NOTICE: This version of the NSF Unidata web site (archive.unidata.ucar.edu) is no longer being updated.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NOTICE: This version of the NSF Unidata web site (archive.unidata.ucar.edu) is no longer being updated.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.
Hello, I wrote a small utility for parsing NetCDF files, and am not understanding how the variable parameters [vsize] and [begin] are used when a record dimension is present. A typical file containing monthly precipitation has dimensions lat, lon and time, with time being unlimited. The [begin] values for lat and lon are as expected. The [begin] for precip is what I would expect -- since time values do not seem to be stored, but rather are derived from the metadata, I would expect the precip values to begin immediately after lat and lon. The [vsize] is not at all what I'd expect. I would think it should be n-lat * n-lon * n-time-records * sizeof(precip), but instead it is n-lat * n-lon * sizeof(precip). That leads to the [begin] value for time, which I would not think should even exist. The [vsize] for time is what the documentation explains, which is sizeof(time). In short, the header suggests that the precip consists of a single time interval, and that there is a block of time values. Obviously I'm misinterpreting the values. I have not been able to find any documentation that makes this clear. Any direction would be appreciated. Duane
netcdfgroup archives: