No access to time intervals when they're not enabled via 'show:true'

[wimbarelds]

No access to time intervals when they're not enabled via 'show:true'

[oucil]

On second (sober) thought, rather than put the pref in the .TimeCircles() instantiation, it should be in the instantiation of the callback function.

[wimbarelds]

Seconds is always being shown (in the "total" attribute), You're absolutely right about the callback needing to be triggered for units that are not shown. Should be a relatively easy fix.

[oucil]

You're right that it's included as total but the problem is it remains static at 119 for the entire period between 120 and 60 when seconds is set to show:false. If you could make an active value available, that would be excellent!

[wimbarelds]

I'm making it so it will trigger the callback for units not being shown.

[wimbarelds]

This might actually be more difficult than I had hoped, because the value of units not shown / shown are a bit errr... Strange? The reason is that, when you show hours, hide minutes, show seconds, then seconds goes up to 3600 instead of 60 (since there's 3600 seconds in an hour).

So, I'm not sure which values to throw into the listeners right now. One thing I am noticing so far is that everyone is using the "total" value almost exclusively (and understandably so).

[oucil]

Is there a reason for sending anything but seconds to the callback? I know it's a pain to break backwards compatibility, but a standard unit would really make it simpler.

[wimbarelds]

The intent was that you could do something like:

$("#timer").TimeCircles().addListener(function(unit, value, total) {
    if(unit === "Hours") $("ClockHourSound").play();
});

But truthfully you could do the same with: if(total % 3600 == 0), true it would get a little bit more complex to get the current number of hours, but that's not a major issue. That could be a request / call.

At this point I'm not entirely sure if I should change it or not. I'll need a little while to figure out if I can think of an elegant fix.

[wimbarelds]

I've put out a new release, version https://github.com/wimbarelds/TimeCircles/releases/tag/1.1.0

It resolves this issue by adding a new parameter to .addListener(). You can use it as follows:

function callbackFunction(unit, value, total){
    // Your code here
}

$("#timer").TimeCircles().addListener(callbackFunction, "all");

Please let me know if this new release has what you're looking for and allows you to do what you want. For now I will close this, but please re-open it if you still have problems.

[oucil]

I'll give it a shot and let you know how it goes, thanks for this!

[oucil]

I'm not getting anything now, neither 1.1.0 or 1.1.1 are displaying anything now, and I'm not getting any debug info in firebug either :( I've reduced the complexity right down to a pretty basic call:

var idle_timer = $('div#idle_timer');
if (idle_timer.TimeCircles != null) { idle_timer.TimeCircles().destroy(); }
idle_timer.data('timer',1800);
idle_timer.TimeCircles({ time: { Days: { show:false }, Hours: { show:false }, Minutes: { color: '#FFF' }, Seconds: { color: '#FFF' } }, circle_bg_color: '#4D8DC1' });

In 1.0.5 I'm at least seeing the two circles (although they're counting up and not seeing the dynamically added data-timer attribute, I'm guessing I was probably drawing from the master as you were committing and the support didn't go back that far maybe?), but as soon as I replace with either 1.1.0 or 1.1.1, I'm getting nothing, no circles, no errors, no debug info.

Any ideas?

[oucil]

BTW, I can't reopen the issue, doesn't seem I've got access anymore, did you tighten things down?

[wimbarelds]

Nope. I didn't tighten it down. It looks like the .destroy line is causing problems, I'm looking at why.

[oucil]

Ok, not sure why I can't re-open, but option is no longer there. In any case, I tried commenting out the .destroy() line and it had no effect.

[oucil]

Hmmm, strike that, thought I'd cleared the cache but apparently not, the .destroy() commented out brought it back this time.

[wimbarelds]

I see the problem.

I used .attr("data-tc-id", cur_id) instead of .data('tc-id', cur_id). However upon destroy it should remove the tc-id. I'd not changed it to remove the attribute instead of the data, so it thought there was still something there.

See 21b7212 for the fix.

[oucil]

Doesn't seem to have worked for me. As long as I include that check and destroy in place, it's not displaying anything, and not providing any error to debug from. Trying some different stuff, and will let you know if I find anything, but I may have to fall back on manually removing the elements themselves for the time being, haven't gotten to the point where I can actually test the "all" feature you added above.

[wimbarelds]

Can you give me a url where I can see this? Because I tested it locally and it solved the problem.

[oucil]

I'm creating another demo and will make it available shortly, I've run into another issue. I temporarily added something that manually kills the children of my main container rather than using destroy, which finally let me test the "all" functionality, and I'm afraid that somewhere in there I'm experiencing a major memory leak now, my memory usage for Firefox jumps from about 700mb on average, up and down every couple seconds, adding about 200mb each time, and finally crashing it at around 2500mb usage. This obviously has something to do with how I've nested the 4 tiers however this wasn't something I was experiencing before.

[wimbarelds]

If you do not destroy the instance, but do remove the elements- then yes there will probably be a memory leak. But that's not wrong, because you're creating a dozen instances without removing old ones properly.

[oucil]

Wasn't doing that in earlier builds, but maybe I just didn't notice or it wasn't as severe, I'd just assumed that the events tied to an element would be referentially destroyed but if not, so be it, that explains that.

[wimbarelds]

Looks like I've found out why it wasn't properly working. I was setting $("#idle_timer").attr("data-tc-id", id), and also removing it. However, I was also trying to read it with .data('tc-id') which makes it a data attribute as well as an html attribute. So now I removeAttr() as well as removeData() on it, and it works. If this works, please let me know and I'll push a new version. 9dcca68

[wimbarelds]

Since you haven't replied, I assume that it's working. I'm going to push the new version now.