Jump to content
GreenSock

Search the Community

Showing results for tags 'bug'.

  • Search By Tags

    Type tags separated by commas.
  • Search By Author

Content Type


Forums

  • GreenSock Forums
    • GSAP
    • Banner Animation
    • Jobs & Freelance
  • Flash / ActionScript Archive
    • GSAP (Flash)
    • Loading (Flash)
    • TransformManager (Flash)

Product Groups

  • Club GreenSock
  • TransformManager
  • Supercharge

Categories

There are no results to display.


Find results in...

Find results that contain...


Date Created

  • Start

    End


Last Updated

  • Start

    End


Filter by number of...

Joined

  • Start

    End


Group


Personal Website


Twitter


CodePen


Company Website


Location


Interests

  1. Hi guys! I have the following problem: There is a draggable with attached methods onPress and onRelease. The draggable component has a button child. When I click on the draggable but outside of the button both events get fired. But when i click on the button only the onPress event gets fired (you can check this in the browser console). Same result happens if I put an input as a child and click on it. Codepen: http://codepen.io/anon/pen/jqqyEQ I need a way to fire both onPress and onRelease events when I click on buttons and inputs that are children of the draggable component. Thanks in advance!
  2. Hy guys, I love your library since the Flash times I just found what I think is a bug. In the codepen link there is an example. If you tween a class which name is contained in another class, the part of name of tweened class was subtracted on the other class <div class="open-sub" id="tween"></div> if I tween a class with the name "open" TweenMax.to("#tween", 1, {className: "+=open"} ); // +=open or -=open produce same bug the "open-sub" class will become "-sub" Nothing breaking, but could lead in a very tricking result
  3. I had this strange bug in a game I was making that caused the player to be able to click twice on a button, even though the event listener removed itself which should have prevented the second click. After spending many hours on this I've managed to boil it down to what I think is a bug in how TimelineLite handles call() Here's an example: var obj = {value: 0}; var timeline_1 = new TimelineLite({delay: 1, onStart: function() { console.log("timeline_1 start") }, onComplete: function() { console.log("timeline_1 complete"); }}) .to(obj, 0.5, {value: 1}) .add(function() { console.log("function 1"); timeline_2.play(0); }, 0.4) var timeline_2 = new TimelineLite({paused: true, onStart: function() { console.log("timeline_2 start") }, onComplete: function() { console.log("timeline_2 complete"); }}) .to(obj, 0.5, {value: 0}) .add(function() { console.log("function 2"); timeline_1.play(0); }, 3); The first timeline tweens an object's value to 1 during 0.5 seconds, and calls a function after 0.4 seconds. That function goes on to play the second timeline (previously paused). The second timeline starts tweening the same object's value back to 0 (0.1 seconds before the first tween is done, causing an overlap), and after 3 seconds calls a function that plays the first timeline from 0 again, effectively creating a loop. The expected behaviour (for me at least!) is that the tween on the second timeline will overwrite the first, cutting it a little short. What actually happens is that the first timeline's call() function runs twice right after each other (logging "function_1" twice to the console), even though the first timeline is only started once! This only happens when the two tweens overlap. If I nudge the function calls and/or tweens so the second tween doesn't overwrite the first, the problem is gone. The same goes for when I have the second tween affect some other unrelated value instead: no problem. Have I misunderstood something here, or is this in fact a bug?
  4. I've come come across a strange behaviour that I would like to understand. It might be bug, but I'm not sure. When setting immediateRender to true I would have guessed that onUpdate would also fire once during that initial render, but it seems to depend on the type of tween involved. .to fires the onUpdate while .from and .fromTo doesn't. Example: var obj1 = {a: 0}; var obj2 = {a: 0}; var obj3 = {a: 0}; // logs "update1" TweenLite.to(obj1, 1, {a: 1, onUpdate: function() { console.log("update1"); }, immediateRender: true, paused: true}); // logs nothing TweenLite.from(obj2, 1, {a: 1, onUpdate: function() { console.log("update2"); }, immediateRender: true, paused: true}); // logs nothing TweenLite.fromTo(obj3, 1, {a: 0}, {a: 1, onUpdate: function() { console.log("update3"); }, immediateRender: true, paused: true});
  5. I have found what seems to be an inconsistency when cancelling tweens. Please see the attached codepen for a demo. Usual Behavior: When TweenMax.set is called while another tween is running, the already running tween is cancelled (when they affect the same values) The Possible Bug: When that initial tween is called, but it hasn't "started", the cancellation does not occur and TweenMax.set is effectively called before the former tween even though it occurs after. Edit: I updated the codepen to work and to make more sense.
  6. OxXxigen

    GSAP VS MooTools

    sorry for bad english. If use mootools and try to change "position" attribute, autoCss function GSAP detect "position" in target and don't add it to vars.css, becouse mootool add "position" method to Node prototype. i fix it in my local machine, but... where can i write, to fix it in future?
  7. Hello all. I have been using your fine library to animate a set of banner ads for a client going live soon. Fairly late in this process, a team member discovered a 1 pixel jitter or wiggle only in Safari. We are using Google DoubleClick, and we also have a private viewer to show clients ads. In both cases, after uploading these animated ads, in Safari the ad will shift back and forth by about one pixel while elements in it are animating. To phrase it in a different way, when the ad is on a page within an iFrame, the iframe or possibly the container it's in shifts side to side in a wiggle by about one pixel. The animations display slightly differently in Safari as opposed to Chrome. This is not as significant as the jitter. This has been found in Safari Version 9.0 in OSX El Capitan and Yosemite. I stripped the ad of any client-specific images, text, or information and uploaded it to CodePen to try and isolate what's going on. I also had to strip out all the DoubleClick specific code because CodePen would not allow it to run with that code. I really have to apologize because this is very ugly at the moment as I try to debug this. So far I have not been able to get the iframe to jitter in CodePen. However in the "Details" view, the animation is affecting the font weight of the page header text! I feel like this is related. I am at wits end. I'm hoping that if you look at it here and can understand why the font is changing during animation, it will help me understand why the ad is affecting its container during animation in Safari. EDIT: To be clear, right now the behavior I'm asking for help debugging is in CodePen, that the font weight of the header in the top left is flickering during the animation of my banner ad. I think this must be caused by the same thing that is making it jitter in Google DoubleClick preview and in our internal adviewer page, which also uses iframes. Any help would be much appreciated. Best, Stephen Link: http://codepen.io/stevium/details/WQdEyG/
  8. I think I stumbled on a bug trying to rotate text elements in SVG with Greensock... It completely ignores the rotation of the text for both the house and the sun... TweenMax.set(sunDegreesText,{rotation:degrees,transformOrigin:"50% 50%"}); TweenMax.set(houseDegreesText,{rotation:degrees,transformOrigin:"50% 50%"}); Any ideas?
  9. I have been working on a 2d/3d transition effect module, and I have been running into some issues integrating with gsap (TweenLite). Most of the Tweens require updating on every $("window").resize() function if they have any media queries. Unfortunately I have been running into issues overwriting TweenLite.to's that involve rotations. A good example of this is a simple "hamburger" icon animation. In the below codepen is a decent example of the issue. If you resize the window while the button is in the X state (or navState = 1) the rotation tween acts as though its stuck in the tween.progress=0 state; http://codepen.io/anon/pen/MwYmVW By changing rotationZ: to rotation: the tween rotation get stuck in the tween.progress=1 state. http://codepen.io/anon/pen/MwYmxa If anyone can shed some light on this issue it would be much appreciated. Thanks, Daniel
  10. Hi, This issue has been bugging me for the past few hours, but I've managed to point it down to the new lines Illustrator adds to SVG files when there are a lot of points and the lines become too long. Maybe it's an option I've been overlooking and I can turn this behaviour off, anyhow these new lines Illustrator adds on points changes the start/ending positions on SVG paths. I've added a codepen, the SVG file gets injected on the place of the img tag with an modified version of SVG Injector (https://github.com/iconic/SVGInjector) I'm not sure if this is a GSAP problem or a bug in Chrome. But it might be worth looking into. Kind regards, Mark
  11. So I've been struggling to figure out how to resolve this issue and figured there has to be some kind of resolution to this by now as I've seen people reporting this bug with chrome for years it seems like. I managed to get it working in Firefox and IE, haven't checked anything else. Although Ie has issues with unset SVG Widths and Heights so it seems like I need to setup a backup script to set that on load for IE. But this is driving me nuts. I just want to animate some simple clouds through a background. The only problem is the clouds are using an alpha mask which won't seem to animate. Also this isn't the final animation, I am just trying to get it to work before I spend more time on it. Any help would be greatly appreciated! Thanks.
  12. Hello! I noticed that if I use SplitText on a div containing text, and that div has a class defining a font, than Splittext will fail in creating the lines array, making an array of lines that is pretty much wrong: the lines are always too long or too short, never correct. I've made a small example in codepen up there: can you tell me if I'm doing something wrong? Cheers! Alex
  13. We're encountering an issue using Draggable on a multitouch table (WIndows 8 + Chrome), but it also seems to happen on Android Lollipop, so it's probably a general issue. When dragging an item and you use another finger to tap shortly on the same item, it will be suspended in place, not firing the proper events for ending a drag (onDragEnd, onRelease). Snapping and throwprops will also not work. The issue is easy to reproduce using the sample on https://greensock.com/draggable. Enable the checkbox for "snap end position to grid". Start dragging one of the squares and then shortly tap on it with another finger. When you release, the square will not snap to a position on the grid. Has anyone encountered this and found a way to fix it? Should I file an issue somewhere?
  14. I've come across a bug when applying a transformation on an already rotated element with 3d rendering enabled: You can see the codepen for the live preview I have an element with 180deg rotation and Z transform: <div id="aa" style="transform:rotate(180deg) translateZ(10px)">I'm rotated 180 Deg</div> // On Webkit/Gecko this happens only with z > 0, On IE it happens also with z == 0 And then I run some transform animation on it TweenMax.from('#aa', 2, {x:'+=100'}) The result is that the rotated div flips back to 0 degrees, or, from matrix(-1,0,0,-1,0,0) it transforms to matrix(1,0,0,1,0,0) This does not happen on other rotation angles.
  15. Bug or feature? When I tl.seek() a paused timeline the animation is "updating" yet the onUpdate function is never called. See plunk.
  16. When reverting a SplitText operation any trailing line breaks are removed from the text, causing a different result: http://codepen.io/patrickklug/pen/Deiba
  17. Adding any < or > characters to a text causes these characters and anything enclosed between them to be omitted from the resulting text. There is also a related issue where <br> causes an exception in SplitText.js:206.
  18. I stumbled upon a weird behaviour of clearProps, and couldn't find any reference to whether it is a bug or not. When using clearProps with explicit values the order of the parameters passed has importance when clearing mixed transforms/properties values. This will clear x and y and skip opacity - clearProps:'x, y, opacity' This will clear only "opacity" and skip x and y. clearProps: 'opacity, x, y' Same goes with other non transform values like "clip" See the codepen for clearer example.
  19. Hi guys, At the moment I'm working on a rotator that rotates between 3 tabs with a 3D perspective (middle tab in the front and 2 on the sides that are in the back). It's all working fine on the platforms and browsers I've tried except for Chrome (both Windows and Mac). The problem there is that one of the side tabs is not rendering on the Z axis while if you inspect it in dev tools the overlay shows the tab in the right position. Here is how it is supposed to look: http://puu.sh/bAamR/98bab7014e.png Here is how it actually looks: http://puu.sh/bAaod/da2a27ac77.png And here is a working demo file I've put online: http://lwigchert.nl/test/demo_issue.html It might be an issue with the z-index because when you play around with that the tabs seem to switch. If there is anything else needed that may help, let me know!
  20. CodePen (I don't think this site supports IE8, so this may not be useful): http://codepen.io/anon/pen/kxhfD Problem Opacity isn't animated when there is an x-translation in Internet Explorer 8. Code // no opacity animation in IE8, works in modern browsers var element = document.getElementById('box'); var animation = { x: -50, opacity: 0 }; TweenMax.from(element, 1, animation); a Extra note If you replace the x with marginLeft, the opacity correctly animates. > Thank you so much for all the hard work, guys. It's really, really appreciated. <
  21. So i'm currently working on a page that has animation when you scroll down the page using scrollmagic and tweenmax. I've pinned the appropriate sections and created the animation, and everything works perfectly. I then added some javascript to have a show/hide function for the different features in a sticky nav (instead of having a long vertical scrolling page, it swaps out content within the same area). I noticed a very weird bug when I switch between the different sections. Here's a screenshot of what the bug looks like when I go back to the original section with the animation. It seems as though the tweenmax plugin is ignoring all of my css and completely breaks. What's even weirder is that when you start to scroll down the css is restored. Any ideas on what this is? I can't seem to figure out what's causing it.
  22. Hi, I don't know if someone can help me but here's my bug. I used TweenLite to make a small 3d transition to show a box. The box itself is slightly rotated using css3 transform:rotate properties. When the tween finish, the box has a weird position glitch. It move down a little bit. If i remove the rotate properties, everything is fine. Here's the CodePen : http://codepen.io/anon/pen/JyLAt I've seen this glitch in all Major Browser Sorry for my english, is not my native language. Thanks in advance for your help
  23. Hi guys, I have a problem with The TweenMax.killAll(true) method. Here is my example: TweenMax.to(targetQ, 2, { scaleX : 2, scaleY : 2, reversed : true, yoyo : true, repeat : 1, x : targetQ.x + (targetQ.width - 2 * targetQ.width) / 2, y : targetQ.y + (targetQ.height - 2 * targetQ.height), ease : Back.easeOut, onReverseComplete : function onReverseComplete():void { trace("complete this tween ;)"); } } ); If you make a reversed tween, it's never call the onComplete function. The tweenax documentation write this: " onComplete : Function – A function that should be called when the tween has completed. To sense when a tween has reached its starting point again after having been reversed, use onReverseComplete." If I call TweenMax.killAll(true), tween not killed!!!!! As I see, the problem is in the killAll method. Inside this, here's a line: isDC = (a.target == a.vars.onComplete); Unfortunately, if you set a reverse to a tween, onComplete callback ignored! cheers, jax
  24. I'm using the greensock animation library to animate the translateZ property of an image. This is my code: $("<img />") .attr({ src: piece, class: "pieces" }) .hover(function() { tl.to($(this), 1, {transform:"translateZ(1px)"})}, function() { tl.fromTo($(this), 1, {transform: "translateZ(0px)"}) }) .load(function () { _stageElement.append($(this)); }) } But instead of stopping at 5px it flies completely off the screen. Does anyone know what's happening here and why? Also if it's important after the image flies off the screen, it comes back to its original location, but the hover property doesn't have any effect anymore. I'm not sure if this is a greensock issue or a jquery issue. Here's an example to see what I mean http://richbaird.net/3dFlip/fip.html
  25. When calling TweenLite.killDelayedCallsTo on a method of an object, all objects with the same prototype will have there delayed calls to that method killed. Cause: When no scope is bound to the method, the method is really the same function object shared by all instances of the prototype. So they will all be killed by 1 call. Solution: Bind your callbacks using .bind(this) or use a closure and place the bound function/closure in a variable. Then, later you can call TweenLite.killDelayedCallsTo on the bound callback. A second argument in TweenLite.killDelayedCallsTo for specifying scope would also be nice, just like with the onCompleteScope etc in TweenLite.to/.from. That would save us a lot of binding and storing functions, and would be in line with the other static TweenLite methods.
×