.. _python_sdk:

Python SDK
========


===========
Installation
===========

python3 is required for the installation. Either setup python3 as the default or use virtualenv. 

Use pip to install the taskmonk SDK ::

    pip3 install taskmonk-sdk

==========
QuickStart
==========
The quickest way to get started with Taskmonk is use the upload file and download of output features as shown below. The projectId will be provided by the annotation partner

.. code-block:: python
   :linenos:

    from taskmonk import TaskMonkClient
    import json
    import logging
    from time import sleep

    level    = logging.DEBUG
    format   = '  %(message)s'
    handlers = [logging.FileHandler('logger.log'), logging.StreamHandler()]

    logging.basicConfig(level = level, format = format, handlers = handlers)

    project_id = "project_id"

    client = TaskMonkClient("https://preprod.taskmonk.io", 
        project_id = project_id, 
        client_id = 'client_id',
        client_secret= 'client_secret')


    #
    # Create the taxonomy
    #
    taxonomy_id = client.create_taxonomy('taxonomy_name') 
    logging.debug('Created taxonomy %s', taxonomy_id)

    #
    # Upload taxonomy categories
    #
    upload_result = client.import_taxonomy(taxonomy_id, '/Users/taskmonk/taxonomy.xlsx')
    logging.debug('Import taxonomy result = %s', upload_result)

    #
    # Create the batch with the default parameters
    #
    batch_id = client.create_batch("batchname")
    logging.debug('Created batch %s', batch_id)

    #
    # Upload the tasks from the csv file
    #
    upload_job_id = client.upload_tasks(batch_id, '/Users/taskmonk/tmp.csv', 'CSV')
    logging.debug("upload job_id = %s", upload_job_id)

    #
    # Check the progress of the upload job
    #
    upload_job_progress = client.get_job_progress(upload_job_id)
    logging.debug("Job Progess = %s", upload_job_progress)

    #
    # Wait till the upload job is complete
    #
    while (not client.is_job_complete(upload_job_id)):
        logging.debug("waiting for job to complete")
        sleep(1)
    logging.debug("Upload tasks is complete")

    #
    # Check the status of the batch
    #
    batch_status = client.get_batch_status(batch_id)
    logging.debug("Batch Status = %s", batch_status)

    #
    # Wait till the batch is completed by the annotation partners
    #
    while (not client.is_batch_complete(batch_id)):
        logging.debug("waiting for batch to complete")
        sleep(1)

    logging.debug("Batch annotation is complete")

    #
    # Get batch output
    # 
    batch_output = client.get_batch_output(batch_id, '/tmp/output.csv', output_format = 'CSV')
    logging.debug('batch_output = %s', batch_output)


=============
Documentation
=============

SDK documentation is available at :ref:`api`.