Natural Language Report Generator for ERPNext

This Frappe app allows users to query ERPNext reports using natural language commands such as:

“Show approved leave of Aruna D” 

Using simple, intuitive phrases, users can generate Leave Applications, Attendance Logs, Expense Claims, and Employee details. Features

Query ERPNext data using plain English. Auto-generates dynamic reports from Leave Application, Attendance, Expense Claim, and Employee doctype. Built-in natural language parser. User-friendly web interface. 

Project Structure

nlp_report/ ├── nlp_report/ │ ├── api.py # Main logic to parse commands & return data │ ├── www/ │ │ ├── natural-language-report.html # Frontend UI │ │ ├── natural_language_report.py # Controller for the UI

Installation

Clone the app into your frappe-bench/apps folder: 

cd ~/frappe-bench/apps git clone https://github.com/ArunaDevraj/nlp_report.git

Install the app: 

cd ~/frappe-bench bench --site your-site-name install-app nlp_report

Restart bench: 

bench restart

Access the Interface

After installation, open the page:

http:///natural-language-report

Enter a natural language command like:

Show all leave applications of Aruna D

Supported Commands

Examples of valid input:

Show approved leave of Aruna D List all expense claims for John Show attendance status Work From Home for Meera Show employee details 

Backend Logic (api.py)

Uses parse_command_with_llm() to convert input into a structured ERPNext query. Supports basic NLP using Python regex. Fetches relevant data using frappe.db.get_all(). 

License

MIT License

Contributors

Aruna D - Initial Developer