* Rename ParentOrElse to ParentBased
* Update samplers to match spec.
* Fix check style
* Fix check style
* Fix review
* Separate test for parentOrElse sampler
* Provide ParentBased.Builder
* Fix code style
* Fix java doc.
* Update javadoc
* Builder use set instead of with
* Rename newParentBasedBuilder to parentBasedBuilder
* More aggresive version of BatchSpanProcessor
* Add benchmark
* Polish
* Polish
* Incorporated some changes from #1571
* Rollback one test change
* Polish
* Polish
* start working on a ReadWriteSpan for startSpan
* Finish converting the disruptor and the unit tests to the new model.
* Add some simple javadoc to the ReadWriteSpan interface
* Split single "sdk" module into separate pieces for common, correlationcontext, metrics, tracing, and the combination of them all
* Modify comment to not reference implementation class
* Moved JMH code to /sdk/all
* Move new files to adjusted location
* Move new files from rebase
* Add documentation to make it clear that the MetricProducer cannot be shared across exporters.
* Update sdk/src/main/java/io/opentelemetry/sdk/metrics/MeterSdkProvider.java
Co-authored-by: Anuraag Agrawal <anuraaga@gmail.com>
* Update sdk/src/main/java/io/opentelemetry/sdk/metrics/MeterSdkProvider.java
Co-authored-by: Anuraag Agrawal <anuraaga@gmail.com>
Co-authored-by: Anuraag Agrawal <anuraaga@gmail.com>
* Makes Export methods async
This commit recognises that the export and flush methods of span and trace exporters can be, and often are, implemented with long-lived operations over networks. We therefore see that these method return types are represented using Java's Future type to account for this common behaviour.
* PR feedback
* write a test that demonstrates a bound counter reporting zeros with no observations.
* Have the metric Aggregator keep track of if there have been recordings since a reset has been done.
Have the batcher skip batching if no recordings have been made.
* Add a test to assert the NoopAggregator has no recordings.
* Tweaks from PR review
* Fix bug where Sampling status would not propagate from parent to child correctly
* use generic parentorelse sampler instead
* bring back alwayson as default sampler
* Refactor decision to sampling result
* Fix bad refactor samplingResult => decision
* Fix decision not sampling result
* Fix comments with divergent explanations of same entity
* Fix make comments clearer
* Add links to SamplingResult
* Add overloads for setParent(Context context).
* Fix the original version.
* Apply feedback.
* Update version for CorrelationContext.
* Add simple extra test for DefaultCorrelationContextManagerTest.
* Refactor the creation of instrument batchers to be in a central location.
This should allow us to build a Views API into the SDK.
* Fill in some missing javadoc.
* move the ViewRegistry up one package, and clean up the visibility of other classes
John Watson
dengliming
Anuraag Agrawal
Nikita Salnikov-Tarnovski
jarebudev
Ken Finnigan
Giovanni Liva
Christopher Hunt
Trask Stalnaker
Christian Neumüller
MitchellDumovic
Bogdan Drutu
Peter Travers
Carlos Alberto Cortez
Lei Wang