ref: Use optional chaining where possible #14954
Conversation
size-limit report 📦
|
23e3a85 to
aaae4ab
Compare
| @@ -57,7 +57,7 @@ export function reactRouterV3BrowserTracingIntegration( | |||
| afterAllSetup(client) { | |||
| integration.afterAllSetup(client); | |||
|
|
|||
| if (instrumentPageLoad && WINDOW && WINDOW.location) { | |||
| if (instrumentPageLoad && WINDOW.location) { | |||
There was a problem hiding this comment.
ah ok, I'm seeing there's a pattern where we no longer check on the definedness of WINDOW. I guess that's alright then? We should always get some kind of global object 🤔
There was a problem hiding this comment.
yeah, I think some global object will always be there, so it should not be necessary to check this - or can we think of a scenario where this is necessary? 🤔
There was a problem hiding this comment.
As long as we don't assume any browser-specific attributes to be present (which we don't in these changes) I think we're good!
|
I would have expected the bundle size decrease to be a bit larger but every lost byte is a good byte 😅 |
Co-authored-by: Lukas Stracke <[email protected]>
Yeah, had the same thought 😅 |
|
Is there a prefer optional chaining lint rule we could turn on? |
There is a rule actually, I will make a follow up PR adding this! |
This PR updates our code to use optional chaining, where possible.
This was mostly search-and-replace by
xx && xx.regex, so there may def. be some remaining places, but this should cover a good amount of usage.