Authentication

Authentication works just the same as shotgun_api3, with the only difference that you are passing in any of the parameters into the pyshotgrid.new_site function. If you already have a valid instance of shotgun_api3.Shotgun you also pass that into the pyshotgrid.new_site function directly. Or if you put that into code:

# Method A
sg_site_b = pyshotgrid.new_site(base_url='https://example.shotgunstudio.com',
                                script_name='Some User',
                                api_key='$ome_password')

# Method B
sg = shotgun_api3.Shotgun(base_url='https://example.shotgunstudio.com',
                          script_name='Some User',
                          api_key='$ome_password')
sg_site_a = pyshotgrid.new_site(sg)

# Both methods lead to the same outcome.
assert sg_site_a == sg_site_b

Use tk-core for authentication

You can piggy back SGTK’s authentication procedures to make your live a bit easier to start out with pyshotgrid. The basic idea is to reuse the session token from either SG Desktop or SG Create, so you do not need to use another FPTR API key.

from tank.authentication import ShotgunAuthenticator
import pyshotgrid

# create an authenticator
sa = ShotgunAuthenticator()

# Get a user object. If the authenticator system has already
# stored a default user belonging to a default shotgun site,
# it will simply return this. Otherwise, it will pop up a UI
# asking the user to log in.
user = sa.get_user()

# now the user object can be used to generate an authenticated
# Shotgun connection.
sg = user.create_sg_connection()

# With this you can generate a new pyshotgrid.SGSite and start working.
sg_site = pyshotgrid.new_site(sg)

Warning

Be aware of the following caveats:

• You are running the all Shotgrid CRUD operations with the permissions of the current user in SG Desktop/SG Create.

• The session token can expire midway through your script and SGTK will prompt you for your credentials again. (Happens roughly once a day.)

Due to this I recommend to use this approach mainly for Developers. It is great for starting out with the library or quickly running a simple scripts to fix up some things in Flow Production Tracking. Depending on your goal it can also be used in production environments, but can be limiting.