ValiotWorker 1.4.0

Enables running Python functions as standalone jobs based on interaction with valiot-jobs API

python-gql-worker

Enables running Python functions as standalone jobs based on interaction with valiot-jobs API

installation

pipenv install valiotworker

Deployment:

python setup.py sdist bdist_wheel && python -m twine upload dist/* --skip-existing # UPLOAD TO PYPI

API < v0.4.0

Job footprint:

@valiotWorker.job(...)
def my_job(job_id, update_job, kwargs)
  ...

Parameters:

• job_id: Number representing the Identifier for this job
• update_job: function to update status of current job (FINISHED, ERROR, etc)
• kwargs: dictionary aliasing the custom params (from job's "input" attribute, for retrocompatibility)

API >= v0.5.0

@valiotWorker.job(...)
def my_job(my_arg1, my_arg2, ... my_arg_n, **_)
  ...

Starting from v0.5.0, every argument is accessed by its keyword.

You may have any arbitrary number of arguments which get loaded from the job input parameter, and the special arguments:

• update_job: function to update status of current job (FINISHED, ERROR, etc)
• job: dictionary containing job information (id, status, queue, insertion date, etc)
• queue: dictionary containing Job's parent Queue information (name, type, frequency, etc)
• kwargs: dictionary aliasing the custom params (from job's "input" attribute, for retrocompatibility)

Worker configuration:

This configurations must run before worker.run() (event loop)

# Poll with subscription
worker.setPollingMode(PollingMode.SUBSCRIPTION)
# or Poll with Queries
worker.setPollingMode(PollingMode.QUERY)

worker.setJobConfigMode(JobConfigMode.KEEP)

# Config remains untouched (Allow persistence of config on DB)
worker.setJobConfigMode(JobConfigMode.KEEP)
# Config updates with job params (Allow quick iteration of job setups)
worker.setJobConfigMode(JobConfigMode.SYNC)