More consistent behavior of 'monocle:loaded'

[aronwoost]

This is somehow related to this ticket.

As I need to hook up with the components content (a.k.a. iFrame.contentDocument) directly, the 'monocle:loaded' event has special importants to me. I expect the event to tell me "all components displayed and rendered complete".

However I noticed that this doesn't work reliable. Sometimes pageDiv.m.activeFrame.m.component.properties.pageLength is still undefined when the event is fired.

Also another 'monocle:componentchange'is fired sometimes before and sometimes after 'monocle:loaded'.

In my perfect world 'monocle:loaded' would be fired when both components (in case of using slider) are initialized, measured and rendered. After that I would subscribe to 'monocle:componentchange' and can respond to user interactions.

[joseph]

That sounds like how things work in the ios5-unstable branch. Want to check it?

[aronwoost]

Ah, ok. Yes, will do.

[aronwoost]

First let me say thanks for implementing the "componentchange" changes. Indeed, the components is rendered and measured when "componentchange" is fired:

    function _onComponentChange(evt) {
        console.log(evt.m.component.lastPageNumber()); // cool!
    }

However, the "monocle:loaded" behavior remains unchanged compared to the master branch.

I have either this order: or that: In the first scenario both components are rendered on "monocle:loaded", in the later only one (which makes sense fron the logs, I guess). Also I can confirm, that this is not cache related.

[joseph]

This is quite odd, since the callback that fires loaded is only invoked when we hear two componentchange events (see Reader#setBook). I'll set up a test to try to reproduce it.

[aronwoost]

Let me know if I can provide you with something.

[joseph]

My commit presumptuously closed this. :)

I haven't been able to reproduce it. In my test (test/events/index.html) in the ios5-unstable branch, monocle:loaded always fired after all initial monocle:componentchange events. But I tweaked it so that it would fire even later — after the monocle:turn event when all iframes have moved to the correct page.

Can you check whether this addresses the problem?