Skip to main content

​
Session.endpoint(route, *, method=‘GET’)

Create a custom endpoint at f"/proxy/{route}".
Calling this method overrides atoti.Session.proxy’s url.
The endpoint logic is written in Python but the endpoint is exposed by Atoti Server. This allows to deploy the project in a container or a VM with a single opened port (the one of Atoti Server) instead of two. The decorated function must:
  • take three parameters with respective types:
  • return a response body as JSON convertible data
  • Parameters:
    • route (str) – The template of the path after "/proxy". For instance, if "foo/bar" is passed, a request to "/proxy/foo/bar?query=string" will match. Path parameters can be configured by wrapping their name in curly braces in the template.
    • method (Literal [ ‘DELETE’ , ‘GET’ , ‘PATCH’ , ‘POST’ , ‘PUT’ ]) – The HTTP method the request must be using to trigger this endpoint. DELETE, PATCH, POST, and PUT requests can have a body but it must be JSON.
  • Return type: Callable[[Callable[[Request, User, Session], JsonValue]], Callable[[Request, User, Session], JsonValue]]

​
Example

>>> df = pd.DataFrame(
...     columns=["Year", "Month", "Day", "Quantity"],
...     data=[
...         (2019, 7, 1, 15),
...         (2019, 7, 2, 20),
...     ],
... )
>>> table = session.read_pandas(
...     df, keys={"Year", "Month", "Day"}, table_name="Quantity"
... )

>>> @session.endpoint("tables/{table_name}/count", method="GET")
... def get_table_row_count(request, user, session, /):
...     table_name = request.path_parameters["table_name"]
...     table = session.tables[table_name]
...     return table.row_count
>>> response = session.client.http_client.get(
...     f"proxy/tables/{table.name}/count"
... )
>>> response.raise_for_status().json()
2

>>> @session.endpoint("tables/{table_name}/rows", method="POST")
... def append_rows_to_table(request, user, session, /):
...     rows = request.body
...     table_name = request.path_parameters["table_name"]
...     table = session.tables[table_name]
...     dataframe = pd.DataFrame(rows, columns=list(table))
...     table.load(dataframe)
>>> response = session.client.http_client.post(
...     f"proxy/tables/{table.name}/rows",
...     json=[
...         (2021, 5, 19, 50),
...         (2021, 5, 20, 6),
...     ],
... )
>>> response.status_code
200
>>> response = session.client.http_client.get(
...     f"proxy/tables/{table.name}/count"
... )
>>> response.raise_for_status().json()
4
>>> table.head()
                Quantity
Year Month Day
2019 7     1          15
           2          20
2021 5     19         50
           20          6
The full flow is: 
>>> with TRACER.start_as_current_span("query endpoint") as span:
...     response = session.client.http_client.get(
...         f"proxy/tables/{table.name}/count"
...     )
...     _ = response.raise_for_status()
>>> _print(span)
⏺ query endpoint                                         [atoti.python_sdk]
└── GET /proxy/**                                        [atoti.server]
    └── unsecured request                                [atoti.server]
        └── GET                                          [atoti.server]
            └── GET /tables/{table_name}/count           [atoti.python_sdk]
                ├── Validate authorization               [atoti.python_sdk]
                │   └── POST /graphql                    [atoti.server]
                │       └── unsecured request            [atoti.server]
                │           └── query                    [atoti.server]
                │               └── currentUser          [atoti.server]
                │                   ├── name             [atoti.server]
                │                   └── roles            [atoti.server]
                └── Execute user callback                [atoti.python_sdk]
                    ├── Tables._get_unambiguous_keys     [atoti.python_sdk]
                    │   └── POST /graphql                [atoti.server]
                    │       └── unsecured request        [atoti.server]
                    │           └── query                [atoti.server]
                    │               └── dataModel        [atoti.server]
                    │                   └── database     [atoti.server]
                    │                       └── table    [atoti.server]
                    │                           └── name [atoti.server]
                    └── Table.row_count                  [atoti.python_sdk]
See also: atoti.Session.proxy.