Introduction

This is the code and data for the 'Digital Humanities meets Medieval Financial Records: The Receipt Rolls of the Irish Exchequer' project.

Setup the environment

Create a virtual environment and activate it, e.g.

python -m venv .myenv
source .myenv/bin/activate

Install the dependencies:

pip install pip --upgrade
pip install -r requirements.txt

Generate the data

On the first run this can take a while as NLTK downloads various corpora. The script will process the text of a transcript, create various CSV files and, ultimately, an Excel spreadsheet. These files are found in the data directory.

python generate_data.py

Generating plots

This will generate the plots used in the blog posts and the paper published in Irish Economic and Social History.

Project Structure

data

Location of the source transcript (roll_1301.txt) and file that hold lists of stop words for various aspects of data processing. CSV and Excel files are also written to here by the Python scripts albeit the .gitignore omits them from the repository.

receipt_roll

A module that contains all the Python files for processing the data. These files are found in the plots directory.

tests

A number of unit tests with particular emphasis on areas that extract monetary values and entities like people and places.

python -m unittest tests/test_extract_person.py
python -m unittest tests/test_extract_place.py
python -m unittest tests/test_pence_to_psd.py
python -m unittest tests/test_psd_to_pence.py 

Generating data

The generate_data.py is a wrapper that runs all the scripts below.

The create_data_csv.py script parsing the transcript of the roll (roll_1301.txt) and created a CSV file of the individual payments (roll_1301.csv) and a list of daily sums calculated by the Exchequer clerks. It converts pound, shilling, pence and mark values into pennies for easy comparison.

The compare_sums_csv.py script compares the daily sums calculated by the clerks with those calculated programmatically. This is useful for spotting parsing errors by the scripts and, more rarely, issues in the transcript or clerical mistakes.

The extract_entities.py scripts processes roll_1301.csv to extract people, places and keywords and store the data in roll_entities_1301.csv.

The create_excel_report.py creates three sheets in a single Excel file containing the CSV data created by the other scripts.

Project funding

• Seed corn funding from the Jean Golding Institute, University of Bristol (2019–2020)