On 2014-08-31 15:24-0700 phil rosenberg wrote: > Re testing with the other settings. I will try
Thanks. That would be worthwhile now just to satisfy my curiosity about whether you get a limited PLplot version in that case. I predict you won't (see below), but seeing is believing.... > - however it will be a very minimal build as I only have static libs linked against the static runtime for all the libraries I have built from source. I actually don't know how the Tcl libraries were built as I downloaded a binary installer, so I'm not sure how they will link. This is case (2), external libraries other than the run-time. In this case for both the -DBUILD_SHARED_LIBS=ON and OFF cases, CMake should just link plplot and the plplot device driver DLL's with the external libraries that CMake finds (whether they are shared or static). So for the default -DBUILD_SHARED_LIBS=ON case you should have just as complete a PLplot as for your current -DBUILD_SHARED_LIBS=OFF case, but if not, let me know because I would be extremely surprised by that result. Just FYI, if CMake has a choice of shared or static library to be found, by default it will pick the shared one, and the result will be our build system will link plplot and the device drivers with shared external libraries. However, it doesn't sound like that is relevant to your case where only the static version of most/all external libraries needed by PLplot is available. Meanwhile, I have done some testing of the "blank-in-pathname" case on Linux and some additional changes beyond what you did are needed even in the Tcl/Tk case. For example, Itcl/Itk version testing was giving incorrect results until I found a fix. I am still in the middle of this and will likely not finish today, but I will give you some notice when I think all Tcl-related "blank-in-pathname" issues are fixed, and it is that stage repeating all your tests (both -DBUILD_SHARED_LIBS=ON and -DBUILD_SHARED_LIBS=OFF) on the VS/Windows platform would be quite valuable as well. But that is just the deal with the testing game; tests have to be repeated again and again. So the best advice I can give you is automate everything so you know you are doing the exact same test each time, and also as a matter of convenience so the repeat tests are trivial for you. 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 __________________________ ------------------------------------------------------------------------------ Slashdot TV. Video for Nerds. Stuff that matters. http://tv.slashdot.org/ _______________________________________________ Plplot-devel mailing list Plplot-devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/plplot-devel
