gradeplotter

This repository now contains two implementations:

• Legacy scripts:
  • grade-analyzer.py
  • grade_analyzer_globals.py
  • grade_analyzer_plotter.py
• New rewrite scaffold (v2):
  • gradeplotter_v2/
  • run_v2_generate.py
  • run_v2_app.py

v2 goals

• Replace monolithic global-state script with modular services.
• Preserve behavior parity for graph generation (histograms and multiyear trends).
• Add run manifests and artifact indexing.
• Provide simple role-based local web app (Admin vs Viewer).

Install dependencies

python3 -m venv .venv
source .venv/bin/activate
python -m pip install --upgrade pip
python -m pip install -r requirements-v2.txt

Generate artifacts (v2)

source .venv/bin/activate
python run_v2_generate.py \
	-i 2020AUT-2021SUM.csv \
	--dest gradeplotter_output \
	--curriculum CSS

Optional flags:

• --instructor '.*'
• --sections '.*'
• --after 2024AUT
• --before 2025AUT
• --no-histograms
• --no-multiyear
• --multiyear-instructor
• --combosections (single histogram for all sections per course)
• --comboinstructor (single histogram for all sections per course per instructor)
• --dfwreport <course> (generate drop/fail/withdraw/repeat report for a course)

Artifacts and manifests are written to:

• gradeplotter_output/runs/<RUN_ID>/...
• gradeplotter_output/runs/<RUN_ID>/manifest.json
• gradeplotter_output/latest-manifest.json

Run local web app (v2)

source .venv/bin/activate
python run_v2_app.py

Open http://127.0.0.1:5000.

Default credentials (override via env vars):

• Admin: admin / admin
• Viewer: viewer / viewer

Supported environment variables:

• GRADEPLOTTER_SECRET_KEY
• GRADEPLOTTER_OUTPUT_ROOT
• GRADEPLOTTER_ADMIN_USER
• GRADEPLOTTER_ADMIN_PASSWORD
• GRADEPLOTTER_VIEWER_USER
• GRADEPLOTTER_VIEWER_PASSWORD

Current rewrite status

Implemented:

• Header-based CSV loading and typed domain models.
• Filtering/query repository for curriculum/instructor/section/term bounds.
• Histogram and multiyear rendering pipeline with manifest output.
• Legacy-style graph output layout for parity: byYear, byInstructor, byCourse, and stats_per_course paths inside each run.
• Legacy-style section filename formatting and --sections matching against compact names.
• Quarter-aware date filtering parity for --after/--before semantics.
• Flask app with login, gallery, run history, and admin generation form.
• Decision/questions tracking in Todo.md.

Not yet implemented:

• Full parity for every legacy report mode (--func, --dfwreport, progression/workload outputs).
• Legacy path-duplication compatibility layer.
• Production-ready auth backend and audit logging.