.. Copyright SAS Institute ========================= Installing the SAS kernel ========================= The SAS kernel package installs just like any other Python package. It is a pure Python package and works with Python 3.X installations. To install using ``pip``, you execute one of the following commands. :: pip install sas_kernel pip install http://github.com/sassoftware/sas_kernel/releases/sas_kernel-X.X.X.tar.gz ****************************************************************** Linux install for Anaconda Python (assuming SAS already installed) ****************************************************************** #. Go to https://www.continuum.io/downloads and install Anaconda Python (make sure you get Python3.X). If you install Anaconda without Superuser privileges (root or sudo), then other users on the system will not be able to access the SAS kernel. Consider the following: * The default installation location is your home directory. This is fine for a single user install. If you want a system-wide installation, then use a common location such as ``/opt``. * One installation prompt is to add the path to your environment. SAS recommends that you answer 'yes' to the prompt so that you get the executables in your path automatically. If you are performing a system-wide installation (using root or sudo), then all the other users must add the path to their environmental variables. #. Install the SAS kernel package. The package has a dependency on the SASPy package. The SASPy package is available from https://github.com/sassoftware/saspy. If the ``pip`` command in your path does not map to Python3, then use ``pip3`` instead. :: pip install sas_kernel #. Verify that the package is installed. :: jupyter kernelspec list If you installed as a Superuser, your output should look similar to the following: :: Available kernels: python3 /opt/Anaconda3-2.5.0/lib/python3.5/site-packages/ipykernel/resources sas /usr/local/share/jupyter/kernels/sas If you installed as a regular user (the sas user account, in this case), your output should look similar to the following: :: Available kernels: python3 /home/sas/anaconda3/lib/python3.5/site-packages/ipykernel/resources sas /home/sas/.local/share/jupyter/kernels/sas #. Configure the SAS executable for your system. The connection of the Jupyter notebook to SAS is made by the SASPy Python package (which was installed as a dependency for you). Use the `SASPy configuration documentation`_ to complete you set up. If you are using Linux or Unix the STDIO_ or `STDIO over SSH`_ will be the fastest connections, but IOM is also available All other systems will use *only* the IOM_ connection. If you want to connect to PC SAS on your local system, use `IOM Local`_ .. _STDIO: https://sassoftware.github.io/saspy/install.html#stdio .. _STDIO over SSH: https://sassoftware.github.io/saspy/install.html#stdio-over-ssh .. _IOM: https://sassoftware.github.io/saspy/install.html#iom .. _IOM Local: https://sassoftware.github.io/saspy/install.html#local .. _SASPy configuration documentation: https://sassoftware.github.io/saspy/install.html#configuration * Configuration documentation: https://sassoftware.github.io/saspy/install.html#configuration * GitHub version of the file: https://github.com/sassoftware/saspy/blob/master/saspy/sascfg.py * More information about SASPy: https://github.com/sassoftware/saspy *********************************************************** Linux install for Centos 6 (assuming SAS already installed) *********************************************************** These instructions describe hot to perform a system-wide installation for all users. You must have Superuser privileges (root or sudo). #. You can use the ``yum`` command to install from RPM packages. :: sudo yum install https://centos6.iuscommunity.org/ius-release.rpm sudo yum install python35u gcc-c++ python35u-devel python35u-pip python35u-tools #. You can use the ``pip`` command. :: wget https://bootstrap.pypa.io/get-pip.py python3.5 get-pip.py pip3 --version`` #. Install Jupyter and the SAS kernel package. The package has a dependency on the SASPy package. The SASPy package is available from https://github.com/sassoftware/saspy. :: pip3.5 install jupyter pip3.5 install sas_kernel #. Verify that the SAS kernel package is installed. :: jupyter kernelspec list Your output should look similar to the following: :: Available kernels: python3 /usr/lib/python3.5/site-packages/ipykernel/resources sas /usr/local/share/jupyter/kernels/sas #. Configure the SAS executable for your system. The connection of the Jupyter notebook to SAS is made by the SASPy Python package (which was installed as a dependency for you). Use the `SASPy configuration documentation`_ to complete you set up. If you are using Linux or Unix the STDIO_ or `STDIO over SSH`_ will be the fastest connections, but IOM is also available All other systems will use *only* the IOM_ connection. If you want to connect to PC SAS on your local system, use `IOM Local`_ .. _STDIO: https://sassoftware.github.io/saspy/install.html#stdio .. _STDIO over SSH: https://sassoftware.github.io/saspy/install.html#stdio-over-ssh .. _IOM: https://sassoftware.github.io/saspy/install.html#iom .. _IOM Local: https://sassoftware.github.io/saspy/install.html#local .. _SASPy configuration documentation: https://sassoftware.github.io/saspy/install.html#configuration * Configuration documentation: https://sassoftware.github.io/saspy/install.html#configuration * GitHub version of the file: https://github.com/sassoftware/saspy/blob/master/saspy/sascfg.py * More information about SASPy: https://github.com/sassoftware/saspy ************************************************ Windows install (assuming SAS already installed) ************************************************ #. Go to https://www.continuum.io/downloads and install Anaconda Python (make sure you get Python3.X). If you install Anaconda without Administrator privileges, then other users on the system will not be able to access the SAS kernel. Consider the following: * Install in the default location unless you have a good reason to change it. Using the default location simplifies administration. .. image:: ./images/ap3.PNG :scale: 50% * One installation prompt is to make Python accessible for just your account or for all users. Select the best response for your situation. .. image:: ./images/ap2.PNG :scale: 50% * Another installation prompt is to add the path to your environment. SAS recommends that you answer 'yes' to the prompt so that you get the executables in your path automatically. Adding the path your environment simplifies starting Python and Jupyter. .. image:: ./images/ap4.PNG :scale: 50% .. IMPORTANT:: This next group of steps is performed from a Windows command prompt ( :menuselection:`Start --> Run --> cmd`) #. Install the SAS kernel package. The package has a dependency on the SASPy package. The SASPy package is available from https://github.com/sassoftware/saspy. If the ``pip`` command in your path does not map to Python3, then use ``pip3`` instead. :: pip install sas_kernel #. Verify that the package is installed. :: jupyter kernelspec list Your output should look similar to the following: :: Available kernels: python3 C:\Users\sas\AppData\Local\Continuum\Anaconda3\lib\site-packages\ipykernel\resources sas C:\ProgramData\jupyter\kernels\sas #. Configure the SAS executable for your system. The connection of the Jupyter notebook to SAS is made by the SASPy Python package (which was installed as a dependency for you). Use the `SASPy configuration documentation`_ to complete you set up. If you are using Linux or Unix the STDIO_ or `STDIO over SSH`_ will be the fastest connections, but IOM is also available All other systems will use *only* the IOM_ connection. If you want to connect to PC SAS on your local system, use `IOM Local`_ .. _STDIO: https://sassoftware.github.io/saspy/install.html#stdio .. _STDIO over SSH: https://sassoftware.github.io/saspy/install.html#stdio-over-ssh .. _IOM: https://sassoftware.github.io/saspy/install.html#iom .. _IOM Local: https://sassoftware.github.io/saspy/install.html#local .. _SASPy configuration documentation: https://sassoftware.github.io/saspy/install.html#configuration * Configuration documentation: https://sassoftware.github.io/saspy/install.html#configuration * GitHub version of the file: https://github.com/sassoftware/saspy/blob/master/saspy/sascfg.py * More information about SASPy: https://github.com/sassoftware/saspy .. NOTE:: For PC SAS (connecting to your local system), use the `IOM Local`_ connection instructions. ***************** OSX (Mac) install ***************** #. Go to https://www.continuum.io/downloads and install Anaconda Python (make sure you get Python3.X). If you install Anaconda without Administrator privileges, then other users on the system will not be able to access the SAS kernel. Consider the following: * Install in the default location unless you have a good reason to change it. Using the default location simplifies administration. * One installation prompt is to make Python accessible for just your account or for all users. Select the best response for your situation. * Another installation prompt is to add the path to your environment. SAS recommends that you answer 'yes' to the prompt so that you get the executables in your path automatically. Adding the path your environment simplifies starting Python and Jupyter. #. Install the SAS kernel package. The package has a dependency on the SASPy package. The SASPy package is available from https://github.com/sassoftware/saspy. If the ``pip`` command in your path does not map to Python3, then use ``pip3`` instead. :: pip install sas_kernel #. Verify that the package is installed. :: jupyter kernelspec list Your output should look similar to the following: :: Available kernels: python3 /Users/sas/anaconda3/lib/python3.5/site-packages/ipykernel/resources sas /usr/local/share/jupyter/kernels/sas #. Configure the SAS executable for your system. The connection of the Jupyter notebook to SAS is made by the SASPy Python package (which was installed as a dependency for you). Use the `SASPy configuration documentation`_ to complete you set up. If you are using Linux or Unix the STDIO_ or `STDIO over SSH`_ will be the fastest connections, but IOM is also available All other systems will use *only* the IOM_ connection. If you want to connect to PC SAS on your local system, use `IOM Local`_ .. _STDIO: https://sassoftware.github.io/saspy/install.html#stdio .. _STDIO over SSH: https://sassoftware.github.io/saspy/install.html#stdio-over-ssh .. _IOM: https://sassoftware.github.io/saspy/install.html#iom .. _IOM Local: https://sassoftware.github.io/saspy/install.html#local .. _SASPy configuration documentation: https://sassoftware.github.io/saspy/install.html#configuration * Configuration documentation: https://sassoftware.github.io/saspy/install.html#configuration * GitHub version of the file: https://github.com/sassoftware/saspy/blob/master/saspy/sascfg.py * More information about SASPy: https://github.com/sassoftware/saspy .. NOTE:: For OSX, the only supported configuration is an IOM_ connection. =========================== Installing SAS NBextensions =========================== ******************** Installing from PyPi ******************** With the release of Jupyter 4.2 (SAS kernel package version 1.2) the installation and enabling of nbextensions is improved. To install and enable the showSASLog extension use the following commands. :: jupyter nbextension install --py sas_kernel.showSASLog jupyter nbextension enable sas_kernel.showSASLog --py To install and enable the theme extension use the following commands. :: jupyter nbextension install --py sas_kernel.theme jupyter nbextension enable sas_kernel.theme --py To verify the nbextensions that you installed use the following commands. :: jupyter nbextension list If the extensions are correctly installed, you will see output similar to the following: :: Known nbextensions: config dir: /root/.jupyter/nbconfig notebook section showSASLog/main enabled - Validating: OK theme/theme_selector enabled - Validating: OK *********************************** Installing from a cloned repository *********************************** The cloned repository has a directory for each nbextension within the file structure as shown below: :: sas_kernel | +-- showSASLog +-- theme You can install the extensions from the command line. To install an extension system-wide, use the following command with Superuser privileges (root or sudo). The following command assumes that you are in the nbextensions directory. Adjust the path if you are not. :: jupyter nbextension install ./showSASLog Your output should look similar to the following (installed with Superuser privileges): :: copying showSASLog/main.js -> /usr/local/share/jupyter/nbextensions/main.js To install for your user account only, use the following command. Again, the sample command assumes that you are in the nbextensions directory. Adjust the path if you are not. :: jupyter nbextension install ./showSASLog --user Your output should look similar to the following (installed for your user account only): :: copying showSASLog/main.js -> /home/sas/.local/share/jupyter/nbextensions/showSASLog/main.js Then enable the notebook extension with the following command. :: jupyter nbextension enable showSASLog To disable the extension, you can run the following command. :: jupyter nbextension disable showSASLog Example ======= There is a `notebook`_ that walks through the steps to install and enable the extensions. .. _notebook: https://github.com/sassoftware/sas_kernel/blob/master/notebook/loadSASExtensions.ipynb