Beiträge von loki

Nice take on the 1.21 features

Coolbox

Thanks for sharing Tuur, great updates.

It would be very kind to give a lead.
Agoras payment plans put off my clients.

Hi Hamid, could you add synced scrolling of iframe content in sync view, so everyone sees the same content when iframe-popup gets scrolled by the host?

Dear Klaus,

I noticed when control.zoomtocursor is set to true the 2-fingers gesture zoom / pinch-zoom will not work with laptop touchpad/trackpad (on chrome, firefox & safari).
Is there any workaround I could apply?

Best,
Max

Not tablet but ios specific:
Using ios app GoCoEdit to do some code reading and minor coding/bugfixing is pretty useful. Sometimes helpful to do so in a different environment than at the same old workstation/desk.

Thanks for the fast fix - works

Same here. Please do send an update.

Is it possible to extend the plugin to jump to a backup host, in case the first provider throws errors?

Do as Klaus said inside your .html file.
The embedpano command needs the addition
„consolelog: true“
to display all krpano console messages also in the browser console.

Hi all,

I'm looking for an experienced krpano developer who is interested in creating a handheld UX (mobile & tablet) of my museum application withing the next 4 weeks.

The tour uses custom Javascript actions to do some custom drag- and scaling-css stuff and needs action adjustments, e.g. including touchstart, touchmove, touchend to work on mobile devices.
Also some basic work is needed designing and creating layers and hotspots for handheld, which only exist for desktop right now.

If anyone is interested please send a private message and lets talk about budget and your references. Then we can look at the existing tour in a screen sharing session and assess effort and further steps.

Thanks for taking the time to read my request.

Best regards,
Max

Thank you Hamid for the quick fix.
Through muazkhans server it works again. I wonder if there will be official cdns for webrtc some day.

I get one error msg in chrome on win10 (firefox doesn't show this error):

DevTools failed to load source map: Could not load content for mydomain.com/socket.io.js.map: HTTP error: status code 500, net::ERR_HTTP_RESPONSE_CODE_FAILURE

Kind regards
Max

Hi Hamid,

a server used by your plugin is down and there is no way to connect to a tour with other devices and video/audio does not work.

In console it says cors-error but I guess that's due to them being down.

Access to XMLHttpRequest at 'https://rtcmulticonnection.herokuapp.com/socket.io/?userid=...' from origin 'https://...' has been blocked by CORS policy: No 'Access-Control-Allow-Origin' header is present on the requested resource.

https://rtcmulticonnection.herokuapp.com says "Offline for maintenance".

How can my clients manage to use this plugin again?

Hi Klaus,

while in Firefox the hotspots will be set to display:"none" and pointer-events:"none", I noticed in Chrome the display setting is not changed but also pointer-events:"none" gets set from "auto" to "none".
I believe it has something to do with either tx/ty/tz or ox/oy values. If I change theses values in a drag action, the display:"none" will be set sooner when changing the view (or later depending on final drag values).
I sent the project by mail some weeks ago, but I totally understand should the big update be more important right now.

Best regards,
Max

//edit:
There is a dirty workaround by adding a css class containing "display:block !important;" and "pointer-events: auto !important;" to these hotspots with a for loop in the onviewchange event. I limited the event so it would not fire with each redraw but I guess the performance still suffers a lot?

This helped me (there is translators from haml to html and scss to css if you need):
https://codepen.io/Quelltextfabrik/pen/LYbrBa

Hi all,

scale=".5" is debugged as string, scale="0.5" as value. I didn't spot problems with it yet, but wanted to make sure it is ok to use both?

Thanks for info.

Hi,

I also have this issue with css3d hotspots on firefox only and suspect cube faces, not bounding-box, see video.
All disappearing hotspots get overridden with "display:none" as vincent also reported.

Is there a way to fix this?

Thanks,
Max

The menu in this example is made with dynamically created hotspots in an array for VR and helped a lot creating my own menu:

https://krpano.com/releases/1.20.…tml5=only+webgl
https://krpano.com/viewsource.htm…r/krpano_vr.xml

This example/tutorial might help: https://krpano.com/forum/wbb/inde…ad&postID=38120

Nice one Tuur. My brain hurts thinking about the possibilities..

Awesome that the dolly is showing in all panos.

In p5 gympano the balls and dolly are not showing. Is it intentional?

Hi all,

could someone help solve my math puzzle for dragging hotspots in 3d space like in Klaus' Abu Simbel example, but along "flat" hittestareas, not in spherical motion?

Based on Klaus' example for vrhittesting (xml) I'm adding hotspots at the hitpoint position (red dot) in 3d space and get tx, ty, tz, and ath values of the hittestareas for the hotspots. I then would like to drag the hotspots along the corresponding hittestarea (considering the tx, ty, tz and ath values).

How does the math need to be adjusted in Klaus' Abu Simbel dragging code below?
Is it a different coordinate transformation than in this example or a different tx/ty/tz calculation?

<action name="drag" scope="local">
		<!-- drag the spots around in 3D space -->
		copy(last_mouse_x, mouse.x);
		copy(last_mouse_y, mouse.y);
		copy(last_view_tx, view.tx);
		copy(last_view_ty, view.ty);
		copy(last_view_tz, view.tz);


		asyncloop(caller.pressed,
			<!-- kind of working condition for four-wall room -->


			if((caller.ath == 90) OR (caller.ath == '-90')
			, 
				trace('wall 1 or 3);
				copy(x, tempwalltx);
				calc(y, caller.ty - view.ty);	
				calc(z, caller.tz - view.tz);
			,
				trace('wall 2 or 4');
				calc(x, caller.tx - view.tx);
				calc(y, caller.ty - view.ty);	
				copy(z, tempwalltz);
			);
		
			spacetosphere(x,y,z, h,v,d);
			spheretoscreen(h,v, screen_x,screen_y);


			calc(screen_x, screen_x + (mouse.x - last_mouse_x));
			calc(screen_y, screen_y + (mouse.y - last_mouse_y));
			copy(last_mouse_x, mouse.x);
			copy(last_mouse_y, mouse.y);


			screentosphere(screen_x,screen_y, h,v);
			spheretospace(h,v,d, x,y,z);


			calc(caller.tx, x + view.tx + (view.tx - last_view_tx));
			calc(caller.ty, y + view.ty + (view.ty - last_view_ty));
			calc(caller.tz, z + view.tz + (view.tz - last_view_tz));
			copy(last_view_tx, view.tx);
			copy(last_view_ty, view.ty);
			copy(last_view_tz, view.tz);	
		);	
	</action>

I get good results for dragging along the hittestarea, but only close to the initial hotspot position, if I put the hittestarea tz or tx value as "z" or "x" in the beginning of the asyncloop, depending on ath (for 0/180 or for -90/90) of the hittestarea, but the further away the hotspot gets dragged from its initial position, the more the dragging effect decreases, because it still tries to go spherical I guess.

Does someone have an advise please?