RCAccelerator Tools

This repository contains useful tools that can be useful when deploying the RCAccelerator/chatbot. As of now, this repository contains:

• data_scraper: a tool that scrapes jira issues and more to store the collected data in a vector database.
• feedback_exporter: a tool to fetch user feedback from Chainlit DB and write it in to a Google Spreadsheet.
• openstack_retriever: set of scripts to pull openstack documentation (requires Python 3.11)
  • openstack_retriever/get_openstack_plaintext_docs.sh script for pulling upstream Openstack documentation.
  • openstack_retriever/get_rhoso_plaintext_docs.sh script for pulling downstream Openstack documentation.
  • openstack_retriever/rhoso_adoc_docs_to_text.py script for conversion of asciidoc files to plaintext.
• osp_doc_scraper: a tool for turning upstream OSP documentation into vector database. Requires documents retrieved by openstack_retriever tools.

Getting Started

1. Run a tool:

   pip install .
   data_scraper --help

Feedback Exporter Tool

This tool fetches user feedback from the Chainlit PostgreSQL database and writes it into a Google Spreadsheet.

Usage

1. Set the following environment variables:

export DATABASE_URL=postgresql://user:pass@host:port/dbname
export APP_BASE_URL=https://chainlit.example.com/thread/
export GOOGLE_SPREADSHEET_ID=your_google_sheet_id
export GOOGLE_CREDENTIALS_JSON='{"type": "service_account", ...}'  # raw JSON string

2. Run the tool:

python feedback_exporter/export_feedback.py

This will populate the Google Spreadsheet with columns: score, thread URL, input, output, comment, and user name.