NOTICE: This version of the NSF Unidata web site (archive.unidata.ucar.edu) is no longer being updated.
Current content can be found at unidata.ucar.edu.
To learn about what's going on, see About the Archive Site.
NOTICE: This version of the NSF Unidata web site (archive.unidata.ucar.edu) is no longer being updated.
Current content can be found at unidata.ucar.edu.
To learn about what's going on, see About the Archive Site.
Hi Barron, > I have found that combining the use of nc_redef and a coordinate > variable causes NETCDF4 formatted files to behave strangely. I have > reduced the problem to its most basic parts and attached a c file and > a bash script that reproduce the problem. > > Conceptual steps to reproduce the problem: > 1. Create a NETCDF4 formatted file > 2. Define coordinate variable dimension > 3. Define another dimension > 4. Define a non-coordinate variable > 5. End definition mode > 6. Reopen definition mode > 7. Define a coordinate variable sharing the dimension name from step 2 > 8. (optional) write data > > Expected result: File with two dimensions and two variables, each > variable having a different dimension. > Actual result: File with two dimensions that, according to ncdump, > share a name and size and two variables with the same dimension > > In case this is platform specific, I am working on a Mac using gcc compilers. Thanks for reporting the problem. We have verified it and also determined that it's not platform-specific. We'll let you know when we have a fix. --Russ Russ Rew UCAR Unidata Program address@hidden http://www.unidata.ucar.edu Ticket Details =================== Ticket ID: WNF-922004 Department: Support netCDF Priority: Normal Status: Closed