[External] : Re: Prioritized event handlers

[John Hendrikx]

On 31/10/2023 21:03, Andy Goryachev wrote:
>
> Dear Martin:
>
> I understand John’s point: you want all the user installed filters and 
> handlers across all levels to be processed before switching to system 
> level processing. I also understand this prioritization proposal is 
> designed to push our existing set of system handlers to a separate 
> phase. But is that all we’re talking about here? I need some 
> clarification.
>
> If I understand this correctly, this proposal advocates making drastic 
> changes to the event handling mechanism.  Not only it introduces the 
> prioritization scheme (which I support), but it also alters the way 
> events are bubbled up, by introducing waves/phases.  I am afraid I 
> don’t see the use case for doing that.  In my opinion, the 
> prioritization scheme should work on handlers added to the same 
> EventTarget.  There should be no multiple waves - if an event gets 
> consumed by an event handler, the dispatching should stop (that change 
> I also support), if not - it bubbles up the hierarchy.
>
The waves (or calling of registered default handlers at the end) are 
needed to also allow users to install handlers at ancestors that would 
take priority over system handlers.  System handlers really shouldn't 
touch events unless they fully bubbled up, so the illusion that user 
event handlers are the sole event handlers in existance is being maintained.

The fact that event handler lists are being shared currently is 
incredibly confusing, so that when I do 
`scene.addEventHandler(KeyEvent.KEY_PRESSED)` I won't see navigation 
keys because they are all consumed by a focused control.  Instead, I 
should be seeing them, and if I choose not to act on them, only then 
should those events be used for navigation.  The current scheme is very 
business orientated (why would you ever not want to have navigation keys 
consumed?), while other use cases may simply want to disable keyboard 
control altogether (games) or have other more important uses for the 
cursor keys (like scrolling a map that does have clickable controls).

Of course, this would need to be done in a backwards compatible way as 
it has been this way for quite a while, so sure it you might see it as 
drastic, but its compatible, fixes a real flaw in FX event handling 
(unpredictability), allows several requested use cases (albeit at a low 
level), and its flexiblity will open up new ways of solving problems we 
haven't even thought of yet.

> Within a given control the order of event processing gets involved. If 
> a Control is subclassed the subclass should get first shot at the 
> event. The same is true for Behaviors and Skins. Beyond that I’m still 
> not clear if the behavior or skin should get the event first or if the 
> skin should get it via the behavior or the other way around. In any 
> case, you’ve got the control, the behavior, the skin, and all of their 
> subclasses trying to sort out the execution order.
>
> Hmmm.  I am not sure I understand exactly what you are saying.
>
> The way I understand this prioritization proposal is that we introduce 
> a priority associated with the handler (and not the filter).  The 
> proposal specifies 3 priority levels, I think there might be more, or 
> perhaps we even have an int priority (might be too much freedom, let’s 
> discuss).
>
> Specifically, I think there might be 5 levels, from high to low (the 
> names are just for the purposes of discussion):
>
> { AppHigh, SkinHigh, AppMedium, SkinLow, AppLow }
>
> This way there is absolutely no ambiguity in deciding which handler 
> gets invoked first.  I must note that SkinHigh, SkinLow should not be 
> available to the application code.  At the same time, 
> AppHigh,AppMedium,AppLow levels should not be available to 
> skins/behaviors.
>
> Side question: should filters also have priority?
>
I think we're over thinking it. Such fine grained priorities are not 
needed.  There are two parties involved, the user and JavaFX, hence two 
priorities should be enough.  FX can be responsible for ensuring its 
Skin/Behavior handlers are installed in the right order (they already 
have to do that), and users can be responsible for sorting out their own 
event handler order (they are after all in full control of that).

Alternatively, we can have no priorities and instead only have a 
mechanism that allows event handlers to forego handling an event, but 
register an interest in handling it if it is unconsumed (after 
dispatching completes for filters, or after bubbling completes for 
handler if we're inclined to offer both) -- I call these default 
handlers, similar to how there is a default exception handler on each 
Thread, each Event can have a (list of) default handlers.


> Based on this discussion (and I might be mistaken on this) it sounds 
> like you're trying to handle all this using this proposal, namely 
> registering event handlers with a prioritization scheme.
>
> Wait, I though that’s what you are proposing, based on the doc 
> https://gist.github.com/mstr2/4bde9c97dcf608a0501030ade1ae7dc1
>
> Are you proposing something else, or is something missing from the doc?
>
> Wouldn’t it be easier to just grab the event and pass it around using 
> Java method calls? Perhaps the call is handleEvent(). A control 
> implements handleEvent() by passing the event off to the behavior’s 
> handleEvent() which passes it off to the skin’s handleEvent(). The 
> skin sends it up the superclass chain by calling super.handleEvent(), 
> etc. so on.
>
> Wouldn’t this be a drastic departure from the established event 
> handling method?  Why call an empty method if we are not interested in 
> the event?  Or maybe I misunderstood you here, what are you asking?
>
> To summarize, I think the idea of explicit priority is a good idea as 
> it solves the current issue or exact ordering of handlers in the event 
> of skin change, to give one example.  I think we might also benefit 
> from a limited set of priorities (5) that reflect the reality of fx 
> having two sides - the application side and the “system”, or 
> skin/behavior, side.
>
> Personally, I find some other ideas problematic: I don’t see a good 
> use case for multiple waves in dispatching, as this represent a major 
> departure from the current mechanism, unless I am missing something.
>
> I don’t understand the paragraph about subclassing.  Perhaps you mean 
> that if we have a situation where one class extends the other, they 
> should coordinate the event handling.  For example, the base class 
> would declare the handling method, register it as a listener, for the 
> child class to override and get the events?  Or the base class should 
> not add any handlers, instead leaving it up to the child class?  But 
> that’s implementation detail.
>
> What do you think?
>
> -andy
>
> *From: *Martin Fox <martin at martinfox.com>
> *Date: *Tuesday, October 31, 2023 at 10:13
> *To: *Andy Goryachev <andy.goryachev at oracle.com>
> *Cc: *Michael Strauß <michaelstrau2 at gmail.com>, openjfx-dev 
> <openjfx-dev at openjdk.org>
> *Subject: *[External] : Re: Prioritized event handlers
>
> I understand John’s point: you want all the user installed filters and 
> handlers across all levels to be processed before switching to system 
> level processing. I also understand this prioritization proposal is 
> designed to push our existing set of system handlers to a separate 
> phase. But is that all we’re talking about here? I need some 
> clarification.
>
> Within a given control the order of event processing gets involved. If 
> a Control is subclassed the subclass should get first shot at the 
> event. The same is true for Behaviors and Skins. Beyond that I’m still 
> not clear if the behavior or skin should get the event first or if the 
> skin should get it via the behavior or the other way around. In any 
> case, you’ve got the control, the behavior, the skin, and all of their 
> subclasses trying to sort out the execution order.
>
> Based on this discussion (and I might be mistaken on this) it sounds 
> like you're trying to handle all this using this proposal, namely 
> registering event handlers with a prioritization scheme. Wouldn’t it 
> be easier to just grab the event and pass it around using Java method 
> calls? Perhaps the call is handleEvent(). A control implements 
> handleEvent() by passing the event off to the behavior’s handleEvent() 
> which passes it off to the skin’s handleEvent(). The skin sends it up 
> the superclass chain by calling super.handleEvent(), etc. so on.
>
> This would make for an easy sell to outside developers. We can tell 
> them that if they subclass a Control and implement handleEvent() they 
> will get events first during the system phase. The same is true if 
> they subclass a behavior or skin. They don’t need to buy into or even 
> see a complicated event prioritization scheme to get exactly what they 
> expect, namely first access to events.
>
> But, again, maybe I’m off base here. Let me know.
>
> Martin
>
>
>
>     On Oct 30, 2023, at 12:53 PM, Andy Goryachev
>     <andy.goryachev at oracle.com> wrote:
>
>     Dear Michael:
>
>     Thank you, this is very helpful.
>
>     Questions/Comments:
>
>     1. Does this proposal changes the way events are dispatched with
>     respect to priority?  In other words, does it first go through the
>     list of all handlers registred on the leaf Node (high priority
>     first, then lower, then lowest), then bubble up?  Or do they
>     propagate upwards looking for high priority handlers first, then
>     the process restarts for lower priorities, as I saw in some
>     previous emails?  (I could be mistaken)
>
>     2. Do you propose to abort event dispatching immediately after the
>     event is consumed?  This probably should be mentioned earlier in
>     the Motivation (the problem statement) section.
>
>     3. I wonder if three priority levels are sufficient.  Let me
>     explain.  We have two possible actors who can register an event
>     listener: the application code and the FX (or, rather more
>     specifically, the skin and its behavior, whatever that might be).
>
>     Application code might want to add handlers at three possible
>     priorities:
>
>       * App handler must always be called before any fx handler
>       * App hander does not care
>       * App handler must always be called after any fx handlers
>
>     For fx/skin handlers we might have fewer levels:
>
>       * Skin handler does not care
>       * Skin handler must be called after all other skin handlers
>
>     This situation maps to 5 priorities and 4 effective levels (or 5).
>
>     We should also mention the fact that when any actor adds two or
>     more handlers for the same event with the same priority, they get
>     invoked in the order added.
>
>     Would you agree, or am I missing some critical aspect of the
>     proposed solution?
>
>     Thank you
>
>     -andy
>
>     *From:*openjfx-dev <openjfx-dev-retn at openjdk.org> on behalf of
>     Michael Strauß <michaelstrau2 at gmail.com>
>     *Date:*Friday, October 27, 2023 at 19:41
>     *To:*openjfx-dev <openjfx-dev at openjdk.org>
>     *Subject:*Re: Prioritized event handlers
>
>     Here is the proposal:
>     https://gist.github.com/mstr2/4bde9c97dcf608a0501030ade1ae7dc1
>     <https://urldefense.com/v3/__https:/gist.github.com/mstr2/4bde9c97dcf608a0501030ade1ae7dc1__;!!ACWV5N9M2RV99hQ!NvO4B-fpHrjczoDGCoctorfNPX48w38MvW-LOf6ElCk0dBqFX_xPlETcr56POnEaBcwENrIOsX4OKDM0OGc07A$>
>
>     Comments are welcome.
>
>
>     On Fri, Oct 27, 2023 at 8:21 PM Andy Goryachev
>     <andy.goryachev at oracle.com> wrote:
>     >
>     > Would it be possible to create a proposal in the JEP format
>     outlining the proposed public API?
>     >
>     >
>     >
>     > Thank you
>     >
>     > -andy
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://mail.openjdk.org/pipermail/openjfx-dev/attachments/20231031/15b45d02/attachment-0001.htm>