As of revision 11964 I have done a number of build-system improvements and minor bug fixes.
I have implemented much more comprehensive OUTPUT attributes for the custom commands run by the test_noninteractive target. As a result "make clean" now removes virtually all plot files generated by that target. I have sorted out some issues (incorrect file names) with running the special 16a fortran examples that "make clean" revealed. I have implemented the clean_ctest_plot_files target which does exactly what it says. I have reorganized scripts/comprehensive_test.sh to take advantage of this new functionality. By default (--do_clean_as_you_go yes) all plot files (including those generated by ctest) are cleaned right after they are generated. This default option now reduces the disk space used by a fully comprehensive test from ~35GB at the end and a somewhat larger high watermark during the running of the script to ~300MB at the end and a high watermark during the running of the script of ~4GB. I hope this very large reduction in disk space usage will encourage lots of people here to use this script. Of course, the default --do_clean_as_you_go yes assumes you are only interested in possible error messages in the *.out files that are output by the commands that generate the plot files. So those plot files are removed by that default option. If you want to look at those plot files in addition to the *.out files then use --do_clean_as_you_go no. However, in that case you should be prepared for something like 40GB (!) disk usage by a complete run of scripts/comprehensive_test.sh. I have no further development plans for PLplot for this release cycle although I do plan to participate in nightly testing using ctest and our cdash server dashboard used to collect and report such tests once we have had a couple of pioneers try out the service. ctest only does a small fraction of what scripts/comprehensive_test.sh does. Nevertheless, it is a useful quick check, and the point of such nightly testing on a number of our developers different platforms is it should greatly reduce the last-minute nasty surprises we get near the end of our release cycles as a result of the much more intensive testing (e.g., with scripts/comprehensive_test.sh) that we do then. 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 __________________________ ------------------------------------------------------------------------------ All the data continuously generated in your IT infrastructure contains a definitive record of customers, application performance, security threats, fraudulent activity and more. Splunk takes this data and makes sense of it. Business sense. IT sense. Common sense. http://p.sf.net/sfu/splunk-d2d-oct _______________________________________________ Plplot-devel mailing list [email protected] https://lists.sourceforge.net/lists/listinfo/plplot-devel
