Angular and reactNative data callback should return updated data.

[graingert]

No description provided.

[graingert]

@benvinegar I don't think this stops adding any of the proposed changes from #803 but it fixes a real problem that exists now.

[graingert]

@benvinegar ping ^

[benvinegar]

The API is really bloated and I'd prefer not to add wrapDataCallback (because then it follows that we'd also need wrapShouldSendCallback and the others).

Can this be done without that?

[graingert]

@benvinegar I did add those, and it's not that much extra code, in exchange I re-factored the duplicate setXCallback APIs

the setXCallback APIs should be deprecated.

[benvinegar]

the setXCallback APIs should be deprecated.

Unfortunately, we have a goal with this project which is to change the APIs as infrequently as possible. And if we're going to do it, it better really be worth it. In this case, I'm not ready to change these APIs until we commit to moving towards the middleware style advocated in #803.

It's also not clear to me what problem this PR is solving, because there are no tests or example failure cases.

[graingert]

@benvinegar the issue is, setDataCallback supports two ways of changing the data sent:

1. The Good way: return a new object with the new data
2. The Naughty way: alter the internals of the object 'mutation' 🤮

however the angular and react-native clobber method 1 of the original functions by not returning the data through the stack of functions.

[graingert]

@benvinegar happy to rename wrapXCallback methods to _wrapXCallback methods

[benvinegar]

Okay, after a long conversation on IRC, I think this closes up some holes in the callback system.

Can you maybe add a cursory set of utils tests for wrappedCallback, and I'll accept.

[graingert]

@benvinegar any update on this?

[graingert]

@benvinegar any update on this?

[graingert]

@benvinegar any update on this?

[benvinegar]

@HazAT – just verifying, but this shouldn't conflict with getsentry/react-native-sentry, correct? (It has its own Raven.js plugin.)

@graingert – thanks and sorry for the slowness.

[graingert]

@benvinegar hey no worries I'm just glad for the existence of the project!

[graingert]

@benvinegar @HazAT yeah it looks like that should be updated too

[graingert]

It won't break it but now it will behave slightly different.