Quickstart

Scripts (Recommended)

Creating a script gives you the full functionality of DisruptionPy.

Examples

Use basic_example_1.py or basic_example_2.py for a simple way to get started. Check out all_defaults_example.py to see the all the default settings for retrieving data.

Creating a DisruptionPy script

1. Create the shot data retrieval settings The retrieval settings allow you to specify the settings for retrieving data for a single shot. You can specify details like the columns you want filled by the physics methods and the timebase domain you want used. See RetrievalSettings for more information and a complete list of settings options.

   from disruption_py.settings.retrieval_settings import RetrievalSettings
   
   retrieval_settings = RetrievalSettings(
       # Use the efit timebase when returning data
       time_setting="efit",
       # Run all available methods
       run_tags=["all"],
   )
   

2. Call get_shots_data. workflow.py is the main entrypoint for retrieving data from DisruptionPy. The get_shots_data method will be the primary method you use to get data, but workflow.py can also help you set up connections to the SQL database and the MDSplus server. See workflow.py for more details.

   from disruption_py.settings.retrieval_settings import RetrievalSettings
   from disruption_py.workflow import get_shots_data
   retrieval_settings = RetrievalSettings(
       ...
   )
   shot_data = get_shots_data(
       tokamak="cmod",
       # Retrieve data for the desired shots
       shotlist_setting=[1150805012, 1150805013, 1150805014],
       # Use the created retrieval_settings
       retrieval_settings=retrieval_settings,
       # Automatically stream retrieved data to a csv file by passing in a file path ending in .csv
       output_setting="data.csv",
       # Use a single process to retrieve the data
       num_processes=1,
   )
   

   I don't want to connect to the SQL database.

   DisruptionPy provides a dummy database class that allows users to retrieve data from MDSplus without having to connect to the associated SQL database. See the example or simply change the code above to pass the dummy database's default inititalizer as the database initializer to the handler class. note: this will result in data retrieval being incorrect for parameter methods that depend on data retrieved from the SQL table eg. time_until_disrupt

   #!/usr/bin/env python3
   
   """
   Example usage of `get_shots_data` demonstrating the use of a dummy sql database.
   """
   
   
   import logging
   import os
   
   from disruption_py.inout.sql import DummyDatabase
   from disruption_py.settings.log_settings import LogSettings
   from disruption_py.settings.retrieval_settings import RetrievalSettings
   from disruption_py.workflow import get_shots_data
   
   retrieval_settings = RetrievalSettings(
       # uses the efit timebase when returning data
       time_setting="ip",
       # run all available methods
       run_tags=["all"],
   )
   shot_data = get_shots_data(
       tokamak="d3d",
       # Retrieve data for the desired shots
       shotlist_setting=[161228, 161237, 166177, 166253],
       retrieval_settings=retrieval_settings,
       # automatically stream retrieved data to a csv file by passing in a file path ending in .csv
       output_setting=f"/tmp/{os.environ['USER']}/data.csv",
       num_processes=1,
       database_initializer=DummyDatabase.initializer,  # use dummy database
       log_settings=LogSettings(
           console_log_level=logging.DEBUG,
       ),
   )