On 2013-10-18 07:07-0000 Arjen Markus wrote: > [...]As I am trying to build it on bare Windows, that macro is not defined > and so I get into the #else > branch. [...]
Sorry, Arjen. I missed before that you were doing a build with bare Windows. I assume in the rest of this that means you having been using Microsoft's MSVC compiler for the issues you have recently been reporting? If so, my advice would be to suspend that work, and first confirm python on MinGW/MSYS works for you (like it should because of my Python success with that combination). And that would immediately give the result I am looking for which is does 32-bit python for MinGW/MSYS always give iffy numerical accuracy for Windows regardless of whether it is the Wine version or Microsoft version of that? Of course, after that MinGW/MSYS python success, you could go back to investigating the MSVC case further. But the point is that MinGW success for you would remove all doubt that your Python/NumPy installation was suitable for PLplot builds and provides a much better comparison for my Python results on Wine. > I removed the #define NPY_NO_DEPRECATED_API ... line from plplotcmodule.i, even though this has been in there for a long time, and that helped a bit: the complaints about PyArray_FLOAT are gone. You will probably want to restore #define NPY_NO_DEPRECATED_API NPY_1_7_API_VERSION later once you figure everything out for the MSVC case. See http://docs.scipy.org/doc/numpy-dev/reference/c-api.deprecations.html for why we will want that #define in place. Since this #define causes no issues for me on MinGW/MSYS (and shouldn't for you as well when you try MinGW/MSYS) I expect in your current MSVC case you are actually compiling slightly different code (due to conditional compilation) and thus running into some old deprecated ways of using numpy that have not been exercised by any of our MinGW/MSYS tests. Alan __________________________ Alan W. Irwin Astronomical research affiliation with Department of Physics and Astronomy, University of Victoria (astrowww.phys.uvic.ca). Programming affiliations with the FreeEOS equation-of-state implementation for stellar interiors (freeeos.sf.net); the Time Ephemerides project (timeephem.sf.net); PLplot scientific plotting software package (plplot.sf.net); the libLASi project (unifont.org/lasi); the Loads of Linux Links project (loll.sf.net); and the Linux Brochure Project (lbproject.sf.net). __________________________ Linux-powered Science __________________________ ------------------------------------------------------------------------------ October Webinars: Code for Performance Free Intel webinars can help you accelerate application performance. Explore tips for MPI, OpenMP, advanced profiling, and more. Get the most from the latest Intel processors and coprocessors. See abstracts and register > http://pubads.g.doubleclick.net/gampad/clk?id=60135031&iu=/4140/ostg.clktrk _______________________________________________ Plplot-devel mailing list [email protected] https://lists.sourceforge.net/lists/listinfo/plplot-devel
