Multichannel instrument consistency for optical spectrum analyzer #431
Conversation
…onsistency fix tests to adjust to new metaclass refractoring Finish optical_spectrum_analyzer channel consistency
|
|
||
|
|
||
| from enum import IntEnum, Enum | ||
| import hashlib |
There was a problem hiding this comment.
cleanup unused import
|
|
||
| def __init__(self, *args, **kwargs): | ||
| super().__init__(*args, **kwargs) | ||
| self._channel_count = len(self.Traces) |
There was a problem hiding this comment.
now required from metaclass. see channel property below for range
Codecov ReportAll modified and coverable lines are covered by tests ✅
Additional details and impacted files@@ Coverage Diff @@
## main #431 +/- ##
=======================================
Coverage 99.14% 99.14%
=======================================
Files 91 91
Lines 9240 9251 +11
=======================================
+ Hits 9161 9172 +11
Misses 79 79
Flags with carried forward coverage won't be shown. Click here to find out more. ☔ View full report in Codecov by Sentry. 🚀 New features to boost your workflow:
|
| with pytest.raises(NotImplementedError): | ||
| _ = inst.channel | ||
| ch = inst.channel[0] | ||
| assert isinstance(ch, ik.abstract_instruments.OpticalSpectrumAnalyzer.Channel) |
There was a problem hiding this comment.
the channel itself is automatically implemented now and - if a single channel instrument - will forward to the parent's channel. so this needed to be changed
|
|
||
|
|
||
| def test_osa_channel_wavelength(osc): | ||
| @pytest.mark.parametrize("num_ch", [1, 5]) |
There was a problem hiding this comment.
take care of single and multichannel instruments
| with expected_protocol(osa, [], []) as inst: | ||
| inst._channel_count = num_ch | ||
| ch = inst.channel[0] | ||
| with pytest.raises(NotImplementedError): |
There was a problem hiding this comment.
call single/mutli channel instrument through the inst.channel ProxyList function and ensure proper forwarding
| with pytest.raises(NotImplementedError): | ||
| ch.wavelength() | ||
| with pytest.raises(NotImplementedError): | ||
| inst.wavelength() # single channel instrument |
There was a problem hiding this comment.
use the single channel function to call (directly from instrument class) and ensure it exists and raises appropriate error
|
|
||
|
|
||
| def test_osa_channel_data(osc): | ||
| @pytest.mark.parametrize("num_ch", [1, 5]) |
|
|
||
| def test_channel_is_channel_class(): | ||
| inst = ik.yokogawa.Yokogawa6370.open_test() | ||
| assert inst._channel_count == len(inst.Traces) |
There was a problem hiding this comment.
ik.yokogawa.Yokogawa6370.Traces is 5 channels long, this is provided to the proxy list. Making sure here we set this now, as it is a new variable defined in the metaclass.
|
seems reasonable to me! |
|
Sweet, there's still some cleanup to do though. |
|
What do you think / was your reasoning about how the |
|
@scasagrande Happy new year! Wanted to check in if you have any info on above question on how |
|
Sorry! What was my reasoning? Oh goodness I don't remember the specifics. I think I removed them from the functions on the instrument class because I wanted the concrete implementation to be in the channel specific class. |
|
Thanks @scasagrande for the answer, well, I'm not that fast these days either ;) Anyway, tried to make it consistent with decorators to keep the ones that are channel associated in the channel class and the others outside in the main class. |
scasagrande
left a comment
scasagrande
left a comment
There was a problem hiding this comment.
looks reasonable to me! mostly test changes, and others seem to be in line with what we've done with other multichannel classes
trappitsch
deleted the
optical_spectrum_analyzer_single_ch_consistency
branch
2 participants
This PR provides multichannel consistency for the optical spectrum analyzer metaclass and is associated with the discussion in #156.
It has been a while that we discussed this, but I finally started looking at it again and figured that the optical spectrum analyzer seems straight forward to get my hands on this issue. Changes in the metaclass are modeled after the function generator. Would be good to hear what you think to see if this is in the right direction. Thanks!
The only instrument dependent on this metaclass is currently the Yokogawa 6370. I have only extended the tests, but not changed the instruments test in order to ensure that there are no breaking changes.
Metaclass tests are changing a bit and I will comment in a sec on the important ones.
Let me know what you think!