The goal of the tenint SDK is to provide a simple and reliable way to build the common framework needed to build the connector so that the framework can load & run the connector code.
Currently the code connector code is in a working state, however the commandline
interface is still being actively developed. Currently supported with the tenint
command is the following commands:
# To initialize a project with an example connector.py & pyproject.toml file.
tenint init
# To build the connector image for local launching within docker
tenint build
Connectors require at a minimum a Settings
object, the Connector
object loaded
and pointed to you're main handler, and the app launcher in order to load the
interface.
At the absolute minimum, you would need the following:
from tenint import Connector, Settings
class AppSettings(Settings):
value: str
connector = Connector(settings=AppSettings)
@connector.job
def run_me(config: AppSettings):
print(f"Hello World! Launched with {config.value}")
if __name__ == '__main__':
connector.app()
The Settings class is a pydantic model that you can setup with the settings that you wish to expose to the user within the TIF interface. It will use the model to generate a JSONSchema defintion that is used to paint that page.
To run the code above, there are two commands exposed, config
and run
python connector.py config
python connector.py run -j "{"value": "some string"}"