goodboy

> i still think #1359's approach is better for the simple reason it almost does not touch the internals of the library except for adding some signals and some minor...

> I noticed that many of the existing examples have lost their axis display on this PR. Yeah haven't touched any existing tests or examples yet; just been trying to...

@pijyoi ahh also click drags will be broken for now since the view box has been modified to be broadcast only in this PR. I need to figure out a...

Actually so missing axes is actually due to removal of the `visibleAxes = ['left', 'bottom']` default thing inside `.setAxisItems()`, which I've now fixed and will push shortly.

Lol, ok so I added a method `@eventrelay` [a few commits ago](https://github.com/pyqtgraph/pyqtgraph/pull/2162/commits/64beeaf52ff4531c4e2ce0bb7575f8a12e87a904#diff-39b99d73a5dd4c001f9fe2d6c3e6276ad2bcec3f5493ea65873450be51bed7cbR80) which does work for a pseudo-non-invasive approach for all this but, obviously it requires a permanent change to...

> i think you are really over-complicating things, why not just changing the event handlers to simple signal throwers and then connecting those signals to a almost-copy-paste version of the...

Oh right and alternatively if we did want to *hack around* requiring `Signal` to be defined at class compile time we can use [this recipe for `Signal` implemented as python...

> Note that the file that you linked to has the following license: Yeah I'm honestly not super fond of this approach - I think we can greatly simplify by...

BTW y'all I would say this is ready for experimental testing though probably a small PR to update the `ViewBox` default handlers is going to be needed before this is...

Ok I'm giving y'all until end of week to give feedback on this approach or I'm closing it and AGPL-ing the recent improvements we've made and moving them out into...