* Empty queued spans when ForceFlush called
Update the implementation of ForceFlush() to first ensure that all spans
which are queued are added to the batch before calling export spans.
Create a small ReadOnlySpan implementation which can be used as a marker
that ForceFlush has been invoked and used to notify when all spans are
ready to be exported.
Fixes#2080.
* Add a changelog entry.
* Update CHANGELOG.md
Co-authored-by: Tyler Yahn <MrAlias@users.noreply.github.com>
* Update sdk/trace/batch_span_processor.go
Co-authored-by: Tyler Yahn <MrAlias@users.noreply.github.com>
* Improve test case to verify multiple flushes.
* Refactor code to use enqueue.
* Be more defensive on waiting for queue.
Update the handling of the force flush span so we only wait on the
channel if we were able to enqueue the span to the queue.
* Fix linter.
Co-authored-by: Tyler Yahn <MrAlias@users.noreply.github.com>
Co-authored-by: Anthony Mirabella <a9@aneurysm9.com>
Import path incorrect for signal, added import "os/signal".
Also the current online instructions are broken, ctx is not plumped through, but it is here in the forked markdown, so probably needs to be pushed out.
* Version stable-v1 modset to be 1.0.2 for release
* Prepare stable-v1 for version v1.0.2
* Update changelog
Add new 1.0.2 section and move relevant items that the release will
contain to this section.
* Add article to CHANGELOG entry to correct grammar
* Release 1.1.0 instead of 1.0.2
* Prepare stable-v1 for version v1.1.0
* Apply suggestions from code review
Co-authored-by: Robert Pająk <pellared@hotmail.com>
Co-authored-by: Robert Pająk <pellared@hotmail.com>
This bug was uncovered during spec review here:
https://github.com/open-telemetry/opentelemetry-specification/pull/2008#discussion_r731359448
apply_to_spans section was missing although it is implied to
be present in the description of how span changes section
is supposed to work.
This fixes the omission by adding apply_to_spans section.
This is not a new feature, just fixing a bug in file format description
and in the corresponding implementation.
Co-authored-by: Anthony Mirabella <a9@aneurysm9.com>
Generated from a modified version of the v1.5.0 release of the
specification using a modified version of the semconvgen tools. The
specification contained capitalized IDs that caused errors from the
semconvgen tool. These errors were manually resolved and the rest of the
Go formatting tools were used to provide consistent code.
Important to note, the Go semconvgen tooling includes new name
capitalization rules for ReplicaSet, StatefulSet, and DaemonSet that
mean code names are not backwards compatible. This is included in the
changelog to help users perform the upgrade.
Generated from the v1.6.1 release of the specification using the
semconvgen tool manually.
This skips the v1.6.0 release of the specification, with no plans to add
support in the future, due to a backwards in compatibility bug in
included.
* Add semconv/v1.7.0 package
Generated from the v1.7.0 release of the specification using the
semconvgen tool manually. This also updates the project to use this
version of the semconv package.
This does not include the prior, and missing, v1.5.0 and v1.6.0 versions
of the semconv package. They are planned to be added in a follow-on PR.
* Update CHANGELOG.md
Explicitly return nil if the wait time and the context deadline expire
simultaneously, instead of indeterminately the context error or nil, for
the retry wait function in the otlptrace/internal package.
* Fix IPv6 handling errors in semconv.NetAttributesFromHTTPRequest
fixes#2283
* Enter PR number in CHANGELOG
* Remove unnecessary creation and then assignment
Standardize order of checks for IP, Name, Port
* Assume happy path when parsing host and port
i.e. assume net.SplitHostPort(input) will succeed
* Get rid of uint64 for port
* Fix git merge of main by adding back strings import
Co-authored-by: Tyler Yahn <MrAlias@users.noreply.github.com>
Co-authored-by: Tyler Yahn <codingalias@gmail.com>
* Fix: http.client_ip vs multiple addresses #2282
* Split only the necessary number of values.
As suggested by @pellared. Good suggestion, that.
Co-authored-by: Robert Pająk <pellared@hotmail.com>
Co-authored-by: Robert Pająk <pellared@hotmail.com>
Co-authored-by: Tyler Yahn <MrAlias@users.noreply.github.com>
* Add ability to parse Schema files according to OTEP 0152
The parser and parsed representation (AST) are placed in a separate
Go module so that they are can be consumed independently without
the need to bring the rest of the SDK.
Ability to use the parsed representation for schema conversions
can be added later.
* Fixes based on PR comments
* Rename "label" to "attributes"
See https://github.com/open-telemetry/oteps/pull/181
* Fixes based on PR comments
* Add README.md
* Wrap the error in Parse()
* Add docs for exporter types
* Use yaml.NewDecoder
* Verify parsed content in the test
* Fix indentation in README example
* Fix README spaces vs tabs
* Correctly space imports
* Add heading to README
Co-authored-by: Tyler Yahn <MrAlias@users.noreply.github.com>
* move Descriptor to sdkapi, add test
* rename Descriptor to sdkapi
* precommit
* Move the rest of the sdkapi
* use alias for Observation and Measurement
* Changelog
* pr num
* comment Measurement and Observation
* swap comments
* move->moved
* Bump minimum Go version to 1.16
Signed-off-by: Anthony J Mirabella <a9@aneurysm9.com>
* Revert "Bump minimum Go version to 1.16"
This reverts commit bcb390405e.
* Add testing for Go 1.17
Signed-off-by: Anthony J Mirabella <a9@aneurysm9.com>
* Update build tools
Signed-off-by: Anthony J Mirabella <a9@aneurysm9.com>
* Update CHANGELOG and versions.yaml
Signed-off-by: Anthony J Mirabella <a9@aneurysm9.com>
* Prepare stable-v1 for version v1.0.1
* Prepare experimental-metrics for version v0.24.0
* Prepare bridge for version v0.24.0
* Update releasing documentation and create add-tags make target
Signed-off-by: Anthony J Mirabella <a9@aneurysm9.com>
* use multimod from the .tools directory
Signed-off-by: Anthony J Mirabella <a9@aneurysm9.com>
* precommit
Signed-off-by: Anthony J Mirabella <a9@aneurysm9.com>
* factor instrumentation library out of the instrument descriptor
* SDK tests pass
* checkpoint work
* otlp and opencensus tests passing
* prometheus
* tests pass, working on lint
* lint applied: MetricReader->Reader
* comments
* Changelog
* Apply suggestions from code review
Co-authored-by: alrex <alrex.boten@gmail.com>
* remove an interdependency
* fix build
* re-indent one
* Apply suggestions from code review
Co-authored-by: Tyler Yahn <MrAlias@users.noreply.github.com>
* Lint&feedback
* update after rename
* comment fix
* style fix for meter options
* remove libraryReader, let Controller implement the reader API directly
* rename 'impl' field to 'provider'
* remove a type assertion
* move metric/registry into internal; move registry.MeterProvider into metric controller
* add test for controller registry function
* CheckpointSet->Reader everywhere
* lint
* remove two unnecessary accessor methods; Controller implements MeterProvider and InstrumentationLibraryReader directly, no need to get these
* use a sync.Map
* ensure the initOnce is always called; handle multiple errors
* Apply suggestions from code review
Co-authored-by: Anthony Mirabella <a9@aneurysm9.com>
* cleanup locking in metrictest
* Revert "ensure the initOnce is always called; handle multiple errors"
This reverts commit 3356eb5ed0.
* Revert "use a sync.Map"
This reverts commit ea7bc599bd.
* restore the TODO about sync.Map
Co-authored-by: alrex <alrex.boten@gmail.com>
Co-authored-by: Tyler Yahn <MrAlias@users.noreply.github.com>
Co-authored-by: Anthony Mirabella <a9@aneurysm9.com>
* Update versions file for 1.0.0 release
* Prepare stable-v1 for version v1.0.0
* Update trace signal status in documentation
* Update changelog
* Update CHANGELOG.md
Co-authored-by: Robert Pająk <pellared@hotmail.com>
Co-authored-by: Robert Pająk <pellared@hotmail.com>
This method did not have any effect. had been used back
when exporters were providing utility methods for setting up full
pipelines. These utilities went away in
4883cb119d.
We previously were recording the Schema URL but were not setting
the recorded value when exporting. This change now uses the recorded
value when exporting to OTLP.
* Fix slice-valued attributes when used as map keys
Signed-off-by: Anthony J Mirabella <a9@aneurysm9.com>
* store pointers to slice values to make them usable as map keys
Signed-off-by: Anthony J Mirabella <a9@aneurysm9.com>
* Emit slice-typed attribute values as slices, not pointers to slices
Signed-off-by: Anthony J Mirabella <a9@aneurysm9.com>
Co-authored-by: Tyler Yahn <MrAlias@users.noreply.github.com>
