fix(node): Record local variables with falsy values (v7) #11189
Closed
Conversation
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Fixes #10525 When writing the rollup config, we didn't include the debug build plugin. This led to things not be replaced properly as `profiling-node` bundles everything into a single file. This was also causing issues in our CI: https://github.com/getsentry/sentry-javascript/actions/runs/7804351046/job/21287026518?pr=10527 Backporting this fix to v7 so we can do a `7.100.1` release after we merge this in.
In CI currently on develop, we are stuck in a situation where we don't build bindings which means that the e2e tests always fail. Let's only run the profiling e2e tests whenever we change bindings, and make it a little more liberal for when we do run CI for changing bindings.
…10536) We are creating a replay breadcrumb when user feedback was submitted, however, the it was not typed correctly, which the timestamp not being included in the proper location.
Backport fixes to v7
This was apparently left over from some in-between state, users should actually use `spanToJSON(span)` there to get attributes. Backporting this (#10604 ) to v7!
As pointed out here, and I also did notice that myself, it is not super
nice - as you need to provide a generic integration to
`getIntegrationByName`, which is annoying to do in a type safe way, esp.
if you want to avoid deprecations:
```ts
const client = getClient();
const replay = client && client.getIntegrationByName && client.getIntegrationByName<ReturnType<typeof replayIntegration>>('Replay');
```
So IMHO a small utility `Sentry.getReplay()` is not unreasonable for
this 🤔
Co-authored-by: Abhijeet Prasad <[email protected]>
We didn't properly await sourcemaps flattening via sorcery before proceeding to upload them. The reason is that the async callbacks in `forEach` weren't awaited. A `for` loop is the better approach here. Wondering if we should lint against async `forEach` callbacks. This behaviour could have caused various inconsistencies. My suspicion is that the timing worked _well enough_ in most cases but we definitely want to properly await this step. Thanks to @MSDev201 for bringing this up! Unfortunately this likely won't fix #10589 as a whole :(
Add the Http 400 avoidance logic from our server-side `load` function instrumentation to the client-side wrapper. Didn't know that these errors were a thing on the client side but now that we know, we definitely don't want to capture them. Co-authored-by: Francesco Novy <[email protected]>
Add an `exports` field to the `package.json` for `@sentry/angular-ivy`. While it seems like regular Angular apps didn't need it, tools like `vitest` expect the field as soon as `type: "module"` is specified. --- Co-authored-by: Andrei Alecu <[email protected]>
Add exports for our semantic attributes keys from the SDK packages so that users don't have to install `@sentry/core` explicitly.
…#10633) This should make using these much easier: * add a default `op` to the spans, so users don't need to specify them. * Return the created span (or undefined), ensuring users don't need to do all the checking for op etc. themselves. Also make some small adjustments to ember & angular instrumentation to leverage some of these changes.
Backport changes to v7
Updates the changelog for v7.101.0. Note, we need to also pick the changelog updates onto develop (also 7.100.x changes) after we published.
…`sources` field (#10641)
…e` property (#10946) `Attachment.attachmentType` was changed to use a string union. The `attachment_type` header property should match so I added an extra type for this.
resolves #6880 Using `const chrome = ...` or similar breaks certain setups based on their bundler config. This makes changes to browser code so that we never declare a variable named `chrome`. I tried setting up https://eslint.org/docs/latest/rules/id-denylist to enforce this, but this unfortunately also looks as types declarations (so `type K = { chrome: ... }` is problematic.
chore: Backport Release for `7.106.0`
…ng (#11085) Adds INP span instrumentation to the old `BrowserTracing` integration.
`@sentry/types` does not align with `@sentry/core`, and it causes build errors when `exactOptionalPropertyTypes` is `true`. This patch fixes the type error by explicitly specifying `undefined`.
…gainst empty metric value Adds more interaction event entry names to the INP handler, and distinguish op between click, hover, drag, and press. Also adds a check to `metric.value` to drop any spans that would have empty exclusive time
|
Oops, wrong target branch |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
Backport to v7 of #10821