introduction of View and first version of LabelsProessor

[as-polyakov]

• Added View as a wrapper around AggregatorFactory and LabelsProcessorFactory
• Added LabelsProcessor interfaces

[bogdandrutu]

Initial feedback about the new interfaces

[bogdandrutu]

what about we make create return "nullable" value, and return null by default? That way we know if no processor is installed and we can optimize more in the accumulator (not even retrieving the current context, etc).

[as-polyakov]

I am not so sure about this one. Having nulls in public methods is usually considered an anti pattern, I do understand though that we have this call on a critical path for every metric recording. Seems like getting current context is pretty lightweight - just pulling it from thread local most of the time. Unless you insist I would avoid null's. If you do insist I am happy to change though

[bogdandrutu]

you are correct, probably we need to expose just the noop instance, so at least we can do the trick if we want to compare with noop reference which should be same as comparing with null.

[codecov]

Codecov Report

Merging #2800 (aeb5ea0) into main (339bbb8) will increase coverage by 0.13%.
The diff coverage is 78.37%.

@@             Coverage Diff              @@
##               main    #2800      +/-   ##
============================================
+ Coverage     89.50%   89.64%   +0.13%     
- Complexity     2613     2637      +24     
============================================
  Files           308      312       +4     
  Lines          8446     8481      +35     
  Branches        855      856       +1     
============================================
+ Hits           7560     7603      +43     
+ Misses          616      614       -2     
+ Partials        270      264       -6     

Impacted Files	Coverage Δ	Complexity Δ
.../sdk/metrics/processor/BaggageLabelsProcessor.java	0.00% <0.00%> (ø)	0.00 <0.00> (?)
.../sdk/metrics/processor/LabelsProcessorFactory.java	33.33% <33.33%> (ø)	1.00 <1.00> (?)
...opentelemetry/sdk/metrics/AbstractAccumulator.java	100.00% <100.00%> (ø)	3.00 <1.00> (+1.00)
...io/opentelemetry/sdk/metrics/SdkMeterProvider.java	100.00% <100.00%> (ø)	11.00 <1.00> (ø)
.../sdk/metrics/SynchronousInstrumentAccumulator.java	84.21% <100.00%> (+1.35%)	8.00 <1.00> (ø)
...ava/io/opentelemetry/sdk/metrics/ViewRegistry.java	95.00% <100.00%> (ø)	10.00 <5.00> (ø)
...try/sdk/metrics/processor/NoopLabelsProcessor.java	100.00% <100.00%> (ø)	2.00 <2.00> (?)
...n/java/io/opentelemetry/sdk/metrics/view/View.java	100.00% <100.00%> (ø)	2.00 <2.00> (?)
...metry/sdk/extension/resources/ProcessResource.java	78.94% <0.00%> (-3.91%)	5.00% <0.00%> (+1.00%)	⬇️
.../java/io/opentelemetry/sdk/resources/Resource.java	85.45% <0.00%> (-3.64%)	16.00% <0.00%> (ø%)
... and 14 more

---

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 339bbb8...aeb5ea0. Read the comment docs.

[jkwatson]

do we still need this PR now that the others have been merged?

[as-polyakov]

no

…

On Fri, Mar 5, 2021 at 9:40 AM John Watson ***@***.***> wrote: do we still need this PR now that the others have been merged? — You are receiving this because you authored the thread. Reply to this email directly, view it on GitHub <#2800 (comment)>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/AC7E772ZMYN2AUZ6KRH7MHDTCEJQ3ANCNFSM4XODGJZQ> .