deploy_package.md

Using StarThinker As A Python Package

This deployment is for non-developers who simply wish to use StarThinker.

Install StarThinker

To run StarThinker you will need the folowing:

1. A unix / linux command line.
2. A version of Python 3.7 or greater.

From the command line on any unix/linux machine install StarThinker. All following examples begin with this step. Choose only one of the following, the first one is recommended for most users:

python3 -m pip install starthinker
python3 -m pip install git+https://github.com/google/starthinker

• If you get a permission error add: --user to the above command.
• If you get a PATH WARNING:
  1. Copy the new command PATH from the warning.
  2. Activate the new command path now, run: export PATH="[PASTE THE PATH HERE]":$PATH;
  3. Activate for all future logins, run: echo 'export PATH="[PASTE THE PATH HERE]":$PATH;' >> ~/.bash_profile;

Optional Setup User Credentials

To run some recipes, functions, or helpers, user credentials are required. To acquire user credentials run the following utility and follow instructions on screen:

st_auth -h

Use A Function

The functions in util or task are directly usable. This example creates a Google Sheet by name and lists all the DV360 reports under a user account. Simply PIP install it and run recipes...

import json

# import StarThinker functions
from starthinker.util.configuration import Configuration
from starthinker.util.sheets import sheets_create
from starthinker.util.google_api import API_DBM

# initialize credentials
config = Configuration(user='[USER CREDENTIALS JSON STRING OR PATH]')

# create a sheet
sheets_create(config, 'user', 'Test Sheet', 'Test Tab')

# list all your reports in DV360
for report in API_DBM(config, 'user', iterate=True).queries().listqueries().execute():
  print(json.dumps(report, indent=2, sort_keys=True))

For more stand alone examples see tools directory.

Run A Recipe

You can use the StarThinker module directly in any python project to run a recipe. The following example will execute two tasks in sequence in a single process.

from starthinker.util.onfiguration import execute, Configuration

if __name__ == "__main__":
  TASKS = [
    { "hello":{
      "auth":"user",
      "say":"Hello World"
    }},
    { "dataset":{
      "auth":"service",
      "dataset":"Test_Dataset"
    }}
  ]

  execute(
    config=Configuration(
      client='[CLIENT CREDENTIALS JSON STRING OR PATH]',
      user='[USER CREDENTIALS JSON STRING OR PATH]',
      service='[SERVICE CREDENTIALS JSON STRING OR PATH]',
      project='[GOOGLE CLOUD PROJECT ID]',
      verbose=True
    ),
    tasks=TASKS,
    force=True
  )

Additional Resources

Try this in a few seconds using Google Cloud Shell...

Next, review list of available tasks in the Recipe Gallery, view Scripts, or check Command Line Helpers.

---

© 2019 Google LLC - Apache License, Version 2.0