Skip to content
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

meta(changelog): Update changelog for 8.19.0 #12987

Merged
merged 12 commits into from
Jul 19, 2024
Merged

Conversation

andreiborza
Copy link
Member

The previews release failed due to node 22.5.0 breaking CI. Here's another try, with new fixins and features :)

AbhiPrasad and others added 12 commits July 18, 2024 08:54
Before: `Time: 3.935 s`

After: `Duration 1.55s (transform 961ms, setup 0ms, collect 3.56s, tests
127ms, environment 1.73s, prepare 367ms)`

If we remove the jsdom based tests it'll probably get even faster.
This PR adds basic scaffolding for the cloudflare workers SDK. Most of
this is based on `@sentry/vercel-edge`.

This adds:

1. A basic cloudflare workers client
2. A set of default integrations for the cloudflare sdk (including a
fetch based one)
3. A cloudflare transport that uses the vercel-edge transport
4. An async context strategy powered by AsyncLocalStorage

You'll notice that there is no `init` method defined or exported from
the SDK. This is on purpose! `init` for cloudflare workers will work a
bit differently than the other SDKs, so I wanted to address it
differently on purpose. You'll see what that looks like in the next PR!
As per https://vitest.dev/config/#globals

> By default, vitest does not provide global APIs for explicitness

I think we should follow vitest defaults here and explicitly import in
the APIs that we need. This refactors our SvelteKit SDK tests to do so.

This change also removes `environment: 'jsdom'` from the sveltekit
global vite config in favour of explicitly adding jsdom environment via
the `@vitest-environment` pragma to the specific test file that needs
it. This should means that our server tests are not polluted with jsdom
globals, and that future writers have to explicitly opt-in to the
behaviour.
Previously, we passed the span name directly to the `tracesSampler`
as-is. However, this is not ideal because this does not match the name
we actually send to sentry later, making this confusing. The reason is
that we infer the name from the attributes for some types of spans, but
we do that at export-time.

This PR adjust this so that we send the inferred name & attributes to
the `tracesSampler`. This works reasonably enough. However, there is a
big caveat: This still only has access to the initial attributes that a
span has at creation time. This means that we'll never have e.g. the
`http.route` correctly there, because all `http.server` spans originate
from the http instrumentation where they do not have a route yet, and
then more specific instrumentation (e.g. express or fastify) add the
`http.route` to that span later. So the name will never be parametrized,
for example, for `http.server` spans, which is not ideal (as it will
still not match the final span exactly) but should still be better than
what we had so far (where the name would always just be e.g. `GET`).

Fixes #12944
Let's pin to Node `22.4` - there's a regression with `22.5` that affects us.

We can revert this pin after they do a Node.js release.
Add the dedupe integration as a default integration to `vercel-edge`.

Saw this discrepancy while working on the cloudflare sdk.
Removes `packages/browser/test/package/npm-build.js`.

`packages/browser/test/package/npm-build.js` contained three tests:

1. Test against `tslib_1__default` appearing in the bundle. I added a
browser integration test to replicate this
2. Test that adding a bundled package would work, our integration + e2e
tests pretty much cover this
3. Test that you can add bundle packages multiple times - our
integration + e2e tests also cover this

I also migrated `packages/utils/test/types/index.js` to use jest instead
of a plain script.
Something I noticed while doing my jest -> vitest conversions, we don't
need this anymore.
[Gitflow] Merge master into develop
@andreiborza andreiborza requested review from Lms24 and mydea July 19, 2024 08:30
@andreiborza andreiborza changed the base branch from develop to master July 19, 2024 08:37
@andreiborza andreiborza requested a review from a team as a code owner July 19, 2024 08:37
@andreiborza andreiborza merged commit 39f9e2a into master Jul 19, 2024
29 checks passed
@andreiborza andreiborza deleted the prepare-release/8.19.0 branch July 19, 2024 08:37
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

Successfully merging this pull request may close these issues.

5 participants