main.py and Schematise.py

Table of Contents

1. Introduction
2. "import" statements
3. Functions
4. Checklist of features across Streamlit and LocalWorkflow

1. Introduction

Schematise.py

This is used in the Streamlit version of the app to create a GUI interface, currently hosted on schematise.streamlit.app.

main.py

This is used in the LocalWorkflow to create a CLI interface for the application. The LocalWorkflow, most notably implements a HuggingFacePipeline, which can be used to run the application on models that don't pose restrictions on utilisation for creation of datasets.

2. "import" statements or dependencies

• .env file with IK API Key and Open AI Key if needed

3. Functions and classes

Schematise.py - main()

• creates the app's main page which allows users to:
• pass their IndianKanoon key or upload the CSV file
  • If IndianKanoon is selected, pass the document number
• Allows user to specify which format they want (LegalRuleML or LegalDocML)
• Specify if a certain range of the csv file is to be converted only
• Allows users to see a dataframe for the range selected
• Writes the XML generated in .TXT format for the user
• This allows the user to verify the XML on their own
• Passes all the required values by other pages in the app to the session variables

Schematise.py - set_env(), load_data()

• used to create a cache of the data which is created or stored by these functions
• For environment variables holding the IK API key and OpenAI Key, so that user does not have to re-enter them
• and for the dataframe selected

Schematise.py - get_XML()

• makes the call to combinedProcess.py for generating the XML at the backend

main.py - main()

• Part of the LocalWorkflow
• Creates a workflow that prompts user for each of the conditions they want to be implemented
• Asks user for document number, if IndianKanoon is selected, and thereafter creates a csv file out of the sections in a standard format.
• passes the csv_file created to utils.csv_parser
• XML_arg and num_interval are still in progress, and are intended to contain the parameters for metamodel alignment if required.

4. Checklist of features across formats

(TBD = To Be Developed)

Feature	Streamlit	LocalWorkflow	LegalDocML	LegalRuleML
OpenAI option	Yes	Yes	Yes	Yes
IndianKanoon option	Yes	Yes	Yes	Yes
CSV upload option	Yes	Yes	Yes	Yes
Download XML file	Yes	Yes	Yes	Yes
Dataframe viewing after selection	Yes	No	Yes (in Streamlit)	Yes (in Streamlit)