# Import standard Libraries 
import os
import argparse
import logging

# Model Libraries
import pandas as pd
import re
import numpy as np
import datetime

# AzureML Libraries
from azureml.core import Dataset, Datastore, Workspace
from azureml.data.datapath import DataPath
from azureml.core import Experiment, Run


### SETUP AZURE ML - ADLS CONNECTION ###
# Get the workspace from config
ws = Workspace.from_config()


config_path = 'config/ml-register-config.csv'
datastore_name = 'cub_pb'



def main(datastore,input_file_name, input_file_path, tags, partition_format,dataset_name):
    
     
    input_file_path = input_file_path.split('/')
    file_path = '/'.join(input_file_path[0:])
    path_on_datastore=''.join([file_path, input_file_name])

    """
    Reads in raw dataset or data from inputs before transforming and registering new datasets in ML workspace

    :param input_file_name: Name of file or dataset to import
    :param datastore : Name of the datastore
    :param input_file_path: If raw file, parameter used to provide path to file where first directory is assumed to be name of datastore
    :param tags: Tags to use when registering dataset
    """
    logging.info("Datastore Name: %s", datastore)
    
    dataset = Dataset.Tabular.from_parquet_files(
		path=(datastore,path_on_datastore),
		validate=False,
		partition_format = partition_format
     )
    
    dataset.register(
        workspace = ws,
        name = dataset_name,
        tags =  {"name": "tapshare", "project": "data-prep1"},
        create_new_version = True,
        description='version 2'
        
    )
    
def get_config_file(datastore_name,config_path):
    """
    Reads in config-file given in input and saves the file as Dataframe 

    :param datastore_name: Name of file or dataset to import
    :param config_path : Name of the config_path where config-file is available
    :param input_file_path: If raw file, parameter used to provide path to file where first directory is assumed to be name of datastore
    :param tags: Tags to use when registering dataset
    """
    datastore = Datastore.get(ws, datastore_name)
    config_dataset = Dataset.Tabular.from_delimited_files(
        path=(datastore,config_path),separator=',',header=True).to_pandas_dataframe()
    return config_dataset
	
    
if __name__ == "__main__":
    os.makedirs('logs', exist_ok=True)
    logging.basicConfig(filename='./logs/processdata.log', level=logging.INFO)

    # Get run context so we can log back to workspace
    run = Run.get_context()

    # Get the workspace this job is running under
    try:
        ws = run.experiment.workspace
    except AttributeError:
        logging.info("Offline run, getting workspace from context")
        ws = Workspace.from_config()

 #   args = parse_args()
  #  logging.info(f"input_file_name argument: {args.input_file_name}")
#  logging.info(f"input_file_path argument: {args.input_file_path}")
  #  logging.info(f"tags argument: {args.tags}")
 #   logging.info(f"partition_format argument: {args.partition_format}")
    
    
    for index,row in get_config_file(
        config_path=config_path,
        datastore_name=datastore_name).iterrows():
        print(row[0],row[1],row[2],row[3],row[4],row[5])
        active=row[5]
        if active:
            main(
	        datastore = Datastore.get(ws, datastore_name=str(row[0])),
             # name of file (or registered dataset)
            input_file_name=str(row[1]),
            # path to dataset
            input_file_path=str(row[2]),
             # tags to use - if any
            tags = None,
            # partition format - if any
            partition_format=str(row[3]),
            # Dataset Name to be registered as
            dataset_name=str(row[4])

            )