py-analytics

This repo contains a python framework with Data Enginner, Data Scientist and 3rd party integration tools capabilities

Installation

py-test-utility can be installed via pip

pip install py-analytics

tdd_utility - module

class load_csv(csv,schema)

Contains methods to extract the equivalent json from csv with nested and repeated records structures

Args

• csv
  • path and file name of the csv
  • mandatory
  • nested fields shall be separated by a dot "." (i.e. item.id, item.quantity)

• schema
  • path and schema file name of the table schema
  • required if the CSV contain nested and repeated records
  • json format i.e.

[  
    {
      "mode": "NULLABLE", 
      "name": "order", 
      "type": "STRING"
    },  
    {
      "fields": [
        {
          "mode": "NULLABLE", 
          "name": "id", 
          "type": "STRING"
        },
        {
          "mode": "NULLABLE", 
          "name": "quantity", 
          "type": "STRING"
        }
      ], 
      "mode": "REPEATED", 
      "name": "item", 
      "type": "RECORD"
    }, 
    {
      "fields": [
        {
          "mode": "NULLABLE", 
          "name": "address", 
          "type": "STRING"
        }, 
        {
          "mode": "NULLABLE", 
          "name": "postcode", 
          "type": "STRING"
        }
      ], 
      "mode": "NULLABLE", 
      "name": "delivery", 
      "type": "RECORD"
    }
  ]

Methods

• to_json()

  • if successfuls return the json extracted from the csv

• to_new_line_delimiter_file(output_file_name)

  • return 0 if successfuls
  • create new line delimiter "output_file_name" file

Usage

>>> from data_prep import tdd_utility as  tu
>>> mockdata_csv = tu.load_csv(
...     csv="path/to/filename/file.csv", 
...     schema="path/to/schema/schema.json") # initialise the object
>>> mockdata_json = mockdata_csv.to_json() # return the equivalent json
>>> mockdata_json = mockdata_csv.to_new_line_delimiter_file(output="path/output_file_name.json") # return output_file_name