[Development] Why we *have to* remove codecFor... ?

[Loaden]

To clear for my previous post.
Under Linux + GCC4.6.x, We can use UTF-8 BOM in *source* files, not in
*header* files.
So if use UTF-8 encoding and with BOM, the source file should can
cross-platform.
In Creator, see https://codereview.qt-project.org/11851
It's easy to add/delete UTF-8 BOM now.

2012/4/23 Loaden <loaden at gmail.com>

>  Seems you still use two different sets of source code.
>>  * Under MSVC 2010, source files saved as UTF8 with BOM.
>>  * Under Linux GCC, source files saved as UTF8 without BOM.
>
> No, Under Linux GCC, the source files saved as UTF-8 *with* BOM too. It's
> works fine when it's Windows either Linux.
> So, That make sure my application is cross-platform.
>
> For now, the code will works fine with Qt4 either Qt5, under Windows or
> Linux.
>
>> #include <QApplication>
>> #include <QTextStream>
>> #include <QLabel>
>> int main(int argc, char *argv[])
>> {
>> #ifdef Q_CC_MSVC
>>     #pragma execution_character_set("UTF-8")
>> #endif
>>     QApplication app(argc, argv);
>>     QString str = QString::fromUtf8("Hello世界World!你好！");
>>     QTextStream out(stdout);
>>     out << str << ", strlen=" << strlen("中文") << ", sizeof=" <<
>> sizeof("中文") << endl;
>>     QLabel *label = new QLabel(str + QObject::trUtf8("Hello世界World!你好！"));
>>     label->show();
>>     return app.exec();
>> }
>>
>
> The current question is: Shall we need or no-need to use UTF-8 encoding as
> default?
> I thought we should use UTF-8 as default, Because it's the better way for
> cross-platform.
> What do you think?
> Has anyone working for this?
>
>
> 2012/4/23 1+1=2 <dbzhang800 at gmail.com>
>
>> Hi Loaden,
>>
>> Seems you still use two different sets of source code.
>>
>>  * Under MSVC 2010, source files saved as UTF8 with BOM.
>>  * Under Linux GCC, source files saved as UTF8 without BOM.
>>
>> which means your application is not corss-platform.
>>
>> And of course, this is a defect of C++ instead of Qt. You should be
>> aware that there are two character set related to the issue.
>>
>>  * the set in which source files are written (the source character set),
>>  * and the set interpreted in the execution environment (the execution
>> character set).
>>
>> But unfortunately, even in the era of the C++11, by introducing u8" "
>> / u" "/ U" " , only execution character set was solved.
>>
>> However, if you still use different source codes for different
>> platforms, like you have done with Qt4, this is still not a problem
>> for you.
>>
>> Debao
>>
>> 2012/4/22 Loaden <loaden at gmail.com>:
>> > Thanks for help!!
>> > I just test on Windows 7 (MSVC2010) and Linux (GCC 4.6.2), both (Win
>> > / Linux, Qt4 / Qt5)  works well use below code.
>> >
>> >>
>> >> #include <QApplication>
>> >> #include <QTextStream>
>> >> #include <QLabel>
>> >> int main(int argc, char *argv[])
>> >> {
>> >> #ifdef Q_CC_MSVC
>> >>     #pragma execution_character_set("UTF-8")
>> >> #endif
>> >>     QApplication app(argc, argv);
>> >>     QString str = QString::fromUtf8("Hello世界World!你好！");
>> >>     QTextStream out(stdout);
>> >>     out << str << ", strlen=" << strlen("中文") << ", sizeof=" <<
>> >> sizeof("中文") << endl;
>> >>     QLabel *label = new QLabel(str +
>> QObject::trUtf8("Hello世界World!你好！"));
>> >>     label->show();
>> >>     return app.exec();
>> >> }
>> >
>>
>
>
>
> --
> Best Regards
> Yuchen
>
>


-- 
Best Regards
Yuchen
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.qt-project.org/pipermail/development/attachments/20120423/b4c5805d/attachment.html>