You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
This would provide an excellent mechanism for testing Wheatley's performance, since if we mocked the Tower class and make the sleep function just increment a counter then we can simulate Wheatley ringing for huge amounts of time.
If someone else wants to have a crack at this then they're very welcome, but if I had to do it I'd probably do something like:
Move all the knowledge of timing (i.e. time.time() and time.sleep()) into the Tower class so that it can be mocked easily.
Make some kind of running mode which records the events passing through the Tower class and records them to some format (probably JSON)
Make a mocked version of Tower which never interacts with RR but instead replays any given simulation using mocked sleep and time functions to go wayyyy faster than real time.
The text was updated successfully, but these errors were encountered:
A record-and-playback feature could also be used for other purposes than just testing, such as for inclusion in pre-recorded worship services, or even semi-live demonstrations (playbacks) for (spatially-) distributed listeners of a single online tower.
It might be useful to somehow indicate visually to onlookers (casual, non-participating listeners, watching the playback ringingroom) a status message that a recording or playback is in session, when it is.
A record-and-playback feature could also be used for other purposes than just testing, such as for inclusion in pre-recorded worship services, or even semi-live demonstrations (playbacks) for (spatially-) distributed listeners of a single online tower.
Agreed. Real-time playback wouldn't be super useful for testing Wheatley, but once we have some kind of recording format it would be really easy to play them back to Ringing Room (we probably don't even need Wheatley - we could just ring the bells directly using WebSocket or a library like belltower).
This would provide an excellent mechanism for testing Wheatley's performance, since if we mocked the
Tower
class and make thesleep
function just increment a counter then we can simulate Wheatley ringing for huge amounts of time.If someone else wants to have a crack at this then they're very welcome, but if I had to do it I'd probably do something like:
time.time()
andtime.sleep()
) into theTower
class so that it can be mocked easily.Tower
class and records them to some format (probably JSON)Tower
which never interacts with RR but instead replays any given simulation using mockedsleep
andtime
functions to go wayyyy faster than real time.The text was updated successfully, but these errors were encountered: