Skip to content

Latest commit

 

History

History
38 lines (21 loc) · 1.5 KB

README.md

File metadata and controls

38 lines (21 loc) · 1.5 KB

Slack Data Access with Sir Perceval

This project automates the process of adding a Slack bot to all channels, extracting messages, and storing them in a PostgreSQL database using Perceval and custom Python scripts.

Prerequisites

  1. Python: Make sure Python is installed on your machine.
  2. psycopg2: Install the psycopg2 library using pip to interact with PostgreSQL.
  3. Sir Perceval: Install Perceval using pip to collect data from Slack.
  4. Slack Bot: Create a bot for the specific Slack workspace and obtain the API token.

Files Included

  • PercevalDatabaseQueries.sql: SQL scripts to create the necessary database tables.
  • AddBotToChannels.py: Script to add the Slack bot to all channels.
  • PercevalDatabaseImplementation.py: Implements the data extraction from Slack and inserts data into the database.
  • PercevalTest.py: Test script to verify the data extraction.

Setup and Execution

Step 1: Database Setup

  1. Run the SQL code from PercevalDatabaseQueries.sql to create all necessary tables in your PostgreSQL database.

Step 2: Adding SlackBot to Channels

  1. Execute AddBotToChannels.py to automatically add your Slack bot to all channels and collect channel IDs.

Step 3: Data Insertion

  1. Run PercevalDatabaseImplementation.py to collect messages and related relevant data.Further, store it in the tables created earlier in your database.

Configuration

Ensure to replace the place-holders in all the scripts with actual database credentials, API token.