Streamline Archives `PUT` request data retrieval

[maxachis]

The PUT method for the /archives endpoint has some odd behaviors that seem to be causing issues with data retrieval.

    def put(self) -> Dict[str, str]:
        #...
        json_data = request.get_json()
        data = json.loads(json_data)
        id = data["id"] if "id" in data else None
        last_cached = data["last_cached"] if "last_cached" in data else None
        broken_as_of = (
            data["broken_source_url_as_of"]
            if "broken_source_url_as_of" in data
            else None
        )

        with self.setup_database_client() as db_client:
            response = update_archives_data(db_client, id, last_cached, broken_as_of)

This requires the dictionary entry to be "dumped" beforehand, as in update_pdap_archives in cache_url.py of the pdap-automatic-archives repository:

def update_pdap_archives(entry: dict):
    """
    Update data in PDAP archives
    :param entry:
    :return:
    """
    entry_json = json.dumps(entry)
    response = requests.put(
        f"{os.getenv('VITE_VUE_APP_BASE_URL')}/archives",
        json=entry_json,
        headers={"Authorization": API_KEY},
    )
    response.raise_for_status()
``

Other resources don't require this. For example, in the `PUT` method for the `DataSourceById` resource class in `DataSources.py`:

```python
    def put(self, data_source_id: str) -> Response:
        # ...
        data = request.get_json()
        with self.setup_database_client() as db_client:
            return update_data_source_wrapper(db_client, data, data_source_id)

And it is thus easier to make a response, as seen in test_data_sources_by_id.py in tests/integration:

    response = client_with_db.put(
        f"/api/data-sources-by-id/SOURCE_UID_1",
        headers={"Authorization": f"Bearer {api_key}"},
        json={"description": desc},
    )

By contrast, if you attempt to do something like this with the archives PUT method, you'll get the error: the JSON object must be str, bytes or bytearray, not dict.

So the Archives PUT method appears to require extra steps which are immediately reversed in the backend. This makes request-making more difficult and makes documentation more challenging: currently, the Gitbook documentation doesn't mention this step.

TODO

• [ ] Revise Archives PUT method to not need the json.loads() step
• [ ] Revise associated Archives tests to not perform the json.dumps() step
• [ ] Revise update_pdap_archives in cache_url.py of the pdap-automatic-archives repository to also not perform json.dumps()

[maxachis]

Marking as post-v2 just because it'll be easier once we're in that phase