vc-auto-tests

Automated test suite for VC, built with Playwright and pytest.

Requirements

• Python 3.13+
• Git

Installation

1. Clone the repository

git clone <repo-url>
cd vc-auto-tests

2. Create and activate a virtual environment

Windows

python -m venv .venv
.venv\Scripts\activate

Linux / macOS

python3 -m venv .venv
source .venv/bin/activate

3. Install dependencies

pip install -e .

4. Install Playwright browsers

playwright install

Seeding test data

The test dataset is seeded into the backend via WebAPI. Run this once against a fresh environment (and whenever the dataset changes):

python -m dataset.dataset_manager --seed --mode ci

Pass entity names to seed only a subset (order is taken from the manifest, not the command line):

python -m dataset.dataset_manager --seed stores pages pages_content

Post-seeding scripts

Some entities need a follow-up action that the create API cannot express directly — a state transition, a workflow trigger, etc. These live in dataset/scripts/ and are run after seeding, on demand.

Script	Purpose	Command
set_builder_page_status	Page-builder pages are always seeded as Draft (publish/archive is a backend workflow, not a settable field). This script transitions one page to a target status.	python -m dataset.scripts.set_builder_page_status --page-id <id> --status <Published|Archived>

Note: seeding leaves every page in Draft — the create API ignores a status field, so publishing and archiving must go through the workflow endpoints. Run set_builder_page_status with the grouped page ids (the stable key the API uses — page names are not guaranteed unique). --page-id accepts several ids at once:

python -m dataset.dataset_manager --seed --mode ci
python -m dataset.scripts.set_builder_page_status --status Published --page-id \
    acme-store-grouped-page-about-store \
    acme-store-grouped-page-spring-sale \
    acme-store-grouped-page-seasonal-lookbook \
    acme-store-grouped-page-wholesale-offers \
    acme-store-grouped-page-partner-portal \
    acme-store-grouped-page-summer-preview
python -m dataset.scripts.set_builder_page_status --status Archived --page-id \
    acme-store-grouped-page-legacy-catalog

That leaves one page per status filter used by the page-builder E2E suite:

Page	Lands in	Exercises
our-team	Draft	draft filter
spring-sale, about-store	Active	plain Published badge
seasonal-lookbook	Active	Published + Scheduled (current date window)
wholesale-offers	Active	Published + Personalized (user group)
partner-portal	Active	Published + Personalized (organization)
summer-preview	Pending	Published + Scheduled (future start date)
legacy-catalog	Archived	archived filter

createdDate / modifiedDate cannot be seeded — the API echoes them back in the POST response but assigns its own values, so the pages all carry seed-time audit dates.

The seed command accepts --mode dev (default, per-item detail) or --mode ci (summary output); set_builder_page_status accepts the same --mode flag.

Running Tests

pytest