This archive contains answers to questions sent to Unidata support through mid-2025. Note that the archive is no longer being updated. We provide the archive for reference; many of the answers presented here remain technically correct, even if somewhat outdated. For the most up-to-date information on the use of NSF Unidata software and data services, please consult the Software Documentation first.
Hi Jeff, We actually think we've figured out the problem and we're testing a fix this afternoon. If our diagnosis is correct, everything is written OK, it's only that the structure offsets that ncdump uses when reading are off. This only occurs in generic programs such as ncdump, which don't have a native structure into which to read the compound value. But the fix would be in the netCDF-4 library to make sure the member offsets are properly adjusted before returning from the nc_inq_compound_field() function that ncdump relies on to figure out where in the void* data it has read it finds the member values. We were able to create test files for this problem with ncgen4 from your CDL and reproduce the problem. Thanks again for reporting this. It's a significant error that our testing doesn't reveal, because we don't do much cross platform testing, but we will add this test and others that catch these sorts of problems. We'll let you know when a release is available with a fix, and after you've verified it fixes the problem to your satisfaction, we'll post a note to the netcdfgroup mailing list. --Russ Russ Rew UCAR Unidata Program address@hidden http://www.unidata.ucar.edu Ticket Details =================== Ticket ID: HPS-623090 Department: Support netCDF Priority: Normal Status: Closed