angles-python-client

A small Python client for the Angles Dashboard REST API, which allows python test frameworks to publish their automated test results.

Install (local / editable)

pip install -e .

Quick usage (singleton reporter)

Use angles_reporter when you want one simple process-wide reporter.

from angles_python_client import angles_reporter
from angles_python_client.models import Artifact, ScreenshotPlatform, Platform

angles_reporter.set_base_url("http://127.0.0.1:3000/rest/api/v1.0/")

build = angles_reporter.start_build(
    name="TestRunName",
    team="Team",
    environment="Environment",
    component="Component",
    phase="optional-phase",
)

angles_reporter.add_artifacts([
    Artifact(groupId="angles-ui", artifactId="anglesHQ", version="1.0.0")
])

angles_reporter.start_test(title="test1", suite="suite1")
angles_reporter.add_action("My first action")

platform = ScreenshotPlatform(
    platformName="Android",
    platformVersion="10",
    browserName="Chrome",
    browserVersion="89.0",
    deviceName="Samsung Galaxy S9",
)

screenshot = angles_reporter.save_screenshot_with_platform(
    file_path="/path/to/screenshot.png",
    view="view_1",
    tags=["smoke", "home"],
    platform=platform,
)

angles_reporter.info_with_screenshot("Checking my view on android", screenshot.get("_id"))

angles_reporter.pass_step("Assertion", expected="true", actual="true", info="Just doing an assertion")
angles_reporter.fail_step("Assertion", expected="true", actual="false", info="Just doing an assertion")

execution = angles_reporter.save_test()

Batch mode

By default every call to save_test() sends the test execution to the Angles API straight away. If you'd rather send the whole test run in a single request at the end (e.g. for large runs), you can enable batch mode. The build is still created up-front and screenshots are still uploaded individually as the tests run (they need the build id), but the executions are gathered by the reporter until you call save_all_tests().

angles_reporter.set_batch_mode(True)
angles_reporter.start_build(name="TestRunName", team="Team", environment="Environment", component="Component")

# run your tests as usual: start_test(), save_screenshot(), pass_step()/fail_step() and save_test()
# save_test() now stores the executions in the reporter rather than sending them.

# once all tests are done, store all the executions against the build in one request.
updated_build = angles_reporter.save_all_tests()

Independent reporters for isolated state

Create AnglesReporter(...) instances directly when you need more than one live reporter at once, for example when publishing multiple builds or environments in parallel. Each instance keeps its own current_build, current_execution, and current_action state.

from angles_python_client import AnglesReporter

reporter_gib = AnglesReporter(base_url="https://angles-api.example/rest/api/v1.0/")
reporter_games_dev = AnglesReporter(base_url="https://angles-api.example/rest/api/v1.0/")

reporter_gib.start_build(
    name="[suite] test-suite integration",
    team="some-team",
    environment="integration",
    component="some-component",
)

reporter_games_dev.start_build(
    name="[suite] test-suite dev",
    team="some-team",
    environment="dev",
    component="some-component",
)

AnglesReporter.get_instance() and the module-level angles_reporter still return the shared convenience singleton.

AnglesReporter.get_instance_with_base_url(...) is kept for backward compatibility, but it now returns a fresh reporter bound to that base URL instead of reusing shared mutable reporting state.

Direct requests usage

from angles_python_client import AnglesHttpClient
from angles_python_client.requests import BuildRequests

http = AnglesHttpClient(base_url="http://127.0.0.1:3000/rest/api/v1.0/")
builds = BuildRequests(http)
build = builds.get_build("your-build-id")

Publish a release to PyPI

1. Bump project.version in pyproject.toml

2. Push a tag like v1.0.1:

   git tag v1.0.1
   git push origin v1.0.1

The workflow .github/workflows/publish.yml will build and publish to PyPI.

Publish to TestPyPI first (optional)

• Push a tag like test-v1.0.1, or run the Publish to TestPyPI workflow manually.

• Install from TestPyPI to validate:

  python -m pip install -i https://test.pypi.org/simple/ --extra-index-url https://pypi.org/simple angles-python-client

Build locally (manual)

python -m pip install -U pip build twine
python -m build
python -m twine check dist/*
python -m twine upload dist/*

Versioning

This project uses a standard pyproject.toml version = "X.Y.Z" field.

Bump version locally (and create a tag)

pip install -U bump2version
bump2version patch   # or minor / major
git push --follow-tags

Bump via GitHub Actions (and publish)

Run the Bump version and tag workflow (Actions tab) and choose:

• target=pypi → creates a tag like v1.0.1 (triggers the PyPI publish workflow)
• target=testpypi → creates a tag like test-v1.0.1 (triggers the TestPyPI publish workflow)

Auto-release on main/master

This repo includes an optional workflow: .github/workflows/release-on-main.yml.

When enabled (it is committed by default), every push to main or master will:

1. Read the current project.version from pyproject.toml
2. Create and push a git tag vX.Y.Z (if it doesn't already exist)
3. Build and publish to PyPI (Trusted Publishing)
4. Bump the version patch for the next release and push that commit back to the default branch

To avoid loops, the post-bump commit includes [skip release] and the workflow ignores commits containing that marker.