krpano 1.0.8.14

Lieber Besucher, herzlich willkommen bei: krpano.com Forum. Falls dies Ihr erster Besuch auf dieser Seite ist, lesen Sie sich bitte die Hilfe durch. Dort wird Ihnen die Bedienung dieser Seite näher erläutert. Darüber hinaus sollten Sie sich registrieren, um alle Funktionen dieser Seite nutzen zu können. Benutzen Sie das Registrierungsformular, um sich zu registrieren oder informieren Sie sich ausführlich über den Registrierungsvorgang. Falls Sie sich bereits zu einem früheren Zeitpunkt registriert haben, können Sie sich hier anmelden.

sachagriffin

Meister

Beiträge: 1 857

Beruf: Virtual Tours - Photography - Krpano developer

Dienstag, 15. November 2011, 22:57

Tablet / Mobile Placeholder %s supported?

So I tried using the new placeholder feature..
It showed only the preview.. Is this supported for html5 devices?
Flash worked fine even with levels place holders.

Quellcode

			<mobile><left  url="%SWFPATH%/images/<?php print $pano[$i] ?>.tiles/tablet_l.jpg" />
				<front url="%SWFPATH%/images/<?php print $pano[$i] ?>.tiles/tablet_f.jpg" />
				<right url="%SWFPATH%/images/<?php print $pano[$i] ?>.tiles/tablet_r.jpg" />
				<back  url="%SWFPATH%/images/<?php print $pano[$i] ?>.tiles/tablet_b.jpg" />
				<up    url="%SWFPATH%/images/<?php print $pano[$i] ?>.tiles/tablet_u.jpg" />
				<down  url="%SWFPATH%/images/<?php print $pano[$i] ?>.tiles/tablet_d.jpg" /></mobile>

became

	Quellcode
1 2 3	<mobile> <cube url="%SWFPATH%/images/<?php print $pano[$i] ?>.tiles/tablet_%s.jpg" /> </mobile>

KRPano Developer: Portfolio ::Gigapixel Tagging Solutions - Porfolio 2 :: Facebook :: Twitter :: reddit.com/r/VirtualTour

Zum Seitenanfang

klaus.krpano

Administrator

Beiträge: 13 499

Donnerstag, 17. November 2011, 15:26

Hi,

Zitat von »sachagriffin«

So I tried using the new placeholder feature..
It showed only the preview.. Is this supported for html5 devices?

yes, it should work in all cases...
can you please post/show a not working example?

best regards,
Klaus

Zum Seitenanfang

sachagriffin

Meister

Beiträge: 1 857

Beruf: Virtual Tours - Photography - Krpano developer

Donnerstag, 17. November 2011, 18:46

PM sent.

KRPano Developer: Portfolio ::Gigapixel Tagging Solutions - Porfolio 2 :: Facebook :: Twitter :: reddit.com/r/VirtualTour

Zum Seitenanfang

rocher

Anfänger

Beiträge: 7

Freitag, 18. November 2011, 21:16

Hi klaus,

I recently updated to last version and discovered that some 'if' actions don't work as expected.
For example, this works perfectly in version 1.0.8.14 Update 3 (2011-09-28)

	Quellcode
1	if( hotspot[ mapping ].loadedbytes GT hotspot[ get( h )].bytes, ...

but not in version 1.0.8.14 Update 6 (2011-10-21).
Do you have any idea?

-- Francesc

Zum Seitenanfang

klaus.krpano

Administrator

Beiträge: 13 499

Samstag, 19. November 2011, 13:57

Hi,

Zitat von »rocher«

I recently updated to last version and discovered that some 'if' actions don't work as expected.

thanks for the report!

the problem was the new if parser that was introduced in build 2011-10-11 to the fix an other if problem,
this parser doesn't assume that whitespaces were possible in variable names/paths,

this will be fixed in the next release,
as workaround you could remove the whitespaces in the array indices,

best regards,
Klaus

Zum Seitenanfang

rocher

Anfänger

Beiträge: 7

Montag, 21. November 2011, 02:35

Hi,

Thanks Klaus, that solved part of the problem.

Here you can see other problems I've encountered:

http://vtour.enacse.com/?c=fundacio-vila…u-solterra&kl=t

and

http://vtour.enacse.com/?c=fundacio-vila…-10-21_update-6

are the same virtual tour using different krpano versions. Please compare the behavior of the buttons at the left.
Its impossible with the latest version of krpano to do anything, so it's absolutely unusable here.

Also, after reloading, look for "Sala 0.2" scene and play with the photo gallery inside it (click one the photographs with a red spot).

Clearly, one of the last three updates introduced a serious bug.
Hope it's easy to solve to not to wait until next release. Any idea on how to solve this?

-- Francesc

Zum Seitenanfang

klaus.krpano

Administrator

Beiträge: 13 499

Montag, 21. November 2011, 12:58

Hi,

this problem is also related to the new if parser (which was introduced in build 2011-10-11),

you were calling your 'openthumbs' action always with 'space/blank' as parameter (this will be resolved to an empty string as parameter):

	Quellcode
1	openthumbs( )

and inside the 'openthumbs' action there is the following if:

	Quellcode
1	if( %1 != NEXT,

which becomes, after resolving the parameter placeholders, to:

	Quellcode
1	if( != NEXT,

the old if parser was able to solve such conditions,
but the new one not, there such conditions always resolve to false,

I have fixed that now, now the new if parser can handle also such conditions,

btw - as workaround just remove the blank when calling actions (I have seen that happens several times in your code),
if there are no additionally bug reports anymore, the new/fixed version might be able to be released very soon (probably today...)

additionally I'm also thinking about if 'empty string' parameters should resolve to 'null'... but this might eventually introduce other compatibly problems so probably not...
are there any opinions to this?

best regards,
Klaus

Zum Seitenanfang

sachagriffin

Meister

Beiträge: 1 857

Beruf: Virtual Tours - Photography - Krpano developer

Montag, 21. November 2011, 15:03

I'd check to see the behavior in js and php and copy that.

KRPano Developer: Portfolio ::Gigapixel Tagging Solutions - Porfolio 2 :: Facebook :: Twitter :: reddit.com/r/VirtualTour

Zum Seitenanfang

Zephyr

Profi

Beiträge: 1 003

Wohnort: Netherlands

Beruf: Web developer

Montag, 21. November 2011, 15:19

Zitat

additionally I'm also thinking about if 'empty string' parameters should resolve to 'null'... but this might eventually introduce other compatibly problems so probably not...
are there any opinions to this?

In javascript an empty string evaluates as a falsy just like null, false, undefined, 0, NaN.

So

	Quellcode
1 2 3	var x = ""; console.log(x == false); //true console.log(x == 0); //true

any other variable that becomes, null, undefined, 0, NaN, should alsoo parse false;

but this doesnt mean that

	Quellcode
1 2 3	var x = ""; console.log(x == null); //false console.log(x === null); //false

what if you did a substring of somesort, and the result becomes an empty string, it would be weird if that validated null.

btw javascript is weird...

Quellcode

var y = "";
var x = null;
if(x == false){
  console.log('x is false');
};

if(y == false){
  console.log('y is false');
};

if(x == y){ //false == false ?
  console.log('so x and y are equal?');
}else {
  console.log('nope');
};

x is not equal to y.

long story short, I think it would better to evaluate it as false, then null. I feel null is special, as something thats not defined. In php you have isset() and empty(). An empty string would return as empty (true) but still isset (true) while null wouldnt be set and empty would fail to parse. I think that php would be the best way to follow for clearness, but perhaps javascript is better since you have krpanoJS,

Zum Seitenanfang

rocher

Anfänger

Beiträge: 7

Montag, 21. November 2011, 16:14

Hi all,
Thank you very much Klaus for your quick response. I'll try to fix that.

But wait, we're talking about WHITE spaces. They are not part of the semantics, just part of the syntax. Please don't consider them as a token, just a separator.
IMHO empty strings should be ALWAYS enclosed between " or ' characters. That's the most usual treatment in almost any programming language, no matter how ancient or modern.
'openthumbs( )' should be exactly the same as 'openthumbs()' and 'openthumbs( )' etc, no matter how many white spaces you put there: that's not an argument at all, and for sure that's not the 'null' argument.
(That's taken from C, C++, ObjectiveC, Java, Javascript, PHP; even Lisp, EmacsLisp, prolog, latin, ancient greek ... ;-)

)

In other words, please don't confuse a single white space with the empty string or the 'null' value or the 'false' boolean.
What about indentation, then? That's a concatenation of empty strings or nulls? What a weird language ...

-- Francesc

Zum Seitenanfang

klaus.krpano

Administrator

Beiträge: 13 499

Montag, 21. November 2011, 18:06

Hi,

Zitat

In javascript an empty string evaluates as a falsy just like null, false, undefined, 0, NaN.

sorry, there is a small misunderstanding, I don't mean the comparing,
I'm mean when passing an 'empty parameter',
e.g. when an action will be called like this:

	Quellcode
1	test(,,,,);

then these empty parameters will be 'empty strings' - the %1,%2,%3 and so on will resolve to empty strings (''),
and I thinking about if they should resolve to null instead in such case...

Zitat

IMHO empty strings should be ALWAYS enclosed between " or ' characters. That's the most usual treatment in almost any programming language, no matter how ancient or modern.
'openthumbs( )' should be exactly the same as 'openthumbs()' and 'openthumbs( )' etc, no matter how many white spaces you put there: that's not an argument at all, and for sure that's not the 'null' argument.

I agree of course,
the detection of white spaces as parameter will be fixed in the next release,
but when there is no argument during call and a argument placeholders (like %1) were used in the action, then this placeholder will be resolve to null (already in the current versions),

best regards,
Klaus

Zum Seitenanfang

sachagriffin

Meister

Beiträge: 1 857

Beruf: Virtual Tours - Photography - Krpano developer

Montag, 21. November 2011, 18:29

I'm pretty sure they resolve to null in php.
Most disciplined programmers fill unused parameters with null.
I think whatever you do, everyone is going to need to check over every null and empty string evaluation when you make your decision for the next version.

"

	Quellcode
1	test(,,,,);

then these empty parameters will be 'empty strings' - the %1,%2,%3 and so on will resolve to empty strings (''),
and I thinking about if they should resolve to null instead in such case...
"

best regards,
Klaus[/quote]

KRPano Developer: Portfolio ::Gigapixel Tagging Solutions - Porfolio 2 :: Facebook :: Twitter :: reddit.com/r/VirtualTour

Zum Seitenanfang

Zephyr

Profi

Beiträge: 1 003

Wohnort: Netherlands

Beruf: Web developer

Montag, 21. November 2011, 19:32

but don't you have something similar already in your code klaus?

for instance:

loadpano('test.xml', , MERGE); is the same as loadpano("test.xml, null, MERGE); no matter the amount of spaces.

But what the others say is true. a space should be quoted if it wants to be passed as a string variable. If an action requires an parameter, and its not given, then the parameter would be null. If I pass null as a parameter, I want to define a defaultvalue in the action.

take this action:

	Quellcode
1 2 3 4 5 6 7	<action name="someaction"> if(%1 === null, set(somevar, 'defaultvalue'); , set(somevar, %1); ); trace(somevar); </action>

given: someaction();
expected: 'defaultvalue'

give: someaction( );
expected: 'defaultvalue'

give: someaction(' ');
expected: ' '

given: someaction(null);
expected: 'defaultvalue'

given: someaction('null');
expected: 'null'

I'm wondering what krpano does with the last 2 examples...

with multiple parameters, like:

someaction(,,,) or someaction( , , , , , ,);
expect all to give null

I think the trouble lies with that you can't set some parameter to be null.

For instance:

Quellcode

	 set(test, 0);
	 set(test, null);
	 trace(test);
	 
	 if(test === null,
		trace('test === null'); 
	 );
	 
	 if(test == null,
		trace('test == null');
	 );
	 
	 if(test == 'null',
		trace('test == null'); 
	 );

the last 2 if statement validate to true, while the strict doesnt.

This means sometimes if a variable doesnt exsit, you have to do a strict comparison, while in other moments, the variable does exist but is set to null, you have to do normal comparison. that would confuse me alot.

plugin[0] === null would work, while %1 === null wouldnt.

Dieser Beitrag wurde bereits 1 mal editiert, zuletzt von »Zephyr« (21. November 2011, 19:42)

Zum Seitenanfang

klaus.krpano

Administrator

Beiträge: 13 499

Dienstag, 22. November 2011, 16:02

Hi,

there a new update for the 1.0.8.14 version available:
http://krpano.com/news/#2011-11-22

Mostly smaller bugfixes, but also some improvements (especially for the Android usage).
Here a listing of the changes:

Changes (Flash and HTML5)

New devices attribute operators AND(+) and NOT(!) for mixing specific device settings - e.g. devices="android+flash".
Documentation update about using variables inside urls via %$VAR%.
The HTML5 plugins are now also included in the tools download package.
Added a new tweenintervalbase setting for the setting the time interval for tweens.
The subtxt action works now also with constant texts.
Videoplayer Plugin - new onvideopaused event.
Fix: videoplayer.swf - the pausedonstart detection was much improved by analyzing the real visible video content instead of relying on the Flash information. Additionally there are workarounds for the multi-threading bugs of the Flashplayer 11.2.
Fix: The if parser was not working correctly when using 'empty strings'.
Fix: White-spaces were detected wrongly as action parameters.

Improvements for the Android usage

New drag2dsmooth control mode (default for Android when using drag2d). In the Android Flashplayer the touch input events were unfortunately dispatched at a lower rate than the screen refreshes - e.g. when the screen refreshes every 20-16ms (50-60 FPS) then the touch events when dragging around the screen occur only every 80-120ms - this means the touch inputs are very jerking and inaccurate. The solve this problem and make the movement smoother the krpano viewer is now 'smoothing' the input.
Updated html templates for correct 1:1 pixel-mapping on Android 3.* tablets.
New default Android skins with zoom buttons.
The krpano Flash viewer supports now the new embedding parameter simulatedevice for simulating the behaviors on Android devices. Possible settings are "android", "androidmobile" or "androidtablet".
e.g. here a link to an example with simulatedevice=android to see the Android skin:
http://krpano.com/krpano.html?pano=panos/rom/marktplatz/marktplatz.xml&simulatedevice=android

krpano HTML5 Viewer (krpanoJS)

To make the Flash/HTML5 usage easier the area.pixel* read-only variables are now also available in the krpanoJS viewer.
Fix: Scaling/size bug when dynamically changing the parent.
Fix: events.onresize will now be only called when the size itself has really changed.
Fix: events.onmouseup was called twice on Desktop.
Fix: events.onclick will be only called when there is no or only a very small movement between mouse-down and mouse-up.
Fix: The url placeholder parsing/resolving was not working for <cube> urls.
Fix: Better Webkit detection (needed for the Google+ App).
Fix: Wrong "loaded/loading" values when setting a plugin again to the same url.

Best regards,
Klaus

Zum Seitenanfang

Tuur

Erleuchteter

Beiträge: 3 839

Wohnort: Netherlands

Beruf: Krpano custom coding / Virtual Tours / Photography / Musician / Recording engineer

Dienstag, 22. November 2011, 17:35

GREAT!!

Tuur *thumbsup*

https://www.virtualtuur.com
Panimator: https://krpano.com/plugins/userplugins/panimator/#top
ToolBox: https://bit.ly/3TTnd6U
Measure in 360 photos

: https://pame.virtualtuur.com
Moving Hotspots V2: https://krpano.com/plugins/userplugins/movinghotspots/#top
Donations
https://bit.ly/3NnpynC

Zum Seitenanfang

klaus.krpano

Administrator

Beiträge: 13 499

Dienstag, 22. November 2011, 22:19

Hi,

just one quick addition to the release from adove:

I have updated the current packages with an updated krpanoJS viewer.
There was still one small bug: (thanks to Sacha Griffin for reporting)

When the url of a plugin or hotspot element was set again to the same url value during loading, then the loading of that element had never finished.

Best regards,
Klaus

Zum Seitenanfang

Nupsi

Profi

Beiträge: 697

Mittwoch, 23. November 2011, 12:10

Hey Klaus,

would you take a look at the videoispaused-problem? The video still wont stop/exit!
And another thing: You said some time ago that the krpanoprotect-tool would support private keys for HTML5-encryption some day. Is that still on your to do-list?

Best regards
Nupsi

P.S.: Oh.....and thanks a ton for your hard work to make krpano better and better!!! Saying that was long overdue!

Zum Seitenanfang

klaus.krpano

Administrator

Beiträge: 13 499

Mittwoch, 23. November 2011, 13:43

Hi,

Zitat von »Nupsi«

would you take a look at the videoispaused-problem? The video still wont stop/exit!

yes, but this is correct so as it is,
see the answers in the thread,

Zitat von »Nupsi«

And another thing: You said some time ago that the krpanoprotect-tool would support private keys for HTML5-encryption some day. Is that still on your to do-list?

yes, of course, why should I have removed that from my todo list?

best regards,
Klaus

Zum Seitenanfang

sachagriffin

Meister

Beiträge: 1 857

Beruf: Virtual Tours - Photography - Krpano developer

Mittwoch, 23. November 2011, 16:53

Thanks for this.
In the future, would you mind adding something to the build when there are multiple updates on a day.
I'm having to triple check to ensure I have the updated updated versions.

Zitat von »klaus.krpano«

Hi,

just one quick addition to the release from adove:

I have updated the current packages with an updated krpanoJS viewer.
There was still one small bug: (thanks to Sacha Griffin for reporting)

When the url of a plugin or hotspot element was set again to the same url value during loading, then the loading of that element had never finished.

Best regards,
Klaus