From: Alan W. I. <ir...@be...> - 2014-08-26 23:00:45
|
On 2014-08-26 13:21-0700 Alan W. Irwin wrote: > Getting back to the principal issue, it appears to me that gcc is > accepting all those "\n\" continuation lines in > bindings/tcl/plplot_parameters.h, but your Portland Group compiler is > not. We will look further at this. Hi Irena: I have now changed the style of plplot_parameters.h to replace one huge string literal by a large number of small string literals that the compiler should automatically concatenate together. I prefer this new style on general principles, and gcc compiles this style just as well as the old style. It's possible your Portland Group compiler will still choke on this style as it attempts to concatenate those string literals together. But on the other hand, it might work. To test that possibility, please try the latest version of our source code using git clone git://git.code.sf.net/p/plplot/plplot plplot.git Then try your normal build (with -DENABLE_tcl=ON or else by removing -DENABLE_tcl=OFF from the cmake command line since ENABLE_tcl is ON by default) referring to this version of the source tree rather than 5.10.0 and starting from an empty build tree. If your compiler is able to compile bindings/tcl/tclAPI.c without issues, then the change in style should be judged a success. Furthermore, I hope your set of Portland Group compilers is able to finish the rest of a default build of PLplot without issues, but if there are such issues we will try to deal with them since we prefer to have PLplot builds be successful on all compilers and not just gcc. By the way, I think the above is the first reference to our new official git repository on this list (although there has been a large amount of traffic about this huge change in PLplot development on the plplot-devel list). The short story is our svn repository at SF is now frozen in read-only mode and as of a few days ago we switched to using git for PLplot development with the above being our official git repo. I am pretty much a git newbie myself, but my use of it for the last several days shows it is an extremely popular version control system for good reasons. Therefore, I think there is an excellent chance this change from subversion to git will not only help our current set of developers but also attract new developers to the project. 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 __________________________ |