Codecov Report

All modified and coverable lines are covered by tests ✅

Comparison is base (bd23d61) 79.56% compared to head (8c5400a) 79.56%.

@@           Coverage Diff           @@
##           master     #567   +/-   ##
=======================================
  Coverage   79.56%   79.56%           
=======================================
  Files          37       37           
  Lines        3083     3083           
  Branches      804      804           
=======================================
  Hits         2453     2453           
  Misses        548      548           
  Partials       82       82           

☔ View full report in Codecov by Sentry.
📢 Have feedback on the report? Share it here.

pytest kasa/tests/test_emeter.py -v -k KP125M

@rytilahti If I run with -k KP125M, all tests are skipped.

If I point it at a device --ip $IP --username "$USERNAME" --password "$PASSWORD", I get:
...
kasa/tests/test_emeter.py::test_emeterstatus_missing_current ✓ 100% ██████████

Results (0.80s):
1 passed
205 skipped

It looks like I'm missing some magic to make WITH_EMETER work properly.

Related https://github.com/python-kasa/python-kasa/pull/566, but I think there's no need to recreate the file as it's just filename changes.
Updated nickname and ssid to "##MASKEDNAME##".
@rytilahti If you nudge me in the right direction, I'll try and get the KP125M emeter tests working. Otherwise, I'm not aiming to add any more to this PR.

@sbytnar ah, sorry, the test harness is currently in a bit of flux due to the new protocols being added. The new fixture is not picked as it's categorized as a "SMART" device while the has_emeter filters only "IOT" devices. This patch will fix that (and cause the tests to fail :-():

diff --git a/kasa/tests/conftest.py b/kasa/tests/conftest.py
index 53e2c49..f77c13b 100644
--- a/kasa/tests/conftest.py
+++ b/kasa/tests/conftest.py
@@ -151,7 +151,7 @@ def parametrize(desc, devices, protocol_filter=None, ids=None):
     )
 
 
-has_emeter = parametrize("has emeter", WITH_EMETER)
+has_emeter = parametrize("has emeter", WITH_EMETER, protocol_filter={"SMART", "IOT"})
 no_emeter = parametrize("no emeter", ALL_DEVICES_IOT - WITH_EMETER)
 
 bulb = parametrize("bulbs", BULBS, protocol_filter={"SMART", "IOT"})

The problem is that the tests are written to test the internal structure which differs from what KP125M reports.

We can leave it as-is (i.e., not testing these for the time being) for this PR and fix that later on. The tests that are failing are checking for the device responses which differ, so we would need to guard those checks by adding a new has_emeter_iot "group" to use for those tests.

This LGTM and ready to merge.

qq. you've masked nickname. Do you think we should mask all nicknames/aliases in dump_devinfo? We currently don't but maybe we should. Especially because if people try to mask nickname themselves they could put a non base64 value in there.

Okay, let's merge this. @sdb9696, yeah, it probably makes sense to mask the name as it might be personalized & not having it masked doesn't bring any benefits that I can think of.