This is the mail archive of the
insight@sources.redhat.com
mailing list for the Insight project.
Re: Why is PASCAL undefined in win/tclWinPort.h?
- From: Keith Seitz <keiths at redhat dot com>
- To: Mo DeJong <supermo at bayarea dot net>
- Cc: Insight <insight at sources dot redhat dot com>
- Date: Thu, 5 Sep 2002 06:20:06 -0700 (PDT)
- Subject: Re: Why is PASCAL undefined in win/tclWinPort.h?
On Wed, 4 Sep 2002, Mo DeJong wrote:
> Why would the sources verson differ? I have no idea. This only matters for
> a VC++ compiled program, so I can't imagine it could effect Insight in any
> way. The only ChangeLog entry I could find on the subject seems to suggest
> that the code is not correct.
According to the CVS log, the #define PASCAL line was touched last by irox
in version 1.5 (2001-09-13). These lines do not appear in version 1.4 of
the file, so irox must have added them... The ChangeLog for it:
2001-09-12 Ian Roxborough <irox@redhat.com>
* Makefile.in: Add 'cygwin' subdirectory to 'make all'
target.
* configure.in: Configure 'cygwin' and 'win' directory
for a cygwin host.
* configure: Regenerated.
* cygwin/configure.in: Change version number to 8.3.
* cygwin/configure: Regenerated.
* cygwin/Makefile.am: Added/removed files to be build
for Tcl8.3.
* cygwin/Makefile.in: Regenerated.
* generic/tcl.h: Don't define __WIN32__ for cygwin or
mwing32 builds. Don't define USE_TCLALLOC when building
for cygwin. Don't use __declspec unless building Tcl or
tk or build with USE_TCL_STUBS.
* generic/tclAlloc.c: For cygwin hosts, don't using
anything in this file unless build with __TCL_UNIX_VARIANT.
* generic/tclClock.c: Declare 'timezone' as an int, if it
hasn't been #defined.
* generic/tclStubInit.c: Don't export any tclAlloc function
when build for a cygwin host.
* win/configure.in: Set DL_LIBS and MATH_LIBS. Create
unix/tclConfig.sh.
* win/configure: Regenerated.
* win/tclWinPort.h: Added missing #endif.
* win/tclWinFile.c (TclpChdir): Don't invert change
directory results on cygwin.
No mention of PASCAL...
> Any objections to removing that undefine? I will do some more testing
> before posting a patch, I was just wondering if this rings a bell with
> anyone.
I don't think we really care, since we would never have _MSC_VER defined
under cygwin. So it's entirely up to you.
Keith