mirror of
https://github.com/BerriAI/litellm.git
synced 2026-09-25 01:02:15 +00:00
37 lines
1.4 KiB
Markdown
37 lines
1.4 KiB
Markdown
# tests/unit
|
|
|
|
In-process. No network, clock or subprocess
|
|
|
|
## What good looks like
|
|
|
|
```python
|
|
def test_send_result_same_version_is_identity_passthrough():
|
|
rpc = _rpc(V03_MESSAGE)
|
|
out = normalize_jsonrpc_response(rpc, "0.3", method="message/send")
|
|
assert out is rpc
|
|
```
|
|
|
|
`is`, because `==` passes on a copy. Many inputs: parametrize
|
|
(`test_an_incomplete_reservation_accrues_nothing`, fifteen cases, fifteen results)
|
|
|
|
No doubles on our own code
|
|
|
|
```python
|
|
with patch.object(streamer, "_group_by_date") as mock_group, patch.object(streamer, "_send_daily_batch") as mock_send:
|
|
mock_group.return_value = {"2025-01-19": pl.DataFrame({"test": ["data1"]}), "2025-01-20": pl.DataFrame({"test": ["data2"]})}
|
|
streamer.send_batched(pl.DataFrame({"test": ["data"]}), "replace_hourly")
|
|
assert mock_send.call_count == 2
|
|
```
|
|
|
|
Green if `send_batched` drops every row. pydantic doubles in 12 of 203 files, fastapi 11 of 594;
|
|
`tests/test_litellm` 59 percent. Exception: the count is the behaviour
|
|
(`test_dual_cache_async_batch_get_cache_coalesces_concurrent_redis_reads`, fifty readers, `call_count == 1`)
|
|
|
|
## Where it goes
|
|
|
|
`tests/unit/<path>` mirrors `litellm/<path>`, so a changed file selects its tests by path, not a mapping
|
|
file. Empty today; new unit tests go here. The examples above live in `tests/test_litellm`
|
|
|
|
## Writing it so a human can read it
|
|
|
|
A class only when tests share an arrange
|