* drop gauge instrument
* Restore the benchmark and stress test for lastvalue aggregator, but remove monotonic last-value support
* Rename gauge->lastvalue and remove remaining uses of the word 'gauge'
Co-authored-by: Krzesimir Nowak <krzesimir@kinvolk.io>
* wip: observers
* wip: float observers
* fix copy pasta
* wip: rework observers in sdk
* small fix in global meter
* wip: aggregators and selectors
* wip: monotonicity option for observers
* some refactor
* wip: docs
needs more package docs (especially for api/metric and sdk/metric)
* fix ci
* Fix copy-pasta in docs
Co-Authored-By: Mauricio Vásquez <mauricio@kinvolk.io>
* recycle unused recorders in observers
if a recorder for a labelset is unused for a second collection cycle
in a row, drop it
* unregister
* thread-safe set callback
* Fix docs
* Revert "wip: aggregators and selectors"
This reverts commit 37b7d05aed5dc90f6d5593325b6eb77494e21736.
* update selector
* tests
* Rework number equality
Compare concrete numbers, so we can get actual numbers in the error
message when they are not equal, not some uint64 representation. This
also uses InDelta for comparing floats.
* Ensure that Observers are registered in the same order
* Run observers in fixed order
So the tests can be reproducible - iterating a map made the order of
measurements random.
* Ensure the proper alignment of the delegates
This wasn't checked at all. After adding the checks, the test-386
failed.
* Small tweaks to the global meter test
* Ensure proper alignment of the callback pointer
test-386 was complaining about it
* update docs
* update a TODO
* address review issues
* drop SetCallback
Co-authored-by: Mauricio Vásquez <mauricio@kinvolk.io>
Co-authored-by: Rahul Patel <rghetia@yahoo.com>
* Update metrics API to match current spec
* update options to match the spec
* drop the global meter API
* more docs
* get rid of leftover mentions about descriptor
* initial metrics work
* rename cumulative to counter
* rename bidirectional to nonmonotonic
* rename unidirectional to monotonic
* rename nonnegative to signed
this changes the default semantics a bit - before the change measure
could record negative values by default, now it can't.
The specification draft currently specifies both NonNegative and
Signed, but I think it's a mistake.
* rename instrument to descriptor
* license
* rework measurement values
* make measurement value a tagged union
* simplify to one kind of metrics
* add observers
* change some interfaces to match the spec
* keep integral measurement separate from floating ones
* remove duplicated measurement type
* add checking for options
* reorder some fields and functions
* rename a function
to avoid confusion between the Handle type and the Measure type
* drop disabled field from descriptor
* add back typed API for metrics
* make metric options type safe
* merge alternatives into a single bool
* make value kind name less stuttery
* fix observation callback prototype
* drop context parameter from NewHandle
* drop useless parameter names
* make descriptor an opaque struct
* use a store helper
* handle comment fixes
* reword Alternate comment
* drop the "any value" metrics
* make measurement value simpler
* document value stuff
* add tests for values
* docs
* do not panic if there is no span ID in the event
