[FEAT] Include file paths as column from read_parquet/csv/json #9143

Workflow file for this run

.github/workflows/python-package.yml at 945cfc7

	# This workflow will install Python dependencies, run tests and lint with a variety of Python versions
	# For more information see: https://help.github.com/actions/language-and-framework-guides/using-python-with-github-actions

	name: daft

	on:
	push:
	branches: [main]
	pull_request:
	branches: [main]
	# `workflow_dispatch` allows CodSpeed to trigger backtest
	# performance analysis in order to generate initial data.
	workflow_dispatch:

	env:
	DAFT_ANALYTICS_ENABLED: '0'
	UV_SYSTEM_PYTHON: 1

	jobs:

	unit-tests-with-coverage:
	runs-on: ${{ matrix.os }}
	timeout-minutes: 30
	strategy:
	fail-fast: false
	matrix:
	python-version: ['3.8', '3.10']
	daft-runner: [py, ray]
	pyarrow-version: [7.0.0, 16.0.0]
	enable-native-executor: [0, 1]
	os: [ubuntu-20.04, windows-latest]
	exclude:
	- daft-runner: ray
	enable-native-executor: 1
	- daft-runner: ray
	pyarrow-version: 7.0.0
	os: ubuntu-20.04
	- daft-runner: py
	python-version: '3.10'
	pyarrow-version: 7.0.0
	os: ubuntu-20.04
	- python-version: '3.8'
	pyarrow-version: 16.0.0
	- os: windows-latest
	python-version: '3.8'
	- os: windows-latest
	pyarrow-version: 7.0.0
	steps:
	- uses: actions/checkout@v4
	- uses: moonrepo/setup-rust@v1
	with:
	cache: false
	- uses: Swatinem/rust-cache@v2
	with:
	key: ${{ runner.os }}-build
	cache-all-crates: 'true'
	- name: Install cargo-llvm-cov
	uses: taiki-e/install-action@cargo-llvm-cov

	- name: install llvm tools
	run: rustup component add llvm-tools-preview

	- name: Set up Python ${{ matrix.python-version }}
	uses: actions/setup-python@v5
	with:
	python-version: ${{ matrix.python-version }}
	cache: pip
	cache-dependency-path: \|
	pyproject.toml
	requirements-dev.txt

	- name: Setup Virtual Env
	run: \|
	python -m venv venv
	echo "$GITHUB_WORKSPACE/venv/bin" >> $GITHUB_PATH
	pip install uv

	- name: Install dependencies
	if: ${{ (runner.os != 'Windows') }}
	run: \|
	activate
	uv pip install -r requirements-dev.txt

	- name: Install dependencies (Windows)
	if: ${{ (runner.os == 'Windows') }}
	run: \|
	.\venv\Scripts\activate
	uv pip install -r requirements-dev.txt

	- name: Override pyarrow
	if: ${{ (matrix.pyarrow-version) && (runner.os != 'Windows') }}
	run: uv pip install pyarrow==${{ matrix.pyarrow-version }}

	- name: Override deltalake for pyarrow
	if: ${{ (matrix.pyarrow-version == '7.0.0') }}
	run: uv pip install deltalake==0.10.0

	- name: Build library and Test with pytest (unix)
	if: ${{ (runner.os != 'Windows') }}
	run: \|
	source activate
	export CARGO_TARGET_DIR=$CARGO_LLVM_COV_TARGET_DIR
	cargo llvm-cov clean --workspace
	maturin develop
	pytest --ignore tests/integration --collect-only -qq # run this to ensure no weird imports that result in `Collector` errors
	pytest --cov=daft --ignore tests/integration --durations=50
	coverage combine -a --data-file='.coverage' \|\| true
	mkdir -p report-output
	coverage xml -o ./report-output/coverage-${{ join(matrix.*, '-') }}.xml
	cargo llvm-cov report --lcov --output-path report-output/rust-coverage-${{ join(matrix.*, '-') }}.lcov
	env:
	# output of `cargo llvm-cov show-env --export-prefix`
	RUSTFLAGS: -C instrument-coverage --cfg=coverage --cfg=coverage_nightly --cfg=trybuild_no_target
	LLVM_PROFILE_FILE: ./target/daft-coverage-%p-%m.profraw
	CARGO_LLVM_COV: 1
	CARGO_LLVM_COV_SHOW_ENV: 1
	CARGO_LLVM_COV_TARGET_DIR: ./target
	CARGO_TARGET_DIR: ./target

	DAFT_RUNNER: ${{ matrix.daft-runner }}
	DAFT_ENABLE_NATIVE_EXECUTOR: ${{ matrix.enable-native-executor }}

	- name: Build library and Test with pytest (Windows)
	if: ${{ (runner.os == 'Windows') }}
	run: \|
	.\venv\Scripts\activate
	# source <(cargo llvm-cov show-env --export-prefix)
	# export CARGO_TARGET_DIR=$CARGO_LLVM_COV_TARGET_DIR
	# export CARGO_INCREMENTAL=1
	# cargo llvm-cov clean --workspace
	maturin develop
	mkdir -p report-output && pytest --cov=daft --ignore tests\integration --durations=50
	coverage combine -a --data-file='.coverage' \|\| true
	coverage xml -o .\report-output\coverage-${{ join(matrix.*, '-') }}.xml
	# cargo llvm-cov --no-run --lcov --output-path report-output\rust-coverage-${{ join(matrix.*, '-') }}.lcov
	env:
	DAFT_RUNNER: ${{ matrix.daft-runner }}

	- name: Upload coverage report
	uses: actions/upload-artifact@v4
	with:
	name: coverage-reports-unit-tests-${{ join(matrix.*, '-') }}
	path: ./report-output

	- name: Send Slack notification on failure
	uses: slackapi/[email protected]
	if: ${{ failure() && (github.ref == 'refs/heads/main') }}
	with:
	payload: \|
	{
	"blocks": [
	{
	"type": "section",
	"text": {
	"type": "mrkdwn",
	"text": ":rotating_light: [CI] Pytest Unit Tests <${{ github.server_url }}/${{ github.repository }}/actions/runs/${{ github.run_id }}\|workflow> FAILED on main :rotating_light:"
	}
	}
	]
	}
	env:
	SLACK_WEBHOOK_URL: ${{ secrets.SLACK_WEBHOOK_URL }}
	SLACK_WEBHOOK_TYPE: INCOMING_WEBHOOK

	unit-tests:
	runs-on: ubuntu-latest
	needs: unit-tests-with-coverage
	if: always()
	steps:
	- name: All tests ok
	if: ${{ !(contains(needs.*.result, 'failure')) }}
	run: exit 0
	- name: Some tests failed
	if: ${{ contains(needs.*.result, 'failure') }}
	run: exit 1

	integration-test-build:
	runs-on: ubuntu-latest
	timeout-minutes: 30
	env:
	package-name: getdaft
	strategy:
	matrix:
	python-version: ['3.8']
	steps:
	- uses: actions/checkout@v4
	with:
	submodules: true
	fetch-depth: 0
	- uses: actions/setup-python@v5
	with:
	python-version: ${{ matrix.python-version }}
	architecture: x64
	- run: pip install -U twine toml maturin
	- uses: moonrepo/setup-rust@v1
	with:
	cache: false
	- uses: Swatinem/rust-cache@v2
	with:
	key: ${{ runner.os }}-integration-build
	cache-all-crates: 'true'

	# NOTE: we don't build with all the actual release optimizations to avoid hellish CI times
	- name: Build wheels
	run: maturin build --release --compatibility linux --out dist
	- name: Upload wheels
	uses: actions/upload-artifact@v4
	with:
	name: wheels-${{ matrix.python-version }}
	path: dist

	integration-test-tpch:
	runs-on: ubuntu-latest
	timeout-minutes: 30
	needs:
	- integration-test-build
	env:
	package-name: getdaft
	strategy:
	fail-fast: false
	matrix:
	python-version: ['3.8']
	daft-runner: [py, ray]
	steps:
	- uses: actions/checkout@v4
	with:
	submodules: true
	fetch-depth: 0
	- uses: actions/setup-python@v5
	with:
	python-version: ${{ matrix.python-version }}
	architecture: x64
	cache: pip
	cache-dependency-path: \|
	pyproject.toml
	requirements-dev.txt

	- name: Download built wheels
	uses: actions/download-artifact@v4
	with:
	pattern: wheels-*
	merge-multiple: true
	path: dist

	- name: Setup Virtual Env
	run: \|
	python -m venv venv
	echo "$GITHUB_WORKSPACE/venv/bin" >> $GITHUB_PATH
	pip install uv

	- name: Install Daft and dev dependencies
	run: \|
	uv pip install -r requirements-dev.txt dist/${{ env.package-name }}-x86_64.whl --force-reinstall
	rm -rf daft
	- uses: actions/cache@v4
	env:
	cache-name: cache-tpch-data
	with:
	path: data/tpch-dbgen
	key: ${{ runner.os }}-build-${{ env.cache-name }}-${{ hashFiles('tests/integration/test_tpch.py', 'benchmarking/tpch/**') }}
	- name: Run TPCH integration tests
	run: \|
	pytest tests/integration/test_tpch.py --durations=50
	env:
	DAFT_RUNNER: ${{ matrix.daft-runner }}

	- name: Send Slack notification on failure
	uses: slackapi/[email protected]
	if: ${{ failure() && (github.ref == 'refs/heads/main') }}
	with:
	payload: \|
	{
	"blocks": [
	{
	"type": "section",
	"text": {
	"type": "mrkdwn",
	"text": ":rotating_light: [CI] TPCH Integration Tests <${{ github.server_url }}/${{ github.repository }}/actions/runs/${{ github.run_id }}\|workflow> FAILED on main :rotating_light:"
	}
	}
	]
	}
	env:
	SLACK_WEBHOOK_URL: ${{ secrets.SLACK_WEBHOOK_URL }}
	SLACK_WEBHOOK_TYPE: INCOMING_WEBHOOK

	integration-test-io:
	runs-on: ubuntu-latest
	timeout-minutes: 30
	needs:
	- integration-test-build
	env:
	package-name: getdaft
	strategy:
	fail-fast: false
	matrix:
	python-version: ['3.8'] # can't use 3.7 due to requiring anon mode for adlfs
	daft-runner: [py, ray]
	steps:
	- uses: actions/checkout@v4
	with:
	submodules: true
	fetch-depth: 0
	- uses: actions/setup-python@v5
	with:
	python-version: ${{ matrix.python-version }}
	architecture: x64
	cache: pip
	cache-dependency-path: \|
	pyproject.toml
	requirements-dev.txt

	- name: Download built wheels
	uses: actions/download-artifact@v4
	with:
	pattern: wheels-*
	merge-multiple: true
	path: dist
	- name: Setup Virtual Env
	run: \|
	python -m venv venv
	echo "$GITHUB_WORKSPACE/venv/bin" >> $GITHUB_PATH
	pip install uv
	- name: Install Daft and dev dependencies
	run: \|
	uv pip install -r requirements-dev.txt dist/${{ env.package-name }}-x86_64.whl --force-reinstall
	rm -rf daft
	- name: Prepare tmpdirs for IO services
	run: \|
	mkdir -p /tmp/daft-integration-testing/nginx
	chmod +rw /tmp/daft-integration-testing/nginx
	- name: Spin up IO services
	uses: isbang/[email protected]
	with:
	compose-file: ./tests/integration/io/docker-compose/docker-compose.yml
	down-flags: --volumes
	- name: Run IO integration tests
	run: \|
	pytest tests/integration/io -m 'integration and not benchmark' --durations=50
	pytest tests/io -m 'integration' --durations=50
	env:
	DAFT_RUNNER: ${{ matrix.daft-runner }}
	- name: Send Slack notification on failure
	uses: slackapi/[email protected]
	if: ${{ failure() && (github.ref == 'refs/heads/main') }}
	with:
	payload: \|
	{
	"blocks": [
	{
	"type": "section",
	"text": {
	"type": "mrkdwn",
	"text": ":rotating_light: [CI] IO Integration Tests <${{ github.server_url }}/${{ github.repository }}/actions/runs/${{ github.run_id }}\|workflow> FAILED on main :rotating_light:"
	}
	}
	]
	}
	env:
	SLACK_WEBHOOK_URL: ${{ secrets.SLACK_WEBHOOK_URL }}
	SLACK_WEBHOOK_TYPE: INCOMING_WEBHOOK

	# Same as integration-test-io but runs the tests that require credentials, only on `main`
	integration-test-io-credentialed:
	if: ${{ github.ref == 'refs/heads/main' }}
	runs-on: ubuntu-latest
	timeout-minutes: 30
	needs:
	- integration-test-build
	env:
	package-name: getdaft
	strategy:
	fail-fast: false
	matrix:
	python-version: ['3.8'] # can't use 3.7 due to requiring anon mode for adlfs
	daft-runner: [py, ray]
	# These permissions are needed to interact with GitHub's OIDC Token endpoint.
	# This is used in the step "Assume GitHub Actions AWS Credentials"
	permissions:
	id-token: write
	contents: read
	steps:
	- uses: actions/checkout@v4
	with:
	submodules: true
	fetch-depth: 0
	- uses: actions/setup-python@v5
	with:
	python-version: ${{ matrix.python-version }}
	architecture: x64
	cache: pip
	cache-dependency-path: \|
	pyproject.toml
	requirements-dev.txt

	- name: Download built wheels
	uses: actions/download-artifact@v4
	with:
	pattern: wheels-*
	merge-multiple: true
	path: dist
	- name: Setup Virtual Env
	run: \|
	python -m venv venv
	echo "$GITHUB_WORKSPACE/venv/bin" >> $GITHUB_PATH
	pip install uv
	- name: Install Daft and dev dependencies
	run: \|
	uv pip install -r requirements-dev.txt dist/${{ env.package-name }}-x86_64.whl --force-reinstall
	rm -rf daft
	- name: Prepare tmpdirs for IO services
	run: \|
	mkdir -p /tmp/daft-integration-testing/nginx
	chmod +rw /tmp/daft-integration-testing/nginx
	- name: Assume GitHub Actions AWS Credentials
	uses: aws-actions/configure-aws-credentials@v4
	with:
	aws-region: us-west-2
	role-to-assume: ${{ secrets.ACTIONS_AWS_ROLE_ARN }}
	role-session-name: DaftPythonPackageGitHubWorkflow
	- name: Assume GitHub Actions GCloud Credentials
	uses: google-github-actions/auth@v2
	with:
	credentials_json: ${{ secrets.ACTIONS_GCP_SERVICE_ACCOUNT_JSON }}
	# NOTE: Workload Identity seems to be having problems with our Rust crate, so we use JSON instead
	# See issue: https://github.com/yoshidan/google-cloud-rust/issues/171#issuecomment-1730511655
	# workload_identity_provider: ${{ secrets.ACTIONS_GCP_WORKLOAD_IDENTITY_PROVIDER }}
	# service_account: ${{ secrets.ACTIONS_GCP_SERVICE_ACCOUNT }}
	- name: Spin up IO services
	uses: isbang/[email protected]
	with:
	compose-file: ./tests/integration/io/docker-compose/docker-compose.yml
	down-flags: --volumes
	- name: Run IO integration tests
	run: \|
	pytest tests/integration/io -m 'integration and not benchmark' --credentials --durations=50
	env:
	DAFT_RUNNER: ${{ matrix.daft-runner }}
	- name: Send Slack notification on failure
	uses: slackapi/[email protected]
	if: ${{ failure() }}
	with:
	payload: \|
	{
	"blocks": [
	{
	"type": "section",
	"text": {
	"type": "mrkdwn",
	"text": ":rotating_light: [CI] IO Integration Tests <${{ github.server_url }}/${{ github.repository }}/actions/runs/${{ github.run_id }}\|workflow> FAILED on main :rotating_light:"
	}
	}
	]
	}
	env:
	SLACK_WEBHOOK_URL: ${{ secrets.SLACK_WEBHOOK_URL }}
	SLACK_WEBHOOK_TYPE: INCOMING_WEBHOOK

	integration-test-iceberg:
	runs-on: ubuntu-latest
	timeout-minutes: 30
	needs:
	- integration-test-build
	env:
	package-name: getdaft
	strategy:
	fail-fast: false
	matrix:
	python-version: ['3.8'] # can't use 3.7 due to requiring anon mode for adlfs
	daft-runner: [py, ray]
	steps:
	- uses: actions/checkout@v4
	with:
	submodules: true
	fetch-depth: 0
	- uses: actions/setup-python@v5
	with:
	python-version: ${{ matrix.python-version }}
	architecture: x64
	cache: pip
	cache-dependency-path: \|
	pyproject.toml
	requirements-dev.txt
	- name: Download built wheels
	uses: actions/download-artifact@v4
	with:
	pattern: wheels-*
	merge-multiple: true
	path: dist
	- name: Setup Virtual Env
	run: \|
	python -m venv venv
	echo "$GITHUB_WORKSPACE/venv/bin" >> $GITHUB_PATH
	pip install uv
	- name: Install Daft and dev dependencies
	run: \|
	uv pip install -r requirements-dev.txt dist/${{ env.package-name }}-x86_64.whl --force-reinstall
	rm -rf daft
	- uses: docker/setup-buildx-action@v3
	- uses: docker/build-push-action@v5
	with:
	context: ./tests/integration/iceberg/docker-compose/
	file: ./tests/integration/iceberg/docker-compose/Dockerfile
	cache-from: type=gha
	cache-to: type=gha,mode=min
	- name: Spin up services
	run: \|
	pushd ./tests/integration/iceberg/docker-compose/
	docker compose -f ./docker-compose.yml up -d
	popd
	- name: Run Iceberg integration tests
	run: \|
	pytest tests/integration/iceberg -m 'integration' --durations=50
	env:
	DAFT_RUNNER: ${{ matrix.daft-runner }}
	- name: Send Slack notification on failure
	uses: slackapi/[email protected]
	if: ${{ failure() && (github.ref == 'refs/heads/main') }}
	with:
	payload: \|
	{
	"blocks": [
	{
	"type": "section",
	"text": {
	"type": "mrkdwn",
	"text": ":rotating_light: [CI] Iceberg Integration Tests <${{ github.server_url }}/${{ github.repository }}/actions/runs/${{ github.run_id }}\|workflow> FAILED on main :rotating_light:"
	}
	}
	]
	}
	env:
	SLACK_WEBHOOK_URL: ${{ secrets.SLACK_WEBHOOK_URL }}
	SLACK_WEBHOOK_TYPE: INCOMING_WEBHOOK

	integration-test-sql:
	runs-on: ubuntu-latest
	timeout-minutes: 30
	needs:
	- integration-test-build
	env:
	package-name: getdaft
	strategy:
	fail-fast: false
	matrix:
	python-version: ['3.8'] # can't use 3.7 due to requiring anon mode for adlfs
	daft-runner: [py, ray]
	steps:
	- uses: actions/checkout@v4
	with:
	submodules: true
	fetch-depth: 0
	- uses: actions/setup-python@v5
	with:
	python-version: ${{ matrix.python-version }}
	architecture: x64
	cache: pip
	cache-dependency-path: \|
	pyproject.toml
	requirements-dev.txt
	- name: Download built wheels
	uses: actions/download-artifact@v4
	with:
	pattern: wheels-*
	merge-multiple: true
	path: dist
	- name: Setup Virtual Env
	run: \|
	python -m venv venv
	echo "$GITHUB_WORKSPACE/venv/bin" >> $GITHUB_PATH
	pip install uv
	- name: Install Daft and dev dependencies
	run: \|
	uv pip install -r requirements-dev.txt dist/${{ env.package-name }}-x86_64.whl --force-reinstall
	rm -rf daft
	- name: Spin up services
	run: \|
	pushd ./tests/integration/sql/docker-compose/
	docker compose -f ./docker-compose.yml up -d
	popd
	- name: Run sql integration tests
	run: \|
	pytest tests/integration/sql -m 'integration or not integration' --durations=50
	env:
	DAFT_RUNNER: ${{ matrix.daft-runner }}
	- name: Send Slack notification on failure
	uses: slackapi/[email protected]
	if: ${{ failure() && (github.ref == 'refs/heads/main') }}
	with:
	payload: \|
	{
	"blocks": [
	{
	"type": "section",
	"text": {
	"type": "mrkdwn",
	"text": ":rotating_light: [CI] SQL Integration Tests <${{ github.server_url }}/${{ github.repository }}/actions/runs/${{ github.run_id }}\|workflow> FAILED on main :rotating_light:"
	}
	}
	]
	}
	env:
	SLACK_WEBHOOK_URL: ${{ secrets.SLACK_WEBHOOK_URL }}
	SLACK_WEBHOOK_TYPE: INCOMING_WEBHOOK


	integration-tests:
	runs-on: ubuntu-latest
	needs:
	- integration-test-tpch
	- integration-test-io
	- integration-test-io-credentialed
	- integration-test-iceberg
	- integration-test-sql
	if: always()
	steps:
	- name: All tests ok
	if: ${{ !(contains(needs.*.result, 'failure')) }}
	run: exit 0
	- name: Some tests failed
	if: ${{ contains(needs.*.result, 'failure') }}
	run: exit 1

	benchmark-codspeed:
	runs-on: buildjet-8vcpu-ubuntu-2204
	timeout-minutes: 30
	env:
	package-name: getdaft
	RUSTFLAGS: -C target-feature=+fxsr,+sse,+sse2,+sse3,+ssse3,+sse4.1,+sse4.2
	strategy:
	fail-fast: false
	steps:
	- uses: actions/checkout@v4
	- uses: moonrepo/setup-rust@v1
	with:
	cache: false
	- uses: Swatinem/rust-cache@v2
	with:
	prefix-key: dev-bench
	key: ${{ runner.os }}-benchmark-build
	cache-all-crates: 'true'
	cache-provider: buildjet
	- uses: actions/setup-python@v5
	with:
	python-version: '3.12' # Codspeed requires Python 3.12
	- name: Setup Virtual Env
	run: \|
	python -m venv venv
	echo "$GITHUB_WORKSPACE/venv/bin" >> $GITHUB_PATH
	pip install uv
	- name: Install dependencies
	run: \|
	uv pip install -r requirements-dev.txt
	- name: Build Rust Library
	run: \|
	source activate
	maturin develop --profile dev-bench
	- uses: actions/cache@v4
	env:
	cache-name: cache-tpch-data
	with:
	path: data/tpch-dbgen
	key: ${{ runner.os }}-build-${{ env.cache-name }}-${{ hashFiles('tests/integration/test_tpch.py', 'benchmarking/tpch/**') }}
	- name: Run benchmarks
	uses: CodSpeedHQ/action@v3
	with:
	run: pytest tests/benchmarks -m benchmark --codspeed
	- name: Send Slack notification on failure
	uses: slackapi/[email protected]
	if: ${{ failure() && (github.ref == 'refs/heads/main') }}
	with:
	payload: \|
	{
	"blocks": [
	{
	"type": "section",
	"text": {
	"type": "mrkdwn",
	"text": ":rotating_light: [CI] Benchmark Local TPCH <${{ github.server_url }}/${{ github.repository }}/actions/runs/${{ github.run_id }}\|workflow> FAILED on main :rotating_light:"
	}
	}
	]
	}
	env:
	SLACK_WEBHOOK_URL: ${{ secrets.SLACK_WEBHOOK_URL }}
	SLACK_WEBHOOK_TYPE: INCOMING_WEBHOOK

	rust-tests-platform:
	runs-on: ${{ matrix.os }}-latest
	timeout-minutes: 30
	strategy:
	fail-fast: false
	matrix:
	os: [ubuntu, Windows]
	steps:
	- uses: actions/checkout@v4
	- uses: moonrepo/setup-rust@v1
	with:
	cache: false
	- uses: Swatinem/rust-cache@v2
	with:
	key: ${{ runner.os }}-rust-build
	cache-all-crates: 'true'

	- name: Install cargo-llvm-cov
	uses: taiki-e/install-action@cargo-llvm-cov
	- name: Generate code coverage
	run: mkdir -p report-output && cargo llvm-cov --no-default-features --workspace --lcov --output-path ./report-output/lcov.info
	- name: Upload coverage report
	uses: actions/upload-artifact@v4
	with:
	name: coverage-reports-rust-tests-${{ runner.os }}
	path: ./report-output
	- name: Send Slack notification on failure
	uses: slackapi/[email protected]
	if: ${{ failure() && (github.ref == 'refs/heads/main') }}
	with:
	payload: \|
	{
	"blocks": [
	{
	"type": "section",
	"text": {
	"type": "mrkdwn",
	"text": ":rotating_light: [CI] Rust Unit Tests <${{ github.server_url }}/${{ github.repository }}/actions/runs/${{ github.run_id }}\|workflow> FAILED on main :rotating_light:"
	}
	}
	]
	}
	env:
	SLACK_WEBHOOK_URL: ${{ secrets.SLACK_WEBHOOK_URL }}
	SLACK_WEBHOOK_TYPE: INCOMING_WEBHOOK

	rust-tests:
	runs-on: ubuntu-latest
	needs:
	- rust-tests-platform
	if: always()
	steps:
	- name: All tests ok
	if: ${{ !(contains(needs.*.result, 'failure')) }}
	run: exit 0
	- name: Some tests failed
	if: ${{ contains(needs.*.result, 'failure') }}
	run: exit 1

	check-unused-cargo-deps:
	runs-on: ubuntu-latest
	steps:
	- uses: actions/checkout@v4
	- uses: moonrepo/setup-rust@v1
	with:
	cache: false
	channel: stable

	- name: Install Machete
	run: cargo +stable install [email protected] --locked
	- name: Run Machete
	run: cargo machete --with-metadata

	doctests:
	runs-on: ubuntu-latest
	continue-on-error: true
	env:
	python-version: '3.10'
	DAFT_BOLD_TABLE_HEADERS: '0'
	steps:
	- uses: actions/checkout@v4
	- uses: moonrepo/setup-rust@v1
	with:
	cache: false
	- uses: Swatinem/rust-cache@v2
	with:
	key: ${{ runner.os }}-build
	cache-all-crates: 'true'
	- name: Set up Python ${{ env.python-version }}
	uses: actions/setup-python@v5
	with:
	python-version: ${{ env.python-version }}
	cache: pip
	cache-dependency-path: \|
	pyproject.toml
	requirements-dev.txt
	- name: Setup Virtual Env
	run: \|
	python -m venv venv
	echo "$GITHUB_WORKSPACE/venv/bin" >> $GITHUB_PATH
	pip install uv
	- name: Install dependencies
	run: \|
	uv pip install -r requirements-dev.txt
	- name: Run doctests
	run: \|
	source activate
	maturin develop
	pytest --doctest-modules --continue-on-collection-errors daft/dataframe/dataframe.py daft/expressions/expressions.py daft/convert.py daft/udf.py

	publish-coverage-reports:
	name: Publish coverage reports to CodeCov
	runs-on: ubuntu-latest
	needs:
	- unit-tests-with-coverage
	- rust-tests-platform
	steps:
	- uses: actions/checkout@v4
	- uses: actions/download-artifact@v4
	with:
	pattern: coverage-reports*
	merge-multiple: true
	path: ./report-output
	- name: Upload coverage reports to Codecov with GitHub Action
	uses: codecov/codecov-action@v3
	env:
	CODECOV_TOKEN: ${{ secrets.CODECOV_TOKEN }}
	files: ./report-output/*
	- name: Send Slack notification on failure
	uses: slackapi/[email protected]
	if: ${{ failure() && (github.ref == 'refs/heads/main') }}
	with:
	payload: \|
	{
	"blocks": [
	{
	"type": "section",
	"text": {
	"type": "mrkdwn",
	"text": ":rotating_light: [CI] Codecov Uploads <${{ github.server_url }}/${{ github.repository }}/actions/runs/${{ github.run_id }}\|workflow> FAILED on main :rotating_light:"
	}
	}
	]
	}
	env:
	SLACK_WEBHOOK_URL: ${{ secrets.SLACK_WEBHOOK_URL }}
	SLACK_WEBHOOK_TYPE: INCOMING_WEBHOOK

	test-imports-platform:
	runs-on: ${{ matrix.os }}-latest
	timeout-minutes: 30
	strategy:
	fail-fast: false
	matrix:
	os: [ubuntu, Windows]
	python-version: ['3.8']
	steps:
	- uses: actions/checkout@v4
	- uses: moonrepo/setup-rust@v1
	with:
	cache: false
	- uses: Swatinem/rust-cache@v2
	with:
	key: ${{ runner.os }}-build
	cache-all-crates: 'true'

	- name: Set up Python ${{ matrix.python-version }}
	uses: actions/setup-python@v5
	with:
	python-version: ${{ matrix.python-version }}

	- name: Unix Build
	if: ${{ (runner.os != 'Windows') }}
	run: \|
	python -m venv venv
	source venv/bin/activate
	python -m pip install maturin
	maturin build --out dist

	- name: Windows Build
	if: ${{ (runner.os == 'Windows') }}
	run: \|
	python -m venv venv
	.\venv\Scripts\activate
	python -m pip install maturin
	maturin build --out dist

	- name: Test Imports in Clean Env (Unix)
	if: ${{ (runner.os != 'Windows') }}
	run: \|
	rm -rf daft
	rm -rf venv
	python -m venv venv
	source venv/bin/activate
	ls -R ./dist
	pip install dist/*.whl
	python -c 'import daft; from daft import *'

	- name: Test Imports in Clean Env (Windows)
	if: ${{ (runner.os == 'Windows') }}
	run: \|
	rd -r daft
	rd -r venv
	python -m venv venv
	.\venv\Scripts\activate
	$FILES = Get-ChildItem -Path .\dist\*.whl -Force -Recurse
	python -m pip install $FILES[0].FullName
	python -c 'import daft; from daft import *'

	- name: Send Slack notification on failure
	uses: slackapi/[email protected]
	if: ${{ failure() && (github.ref == 'refs/heads/main') }}
	with:
	payload: \|
	{
	"blocks": [
	{
	"type": "section",
	"text": {
	"type": "mrkdwn",
	"text": ":rotating_light: [CI] Python Import Checks <${{ github.server_url }}/${{ github.repository }}/actions/runs/${{ github.run_id }}\|workflow> FAILED on main :rotating_light:"
	}
	}
	]
	}
	env:
	SLACK_WEBHOOK_URL: ${{ secrets.SLACK_WEBHOOK_URL }}
	SLACK_WEBHOOK_TYPE: INCOMING_WEBHOOK

	test-imports:
	runs-on: ubuntu-latest
	needs:
	- test-imports-platform
	if: always()
	steps:
	- name: All tests ok
	if: ${{ !(contains(needs.*.result, 'failure')) }}
	run: exit 0
	- name: Some tests failed
	if: ${{ contains(needs.*.result, 'failure') }}
	run: exit 1


	style:
	runs-on: ubuntu-latest
	timeout-minutes: 15
	env:
	python-version: '3.8'
	steps:
	- uses: actions/checkout@v4
	- name: Set up Python ${{ env.python-version }}
	uses: actions/setup-python@v5
	with:
	python-version: ${{ env.python-version }}

	- name: Install pre-commit
	run: \|
	pip install uv
	uv pip install pre-commit
	- uses: moonrepo/setup-rust@v1
	with:
	cache: false
	- uses: Swatinem/rust-cache@v2
	with:
	key: ${{ runner.os }}-build
	cache-all-crates: 'true'

	- uses: actions/cache@v4
	id: pre-commit-cache
	with:
	path: ~/.cache/pre-commit/
	key: ${{ runner.os }}-python-${{ env.python-version }}-pre-commit-${{ hashFiles('.pre-commit-config.yaml') }}
	- name: Python And Rust Style Check
	run: \|
	pre-commit run --all-files

	- name: Send Slack notification on failure
	uses: slackapi/[email protected]
	if: ${{ failure() && (github.ref == 'refs/heads/main') }}
	with:
	payload: \|
	{
	"blocks": [
	{
	"type": "section",
	"text": {
	"type": "mrkdwn",
	"text": ":rotating_light: [CI] Style Checks <${{ github.server_url }}/${{ github.repository }}/actions/runs/${{ github.run_id }}\|workflow> FAILED on main :rotating_light:"
	}
	}
	]
	}
	env:
	SLACK_WEBHOOK_URL: ${{ secrets.SLACK_WEBHOOK_URL }}
	SLACK_WEBHOOK_TYPE: INCOMING_WEBHOOK

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

[FEAT] Include file paths as column from read_parquet/csv/json #9143

Workflow file

[FEAT] Include file paths as column from read_parquet/csv/json #9143

Jobs

Run details

Workflow file for this run