Switch the dev/CI toolchain from conda to uv - #78
Merged
Conversation
The mmseqs2 bioconda binary was the only reason arda needed conda. It ships an official static build, which arda already discovers and auto-fetches, so the dev environment no longer needs conda at all. - setup.sh: uv venv .venv + `uv pip install -e . --no-build-isolation` (build deps in the venv so the scikit-build on-import rebuild finds pybind11), then the IgBLAST release and a static mmseqs2 (skipped if one is on PATH). Portable under bash and zsh. Flags reduced to --build-db / --tests. - ci.yml: setup-uv, `uv pip install --system` for the toolchain + editable install. - README / repo memory: conda -> uv; mmseqs is the static binary, no conda. - Drop the root environment.yml (dev conda env). Conda stays ONLY for the Nextflow integration (integrations/nextflow/arda, the Gamaleya/ISP pipeline), which ships its own environment.yml + Dockerfile. Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
This was referenced Jul 18, 2026
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
1 participant
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Why
The only thing tying arda's dev environment to conda was the mmseqs2 bioconda binary. mmseqs2 ships an official static build, which arda already discovers and auto-fetches (
$ARDA_MMSEQS→bin/mmseqs→ PATH → fetch), so conda buys nothing for development.Changes
setup.sh—uv venv .venv+uv pip install -e . --no-build-isolation(build deps live in the venv so the scikit-build editable on-import rebuild can find pybind11), then fetch the IgBLAST release and a static mmseqs2 (skipped when one is already on PATH). Portable under bash and zsh (${BASH_SOURCE[0]:-$0}). Flags trimmed to--build-db/--tests..github/workflows/ci.yml—astral-sh/setup-uv,uv pip install --systemfor the toolchain and the editable install.memory/— conda → uv; mmseqs is the static binary, no conda.environment.yml(dev conda env).Scope — ISP/Gamaleya integration keeps conda
Conda survives only in
integrations/nextflow/arda/(the Gamaleya/ISP Nextflow pipeline), which ships its ownenvironment.yml+ Dockerfile. Untouched.Validation
Full uv install exercised locally:
.venvbuilds the_markupC++ extension,import arda(2.5.6) works, mmseqs resolves off PATH with no conda /ARDA_MMSEQS.bash -n/zsh -nclean;ROOTresolves identically under both shells. Unit + synthetic + realworld suite green in the uv venv.🤖 Generated with Claude Code