Development

Development

git clone https://github.com/duceppemo/mashID
cd mashID
conda env create -f environment.yml && conda activate mashID
pip install -e ".[test]"
pre-commit install          # optional: ruff on every commit
ruff check .
pytest

The integration tests build a small synthetic database and are skipped when mash is not on PATH.

Layout

Path	Content
mashid/cli.py, makedb.py, download_cli.py	The three command-line entry points.
mashid/pipeline.py	Sample discovery → stats → screening → annotation → tables.
mashid/mash.py	Wrappers around mash screen, mash sketch, mash info.
mashid/samples.py	Input discovery and sample naming.
mashid/seqstats.py	Read/base counting and the --max-reads streamer.
mashid/taxonomy.py	Organism name parsing from fasta headers.
mashid/metadata.py	The metadata sidecar.
mashid/databases.py	Registry and download of pre-built databases.
docs/wiki/	Source of this wiki, published by .github/workflows/wiki.yml.
recipe/	Bioconda recipe.

Continuous integration

.github/workflows/ci.yml runs on every push and pull request: environment from environment.yml on Python 3.10 and 3.12, ruff check, pytest, and the CLIs' --help.

Releasing

1. Update __version__ in mashid/__init__.py and add a section to CHANGELOG.md.
2. Commit, then tag: git tag -a vX.Y.Z -m "mashID X.Y.Z" && git push origin master vX.Y.Z.
3. .github/workflows/release.yml builds the wheel and creates the GitHub release with the changelog section as notes.
4. Update recipe/meta.yaml: version, sha256 of the new tag's tarball (curl -sL https://github.com/duceppemo/mashID/archive/refs/tags/vX.Y.Z.tar.gz | sha256sum), and number: 0. Copy recipe/ to bioconda-recipes/recipes/mashid and open a pull request; once its checks pass, comment @BiocondaBot please add label.

Editing this wiki

Edit the Markdown files in docs/wiki/ and push to master. The wiki.yml workflow copies them to the wiki repository. Page names are file names (Databases.md → Databases); _Sidebar.md is the navigation.