BrainPad / cliboa

application framework for ETL(ELT) pipeline, process
MIT License
20 stars 13 forks source link

PyPI PyPI - Implementation PyPI - Python Version GitHub Actions Code Style:
black Contributions Welcome Repo Size Gitter

Table of Contents

Introduction

What is cliboa

cliboa is an application framework which can implement ETL(ELT) pipeline. It eases the implementation of ETL(ELT) pipeline. In this case, ETL(ELT) pipeline means the processings like fetch, transform and transfer of data between various databases, storages, and other services.

Features

Manual

See MANUAL.md

How to Contribute

See CONTRIBUTING.md

Quick Start

Requirements

Available on macOS and any Linux distributions, like Debian, Ubuntu, CentOS, REL, or etc.

Install cliboa

Python version 3.7 or later and poetry are required. In the environment which pip can be used, execute as below.

sudo pip3 install poetry
sudo pip3 install cliboa

Configuration of a Simple ETL Processing

After installed cliboa, 'cliboadmin' can be used as an administrator command.

Create an executable environment of cliboa by using cliboadmin.

$ cd /usr/local
$ sudo cliboadmin init sample
$ cd sample
$ sudo cliboadmin create simple-etl

Directory Tree

Directory tree which was created aforementioned commands is as below.

sample
├── pyproject.toml
├── bin
│   └── clibomanager.py
├── cliboa
│   └── conf
├── common
│   ├── __init__.py
│   ├── environment.py
│   └── scenario
├── conf
│   ├── cliboa.ini
│   └── logging.conf
├── logs
├── project
│   └── simple-etl
│       ├── scenario
│       └── scenario.yml

Install PyPI packages

$ cd sample
$ poetry install

Write a Scenario of ETL Processing

As a simple ETL processing, write scenario.yml in simple-etl as below.

The following example is just download a gzip file from the local sftp server, decompress it, and upload it to the local sftp server.

See Examples

Set an Environment

To make the above scenario available, set a local machine as a sftp server according to respective environments. Also, put "test.csv.gz" under /usr/local.

Execute a Scenario of ETL Processing

After wrote scenario.yml and set the environment, execute a scenario by as below command.

cd sample
poetry run python3 bin/clibomanager.py simple-etl

YAML Configuration

see yaml_configuration.md

Default ETL Modules

see default_etl_modules.md

How to Implement Additional ETL Modules

see additional_etl_modules.md