datalab command line client
You can install the datalab script (a command-line tool) by cloning from our github repository. Assuming that you have git installed on your machine, simply clone the client:
git clone https://github.com/noao-datalab/datalab-client.git
cd datalab-client/
Install the client into your system:
python setup.py install
If all went well, you should be able to run the datalab command from a terminal on your computer.
Using Data Lab interfaces in local Python notebooks
If you have cloned the datalab client as shown above, you can also use the Python interfaces directly in you notebooks, running on an own Jupyter server (that you launch locally on your computer). You just have to point your system to where the cloned datalab code resides in your file system.
Let's assume that you cloned datalab to a directory called ~/foo/, i.e. the full path to the datalab directory is
/home/you/foo/datalab-client/
The easiest way to let your your system know about it is to add this directory to the PYTHONPATH environment variable. Assuming you are using the bash shell, add (or edit) this line to the file ~/.bashrc
export PYTHONPATH="/home/you/foo/datalab-client:${PYTHONPATH}"Make sure you source the file
source ~/.bashrc
If you now start your local Jupyter notebook server
jupyter notebook
a browser window with your notebooks will pop up, where you can then use the Data Lab interfaces by simply importing any or all of them:
from dl import authClient as ac
from dl import queryClient as qc
from dl import storeClient as sc