Installation Quickstart
-----------------------

.. _linux-installation:

MacOS & Linux User Installation
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^

For non-developers, the :code:`astro-otter` API can be installed using :code:`pip`:

::

   python3 -m pip install astro-otter[standard]

Or, for a more minimal installation without the dependencies for the :code:`DataFinder`
or :code:`plotter` modules:

::

   python3 -m pip install astro-otter

Then, from a python terminal you can do things like

::

   import otter
   db = otter.Otter()
   db.query(names="18hyz")

Note, we do not natively support Windows OS installation but do support WSL2 installation
(see below). This is a change as of v0.7.0 and is because of the new :code:`dustmaps`
package requirement. The :code:`dustmaps` code relies on healpy which does not support
Windows.

Windows Installation
^^^^^^^^^^^^^^^^^^^^
Since we do not natively support Windows, we suggest installing and running Windows
Subsystem for Linux (WSL2). You can follow the instructions on the
`Microsoft docs page <https://learn.microsoft.com/en-us/windows/wsl/install>`_
to install WSL2. Once it is installed
you can start a WSL2 terminal by running the :code:`wsl` command in a powershell terminal.
After running this command and entering the WSL2 terminal, you can return to :ref:`linux-installation`.

Developer Installation
^^^^^^^^^^^^^^^^^^^^^^

For a local installation of everything, typically for developers, follow these steps instead:

1. First, install some other dependencies:
   a. ADS API: https://ads.readthedocs.io/en/latest/#getting-started
   b. Docker (install AND run):

2. Then, set the following environment variables in your .bashrc/.zshrc file:

::

   export OTTER_ROOT=/path/to/where/to/clone
   export ARANGO_URL=<insert url>
   export ARANGO_ROOT_PASSWORD=<insert password>
   export VETTING_PASSWORD=<insert password>
   export ARANGO_USER_PASSWORD=<insert password>
   export ADS_DEV_KEY=<insert api key>
   export TNS_BOT_ID=<bot id>
   export TNS_BOT_NAME=<bot name>
   export TNS_API_KEY=<api key>
   export OTTER_WEB_BASE_URL="/otter"

If you don't know what any of the values should be, reach out to an existing developer

3. Source your .bashrc/.zshrc

::

   source ~/.bashrc

4. Clone the relevant repositories

::

   git clone https://github.com/astro-otter/otter.git $OTTER_ROOT/otter
   git clone https://github.com/astro-otter/otterdb.git $OTTER_ROOT/otterdb
   git clone https://github.com/astro-otter/otter-web.git $OTTER_ROOT/otter-web
   git clone https://github.com/astro-otter/otter-docker.git $OTTER_ROOT/otter-docker

5. Install the otter python API from source and install our pre-commit
   hooks that help maintain code quality.

::

   cd $OTTER_ROOT/otter
   python -m pip install -e .
   pre-commit install


6. Install the web application by running

   ::

      cd $OTTER_DIR/otter-web
      python3 -m pip install -e .

7. Run the web application in developer mode. Note that this will connect to the database at the
   environment variable :code:`ARANGO_URL`. From the otter-web directory

   ::

      python3 start.py

8. Test that the website is running by navigating to http://localhost:8080/otter

9. Assuming everything is working as expected, you are now able to change things as you wish!

Setting up a local copy of the database
"""""""""""""""""""""""""""""""""""""""
Hopefully, a developer should never need to setup a local copy of the database and can instead connect
to the production database when modifying the web applications and API. In the unlikely scenario that
a developer does need to setup a local copy of the database, they can follow these steps.

1. Get the database up and running:

   a. First, make sure docker is running

   b. If you have an arm64 cpu (like a Mac Silicon chip), you will need to build the docker container on
      that cpu architecture. From the otterdb directory run:

      ::

	 docker build -t otterdb:v0.3.6 .

   c. Then run the docker container from dockerhub (if you built it locally, make sure you change the
      name at the bottom of this command)

      ::

	 docker run \
	 -e ARANGO_ROOT_PASSWORD=$ARANGO_ROOT_PASSWORD \
	 -e VETTING_PASSWORD=$VETTING_PASSWORD \
	 -e VETTING_USER="vetting-user" \
	 -e ARANGO_USER_USERNAME="user-guest" \
	 -e ARANGO_USER_PASSWORD=$ARANGO_USER_PASSWORD \
	 -e DB_LINK_PORT_8529_TCP="http://127.0.0.1:8529" \
	 -p 8529:8529 \
	 noahfranz13/otterdb:v0.3.6

   d. Open a new terminal. Then, add a copy of the data to the database from the master copy at SciServer.
      From the otterdb directory, run

      ::

	 python3 import_from_sciserver.py

2. Test that arangodb is running correctly and the data is imported by navigating to http://localhost:8529
   in a browser and logging in as user "root" and the password set by $ARANGO_ROOT_PASSWORD

3. Start the web application locally, connected to your local database, with

   ::

      export ARANGO_URL="http://localhost:8529" && python3 start.py