1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
|
# Todo: rework this file, it is not up to date
# Smoke and Integration tests
## Test for lost client connection:
1. start example.py with log_level='trace'
2. curl http://localhost:8000/endless
3. kill curl
### expected outcome:
all streaming stops, including pings (log output)
## Test for uvicorn shutdown (Ctrl-C) with long running task and multiple clients:
- see also: `test_multiple_consumers.py`
1. start example.py with log_level='trace'
2. curl http://localhost:8000/endless from multiple clients/terminals
3. CTRL-C: stop server
### expected outcome:
1. server shut down gracefully, no pending tasks
2. all clients stop (transfer closed with outstanding read data remaining)
## Test for stream_generator.py
1. start stream_generator.py with log_level='trace'
2. Run http calls in integration_testing.http
### expected outcome:
1. every post from one client should be seen on consuming client
|
