Closed yugangw-msft closed 6 years ago
Merging #89 into master will increase coverage by
0.15%
. The diff coverage is100%
.
@@ Coverage Diff @@
## master #89 +/- ##
==========================================
+ Coverage 28.94% 29.09% +0.15%
==========================================
Files 380 380
Lines 16332 16170 -162
==========================================
- Hits 4727 4705 -22
+ Misses 11605 11465 -140
Impacted Files | Coverage Δ | |
---|---|---|
msrest/serialization.py | 82.29% <100%> (-0.03%) |
:arrow_down: |
test/vanilla/AcceptanceTests/test_url.py | 22.72% <0%> (+0.17%) |
:arrow_up: |
...t/vanilla/AcceptanceTests/test_model_flattening.py | 21.66% <0%> (+1.29%) |
:arrow_up: |
test/vanilla/AcceptanceTests/test_dictionary.py | 13.04% <0%> (+1.88%) |
:arrow_up: |
.../vanilla/AcceptanceTests/test_required_optional.py | 31.25% <0%> (+3.97%) |
:arrow_up: |
Continue to review full report at Codecov.
Legend - Click here to learn more
Δ = absolute <relative> (impact)
,ø = not affected
,? = missing data
Powered by Codecov. Last update 9d81113...bd892c5. Read the comment docs.
@lmazuel, i am closing this out since you are not convinced this is the right thing to do
For https://github.com/Azure/azure-cli/issues/5733 I thought more about pruning at CLI level, but it has perf related drawback of having to go recursively to the leaf level; also, commands can create its own dictionary typed output which includes model instances, so CLI can't just check the root level types. So submit this PR to get conversation going. I am fine with any other options as long as we can mitigate the confusion. Json output is supposed to honestly reflect what is on the wire, and i hope the spirit can continue. //CC: @lmazuel @tjprescott @derekbekoe