Beiträge von jmugarra

I would do in two steps:
-First you should set your initial viewing angle in the "view" tag defining the hlookat vlookat and fov paramaters.
-The change of angle should be called in the scene "onstart", either directly inside it, or by defining an action.
-On the sample link you provided, autorotate gets activated after reaching the target angle. You can set this at the end of the last tween.

Here's a sample code that starts a scene looking up (vlookat=-50), and after it starts, it turns the vewing direction 45 degrees and levels it (vlookat = 0) using the default tween type, during a period of 1 second. After leveling the view, autorotate gets started.

Hope it helps.

<scene name="scene01" title="Sample Scene" onstart="changeAngle();" ...>


		<view hlookat="0 vlookat="-50" fovtype="MFOV" fov="120.000" ... />


		<image>
		...
		</image>


		<hotspot name="spot1" .... />
		...
		</hotspot> 
	</scene>




	<action name="changeAngle">
		tween(view.hlookat, 45, 1);
		tween(view.vlookat, 0, 1, default, set(autorotate.enabled, true));
	</action>

The default behavior is what you need. The bat files used to create your tour automatically added the hlookat="0" and vlookat="0" to your "view" elements, which reset the view with every scene change. Erase those and you’re good to go!

I just came across this post. Congrats to both you and Tessa for a truly engaging and inspiring experience.

I've tried that, but still no luck.
I figured the problem must be related to the processing speed and number of threads. Layer B starts to get created before Layer A has been completed. So I also tried using nexttick in the initial action, trying to first create Layer A, and then Layer B, but I can't figure it out.
I've come up with a somewhat clumpsy solution: I delay the creation of Layer B for a small amount, giving it time to complete the previous layer. It works, but it seems to me there must be a more elegant way to work around this, since I'm afraid it'll become a mess when dealing with many layers that depend on previous ones for their sizing...
This is my temporary solution:

<krpano version="1.19" title="Virtual Tour">


	<action name="startup" autorun="onstart">
		create_layer_A();
		delayedcall(0.03, create_layer_B());
		delayedcall(0.03, check_values());
	</action>


	<action name="create_layer_A">
		addlayer(layer_A);
		set(layer[layer_A].type, container);
		set(layer[layer_A].width, 40%);
		set(layer[layer_A].height, 10%);
		set(layer[layer_A].align, topleft);
		set(layer[layer_A].x, 0);
		set(layer[layer_A].y, 10%);
		set(layer[layer_A].bgcolor, 0xfbfbfb);
		set(layer[layer_A].bgalpha, 1);		
		layer[layer_A].updatepos();
	</action>


	<action name="create_layer_B">
		addlayer(layer_B);
		set(layer[layer_B].type, container);
		set(layer[layer_B].width, 10%);
		set(layer[layer_B].height, 10%);
		set(layer[layer_B].align, topleft);
		set(layer[layer_B].x, get(layer[layer_A].pixelwidth));
		set(layer[layer_B].y, 30%);
		set(layer[layer_B].bgcolor, 0xaafff6);
		set(layer[layer_B].bgalpha, 1);
		layer[layer_B].updatepos();
	</action>


	<action name="check_values">
		trace('Layer A width: ', get(layer[layer_A].width));
		trace('Layer A pixelwidth: ', get(layer[layer_A].pixelwidth));
		trace('Layer A X: : ', get(layer[layer_A].x));
		trace('Layer A pixelX: : ', get(layer[layer_A].pixelx));
		trace('Layer B width: ', get(layer[layer_B].width));
		trace('Layer B pixelwidth: : ', get(layer[layer_B].pixelwidth));
		trace('Layer B X: : ', get(layer[layer_B].x));
		trace('Layer B pixelX: : ', get(layer[layer_B].pixelx));
	</action>


</krpano>

OK, so a weird update after additional testing...
It works properly if viewed in an incognito window on Chrome, returning the right pixel values and positioning the Layer B right after Layer A. It also works on MS Explorer 11 and Edge straight away, and on an Android phone using Chrome. But it doesn't work using a normal tab of Chrome, or Firefox.
Does this make any sense?

Hello everyone.
I now I must be missing something really obvious here, my coding skills are very limited, so I apologize in advance...
I'm trying to recover the pixel width of "Layer A" (which has been created using relative parameters), and then use that value for the x position of "Layer B".

This is my simple test code:

<krpano version="1.19" title="Virtual Tour">


	<action name="startup" autorun="onstart">
		create_layer_A();
		create_layer_B();
		check_values();
	</action>


	<action name="create_layer_A">
		addlayer(layer_A);
		set(layer[layer_A].type, container);
		set(layer[layer_A].width, 40%);
		set(layer[layer_A].height, 10%);
		set(layer[layer_A].align, topleft);
		set(layer[layer_A].x, 0);
		set(layer[layer_A].y, 10%);
		set(layer[layer_A].bgcolor, 0xfbfbfb);
		set(layer[layer_A].bgalpha, 1);
	</action>


	<action name="create_layer_B">
		addlayer(layer_B);
		set(layer[layer_B].type, container);
		set(layer[layer_B].width, 10%);
		set(layer[layer_B].height, 10%);
		set(layer[layer_B].align, topleft);
		layer[layer_A].updatepos();
		set(layer[layer_B].x, get(layer[layer_A].pixelwidth));
		set(layer[layer_B].y, 30%);
		set(layer[layer_B].bgcolor, 0xaafff6);
		set(layer[layer_B].bgalpha, 1);
	</action>


	<action name="check_values">
		trace('Layer A width: ', get(layer[layer_A].width));
		trace('Layer A pixelwidth: ', get(layer[layer_A].pixelwidth));
		trace('Layer A X: : ', get(layer[layer_A].x));
		trace('Layer A pixelX: : ', get(layer[layer_A].pixelx));
		trace('Layer B width: ', get(layer[layer_B].width));
		trace('Layer B pixelwidth: : ', get(layer[layer_B].pixelwidth));
		trace('Layer B X: : ', get(layer[layer_B].x));
		trace('Layer B pixelX: : ', get(layer[layer_B].pixelx));
	</action>


</krpano>

Layer B's x value should be equal to Layer A's width in pixels, but it's "0". And the traced value is also "0" in every "pixel" dimension.
If I change set(layer[layer_B].x, get(layer[layer_A].pixelwidth)); to set(layer[layer_B].x, get(layer[layer_A].width)); it moves to where I want it, because it takes de relative value (40%). But I really need to work with pixel values.
Any help would be much appreciated.

I've tried that since you suggested it, but still no luck. Thanks for the input, though.

Hello everyone.
I'm trying to give my combobox a more "flat" look, and so far I've succeeded in the flash version, which I thought would be the toughest one.
Regarding the HTML5, by reading the documentation, I assumed it would be just a matter of adding the "customstyle" attribute and directing it to a new PNG image created with the provided default combobox base-image as a template. So I created my flat-looking png, dropped it into the skin folder, and added the following code:

<style name="combobox" devices="desktop|flash" url="%SWFPATH%/plugins/combobox.swf" alturl="%SWFPATH%/plugins/combobox.js" native="false" rowcount="20" />
	<style name="combobox" devices="iphone" url="%SWFPATH%/plugins/combobox.js" cbdesignscale="auto" customstyle="%SWFPATH%/skin/combobox_flat.png|0x797979|0x797979|0x797979|0x797979|0xf0f0ee|0xf0f0ee|0xf0f0ee" />
        <style name="combobox" devices="ipad" url="%SWFPATH%/plugins/combobox.js" customstyle="%SWFPATH%/skin/combobox_flat.png|0x797979|0x797979|0x797979|0x797979|0xf0f0ee|0xf0f0ee|0xf0f0ee" />
	<plugin name="cb_ENG" style="combobox" align="righttop" x="20" y="10" keep="true" >
		<item name="title" 		caption="VIEWS"  />
		<item name="1" 	        caption="1"             onclick="loadscene(scene_A01, null, MERGE, BLEND(1));"/>
		<item name="2" 		caption="2"		onclick="loadscene(scene_A02, null, MERGE, BLEND(1));"/>
	         .....
                <item name="9" 		caption="9"		onclick="loadscene(scene_A09, null, MERGE, BLEND(1));"/>
	</plugin>

As a result, the html5 version keeps showing the default look for the combobox.
I have the feeling I'm missing something obvious here... So I apologize if this is the case. Anyway, any hint will be much appreciated.

Hello, everyone.
I like adding a little vignetting to my panos using a mask, like this:

<plugin url="%CURRENTXML%/images/darkmask.png" name="mask" devices="all" origin="lefttop" width="100%" height="100%" enabled="false" keep="true" zorder="-10" visible="true" alpha="0.4"/>

This used to work just fine with Krpano 1.0.8.12, even for html5 devices, such as the Ipad and the Iphone. Unfortunately, after upgrading to 1.0.8.14, this doesn't seem to work 100% anymore. I mean, the vignetting is there, but it covers the hotspots, rendering them useless in html5 devices. I've read in the documentation that for html5 zorder must be in the 0-99 range, no negative values, so I've tried gibing the mask a zorder of 0 and some higher values for everything else, but still no luck (with 1.0.8.12, a zorder of "-10" worked perfectly).
So right now, for the Ipad/Iphone, I must choose between vignetting and Aldo's awesome Gyro plugin. As a temporary (I hope) solution, I've changed "devices" to "flash", so I lose the mask in html5, but keep the gyroscope control.
I don't know if this is a bug, but it seems like zorder, html5 and 1.0.8.14 aren't getting along too well.
Any input will be much appreciated.

Juan

Hi there.
Just a thought... As I understand it, the problem is linking the flyin/flyout action and the togglepause action to the same control (click).
You could link one of them to a different user input.
Aldo Hoeben developed a plugin that adds double click events. I haven't tried it yet, but I'm sure it'll work perfectly, and it could solve your problem.
Cheers

OK, I edit:
After a breifly testing the doubleclick plugin, it seems to work only in the whole viewer, not being able to set that event for a hotspot. Sorry...