Thanks, that code came from a Windows environment (hence the ^M) and we
cleaned it up enough to get it to compile on Linux (the function
prototypes were not up to the Linux standard) but obviously not enough to
make Solaris happy. Fortunately that it only used for Clouds stuff so it
probably won't affect you for now, but we'll clean it up further for the
next release.
Anything we can do to help on the VNMR front let us know. We only have
Brukers here so it's difficult for us to test Varian stuff out, but we
obviously want to support Varian data as much as possible (not sure how
much Wim has in the Format Converter for this).
Wayne
On Tue, 15 Mar 2005, Bruce D. Ray wrote:
> As before, system consists of Sun Ultra 10 with Solaris 7.
> Compiler is fully patched Sun WorkshopPro 5.0.
> Because of previously reported OpenGL and glut problems,
> I tried the ignore GL option to use tk graphics only.
>
> python is 2.3.4 compiled with the modification to socketmodule.c
> as noted on the sourceforge python bugtracker, item 972724 for lack
> of AF_INET6 and INET_ADDRSTRLEN in the Solaris 7 headers. Note that
> the suggestion for python 2.3.3c1, item 854823, does not produce
> the same functionality. Numeric 23.3 is also present in this python.
>
> tcl/tk is 8.3.5 with tix 8.1.4
>
> X11 is in /usr/openwin
> python, tcl, and tk are under /usr/local
>
> As suggested, lines 80 and 81 of py_struct_util.c must be changed to
> read:
>
> static float r0[COORD_NDIMS], r1[COORD_NDIMS], r2[COORD_NDIMS];
> static float *rotation[] = {r0, r1, r2};
>
>
> With these changes, the compilation failed reporting errors in the
> bacus.h and bacus.c source code as follows:
>
> cc -c -I../../memops/global -O bacus.c
> "./bacus.h", line 96: warning: invalid white space character in directive
> "./bacus.h", line 97: warning: invalid white space character in directive
> "./bacus.h", line 99: warning: invalid white space character in directive
> "./bacus.h", line 100: warning: invalid white space character in directive
> "./bacus.h", line 104: warning: invalid white space character in directive
> "bacus.c", line 232: syntax error before or at: /
> "bacus.c", line 242: cannot recover from previous errors
> cc: acomp failed for bacus.c
> make[1]: *** [bacus.o] Error 2
> make[1]: Leaving directory `/export/home/bruce/ccpnmr/ccpnmr1.0/c/ccpnmr/clouds'
> make: *** [clouds] Error 2
>
> The problem in bacus.h is the presence of cntrl-M "^M" characters at
> the end of every line. Apparently Sun Workshop 5 C regards these as
> white space characters and balks at them. After line termination
> correction, these were not a problem.
>
> The problem in bacus.c is the use of C++ style comments ("//") in C
> code. While GNU C accepts this, Sun Workshop 5 C doesn't. Once I
> changed these to C style ("/* ... */") comments, and there are a
> number of these places, compilation went to completion.
>
> With all graphic handling going through tk, analysis behaves quite
> nicely including allowing me to select contour colors and actually
> giving the selected contour color. I have run the tutorial through
> peak picking on the hsqc spectrum and have not experienced any of
> the error messages, or cursor persistence problems I reported with
> various prior builds.
>
> (Now, if I can just figure out how to set the correct reference in
> analysis for my VNMR data so that I get something near the same
> shifts for peaks between VNMR and analysis ...)
>
>
> Sincerely,
>
>
> --
> Bruce D. Ray, Ph.D.
> Associate Scientist, and Operations Director
> NMR Center
> IUPUI
> Physics Dept.
> 402 N. Blackford St.
> Indianapolis, IN 46202-3273
>
|