Re: [Texmacs-dev] macro OS_WIN32 is not defined

[Gubinelli Massimiliano]

On 1 juil. 10, at 10:38, zohooo wrote:

> Thanks very much for your kindly feedback.
>
> I like QT, and agree with you that QT port is better than native  
> Windows port.
>
> Also I think using __MINGW32__ macro is not a good choice, since  
> maybe someone (in the future?) want to compile TeXmacs with  
> Microsoft Visual C++ which there is not

> __MINGW32__ macro there.


I agree and I wanted to change that in the future. However for the  
moment we are not allowed to reuse the OS_WIN32 checks and predicates  
since have still a meaning and purpose in the legacy code (which for  
the moment will not be removed). So for the moment I suggest to stick  
to this less than optimal labeling, or  maybe add some scheme  
predicate to check for Qt/Win and reassign keys accordingly.

best
massimiliano




> 2010-07-01 16:22:40，"Gubinelli Massimiliano" wrote：
> >Hi,
> >
> >There are essentially two flavors of TeXmacs which relates to  
> Windows.
> >The first is an old native port which is not more mantained (it was
> >developed by Dan Mertens if I recall correctly). The other is the
> >recent Qt port which tries not to assume much on the underlying  
> system
> >and do as much as possible via the Qt compatibility layer. The
> >sensible sectors where the underlying system matters are mainly
> >process management, I/O (sockets and other connections), file path
> >handling, locale handling, spell checking, etc.... The OS_WIN32 macro
> >conditionally enable code for the old native windows port. It is not
> >meant to be used in general for Windows related support. When
> >compiling TeXmacs/Qt the underlying system is detected via either the
> >__MINGW32__ macro or via the relevant Qt macro (QT_OSWIN or something
> >like this) which is set up in the Qt library.
> >
> >The old native Windows port cannot be maintained since the known
> >active developers (Joris, David and me) are not able/willing to write
> >Windows code. Support for Windows is via the Qt layer (at least for
> >the moment).
> >
> >I agree that a more integrated Windows interface should be available.
> >
> >If you check the sources at src/Guile/Glue you can find the all the
> >functions exported in Scheme code. In glue_basic.cpp:52 there is the
> >definition of the predicate os_mingw? which returns true if we are
> >compiling under mingw, so a possible way to proceed is to add to tm-
> >server.scm the following lines
> >
> >(define (get-default-look-and-feel)
> >   (if (or (os-win32?) (os-mingw?)) "windows" "emacs"))
> >
> >
> >(I've not check if it works....)
> >
> >Best
> >Massimiliano
> >
> >ps: Currently I'm a bit busy, but I follow all remarks in the list.
> >Please continue the very useful feedback!!! Thanks!
> >_______________________________________________
> >Texmacs-dev mailing list
> >address@hidden
> >http://lists.gnu.org/mailman/listinfo/texmacs-dev
> _______________________________________________
> Texmacs-dev mailing list
> address@hidden
> http://lists.gnu.org/mailman/listinfo/texmacs-dev