-
Notifications
You must be signed in to change notification settings - Fork 832
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
feat(experimental-packages): Update packages to latest SDK Version. #2871
feat(experimental-packages): Update packages to latest SDK Version. #2871
Conversation
experimental/packages/opentelemetry-exporter-metrics-otlp-grpc/package.json
Outdated
Show resolved
Hide resolved
Codecov Report
@@ Coverage Diff @@
## main #2871 +/- ##
==========================================
- Coverage 93.42% 92.22% -1.20%
==========================================
Files 162 68 -94
Lines 5564 1993 -3571
Branches 1175 435 -740
==========================================
- Hits 5198 1838 -3360
+ Misses 366 155 -211
|
e373187
to
f1ba2fb
Compare
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
What about the proto transformer package (introduced in #2746)? Should it be used as some of the proto types in this PR are a bit out of date, i.e. intSum
and doubleSum
metricfields?
64b8e84
to
1f9a214
Compare
True. Since we're now dropping the |
1f9a214
to
ff547ff
Compare
We later discussed changes in this PR after the official part of the SIG meeting. To my understanding, updating anything other than devDeps is the way we want to go.
The problems(mentioned here) elsewhere were caused by an invalid bump in If we merge this we force the whole userbase(that mainly uses caret-ranges for SDKs) update the API and in turn everything else and run into potential issues with implementing packages that do not support the newest API. EDIT Above actually only applies to peer dependency ranges, which, hopefully, only widen from the "top" in the context of one major API version. To conclude, my reconsidered stance is: don't change peer dependency ranges stricter, but updating everything else is fine. |
NPM doesn't dedup everything even ranges would allow it. I haven't fully understood this and it depends on the NPM version. I think the problem is not the API alone.
Maybe it's better to create a dedicated issue to discuss and agree an a versioning schema. This PR is already complex enough to get it somewhat running. Tuning the devDeps/Deps ranges in a followup (before release) can be done). |
Sure. Comments are welcome: #2863 |
1c294db
to
da6e9b3
Compare
A quick update on the progress of this PR: Sorry that this is taking so long, most of these packages are new to me. |
a3078b6
to
df55205
Compare
Not related to your changes, but noticed that exporting of |
Yes, I noticed that too. I just talked to @dyladan and am working on a follow-up PR already. To make it easier to change the exporter packages to use the We could continue the discussion on this in #2774 🙂 |
Yes this was done because the SDK types were still WIP at that time and I wanted to get the package merged with the idea that it could be updated in the future. Now that those types are settled it can be updated and the exports can be added back. |
CHANGELOG.md
Outdated
@@ -8,6 +8,8 @@ All notable changes to this project will be documented in this file. | |||
|
|||
### :rocket: (Enhancement) | |||
|
|||
* [#2871](https://github.com/open-telemetry/opentelemetry-js/pull/2871) feat(experimental-packages): Update packages to latest SDK Version. ([@pichlermarc](https://github.com/pichlermarc)) |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Can you add sub-bullet points to this? This PR handles a lot more work than what is implied by this title.
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
We can shorten this entry to be like:
* [#2871](https://github.com/open-telemetry/opentelemetry-js/pull/2871) feat(experimental-packages): Update packages to latest SDK Version. ([@pichlermarc](https://github.com/pichlermarc)) | |
* feat(experimental-packages): Update packages to latest SDK Version. #2871 @pichlermarc |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
I've submitted #2889 to discuss the verbose format.
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
@legendecas yes, this seems way more concise, good idea. 🙂 I updated the entry to #2889 format in ab80dba, and moved the entry to the experimental changelog. However, this format does not work for anyone viewing the changelog file locally.
...ckages/opentelemetry-exporter-metrics-otlp-http/test/browser/CollectorMetricExporter.test.ts
Outdated
Show resolved
Hide resolved
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Overall LGTM
@open-telemetry/javascript-maintainers I'm going to merge this tomorrow if there are no objections |
… move entry to experimental changelog.
@@ -316,16 +321,24 @@ export class FetchInstrumentation extends InstrumentationBase< | |||
|
|||
function endSpanOnSuccess(span: api.Span, response: Response) { | |||
plugin._applyAttributesAfterFetch(span, options, response); | |||
const spanResponse = { |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Was that meant to be used as a third argument of _endSpan
call? Right now the spanResponse
variable is unused.
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Which problem is this PR solving?
This PR
-wip
suffix fromapi-metrics
andmetrics-sdk-base
, this causeslerna
to link the local version of the packages. This prompts the following updates:exporter-metrics-otlp-grpc
,exporter-metrics-otlp-http
,exporter-metrics-otlp-proto
)opentelemetry-sdk-node
to the latest Metrics SDK.otlp-transformer
to the latest Metrics SDK.1.1.1
which required updates to:instrumentation-*
packages due toparseUrl
fromsdk-trace-web
does not support relative paths anymore #2884I apologize for this huge PR, it was discussed and it is, unfortunately, necessary. The
*-wip
packages pulled in1.0.x
, and leaving it out caused the build of this PR (originally only an update to the OTLP Metrics Exporters) to fail (see also: comments on #2874). This was due to their dependency onsdk-trace-base
viaexporter-trace-otlp-http
.Original Issue #2774
Supersedes #2874
Type of change
How Has This Been Tested?
Checklist: