DOCKER !
pytest panoptes
if you wish to run it against production:
pytest panoptes --ENV=prod
If you don't wish to return Tesults results, perhaps due to local dev, use the following:
pytest panoptes --REPORT=no
to build dockerfile:
from containing api-check
folder:
docker build -t panoptes panoptes
to run dockerfile:
docker run panoptes
The dockerfile uses the panoptes/runScript.py
to run several test sub-suites, mostly using the pytest mark syntax. These tests are run against both QA and Prod, and results can be analyzed using Tesults
Smoke Tests are for before and after a deploy, to verify that the deploy actually deployed what you wanted it to.
Smoke Tests verify:
- that the API ENV is responding
- that API authentication works (both buyer and seller)
- that the database can be accessed (via API object creation, retrieval, delete & verification)
- DevCenter website is responding
- basic message senders are correctly triggering and emails are recieved for them (emails are through mailinator)
- --DevCenter authentication works and redirects correctly--
These tests cover the new search 2.0 material, such as faceted navigation and elastic search queries
These are the "poor man's load tests", where a lot of common calls are run against the api. Unlike the full Locust load tests, they do not run full end to end tests, and they are focused on testing performance-sensitive queries such as /me/products.
- switch all request calls to try/catch
- add sphinx documentation
- add non established buyer user runs to tests -- ex, anon, etc
- repair and improve xp with depths