UPDATE - I just merged the new ExodusII library with libMesh HEAD compiled and ran our suite of tests (hundreds). Here is the preliminary information that I have:
***************************************************************************** There was a very large number of non-whitespace changes in the new API, nearly all of which are not whitespace changes. $ git log --stat ... 172 files changed, 8880 insertions(+), 3547 deletions(-) Compiling inside of libMesh's existing contrib structure produced only 2 warnings and 1 error - int expected but a "size_t" was supplied in a printf statement FIXED with a c-cast - incompatible implicit malloc FIXED with a proper "stdlib.h" include - error with undefined netCDF symbol: NC_EAXISTYPE This symbol was used in a case statement in an error reporting routine. I removed the error by combining that error into the final default case. - ALL libMesh examples "run" with the default configuration options on my workstation - ALL of our application tests passed with the new API in place. ***************************************************************************** This update looks good from my perspective, however maybe we need to update netCDF too? Opinions? I'm happy to commit my changes but I'd like to hear back from the developers before doing so. Thanks, Cody On Jul 11, 2011, at 12:03 PM, Cody Permann wrote: > Greg > Great! Thanks for supporting us on this - we'll get these changes > incorporated into libMesh. Thanks again! > > Roy, I'll download the newest libraries and test our framework against it. > You may want to do you own independent testing to make sure there aren't any > other effects as I believe we are a ways out of date with the version of > ExodusII bundled with libMesh. Once we get the new libraries in libMesh, > I'll submit my patch to the Exodus and Nemesis writers to enable writing zero > ids to those formats. > > Cody > > > On Jul 11, 2011, at 11:50 AM, Gregory Sjaardema wrote: > >> Cody -- >> >> The exodusII libraries available at sourceforge.net/projects/exodusii and >> sourceforge.net/projects/seacas should have the changes to permit 0 as a >> valid entity (block, set, map) id. Let me know if there are any problems. >> >> --Greg >> >> On 6/13/11 11:32 AM, Cody Permann wrote: >>> >>> Greg, >>> >>> Somehow I made a spelling error with one of the variables. Here is the >>> "corrected" patch for ExodusII. >>> >>> >>> >>> >>> Cody >> > > > ------------------------------------------------------------------------------ > All of the data generated in your IT infrastructure is seriously valuable. > Why? It contains a definitive record of application performance, security > threats, fraudulent activity, and more. Splunk takes this data and makes > sense of it. IT sense. And common sense. > http://p.sf.net/sfu/splunk-d2d-c2 > _______________________________________________ > Libmesh-devel mailing list > [email protected] > https://lists.sourceforge.net/lists/listinfo/libmesh-devel ------------------------------------------------------------------------------ All of the data generated in your IT infrastructure is seriously valuable. Why? It contains a definitive record of application performance, security threats, fraudulent activity, and more. Splunk takes this data and makes sense of it. IT sense. And common sense. http://p.sf.net/sfu/splunk-d2d-c2 _______________________________________________ Libmesh-devel mailing list [email protected] https://lists.sourceforge.net/lists/listinfo/libmesh-devel
