mirror of
https://github.com/hanzoai/hanzo.vim.git
synced 2026-11-09 15:19:11 +00:00
github.com has no runner for hanzo-build-linux-amd64, so a caller under .github/workflows is a gate that can never be scheduled. This is the ~7-line caller on the plane that can (git.hanzo.ai git-runner fleet), pinned @v1. Co-authored-by: Hanzo Dev <[email protected]>
29 lines
1.4 KiB
YAML
29 lines
1.4 KiB
YAML
# Canonical CI/CD config for hanzoai/hanzo.vim — the one file both the
|
|
# hanzoai/ci reusable (.hanzo/workflows/cicd.yml) and platform.hanzo.ai read.
|
|
#
|
|
# An editor plugin: a Vim/Neovim tree plus the `neural` Python package that
|
|
# backs it. Nothing is deployed and the Dockerfile here exists only to give
|
|
# ./run-tests a pinned Vim/Neovim matrix to run the Vader suite inside — so no
|
|
# `images:` and no `deploy:`, only a gate.
|
|
#
|
|
# The gate is the Python half, which needs no Vim and no Docker: ruff over the
|
|
# rule set this repo actually selected (pycodestyle, isort, bugbear, builtins,
|
|
# commas, async, boolean-trap, pyflakes, ruff, pyupgrade, mccabe) and pytest
|
|
# over test/python, which pyproject already declares as testpaths. uv builds the
|
|
# dev group from uv.lock, so the gate runs the pinned toolchain and not whatever
|
|
# floats.
|
|
#
|
|
# Not claimed: the Vader suites (test/vim) and the linter leg. Those run in the
|
|
# denseanalysis/neural container that ./run-tests builds, across vim-8.0,
|
|
# vim-9.0 and neovim-0.8 — three interpreters this fleet does not have. They
|
|
# come back when the runner can run that image, not by pretending here.
|
|
test:
|
|
- name: build
|
|
run: |
|
|
set -e
|
|
command -v uv >/dev/null || curl -LsSf https://astral.sh/uv/install.sh | sh
|
|
export PATH="$HOME/.local/bin:$PATH"
|
|
uv sync --group dev
|
|
uv run ruff check
|
|
uv run pytest -q
|