On 2014-10-07 09:05-0000 phil rosenberg wrote: > Just to confirm - everything seems to work fine on Cygwin too
Hi Phil: It was glad to hear you got good results on both the MSVC and Cygwin platforms (while testing with Cygwin's bash in both cases) for my commit 5dd59c. I was pleasantly surprised our series of succesive approximations did not take more iterations. That is contrary to most of my computer experience! :-) To ask the question before Arjen does, is that the 64-bit Cygwin or 32-bit Cygwin? With regard to the blank in build-tree name issue, are you actually getting good results for the "test_tcl" target on both platforms? From my Linux results with a blank in the build-tree pathname, there is more to do there (which I leave to you to sort out since you often use blanks in pathnames). BTW, once the test_tcl target works for you, then that is a good proof of concept that many of the test targets in the test suite will also work for you, and I would encourage you to keep expanding the tests by running (and fixing any issues that might show up on your MSVC and Cygwin platforms the following tests: (i) the test_interactive target (which automatically runs the interactive part of test_tcl plus all other interactive tests), (ii) the test_noninteractive target (which runs the noninteractive part of test_tcl plus all other noninteractive tests), and finally scripts/comprehensive_test.sh (which for our three major configurations runs ctest for the build tree (similar to but not identical to the tests that are run by test_noninteractive target) and the test_interactive and test_noninteractive targets for build tree, install tree, and traditional install tree). The last time I ran that script for MinGW/MSYS/Wine (in March, see the record in https://sourceforge.net/p/plplot/wiki/Testing_PLplot/) it produced perfect results, but it took a lot of pathname manipulations on my part to obtain that MSYS platform result. I am encouraged by your recent test results that those manipulations also basically work for MSVC and Cygwin so I really don't think you are too far from being able to run that script on those two platforms with success. Of course, once you have scripts/comprehensive_test.sh working for any platform, then you should record that good result in https://sourceforge.net/p/plplot/wiki/Testing_PLplot/ for future reference. And you should also run that script again a week or so before releases (and update that wiki page accordingly) to make sure there were no build-system regressions introduced during the release cycle for platforms of interest to you. In sum, you now have a foot in the door leading to comprehensive testing success for both the MSVC and Cygwin platforms, and it is time to kick that door wide open. :-) 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 __________________________ ------------------------------------------------------------------------------ Meet PCI DSS 3.0 Compliance Requirements with EventLog Analyzer Achieve PCI DSS 3.0 Compliant Status with Out-of-the-box PCI DSS Reports Are you Audit-Ready for PCI DSS 3.0 Compliance? Download White paper Comply to PCI DSS 3.0 Requirement 10 and 11.5 with EventLog Analyzer http://pubads.g.doubleclick.net/gampad/clk?id=154622311&iu=/4140/ostg.clktrk _______________________________________________ Plplot-devel mailing list Plplot-devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/plplot-devel
