dataverse:unstable increased the version to 6.3

[shoeffner]

Prerequisites

• [x] Are you running the expected version of pyDataverse? (check via pip freeze).

main branch

Bug report

The :unstable docker image now ships dataverse 6.3, but DV_VERSION in the docker-compose-test-all.yml is pinned to 6.2, causing several tests to fail.

1. Describe your environment

• [x] OS: Ubuntu (I think... in the container)
• [x] pyDataverse: main branch
• [x] Python: 3.11 (in the container)
• [x] Dataverse: 6.3 (!)

2. Actual behaviour:

Some tests fail (here one example):

_______________________ TestApiToken.test_token_missing ________________________

self = <tests.api.test_api.TestApiToken object at 0xffff8b262fd0>

    def test_token_missing(self):
        BASE_URL = os.getenv("BASE_URL").rstrip("/")
        api = NativeApi(BASE_URL)
        resp = api.get_info_version()
>       assert resp.json()["data"]["version"] == os.getenv("DV_VERSION")
E       AssertionError: assert '6.3' == '6.2'
E         
E         - 6.2
E         + 6.3

tests/api/test_api.py:75: AssertionError

3. Expected behaviour:

The tests pass.

4. Steps to reproduce

1. sh run-tests.sh.

❌ Unit tests failed. Printing logs...

5. Possible solution

• Update DV_VERSION in docker-compose-test-all.yml to 6.3.
• Do not use :unstable but pin to a specific image version. Since Dataverse only tags two versions, I guess a hash might work instead.
• Ask if Dataverse's docker images can be tagged with the versions, so that we could even test against different versions easily, for example to handle different API changes.

I will provide a PR to increase the version to 6.3, as described in the first variant. I don't think the second is feasible right now, but I think it might be worth exploring the third option, though I really don't know what the status of the Dataverse development in general is on that matter.