尽管项目为多字节，但在VS2010中定义了UNICODE的问题 [英] Issue with UNICODE being defined in VS2010 despite a project being multibyte

问题描述

我最近在编译以32位模式工作的项目时，VS2010遇到了一个奇怪的问题.我收到许多相同形式的错误:

I've recently run into a strange issue with VS2010 while compiling the project I work on in 32-bit mode.  I get many errors of the same general form:

6>c:\opticks\future\opticks-1288\future\code\application\pluginutilities\systemservicesimp.cpp(101): error C2440: 'initializing' : cannot convert from 'const char *' to 'PCTSTR'
6>          Types pointed to are unrelated; conversion requires reinterpret_cast, C-style cast or function-style cast

此特定错误对应于以下代码块:

This particular error corresponds to this block of code:

      PCTSTR aInsertions[] = {message.c_str()};

其中消息只是一个std :: string

Where message is just a std::string

调查进一步指出，这与项目中的Unicode/多字节设置有关.从 这个主题，我发现在项目中使用Unicode时，PCTSTR必须是wchar_t *而不是char *.但是，我正在构建的项目在其项目设置中设置为使用多字节字符.但同时，当我翻阅 似乎仍在定义UNICODE的源，导致WinNT.h将def PCTSTR键入wchar_t.我还没有找到一种明确的方法来告诉项目未定义UNICODE的方法.我只能通过在WinNT.h中将鼠标悬停在上面来定义它 并看到出现#define UNICODE 1，但在任何实际代码，项目设置等中都找不到该符号.

Investigating further pointed to this being some problem with the Unicode/multi-byte settings in the project.  From this topic, I found that, when using Unicode in a project, a PCTSTR has to be a wchar_t* and not a char*.  However, the project I am building is set to use multibyte characters in its project setting.  At the same time, though, when I looked through the source it appears that UNICODE is still being defined, causing WinNT.h to typedef PCTSTR to wchar_t.  I haven't seen a way to explicitly tell the project to not define UNICODE.  I can only tell it's being defined by hovering over it in WinNT.h and seeing #define UNICODE 1 appear, but I can't find the symbol in any actual code, project settings, etc.

只有在以32位模式进行编译时才会发生这种情况，而我们团队中的其他人都没有遇到过此问题.我不太确定关于VS2010可能安装或更改的内容.或可能导致此问题出现的Windows，但是我 完全重新安装了VS2010，但我仍然遇到此问题.谁能指出我正确的方向?

This only occurs when compiling in 32-bit mode, and no one else on my team has encountered this problem.  I'm not really sure what I may have installed or changed with regard to VS2010  or Windows that may have caused this issue to arise, but I had VS2010 completely reinstalled and I am still running into this problem.  Can anyone point me in the right direction?

推荐答案

     PCTSTR aInsertions [] = {message.c_str()};
其中消息只是一个std :: string

     PCTSTR aInsertions[] = {message.c_str()};
Where message is just a std::string

在那种情况下，您确实应该写出以下任何一个文件:

In that case you really ought to have written either:

    PCSTR而不是PCTSTR

    PCSTR rather than PCTSTR

或使用类似的类型定义的basic_string< TCHAR>而不是字符串.

or used a similar typedefed basic_string<TCHAR> rather than string.

但是，...

但是，我正在构建的项目在其项目设置中设置为使用多字节字符.

However, the project I am building is set to use multibyte characters in its project setting.

也可以按源文件进行设置.你检查过了吗
仅选择一个源文件时的预处理器符号
在解决方案资源管理器中?

Settings can also be made per-source file. Have you checked the
preprocessor symbols when you have just that one source file selected
in the Solution Explorer?

有2个符号UNICODE和_UNICODE起作用.是
最好只通过项目的字符集"进行设置在
上设置 常规页面-未在预处理器符号中明确显示.

There's are 2 symbols UNICODE and _UNICODE that come into play. It's
best to just set it via the project's "Character Set" setting on the
General page - and not explicitly in the preprocessor symbols.

戴夫

这篇关于尽管项目为多字节，但在VS2010中定义了UNICODE的问题的文章就介绍到这了，希望我们推荐的答案对大家有所帮助，也希望大家多多支持IT屋！