Your Data Library Home Directory

If you are here at LDEO/IRI, you can get a home directory in the Data Library which can point to all your datasets. I would recommend that it be in your home directory, and most people call it 'mydata'. For example, if I say

home .benno

in the Data Library, I get the contents of

/home/benno/mydata

which points to all my private datasets.

The next step is to connect this directory to the data you wish to use with the Data Library.

Cleanest way is to create a directory tree of subdirectories beneath mydata that embodies the structure that you want. The leaves on this tree are subdirectories that correspond to data files. In each of those subdirectories you have a file called index.tex which contains instructions for accessing the data. For example, if the data is in a netcdf file (dirname/filename.nc), index.tex contains

  
\begin{ingrid}
(dirname/filename.nc)continuewithCDF
\end{ingrid}

You can also extract variables from files, in case you need to restructure things.

  
\begin{ingrid}
/varnameyouwant {(dirname/filename.nc)readfile .varnameinfile} defasvarsilent
\end{ingrid}

Alternatively, you can describe data that is kept in binary files. In this case, index.tex would contain

  
\begin{ingrid}
continuedataset:
grid:
/name /X def
...
:grid
...
variable:
...
:variable

etc

:dataset
\end{ingrid}

using the usual Ingrid dataset description language (documentation available).