I'm trying to change the color of a fixed element (.logo) when scrolling on top of a dark image (.image). I came across this solution:
Detect when static element overlaps fixed element position on scroll
Which only works for a single image. But what if I want to change the color of the fixed element when scrolling passed all the images with the image class by using querySelectorAll?
I tried to solve this with a forEach but the fixed element only changes color on the last image. Can somebody explain this behaviour, in my mind this should work?
https://codepen.io/bosbode/pen/GaJNKr
HTML
<p class="logo">Logo</p>
<div class="image"></div>
<div class="image"></div>
CSS
body {
text-align: center;
height: 100%;
font-size: 1.5rem;
}
.image {
width: 800px;
height: 600px;
background: blue;
margin: 100px auto;
}
.logo {
position: fixed;
top: 0;
left: 50%;
transform: translate(-50%, 0)
}
JavaScript
const logo = document.querySelector('.logo');
const images = document.querySelectorAll('.image');
window.addEventListener('scroll', function () {
const a = logo.getBoundingClientRect();
images.forEach((item, index) => {
const b = item.getBoundingClientRect();
if (a.top <= (b.top + b.height) && (a.top + a.height) > b.top) {
logo.style.color = 'white';
} else {
logo.style.color = 'black';
}
})
});
for (var i = 0; i < images.length; i++) {
var item = images[i]
let b = item.getBoundingClientRect();
if (a.top <= (b.top + b.height) && (a.top + a.height) > b.top) {
logo.style.color = 'white';
break;
} else {
logo.style.color = 'black';
}
};
Updated content:
1. For this solution, I need using break statement in loop, so I am using for instead of foreach
2. We need break if logo is inside in every image.
WHY?:
If logo is inside image1, then its' color can be white, but next step, its' color can be black, because logo is not inside image2.
Below code is more readable for this solution:
const logo = document.querySelector('.logo');
const images = document.querySelectorAll('.image');
function isInsideInImages(images, logoPos) {
for (var i = 0; i < images.length; i++) {
let imagePos = images[i].getBoundingClientRect();
if (logoPos.top <= (imagePos.top + imagePos.height) && (logoPos.top + logoPos.height) > imagePos.top) {
return true;
}
};
return false;
}
window.addEventListener('scroll', function () {
const a = logo.getBoundingClientRect();
if (isInsideInImages(images, a)) {
logo.style.color = 'white';
} else {
logo.style.color = 'black';
}
});
Related
Im working on a project game and I want to see if a div "rope" hits one of my pictures class='Fish1'.
I managed to get it to work by using document.GetElementsById("fish") but when I change it to document.getElementsByClassName('fish1') it gives me an error saying:
TypeError: Cannot read properties of undefined (reading
'getBoundingClientRect') at checkCollision
How can i fix this error? Also how can I check the height during the transition of a div and have it give me the height that it is at that point in the transition?
I've tried this:
function checkCollision(rope, fishy) {
var line = rope;
var fishy = document.getElementsByClassName('fish1');
var ropeRect = line.getBoundingClientRect();
for (var i = 0; i < fishy.length; i++) {
var fishyRect = fishy[i].getBoundingClientRect(i);
}
return (ropeRect.right >= fishyRect.left &&
ropeRect.left <= fishyRect.right) &&
(ropeRect.bottom >= fishyRect.top &&
ropeRect.top <= fishyRect.bottom);
}
And this is what worked for the id only:
function checkCollision(line, fishy) {
var line = rope;
var fishy = document.getElementById('fish');
var lineRect = line.getBoundingClientRect();
var fishyRect = fishy.getBoundingClientRect();
return (lineRect.right >= fishyRect.left &&
lineRect.left <= fishyRect.right) &&
(lineRect.bottom >= fishyRect.top &&
lineRect.top <= fishyRect.bottom);
}
Checking collision:
You can check if two elements overlap by using conditions that compare their x,y coords also involving their width, height:
//returns true if obj1 and obj2 collided, otherwise false
function checkCollision(obj1, obj2) {
const rect1 = obj1.getBoundingClientRect();
const rect2 = obj2.getBoundingClientRect();
return (rect1.x + rect1.width >= rect2.x && rect1.x <= rect2.x + rect2.width) &&
(rect1.y + rect1.height >= rect2.y && rect1.y <= rect2.y + rect2.height)
}
Moving the elements:
Once you have the function, you can call it inside your loop after changing the position of the elements at each iteration.
You didn't share any other detail apart the not working checkCollision function so I have no idea how did you perform the motion or probably you didn't at all yet.
Here I used Window.requestAnimationFrame()
https://developer.mozilla.org/en-US/docs/Web/API/window/requestAnimationFrame
The window.requestAnimationFrame() method tells the browser that you
wish to perform an animation and requests that the browser calls a
specified function to update an animation before the next repaint. The
method takes a callback as an argument to be invoked before the
repaint.
Here in this example, I animate a group of .fish elements over a sin wave crossing a line in the middle. Every time an element overlaps with the rope, it calls the function collisionOccurred printing on console its id property.
Since you have the exact element partecipating to the event you are free to fetch anything about it like for example the exact position where it is at in that instant (or the height you were expecting to know).
Later I also added the events fishEnter and fishLeave that fire respectively when the element enters and leaves the rope area.
const container = document.querySelector('body');
const rope = document.getElementById('rope');
const fishes = document.getElementsByClassName('fish');
let start = null;
let containerWidth, containerHeight;
beginAnimation();
//begins the fishes animation
function beginAnimation() {
window.requestAnimationFrame(moveFish);
}
//returns the x,y coords of a sin function
function getSin(millisecondsPast, cycleDuration = 2000) {
//ranging 0-1 in the span of cycleDuration (ms)
const progress = (millisecondsPast % cycleDuration) / cycleDuration;
//=~6.28
const twoPi = 2 * Math.PI;
//x ranging from 0-6.28 in the span of cycleDuration
const x = twoPi * progress;
//sin(x) (radius=1)
const y = Math.sin(x);
return {
//[0-1] (in the span of cycleDuration ms)
x: progress,
//[-1,+1]
y: y
}
}
//moves all the fishes following a sin wave
function moveFish(timestamp) {
const containerWidth = parseInt(container.offsetWidth);
const containerHeight = parseInt(container.offsetHeight);
//keep track of the animation progress
if (!start) start = timestamp;
let progress = timestamp - start;
//for all the fishes
for (fish of fishes) {
//calculate the x and y of a sin wave
//spread across the container width in the 0-2pi space
const duration = parseInt(fish.dataset.duration);
const vertical = parseInt(fish.dataset.verticalspan);
const sin = getSin(progress, duration);
let x = sin.x * containerWidth;
let y = sin.y * (vertical/2) + (containerHeight / 2);
//change the position of the current fish
fish.style.left = `${x}px`;
fish.style.top = `${y}px`;
const didcollide = checkCollision(rope, fish);
const overlapping = (fish.dataset.overlapping == 'true') ? true : false;
//calls fishEnter if the fish entered in the space of the rope
if (!overlapping && didcollide) {
fish.dataset.overlapping = 'true';
fishEnter(fish);
}
//calls fishLeave if the fish left the space of the rope
else if (overlapping && !didcollide) {
fish.dataset.overlapping = 'false';
fishLeave(fish);
}
//calls collisionOccurred if the fish collided with the rope
if (didcollide)
collisionOccurred(fish);
}
//render next iteration
window.requestAnimationFrame(moveFish);
}
//returns true if obj1 and obj2 collided, otherwise false
function checkCollision(obj1, obj2) {
const rect1 = obj1.getBoundingClientRect();
const rect2 = obj2.getBoundingClientRect();
const didcollide =
(rect1.x + rect1.width >= rect2.x && rect1.x <= rect2.x + rect2.width) &&
(rect1.y + rect1.height >= rect2.y && rect1.y <= rect2.y + rect2.height);
return didcollide;
}
//gets fired when the fish enters in the space of the rope
function fishEnter(target) {
target.classList.add('crossing');
console.log(`The fish id: ${target.id} entered in the rope space!`);
}
//gets fired when the fish leaves in the space of the rope
function fishLeave(target) {
target.classList.remove('crossing');
console.log(`The fish id: ${target.id} left the rope space!`);
}
//gets fired when the collision event occurs
function collisionOccurred(target) {
//console.log(`The fish id: ${target.id} crossed the rope!`);
}
*{
box-sizing: border-box;
}
body {
position: relative;
height: 100vh;
padding: 0;
margin: 0;
overflow: hidden;
}
#rope {
--size: 80px;
--border: 5px;
position: absolute;
top: calc(50% + var(--border) - var(--size) / 2) ;
width: 100%;
height: var(--size);
background: brown;
display: flex;
justify-content: center;
align-items: center;
font-weight: 600;
outline: solid var(--border) darkorange;
color: darkorange;
font-size: 2rem;
}
.fish {
position: absolute;
width: 80px;
line-height: 2rem;
background: blue;
font-weight: 600;
outline: solid darkblue;
text-align: center;
font-size: 1.5rem;
}
.fish::after {
content: attr(id);
color: white;
}
button {
padding: 1rem;
cursor: pointer;
}
.crossing{
background: yellow;
}
.crossing::after{
color: black !important;
}
<div id="rope">ROPE AREA</div>
<div id="fish1" class="fish" data-duration="5000" data-verticalspan="200"></div>
<div id="fish2" class="fish" data-duration="8000" data-verticalspan="300"></div>
<div id="fish3" class="fish" data-duration="10000" data-verticalspan="100"></div>
<div id="fish4" class="fish" data-duration="3000" data-verticalspan="500"></div>
I have a 11500x11500 div that consists of 400 images, that obviously overflows the viewport.
I would like to pan around the whole div programmatically.
I want to generate an animation and by the time the animation is over, the whole of the div must have been panned across the viewport, top to bottom, left to right.
Right now, I am "splitting" my 11500x1500 div into tiles. The maximum width and height of each tile is the width and height of the viewport.
I store the coordinates of each tile and then I randomly choose one, pan it left-to-right and then move on to the next one.
I would like to know:
whether my method is correct or whether I am missing something in my calculations/approach and it could be improved. Given the size, it is hard for me to tell whether I'm actually panning the whole of the div after all
whether I can make the panning effect feel more "organic"/"natural". In order to be sure that the whole div is eventually panned, I pick each tile and pan it left-to-right, move on to the next one etc. This feels kind of rigid and too formalised. Is there a way to pan at let's say an angle or with a movement that is even more random and yet be sure that the whole div will eventually be panned ?
Thank in advance for any help.
This is the jsfiddle and this is the code (for the sake of the example/test every "image" is actually a div containing its index as text):
function forMs(time) {
return new Promise((resolve) => {
setTimeout(() => {
resolve()
}, time)
})
}
let container = document.getElementById('container')
let {
width,
height
} = container.getBoundingClientRect()
let minLeft = window.innerWidth - width
let minTop = window.innerHeight - height
let i = 0
while (i < 400) {
// adding "image" to the container
let image = document.createElement('div')
// add some text to the "image"
// to know what we're looking at while panning
image.innerHTML = ''
let j = 0
while (j < 100) {
image.innerHTML += ` ${i + 1}`
j++
}
container.appendChild(image)
i++
}
let coords = []
let x = 0
while (x < width) {
let y = 0
while (y < height) {
coords.push({
x,
y
})
y += window.innerHeight
}
x += window.innerWidth
}
async function pan() {
if (!coords.length) {
return;
}
let randomIdx = Math.floor(Math.random() * coords.length)
let [randomCoord] = coords.splice(randomIdx, 1);
console.log(coords.length)
container.classList.add('fast')
// update style in new thread so new transition-duration is applied
await forMs(10)
// move to new yet-unpanned area
container.style.top = Math.max(-randomCoord.y, minTop) + 'px'
container.style.left = Math.max(-randomCoord.x, minLeft) + 'px'
// wait (approx.) for transition to end
await forMs(2500)
container.classList.remove('fast')
// update style in new thread so new transition-duration is applied
await forMs(10)
//pan that area
let newLeft = -(randomCoord.x + window.innerWidth)
if (newLeft < minLeft) {
newLeft = minLeft
}
container.style.left = newLeft + 'px'
// wait (approx.) for transition to end
await forMs(4500)
// move on to next random area
await pan()
}
pan()
html,
body {
position: relative;
width: 100%;
height: 100%;
margin: 0;
padding: 0;
overflow: auto;
}
* {
margin: 0;
padding: 0;
}
#container {
position: absolute;
top: 0;
left: 0;
text-align: left;
width: 11500px;
height: 11500px;
transition: all 4s ease-in-out;
transition-property: top left;
font-size: 0;
}
#container.fast {
transition-duration: 2s;
}
#container div {
display: inline-block;
height: 575px;
width: 575px;
border: 1px solid black;
box-sizing: border-box;
font-size: 45px;
overflow: hidden;
word-break: break-all;
}
<div id="container"></div>
I think following improvements can be made:
Hide overflow on html and body so user can not move scrollbar and disturb the flow.
Calculate minLeft and minTop every time to account for window resizing. You might need ResizeObserver to recalculate things.
Increase transition times to avoid Cybersickness. In worse case RNG will pick bottom right tile first so your container will move the longest in 2seconds! Maybe, you can zoom-out and move then zoom-in then perform pan. Or use any serpentine path which will make shorter jumps.
Performance improvements:
Use transform instead of top, left for animation.
Use will-change: transform;. will-change will let browser know what to optimize.
Use translate3D() instead of translate(). ref
Use requestAnimationFrame. Avoid setTimeout, setInterval.
This is an old but good article: https://www.paulirish.com/2012/why-moving-elements-with-translate-is-better-than-posabs-topleft/
Modified code to use transform:
function forMs(time) {
return new Promise((resolve) => {
setTimeout(() => {
resolve()
}, time)
})
}
let container = document.getElementById('container')
let stat = document.getElementById('stats');
let {
width,
height
} = container.getBoundingClientRect()
let minLeft = window.innerWidth - width
let minTop = window.innerHeight - height
let i = 0
while (i < 400) {
// adding "image" to the container
let image = document.createElement('div')
// add some text to the "image"
// to know what we're looking at while panning
image.innerHTML = ''
let j = 0
while (j < 100) {
image.innerHTML += ` ${i + 1}`
j++
}
container.appendChild(image)
i++
}
let coords = []
let x = 0
while (x < width) {
let y = 0
while (y < height) {
coords.push({
x,
y
})
y += window.innerHeight
}
x += window.innerWidth
}
let count = 0;
async function pan() {
if (!coords.length) {
stat.innerText = 'iteration: ' +
(++count) + '\n tile# ' + randomIdx + ' done!!';
stat.style.backgroundColor = 'red';
return;
}
let minLeft = window.innerWidth - width
let minTop = window.innerHeight - height
let randomIdx = Math.floor(Math.random() * coords.length);
randomIdx = 1; //remove after debugging
let [randomCoord] = coords.splice(randomIdx, 1);
stat.innerText = 'iteration: ' +
(++count) + '\n tile# ' + randomIdx;
console.log(coords.length + ' - ' + randomIdx)
container.classList.add('fast')
// update style in new thread so new transition-duration is applied
await forMs(10)
// move to new yet-unpanned area
let yy = Math.max(-randomCoord.y, minTop);
let xx = Math.max(-randomCoord.x, minLeft);
move(xx, yy);
// wait (approx.) for transition to end
await forMs(2500)
container.classList.remove('fast')
// update style in new thread so new transition-duration is applied
await forMs(10)
//pan that area
let newLeft = -(randomCoord.x + window.innerWidth)
if (newLeft < minLeft) {
newLeft = minLeft
}
xx = newLeft;
//container.style.left = newLeft + 'px'
move(xx, yy);
// wait (approx.) for transition to end
await forMs(4500)
// move on to next random area
await pan()
}
pan()
function move(xx, yy) {
container.style.transform = "translate3D(" + xx + "px," + yy + "px,0px)";
}
html,
body {
position: relative;
width: 100%;
height: 100%;
margin: 0;
padding: 0;
overflow: hidden;
}
* {
margin: 0;
padding: 0;
box-sizing: border-box;
}
#container {
text-align: left;
width: 11500px;
height: 11500px;
transition: all 4s ease-in-out;
transition-property: transform;
font-size: 0;
will-change: transform;
}
#container.fast {
transition-duration: 2s;
}
#container div {
display: inline-block;
height: 575px;
width: 575px;
border: 1px solid black;
box-sizing: border-box;
font-size: 45px;
overflow: hidden;
word-break: break-all;
}
#stats {
border: 2px solid green;
width: 100px;
background-color: lightgreen;
position: fixed;
opacity: 1;
top: 0;
left: 0;
z-index: 10;
}
<div id=stats>iteration: 1 tile# 11</div>
<div id="container"></div>
Note I haven't implemented everything in above snippet.
I have a large background image and some much smaller images for the user to drag around on the background. I need this to be efficient in terms of performance, so i'm trying to avoid libraries. I'm fine with drag 'n' drop if it work's well, but im trying to get drag.
Im pretty much trying to do this. But after 8 years there must be a cleaner way to do this right?
I currently have a drag 'n' drop system that almost works, but when i drop the smaller images, they are just a little off and it's very annoying. Is there a way to fix my code, or do i need to take a whole different approach?
This is my code so far:
var draggedPoint;
function dragStart(event) {
draggedPoint = event.target; // my global var
}
function drop(event) {
event.preventDefault();
let xDiff = draggedPoint.x - event.pageX;
let yDiff = draggedPoint.y - event.pageY;
let left = draggedPoint.style.marginLeft; // get margins
let top = draggedPoint.style.marginTop;
let leftNum = Number(left.substring(0, left.length - 2)); // cut off px from the end
let topNum = Number(top.substring(0, top.length - 2));
let newLeft = leftNum - xDiff + "px" // count new margins and put px back to the end
let newTop = topNum - yDiff + "px"
draggedPoint.style.marginLeft = newLeft;
draggedPoint.style.marginTop = newTop;
}
function allowDrop(event) {
event.preventDefault();
}
let imgs = [
"https://upload.wikimedia.org/wikipedia/commons/6/67/Orange_juice_1_edit1.jpg",
"https://upload.wikimedia.org/wikipedia/commons/f/ff/Solid_blue.svg",
"https://upload.wikimedia.org/wikipedia/commons/b/b4/Litoria_infrafrenata_-_Julatten.jpg"
]
/* my smaller images: */
for (let i = 0; i < 6; i++) {
let sensor = document.createElement("img");
sensor.src = imgs[i % imgs.length];
sensor.alt = i;
sensor.draggable = true;
sensor.classList.add("sensor");
sensor.style.marginLeft = `${Math.floor(Math.random() * 900)}px`
sensor.style.marginTop = `${Math.floor(Math.random() * 500)}px`
sensor.onclick = function() {
sensorClick(logs[i].id)
};
sensor.addEventListener("dragstart", dragStart, null);
let parent = document.getElementsByClassName("map")[0];
parent.appendChild(sensor);
}
<!-- my html: -->
<style>
.map {
width: 900px;
height: 500px;
align-content: center;
margin: 150px auto 150px auto;
}
.map .base {
position: absolute;
width: inherit;
height: inherit;
}
.map .sensor {
position: absolute;
width: 50px;
height: 50px;
}
</style>
<div class="map" onDrop="drop(event)" ondragover="allowDrop(event)">
<img src='https://upload.wikimedia.org/wikipedia/commons/f/f7/Plan-Oum-el-Awamid.jpg' alt="pohja" class="base" draggable="false">
<div>
With the answers from here and some time i was able to get a smooth drag and click with pure js.
Here is a JSFiddle to see it in action.
let maxLeft;
let maxTop;
const minLeft = 0;
const minTop = 0;
let timeDelta;
let imgs = [
"https://upload.wikimedia.org/wikipedia/commons/6/67/Orange_juice_1_edit1.jpg",
"https://upload.wikimedia.org/wikipedia/commons/f/ff/Solid_blue.svg",
"https://upload.wikimedia.org/wikipedia/commons/b/b4/Litoria_infrafrenata_-_Julatten.jpg"
]
var originalX;
var originalY;
window.onload = function() {
document.onmousedown = startDrag;
document.onmouseup = stopDrag;
}
function sensorClick () {
if (Date.now() - timeDelta < 150) { // check that we didn't drag
createPopup(this);
}
}
// create a popup when we click
function createPopup(parent) {
let p = document.getElementById("popup");
if (p) {
p.parentNode.removeChild(p);
}
let popup = document.createElement("div");
popup.id = "popup";
popup.className = "popup";
popup.style.top = parent.y - 110 + "px";
popup.style.left = parent.x - 75 + "px";
let text = document.createElement("span");
text.textContent = parent.id;
popup.appendChild(text);
var map = document.getElementsByClassName("map")[0];
map.appendChild(popup);
}
// when our base is loaded
function baseOnLoad() {
var map = document.getElementsByClassName("map")[0];
let base = document.getElementsByClassName("base")[0];
maxLeft = base.width - 50;
maxTop = base.height - 50;
/* my smaller images: */
for (let i = 0; i < 6; i++) {
let sensor = document.createElement("img");
sensor.src = imgs[i % imgs.length];
sensor.alt = i;
sensor.id = i;
sensor.draggable = true;
sensor.classList.add("sensor");
sensor.classList.add("dragme");
sensor.style.left = `${Math.floor(Math.random() * 900)}px`
sensor.style.top = `${Math.floor(Math.random() * 500)}px`
sensor.onclick = sensorClick;
let parent = document.getElementsByClassName("map")[0];
parent.appendChild(sensor);
}
}
function startDrag(e) {
timeDelta = Date.now(); // get current millis
// determine event object
if (!e) var e = window.event;
// prevent default event
if(e.preventDefault) e.preventDefault();
// IE uses srcElement, others use target
targ = e.target ? e.target : e.srcElement;
originalX = targ.style.left;
originalY = targ.style.top;
// check that this is a draggable element
if (!targ.classList.contains('dragme')) return;
// calculate event X, Y coordinates
offsetX = e.clientX;
offsetY = e.clientY;
// calculate integer values for top and left properties
coordX = parseInt(targ.style.left);
coordY = parseInt(targ.style.top);
drag = true;
document.onmousemove = dragDiv; // move div element
return false; // prevent default event
}
function dragDiv(e) {
if (!drag) return;
if (!e) var e = window.event;
// move div element and check for borders
let newLeft = coordX + e.clientX - offsetX;
if (newLeft < maxLeft && newLeft > minLeft) targ.style.left = newLeft + 'px'
let newTop = coordY + e.clientY - offsetY;
if (newTop < maxTop && newTop > minTop) targ.style.top = newTop + 'px'
return false; // prevent default event
}
function stopDrag() {
if (typeof drag == "undefined") return;
if (drag) {
if (Date.now() - timeDelta > 150) { // we dragged
let p = document.getElementById("popup");
if (p) {
p.parentNode.removeChild(p);
}
} else {
targ.style.left = originalX;
targ.style.top = originalY;
}
}
drag = false;
}
.map {
width: 900px;
height: 500px;
margin: 50px
position: relative;
}
.map .base {
position: absolute;
width: inherit;
height: inherit;
}
.map .sensor {
display: inline-block;
position: absolute;
width: 50px;
height: 50px;
}
.dragme {
cursor: move;
left: 0px;
top: 0px;
}
.popup {
position: absolute;
display: inline-block;
width: 200px;
height: 100px;
background-color: #9FC990;
border-radius: 10%;
}
.popup::after {
content: "";
position: absolute;
top: 100%;
left: 50%;
margin-left: -10px;
border-width: 10px;
border-style: solid;
border-color: #9FC990 transparent transparent transparent;
}
.popup span {
width: 90%;
margin: 10px;
display: inline-block;
text-align: center;
}
<div class="map" width="950px" height="500px">
<img src='https://upload.wikimedia.org/wikipedia/commons/f/f7/Plan-Oum-el-Awamid.jpg' alt="pohja" class="base" draggable="false" onload="baseOnLoad()">
<div>
I'm trying to make an image slider. But as you can see the distance between the first and last element is not consistent. If you keep on dragging to left, the distance decreases and if you keep on dragging to right, the distance increases. Looks like the code is behaving differently on different zoom levels (sometimes?) and hence distance between every elements is changing at times.
//project refers to placeholder rectangular divs
projectContainer = document.querySelector(".project-container")
projects = document.querySelectorAll(".project")
elementAOffset = projects[0].offsetLeft;
elementBOffset = projects[1].offsetLeft;
elementAWidth = parseInt(getComputedStyle(projects[0]).width)
margin = (elementBOffset - (elementAOffset + elementAWidth))
LeftSideBoundary = -(elementAWidth)
RightSideBoundary = (elementAWidth * (projects.length)) + (margin * (projects.length))
RightSidePosition = RightSideBoundary - elementAWidth;
initialPosition = 0; //referring to mouse
mouseIsDown = false
projectContainer.addEventListener("mousedown", e => {
mouseIsDown = true
initialPosition = e.clientX;
})
projectContainer.addEventListener("mouseup", e => {
mouseExit(e)
})
projectContainer.addEventListener("mouseleave", e => {
mouseExit(e);
})
function mouseExit(e) {
mouseIsDown = false
//updates translateX value of transform
projects.forEach(project => {
var style = window.getComputedStyle(project)
project.currentTranslationX = (new WebKitCSSMatrix(style.webkitTransform)).m41
project.style.transform = 'translateX(' + (project.currentTranslationX) + 'px)'
})
}
projectContainer.addEventListener("mousemove", e => {
if (!mouseIsDown) { return };
// adds mousemovement to translateX
projects.forEach(project => {
project.style.transform = 'translateX(' + ((project.currentTranslationX ?? 0) + (e.clientX - initialPosition)) + 'px)'
shiftPosition(e, project)
})
})
//teleports div if it hits left or right boundary to make an infinite loop
function shiftPosition(e, project) {
projectStyle = window.getComputedStyle(project)
projectTranslateX = (new WebKitCSSMatrix(projectStyle.webkitTransform)).m41
//projectVisualPosition is relative to the left border of container div
projectVisualPosition = project.offsetLeft + projectTranslateX
if (projectVisualPosition <= LeftSideBoundary) {
project.style.transform = "translateX(" + ((RightSidePosition - project.offsetLeft)) + "px)"
updateTranslateX(e);
}
if (projectVisualPosition >= RightSidePosition) {
newPosition = -1 * (project.offsetLeft + elementAWidth)
project.style.transform = "translateX(" + newPosition + "px)"
updateTranslateX(e);
}
}
function updateTranslateX(e) {
projects.forEach(project => {
style = window.getComputedStyle(project)
project.currentTranslationX = (new WebKitCSSMatrix(style.webkitTransform)).m41
project.style.transform = 'translateX(' + (project.currentTranslationX) + 'px)'
initialPosition = e.clientX
})
}
*, *::before, *::after{
margin:0px;
padding:0px;
box-sizing: border-box;
font-size:0px;
user-select: none;
}
.project-container{
font-size: 0px;
position: relative;
width:1500px;
height:400px;
background-color: rgb(15, 207, 224);
margin:auto;
margin-top:60px;
white-space: nowrap;
overflow: hidden;
padding-left:40px;
padding-right:40px;
}
.project{
font-size:100px;
margin:40px;
display: inline-block;
height:300px;
width:350px;
background-color:red;
border: black 3px solid;
user-select: none;
}
<div class="project-container">
<div class="project">1</div>
<div class="project">2</div>
<div class="project">3</div>
<div class="project">4</div>
<div class="project">5</div>
<div class="project">6</div>
<div class="project">7</div>
<div class="project">8</div>
</div>
I'm not sure exactly how you would go about fixing your implementation. I played around with it for a while and discovered a few things; dragging more quickly makes the displacement worse, and the displacement seems to happen mainly when the elements are teleported at each end of the container.
I would guess that the main reason for this is that you are looping over all the elements and spacing them individually. Mouse move events generally happen under 20ms apart, and you are relying on all the DOM elements being repainted with their new transform positions before the next move is registered.
I did come up with a different approach using absolutely placed elements and the IntersectionObserver API, which is now supported in all modern browsers. The idea here is basically that when each element intersects with the edge of the container, it triggers an array lookup to see if the next element in the sequence is on the correct end and moves it there if not. Elements are only ever spaced by a static variable, while the job of sliding them is passed up to a new parent wrapper .project-slider.
window.addEventListener('DOMContentLoaded', () => {
// Style variables
const styles = {
width: 350,
margin: 40
};
const space = styles.margin*2 + styles.width;
// Document variables
const projectContainer = document.querySelector(".project-container");
const projectSlider = document.querySelector(".project-slider");
const projects = Array.from(document.querySelectorAll(".project"));
// Mouse interactions
let dragActive = false;
let prevPos = 0;
projectContainer.addEventListener('mousedown', e => {
dragActive = true;
prevPos = e.clientX;
});
projectContainer.addEventListener('mouseup', () => dragActive = false);
projectContainer.addEventListener('mouseleave', () => dragActive = false);
projectContainer.addEventListener('mousemove', e => {
if (!dragActive) return;
const newTrans = projectSlider.currentTransX + e.clientX - prevPos;
projectSlider.style.transform = `translateX(${newTrans}px)`;
projectSlider.currentTransX = newTrans;
prevPos = e.clientX;
});
// Generate initial layout
function init() {
let workingLeft = styles.margin;
projects.forEach((project, i) => {
if (i === projects.length - 1) {
project.style.left = `-${space - styles.margin}px`;
} else {
i !== 0 && (workingLeft += space);
project.style.left = `${workingLeft}px`;
};
});
projectSlider.currentTransX = 0;
};
// Intersection observer
function observe() {
const callback = (entries, observer) => {
entries.forEach(entry => {
if (entry.isIntersecting) {
// Find intersecting edge
const { left } = entry.boundingClientRect;
const isLeftEdge = left < projectContainer.clientWidth - left;
// Test and reposition next element
const targetIdx = projects.findIndex(project => project === entry.target);
let nextIdx = null;
const nextEl = () => projects[nextIdx];
const targetLeft = parseInt(entry.target.style.left);
const nextLeft = () => parseInt(nextEl().style.left);
if (isLeftEdge) {
nextIdx = targetIdx === 0 ? projects.length-1 : targetIdx - 1;
nextLeft() > targetLeft && (nextEl().style.left = `${targetLeft - space}px`);
} else {
nextIdx = targetIdx === projects.length-1 ? 0 : targetIdx + 1;
nextLeft() < targetLeft && (nextEl().style.left = `${targetLeft + space}px`);
};
};
});
};
const observer = new IntersectionObserver(callback, {root: projectContainer});
projects.forEach(project => observer.observe(project));
};
init();
observe();
});
*, *::before, *::after{
margin:0px;
padding:0px;
box-sizing: border-box;
font-size:0px;
user-select: none;
}
.project-container {
font-size: 0px;
width: 100%;
height: 400px;
background-color: rgb(15, 207, 224);
margin:auto;
margin-top:60px;
white-space: nowrap;
overflow: hidden;
}
.project-slider {
position: relative;
}
.project {
font-size:100px;
display: block;
position: absolute;
top: 40px;
height:300px;
width:350px;
background-color:red;
border: black 3px solid;
user-select: none;
}
<div class="project-container">
<div class="project-slider">
<div class="project">1</div>
<div class="project">2</div>
<div class="project">3</div>
<div class="project">4</div>
<div class="project">5</div>
<div class="project">6</div>
<div class="project">7</div>
<div class="project">8</div>
</div>
</div>
There is still an issue here which is how to resize the elements for smaller screens, and on browser resizes. You would have to add another event listener for window resizes which resets the positions and styles at certain breakpoints, and also determine the style variables programmatically when the page first loads. I believe this would still have been a partial issue with the original implementation so you'd have to address it at some point either way.
I am writing an HTML5 board game and I am having issues with jQuery's offset(). The grid of DIVs that make up the game board reside within a wrapper DIV that has CSS that sets overflow:scroll/width and height:100%. The game board itself is quite larger, so the scroll can be quite a bit horizontally and vertically.
The problem is that when I click on a DIV and try to move the player to that board piece, the player shifts around and is never in a consistent place relative to the board piece clicked.
Code:
$(".boardGridPiece").click(function(){
if(!$(this).hasClass("room") && $(this).hasClass("eligibleMove")){
playerStartX = $(this).offset().left;
playerStartY = ($(this).offset().top;
player.css("left", playerStartX);
player.css("top", playerStartY);
determineEligibleMoves($(this).attr("id"));
}
});
You can see that when a board piece is clicked, the offset of the board piece is grabbed and set to the player's X and Y.
CSS:
#boardWrapper {
position:relative;
width:100%;
height:80%;
overflow: scroll;
}
#theGame {
background-color: #fff;
height: 1080px;
width: 1920px;
}
Depending on where the player is relative to the current scroll view, when I click on a board piece he shifts around in a very inconsistent manner. Sometimes he's far left of where I click, or far up, etc.
What am I doing wrong? How do I take into account relative scroll position to get consistent positioning?
Here's a board for you to play with as an example. Also, my stab at the jquery is included as well. Basically, it finds where you clicked, calculated the px distance in float form, and animates the soldier to slide to his new position:
var black = '<td style="background-color:black"><div class="boardGridPiece"></div></td>';
var white = '<td style="background-color:white"><div class="boardGridPiece"></div></td>';
var squares = [black, white];
var grid = "";
for (var i = 0; i < 9; i++) {
grid += "<tr>";
for (var j = 0; j < 16; j++) {
grid += squares[(i + j) % 2];
}
grid += "<\tr>";
}
$('#gameboard').append(grid);
var gridSelected = $('#gameboard').find('tr:nth-child(2)').find('td:first').find('div');
gridSelected.toggleClass('position');
$('.boardGridPiece').click(function () {
$('.position').removeClass('position');
var gridSelected = $(this);
gridSelected.toggleClass('position');
var thisBox = $('.position');
var finalX = 0;
var finalY = 0;
for (var i = 0; i < 9.00; i++) {
for (var j = 0; j < 16.00; j++) {
var aBox = $('#gameboard').find('tr:nth-child(' + (i + 1) + ')').find('td:nth-child(' + (j + 1) + ')').find('div');
if (thisBox.get(0) == aBox.get(0)) {
finalX = j + 1;
finalY = i;
i = j = 16; // soft break
}
}
}
var overX = (finalX * parseFloat(1920))/16.00;
var downY = (finalY * parseFloat(1080))/9.00;
$('#player').animate({ left: overX, top: downY });});
html {
background-color:gray;
}
#gameboard {
margin: 100px;
height:1080px;
width:1920px;
border:1px solid black;
}
.boardGridPiece {
height: 110px;
width: 110px;
}
.position {
box-sizing:border-box;
-moz-box-sizing:border-box;
-webkit-box-sizing:border-box;
border: 1px solid red;
}
#player {
position: absolute;
top: 120px;
left: 125px;
height: 200px;
width: auto;
}
<script src="https://ajax.googleapis.com/ajax/libs/jquery/2.1.1/jquery.min.js"></script>
<table id="gameboard" style=""></table>
<img id="player" src="http://img2.wikia.nocookie.net/__cb20110819034426/halo/images/7/74/ODST_Helljumper.png" />