Prove the rename's stream rebuild on a real Redis: same ids, group, lag and pending entries after RENAME, ENTRIESREAD set after XSETID because Redis caps it; backend CI gets a Redis 8.2 service

This commit is contained in:
2026-11-02 18:51:14 +01:00
parent 08a6376836
commit e6027af8fe
2 changed files with 249 additions and 1 deletions
+10 -1
View File
@@ -25,10 +25,18 @@ jobs:
fail-fast: false
matrix:
python-version: ["3.12", "3.14"]
# The Redis queue and stream tests run against a real server. Same major
# version as the shipped image in docker/compose.yml.
services:
redis:
image: redis:8.2-alpine
env:
# uv would otherwise take a managed interpreter it already has over the
# one set up below.
UV_PYTHON: ${{ matrix.python-version }}
# The job runs in a container next to the service, which answers to its
# name on the default port.
FLUKSIO_TEST_REDIS_HOST: redis
steps:
- name: Checkout
uses: actions/checkout@v4
@@ -46,9 +54,10 @@ jobs:
# The extras too, as `make test-backend` does: the mail, MQTT and Influx
# tests import what only the `server` extra installs.
- run: uv sync --all-extras
# Nothing to start: the suite names its own SQLite file in a temp
# Nothing else to start: the suite names its own SQLite file in a temp
# directory (backend/tests/__init__.py), the mail paths are patched, and
# the flow engine falls back to in-memory state without a Redis host.
# Only the tests gated on FLUKSIO_TEST_REDIS_HOST use the service.
- name: Run tests with coverage
run: uv run bash scripts/test.sh "Coverage for ${{ github.sha }}"
working-directory: backend