToWideChar(cp, 0, mbstr, -1, wcstr, size);
if (ret == 0) { // MultiByteToWideChar returns 0 if it does not succeed.
#else
size_t ret = mbstowcs(wcstr, mbstr, size+1);
if (ret == -1) {
#endif
delete[] wcstr;
wcstr = NULL;
}
s2 = 2*size;
}
return wcstr;
}
char* wc2mb(const wchar_t* wcstr, int& s2, int cp) {
char* mbstr = NULL;
#ifdef WIN32
int size = WideCharToMultiByte(cp, 0, wcstr, -1, NULL, 0, NULL, NULL);
#else
size_t size = wcstombs(NULL, wcstr, 0);
#endif
mbstr = new char[size];
if (mbstr) {
memset(mbstr, 0, size * sizeof(char));
#ifdef WIN32
int ret = WideCharToMultiByte(cp, 0, wcstr, -1, mbstr, size, NULL, NULL);
if (ret == 0) { // MultiByteToWideChar returns 0 if it does not succeed.
#else
size_t ret = wcstombs(mbstr, wcstr, size+1);
if (ret == -1) {
#endif
delete[] mbstr;
mbstr = NULL;
}
s2 = size;
}
return mbstr;
}
int lua_help(lua_State* L) {
const char* s=
"Simple Characters Transformation\n"
" a2w(ansi to unicode)\n"
" u2w(utf8 to unicode)\n"
" w2a(unicode to ansi)\n"
" w2u(unicode to utf8)\n"
" u2a(utf8 to ansi)\n"
" a2u(ansi to utf8)\n"
" bstr(bytes of str)\n"
" help(show this)\n\n"
" example :\n"
" local s = \"I like lua\"\n"
" print(lc.bstr(s, string.len(s)+1))\n"
" local ws, s2 = lc.a2w(s)\n"
"wunoman@qq.com 2012/03/06\n"
;
lua_pushstring(L, s);
return 1;
}
luaL_reg lrg_lc[] = {
{"a2w", lua_a2w},
{"u2w", lua_u2w},
{"w2a", lua_w2a},
{"w2u", lua_w2u},
{"u2a", lua_u2a},
{"a2u", lua_a2u},
{"bstr", lua_bstr},
{"help", lua_help},
{NULL, NULL}
};
extern "C" int luaopen_lc(lua_State* L) {
luaL_register(L, LN_lc, lrg_lc);
return 1;
}

更多 0

上一篇：No1:luacom excel 使用分列设置文本格式

下一篇：No3:luacom WMI monitor process create and close

查看评论

4楼 windtailljj 2012-05-12 22:19发表 [回复]: 我看了下你说的那个函数，貌似没有问题，因为上面的bstr2string()函数有一个默认参数是nullterminated=true，返回的字符串中带有一个'\0'，而他使用的pushlstring（而不是pushstring），确实应该 -1 的啊，你这个修改应该是碰巧解决了问题吧。

3楼 danninggao 2012-04-24 14:44发表 [回复]: 虽然不知道是什么原因但是胡乱尝试下好像解决问题了，再获得EXCEL表中值时 local sRet = oExcel.Activesheet.Cells(Row, Column).Value2 ，些时为UTF-8，然后将字符串长度减2，local nTemp = string.sub(nRet,1,string.len(nRet)-2),再将utf-8转为gbk,这时后就正常了