add tinybird api clients

[thrau]

Motivation

I've wanted better API clients for a while, where classes map to the actual APIs and their API concepts. These can then be used to implement higher-level abstractions like the DataSource class.

With #4 , it's now massively easier to develop and test this, so I went on a quest to implement the most important APIs.

The following APIs have been (partially) implemented, and can be accessed through Client.api:

• /v0/datasources: client.api.datasources
• /v0/events: client.api.events
• /v0/pipes: client.api.pipes
• /v0/tokens: client.api.tokens
• /v0/variables: client.api.variables

Examples

For example, you can query a pipe using the pipes API:

from verdin import tinybird

client = tinybird.Client(...)

response = client.api.pipes.query(
    "my_pipe",
    parameters={"my_param": "..."},
    query="SELECT * FROM _ LIMIT 10",
)

for record in response.data:
    # each record is a dictionary
    ...

Or use the HFI endpoint to append data:

from verdin import tinybird

client = tinybird.Client(...)

response = client.api.events.send("my_datasource", records=[
    {"id": "...", "value": "..."},
    ...
])
assert response.quarantined_rows == 0

Changes

• Introduce verdin.api which contains classes that map to APIs defined in https://www.tinybird.co/docs/api-reference/

[Pive01]

FYI this seems to be a breaking change as you have changed the interface of the query method forcing parameters to be dict[str, str] while before it was less strict accepting dict[str, str | None] and now if params has a None value doing len(None) causes a runtime error.... Not sure if is more correct now or before 😄

[thrau]

thank you! fixed in #8