Stygo

Stygo scrapes, tokenizes, and analyzes Ancient Greek tragedy texts. It normalizes polytonic Greek, removes common noise terms, generates simple lemma families, and counts sentiment-related token groups.

Data visualization: https://clausing-lage-portfolio.de/tragedy

Setup

Use Python 3.11 or newer.

python -m venv .venv
.\.venv\Scripts\Activate.ps1
python -m pip install -r requirements-dev.txt

Analyze Texts

The tokenizer reads greek-nlp/tokenizer/input.json by default and writes to greek-nlp/output.

python greek-nlp/main_tokenizer.py

Custom paths are supported:

python greek-nlp/main_tokenizer.py --input greek-nlp/tokenizer/input.json --output-dir greek-nlp/output

Scrape Perseus

The scraper downloads tragedy texts from Perseus and writes JSON.

python viewer_scraper/scrapeTragedists.py --output viewer_scraper/tragedistsTexts.json

Quality Checks

python -m pytest
python -m ruff check .

Layout

• greek-nlp/main_tokenizer.py: main analysis CLI.
• greek-nlp/tokenizer/: text normalization, noise removal, and lemma generation.
• viewer_scraper/scrapeTragedists.py: Perseus scraper.
• viewer_scraper/Texts/: checked-in source text snapshots.