Mock resampler: Improve variable names #879
Conversation
Since the mock resampler uses a pair of channels, one to receive samples to be (fakely) resampled (input) and one to send the (fakely) resampmled samples (output), we should use clear names to identify both otherwise is really hard to follow which is which when reading the code. Signed-off-by: Leandro Lucarella <luca-frequenz@llucax.com>
github-actions
bot
added
the
part:tests
llucax
added
the
type:tech-debt
|
Enabled auto-merge. |
| input_chan_recv_name = f"{request.component_id}:{request.metric_id}" | ||
| input_chan_recv = self._input_channels_receivers[input_chan_recv_name].pop() | ||
| assert input_chan_recv is not None | ||
| output_chan_sender: Sender[Sample[Quantity]] = ( | ||
| self._channel_registry.get_or_create( | ||
| Sample[Quantity], name | ||
| ).new_sender() | ||
| ) |
There was a problem hiding this comment.
Interesting, this is better than basic_receivers, but input_receivers and output_sender might have even nicer. input_channels_receivers make me think that they receive channels and not input values.
There was a problem hiding this comment.
Yeah, at first I named it without the chan, but for some reason it still seemed ambiguous, but now I don't remember why. Maybe @daniel-zullo-frequenz as he was helping me to understand the code while I did it.
There was a problem hiding this comment.
IIRC we wanted to mainly get rid of basic in the variable name because it was confusing/difficult-to-understand what a basic_receiver was. I do agree having only receivers and sender in the name would have suffice to understand they are channels given the context. However I don't think it causes any misunderstanding if we keep chan/channels explicitly in the variable names. I mean I haven't seen yet any scenarios where we send/receive channels
Since the mock resampler uses a pair of channels, one to receive samples to be (fakely) resampled (input) and one to send the (fakely) resampmled samples (output), we should use clear names to identify both otherwise is really hard to follow which is which when reading the code.