make install doesn't work.The issue is due to a bug in the old version of autoconf and automake. Using a newer version of auto tools(autoconf 2.67 and automake 1.11) to generate the corresponding configuration files
For HDF5 fixed size string, one needs to add another character for the termination.The augmentation tool doesn't add this character. Although h5dump or HDFView shows the full string correctly. h5py picks this up. So fixing the attribute routine in the tool and re-generate the expected files.
Updating the testing fileFor HDF5 fixed size string, one needs to add another character for the termination. The augmentation tool doesn't add this character. Although h5dump or HDFView shows the full string correctly. h5py picks this up. So fixing the attribute routine in the tool and re-generate the expected files.
_
Updating the testing fileFor HDF5 fixed size string, one needs to add another character for the termination. The augmentation tool doesn't add this character. Although h5dump or HDFView shows the full string correctly. h5py picks this up. So fixing the attribute routine in the tool and re-generate the expected files.
Previously for 1-D unlimited dimension extendible dataset,the chunk size is set to be 1, which may cause really bad performance.Given the fact the dimension scale array may not be very large, I set the chunk size to be 100 instead of 1.
1. Add shell scripts and the mapping files and the dimensional scale data files for NASA HDF-EOS5 files.2. The real NASA files are huge so a script of downloading the NASA files will be provided in case users whant to test
1. Add shell scripts to test the fake files for default, file and index options and also to test if the tool correctly handles the augmented file2. Add fake test files(mostly from the HDF5 OPeNDAP handler) and the corresponding mapping files for the file option
3. Test on 64-bit Linux xile-vm 2.6.18-371.1.2.el5 #1 SMP
Enhance code to read dimension scale data from an ASCII filePreviously, the tool only supports reading the dimension scale values from an ASCII file that holds the format 'one number per line'. While testing OMI and TES data, I found that some of their dimension scale data are stored in an attribute. HDFView shows the attribute values with commas. For users to conveniently copy the dimension scale data to an ASCII file,the tool had better support the re...
1. Finish the major coding part to check if the dimension names and the field names in the mapping file match corresponding names in the HDF-EOS5 file2. Test various combinatios of single grid, multiple grids,single swath, multiple swaths, single zonal and multiple zonal cases.
3. Comments etc. still need to be cleaned up
