Skip to content

ci: check that API references do not cause Docusaurus build failures #15

ci: check that API references do not cause Docusaurus build failures

ci: check that API references do not cause Docusaurus build failures #15

name: Core / Check API reference changes
on:
pull_request:
paths:
- "integrations/**/*.py"
- "integrations/**/config_docusaurus.yml"
jobs:
detect-changes:
runs-on: ubuntu-slim
outputs:
integrations: ${{ steps.changed.outputs.integrations }}
steps:
- uses: actions/checkout@v6
with:
fetch-depth: 0
- name: Set up Python
uses: actions/setup-python@v6
with:
python-version: "3.13"
- name: Detect integrations with API reference changes
id: changed
shell: python
env:
BASE_SHA: ${{ github.event.pull_request.base.sha }}
run: |
import json
import os
import subprocess
import sys
from pathlib import Path
sys.path.insert(0, ".github/utils")
from docstrings_checksum import docstrings_checksum
base_sha = os.environ["BASE_SHA"]
runner_temp = os.environ["RUNNER_TEMP"]
def git(*args):
result = subprocess.run(["git", *args], capture_output=True, text=True)
return result.stdout.strip(), result.returncode
# Get all changed files
diff_output, _ = git(
"diff", "--name-only", f"{base_sha}...HEAD", "--", "integrations"
)
changed_files = set(diff_output.splitlines())
# Extract integration names
candidates = sorted({
Path(p).parts[1]
for p in changed_files
if p.startswith("integrations/") and len(Path(p).parts) > 1
})
# Create base worktree for docstring comparison
base_worktree = os.path.join(runner_temp, "base")
_, return_code = git("worktree", "add", base_worktree, base_sha)
has_worktree = return_code == 0
changed_integrations = []
for integration in candidates:
# If pydoc config changed, always include
if f"integrations/{integration}/pydoc/config_docusaurus.yml" in changed_files:
changed_integrations.append(integration)
continue
# Otherwise, check if docstrings changed
pr_docstrings_checksum = docstrings_checksum(Path(".").glob(f"integrations/{integration}/**/*.py"))
base_docstrings_checksum = ""
if has_worktree:
base_docstrings_checksum = docstrings_checksum(Path(base_worktree).glob(f"integrations/{integration}/**/*.py"))
if base_docstrings_checksum != pr_docstrings_checksum:
changed_integrations.append(integration)
if has_worktree:
git("worktree", "remove", base_worktree)
print(f"Integrations with API reference changes: {json.dumps(changed_integrations)}")
with open(os.environ["GITHUB_OUTPUT"], "a") as f:
f.write(f"integrations={json.dumps(changed_integrations)}\n")
test-api-reference-build:
needs: detect-changes
if: needs.detect-changes.outputs.integrations != '[]'
runs-on: ubuntu-latest
strategy:
fail-fast: false
matrix:
integration: ${{ fromJson(needs.detect-changes.outputs.integrations) }}
steps:
- uses: actions/checkout@v6
- name: Set up Python 3.13
uses: actions/setup-python@v6
with:
python-version: "3.13"
- name: Install Hatch
run: pip install --upgrade hatch
- name: Generate API reference
working-directory: integrations/${{ matrix.integration }}
run: |
hatch run docs
# Save the generated file before the Haystack checkout overwrites the working tree
cp ${{ matrix.integration }}.md "${{ runner.temp }}/${{ matrix.integration }}.md"
# Also copy the Copy script to the temp directory
cp ../../.github/utils/copy_file_to_api_reference.py "${{ runner.temp }}/copy_file_to_api_reference.py"
- name: Checkout Haystack Docs Website
uses: actions/checkout@v6
with:
repository: deepset-ai/haystack
ref: main
sparse-checkout: docs-website
- name: Copy file to API reference
run: python ${{ runner.temp }}/copy_file_to_api_reference.py ${{ runner.temp }}/${{ matrix.integration }}.md
- name: Install Node.js
uses: actions/setup-node@v6
with:
node-version: "22"
- name: Build docs-website
working-directory: docs-website
run: |
npm install
npm run build