Set up a development environment¶

Application repository¶

git clone https://github.com/rensutheart/napari-vipp.git
cd napari-vipp
python -m venv .venv
.\.venv\Scripts\Activate.ps1
python -m pip install --upgrade pip
python -m pip install -e ".[dev]"

On macOS/Linux activate with source .venv/bin/activate.

Run the baseline checks before editing:

python -m npe2 validate src/napari_vipp/napari.yaml
python -m ruff check .
python -m pytest

Launch a development widget with python scripts/launch_vipp_sample.py or list review workflows with:

python scripts/launch_vipp_intensity_workflow.py --list

The editable .[dev] environment is the portable CPU development baseline. On an NVIDIA qualification host, create a separate CPython 3.12 environment for exactly one CUDA track; do not add CUDA 12 and CUDA 13 extras to the same environment. Run vipp-compute-doctor --track cuda13 before any opt-in real_cuda tests. A working import or kernel is still not operation admission; the executable environment and workload policy must also match.

Manual repository¶

git clone https://github.com/rensutheart/vipp-mkdocs.git
cd vipp-mkdocs
python -m venv .venv
.\.venv\Scripts\Activate.ps1
python -m pip install -r requirements.txt
mkdocs serve

Before submitting a manual change run:

mkdocs build --strict

Use the manual's exact dependency pins so local warning behavior matches CI.

Useful application maps¶

The application repository's maintained architecture and contributor notes are the source of truth for implementation boundaries:

Read only the relevant plan/history material; do not copy speculative roadmap claims into the public manual as current capability.