This code is from here http://unstoppablerobotninja.com/entry/fluid-images
It takes all the images from a page and makes them fluid. It works great and i am sure many of you know it already.
My question is how can i apply this code only to certain images on the page, either through an ID or through a CLASS. something like this:
<img alt="" src="fluid_image.png" id="makeitfuild"/>
I don't want all my pictures to be fluid but still i want to use the code to certain images.
Please help as i don't know what i need to modify in the code
var imgSizer = {
Config : {
imgCache : []
,spacer : "/path/to/your/spacer.gif"
}
,collate : function(aScope) {
var isOldIE = (document.all && !window.opera && !window.XDomainRequest) ? 1 : 0;
if (isOldIE && document.getElementsByTagName) {
var c = imgSizer;
var imgCache = c.Config.imgCache;
var images = (aScope && aScope.length) ? aScope : document.getElementsByTagName("img");
for (var i = 0; i < images.length; i++) {
images.origWidth = images.offsetWidth;
images.origHeight = images.offsetHeight;
imgCache.push(images);
c.ieAlpha(images);
images.style.width = "100%";
}
if (imgCache.length) {
c.resize(function() {
for (var i = 0; i < imgCache.length; i++) {
var ratio = (imgCache.offsetWidth / imgCache.origWidth);
imgCache.style.height = (imgCache.origHeight * ratio) + "px";
}
});
}
}
}
,ieAlpha : function(img) {
var c = imgSizer;
if (img.oldSrc) {
img.src = img.oldSrc;
}
var src = img.src;
img.style.width = img.offsetWidth + "px";
img.style.height = img.offsetHeight + "px";
img.style.filter = "progid:DXImageTransform.Microsoft.AlphaImageLoader(src='" + src + "', sizingMethod='scale')"
img.oldSrc = src;
img.src = c.Config.spacer;
}
// Ghettomodified version of Simon Willison's addLoadEvent() -- http://simonwillison.net/2004/May/26/addLoadEvent/
,resize : function(func) {
var oldonresize = window.onresize;
if (typeof window.onresize != 'function') {
window.onresize = func;
} else {
window.onresize = function() {
if (oldonresize) {
oldonresize();
}
func();
}
}
}
}
You'll want to change the following line:
var images = (aScope && aScope.length) ? aScope : document.getElementsByTagName("img");
to something that will get your images by a class name. There are a few ways to do that: you can use jQuery, in which case it simply becomes
var images = (aScope && aScope.length) ? aScope : $("#myClassName");
or, you can look up any of the questions about how to get elements simply by class using only javascript (How to getElementByClass instead of GetElementById with Javascript? and How to Get Element By Class in JavaScript? both quickly jump to mind).
Related
I have faced such a problem. I need to make a preloader with a percentage for a page but I don't know how. Actually, I don't need animation or simple preloader. What do I can and what do I have?
window.onload = function() {
var images = document.images,
imagesTotalCount = images.length,
imagesLoadedCount = 0,
preloader = document.getElementById('js_preloader'),
percDisplay = document.getElementById('js_preload__percentage');
for(var i = 0; i < imagesTotalCount; i++) {
image_clone = new Image();
image_clone.onload = image_loaded;
image_clone.onerror = image_loaded;
image_clone.src = images[i].src;
}
function image_loaded() {
imagesTotalCount++;
percDisplay.innerHTML = (((100 / imagesTotalCount) * imagesLoadedCount) << 0) + '%';
if(imagesLoadedCount >= imagesTotalCount) {
setTimeout(function() {
if(!preloader.classList.contains('done')) {
preloader.classList.add('done');
}
}, 1500);
}
}
};
This aproach allows to see all images to be downloaded and and calculate percentage. But how do I can also take in count the download of css and js files?
You could use the same approach, but with document.scripts and document.styleSheets collections.
Basically below is my script for a prototype which uses 128x128 tiles to draw a map on a canvas which user can drag to move around.
Script does work. However I have a few problems to be solved:
1. Poor performance and I can't figure out why.
2. I am missing a method to buffer the tiles before the actual drawing.
3. If you notice any other issues also that could help me to make things run more smoothly it would be fantastic.
Some explanations for the script:
variables
coordinates - Defines the actual images to be displayed. Image file names are type of '0_1.jpg', where 0 is Y and 1 is X.
mouse_position - As name says, is keeping record of mouse position.
position - This is a poorly named variable. It defines the position of the context drawn on canvas. This changes when user drags the view.
Any assistance would be appreciated greatly. Thank you.
var coordinates = [0, 0];
var mouse_position = [0, 0];
var position = [-128, -128];
var canvas = document.getElementById('map_canvas');
var context = canvas.getContext('2d');
var buffer = [];
var buffer_x = Math.floor(window.innerWidth/128)+4;
var buffer_y = Math.floor(window.innerHeight/128)+4;
var animation_frame_request = function() {
var a = window.requestAnimationFrame;
var b = window.webkitRequestAnimationFrame;
var c = window.mozRequestAnimationFrame;
var d = function(callback) {
window.setTimeout(callback, 1000/60);
}
return a || b || c || d;
}
var resizeCanvas = function() {
window.canvas.width = window.innerWidth;
window.canvas.height = window.innerHeight;
window.buffer_x = Math.floor(window.innerWidth/128)+4;
window.buffer_y = Math.floor(window.innerHeight/128)+4;
window.buffer = [];
for (row = 0; row < window.buffer_y; row++) {
x = [];
for (col = 0; col < window.buffer_x; col++) {
x.push(new Image());
}
window.buffer.push(x);
}
}
var render = function() {
animation_frame_request(render);
for (row = 0; row < window.buffer_y; row++) {
for (col = 0; col < window.buffer_x; col++) {
cy = window.coordinates[1]+row;
cx = window.coordinates[0]+col;
window.buffer[row][col].src = 'map/'+cy+'_'+cx+'.jpg';
}
}
for (row = 0; row < window.buffer_y; row++) {
for (col = 0; col < window.buffer_x; col++) {
window.context.drawImage(window.buffer[row][col],
window.position[0]+col*128,
window.position[1]+row*128, 128, 128);
}
}
}
var events = function() {
window.canvas.onmousemove = function(e) {
if (e['buttons'] == 1) {
window.position[0] += (e.clientX-window.mouse_position[0]);
window.position[1] += (e.clientY-window.mouse_position[1]);
if (window.position[0] >= 0) {
window.position[0] = -128;
window.coordinates[0] -= 1;
} else if (window.position[0] < -128) {
window.position[0] = 0;
window.coordinates[0] += 1;
}
if (window.position[1] >= 0) {
window.position[1] = -128;
window.coordinates[1] -= 1;
} else if (window.position[1] < -128) {
window.position[1] = 0;
window.coordinates[1] += 1;
}
render();
}
window.mouse_position[0] = e.clientX;
window.mouse_position[1] = e.clientY;
}
}
window.addEventListener('resize', resizeCanvas, false);
window.addEventListener('load', resizeCanvas, false);
window.addEventListener('mousemove', events, false);
resizeCanvas();
To get better performance you should avoid changing the src of img nodes and move them around instead.
A simple way to minimize the number of img nodes handled and modified (except for screen positioning) is to use an LRU (Least Recently Used) cache.
Basically you keep a cache of last say 100 image nodes (they must be enough to cover at least one screen) by using a dictionary mapping the src url to a node object and also keeping them all in a doubly-linked list.
When a tile is required you first check in the cache, and if it's already there just move it to the front of LRU list and move the img coordinates, otherwise create a new node and set the source or, if you already hit the cache limit, reuse the last node in the doubly-linked list instead. In code:
function setTile(x, y, src) {
var t = cache[src];
if (!t) {
if (cache_count == MAXCACHE) {
t = lru_last;
t.prev.next = null;
lru_last = t.prev;
t.prev = t.next = null;
delete cache[t.src]
t.src = src;
t.img.src = src;
cache[t.src] = t;
} else {
t = { prev: null,
next: null,
img: document.createElement("img") };
t.src = src;
t.img.src = src;
t.img.className = "tile";
scr.appendChild(t.img);
cache[t.src] = t;
cache_count += 1;
}
} else {
if (t.prev) t.prev.next = t.next; else lru_first = t.next;
if (t.next) t.next.prev = t.prev; else lru_last = t.prev;
}
t.prev = null; t.next = lru_first;
if (t.next) t.next.prev = t; else lru_last = t;
lru_first = t;
t.img.style.left = x + "px";
t.img.style.top = y + "px";
scr.appendChild(t.img);
}
I'm also always appending the requested tile to the container so that it goes in front of all other existing tiles; this way I don't need to remove old tiles and they're simply left behind.
To update the screen I just iterate over all the tiles I need and request them:
function setView(x0, y0) {
var w = scr.offsetWidth;
var h = scr.offsetHeight;
var iy0 = y0 >> 7;
var ix0 = x0 >> 7;
for (var y=iy0; y*128 < y0+h; y++) {
for (var x=ix0; x*128 < x0+w; x++) {
setTile(x*128-x0, y*128-y0, "tile_" + y + "_" + x + ".jpg");
}
}
}
most of the time the setTile request will just update the x and y coordinates of an existing img tag, without changing anything else. At the same time no more than MAXCACHE image nodes will be present on the screen.
You can see a full working example in
http://raksy.dyndns.org/tiles/tiles.html
I need to adjust the script from http://javascript.about.com/library/blcvert.htm to change direction of scrolling to DOWN.
Could anybody help?
Of course, it would be also helpful if anybody knows/have some other script which produces the same effect.
Thanx
P.S. the script (in readable format is):
var imgAry1 = ['img1.png','img2.png'];
function startCloud() {
new mq('clouds', imgAry1, 380);
mqRotate(mqr);
}
$(document).ready(function() {
startCloud();
});
var mqr = [];
function mq(id, ary, heit) {
this.mqo=document.getElementById(id);
var wid = this.mqo.style.width;
this.mqo.onmouseout=function() { mqRotate(mqr); };
this.mqo.onmouseover=function() { clearTimeout(mqr[0].TO); };
this.mqo.ary=[];
var maxw = ary.length;
for (var i=0;i<maxw;i++) {
this.mqo.ary[i]=document.createElement('img');
this.mqo.ary[i].src=ary[i];
this.mqo.ary[i].style.position = 'absolute';
this.mqo.ary[i].style.top = (heit*i)+'px';
this.mqo.ary[i].style.height = heit+'px';
this.mqo.ary[i].style.width = wid;
this.mqo.appendChild(this.mqo.ary[i]);
}
mqr.push(this.mqo);
}
function mqRotate(mqr) {
if (!mqr) return;
for (var j=mqr.length - 1; j > -1; j--) {
maxa = mqr[j].ary.length;
for (var i=0;i<maxa;i++) {
var x = mqr[j].ary[i].style;
x.top=(parseInt(x.top,10)-1)+'px';
}
var y = mqr[j].ary[0].style;
if (parseInt(y.top,10)+parseInt(y.height,10)<0) {
var z = mqr[j].ary.shift();
z.style.top = (parseInt(z.style.top) + parseInt(z.style.height)*maxa) + 'px';
mqr[j].ary.push(z);
}
}
mqr[0].TO=setTimeout('mqRotate(mqr)',10);
}
On this line:
x.top=(parseInt(x.top,10)-1)+'px';
it says that you take x.top in pixels, parse out the number, subtract one and add the 'px' again. The element's position from top is decreased by 1 each time, so it goes up. All you need to do for it to go down is to add the one.
x.top=(parseInt(x.top,10)+1)+'px';
I also tested this hypothesis on the page you linked :)
I'm building a navigation bar where the images should be swapped out on mouseover; normally I use CSS for this but this time I'm trying to figure out javascript. This is what I have right now:
HTML:
<li class="bio"><img src="images/nav/bio.jpg" name="bio" /></li>
Javascript:
if (document.images) {
var bio_up = new Image();
bio_up.src = "images/nav/bio.jpg";
var bio_over = new Image();
bio_over.src = "images/nav/bio-ov.jpg";
}
function over_bio() {
if (document.images) {
document["bio"].src = bio_over.src
}
}
function up_bio() {
if (document.images) {
document["bio"].src = bio_up.src
}
}
However, all of the images have names of the form "xyz.jpg" and "xyz-ov.jpg", so I would prefer to just have a generic function that works for every image in the navbar, rather than a separate function for each image.
A quick-fire solution which should be robust enough provided all your images are of the same type:
$("li.bio a").hover(function() {
var $img = $(this).find("img");
$img[0].src = $img[0].src.replace(".jpg", "") + "-ov.jpg";
}, function() {
var $img = $(this).find("img");
$img[0].src = $img[0].src.replace("-ov.jpg", "") + ".jpg";
});
This should work will all image formats as long as the extension is between 2 and 4 characters long IE. png, jpeg, jpg, gif etc.
var images = document.getElementById('navbar').getElementsByTagName('img'), i;
for(i = 0; i < images.length; i++){
images[i].onmouseover = function(){
this.src = this.src.replace(/^(.*)(\.\w{2,4})$/, '$1'+'-ov'+'$2');
}
images[i].onmouseout = function(){
this.src = this.src.replace(/^(.*)-ov(\.\w{2,4})$/, '$1'+'$2');
}
}
Here's an idea in plain javascript (no jQuery):
function onMouseOverSwap(e) {
e.src = e.src.replace(/\.jpg$/", "-ov.jpg"); // add -ov onto end
}
function onMouseOutSwap(e) {
e.src = e.src.replace(/(-ov)+\.jpg$/, ".jpg"); // remove -ov
}
I've got a little problem here. I've been trying to do an image gallery with JavaScript but there's something that I got a problem with. I can get the image to resize when the first image load, but as soon as I load another image, it won't resize anymore! Since the user will be able to upload a lot of different size pictures, I really need to make it work.
I've checked for ready-to-use image gallery and such and nothing was doing what I need to do.
Here's my javascript:
function changeCurrentImage(conteneur)
{
var img = conteneur.getElementsByTagName("img");
var imgUrl = img[0].src;
var imgFirstPart = imgUrl.substring(0, imgUrl.lastIndexOf('.') - 9);
var imgLastPart = imgUrl.substring(imgUrl.lastIndexOf('.'));
var currentImg = document.getElementById('currentImage');
currentImg.src = imgFirstPart + "borne" + imgLastPart;
resize(document.getElementById('currentImage'), 375, 655);
}
function resize(img, maxh, maxw) {
var ratio = maxh/maxw;
if (img.height/img.width > ratio){
// height is the problem
if (img.height > maxh){
img.width = Math.round(img.width*(maxh/img.height));
img.height = maxh;
}
} else {
// width is the problem
if (img.width > maxw){
img.height = Math.round(img.height*(maxw/img.width));
img.width = maxw;
}
}
};
Here's the HTML (using ASP.Net Repeater):
<asp:Repeater ID="rptImages" runat="server">
<HeaderTemplate>
</HeaderTemplate>
<ItemTemplate>
<a href="#">
<div id="thumbnailImageContainer1" onclick="changeCurrentImage(this)">
<div id="thumbnailImageContainer2">
<img id="thumbnailImage" src="<%# SiteUrl + Eval("ImageThumbnailPath")%>?rn=<%=Random()%>" alt="Photo" onload="resize(this, 60, 105)" />
</div>
</div>
</a>
</ItemTemplate>
<FooterTemplate>
</FooterTemplate>
</asp:Repeater>
Most likely the image is not yet downloaded so img.height and img.width are not yet there. Technically you don't need to wait till the whole image is downloaded, you can poll the image in a timer until width and height are non-zero. This sounds messy but can be done nicely if you take the time to do it right. (I have an ImageLoader utility I made for this purpose....has only one timer even if it is handling multiple images at once, and calls a callback function when it has the sizes) I have to disagree with Marcel....client side works great for this sort of thing, and can work even if the images are from a source other than your server.
Edit: add ImageLoader utility:
var ImageLoader = {
maxChecks: 1000,
list: [],
intervalHandle : null,
loadImage : function (callback, url, userdata) {
var img = new Image ();
img.src = url;
if (img.width && img.height) {
callback (img.width, img.height, url, 0, userdata);
}
else {
var obj = {image: img, url: url, callback: callback,
checks: 1, userdata: userdata};
var i;
for (i=0; i < this.list.length; i++) {
if (this.list[i] == null)
break;
}
this.list[i] = obj;
if (!this.intervalHandle)
this.intervalHandle = setInterval(this.interval, 30);
}
},
// called by setInterval
interval : function () {
var count = 0;
var list = ImageLoader.list, item;
for (var i=0; i<list.length; i++) {
item = list[i];
if (item != null) {
if (item.image.width && item.image.height) {
item.callback (item.image.width, item.image.height,
item.url, item.checks, item.userdata);
ImageLoader.list[i] = null;
}
else if (item.checks > ImageLoader.maxChecks) {
item.callback (0, 0, item.url, item.checks, item.userdata);
ImageLoader.list[i] = null;
}
else {
count++;
item.checks++;
}
}
}
if (count == 0) {
ImageLoader.list = [];
clearInterval (ImageLoader.intervalHandle);
delete ImageLoader.intervalHandle;
}
}
};
Example usage:
var callback = function (width, height, url, checks, userdata) {
// show stuff in the title
document.title = "w: " + width + ", h:" + height +
", url:" + url + ", checks:" + checks + ", userdata: " + userdata;
var img = document.createElement("IMG");
img.src = url;
// size it to be 100 px wide, and the correct
// height for its aspect ratio
img.style.width = "100px";
img.style.height = ((height/width)*100) + "px";
document.body.appendChild (img);
};
ImageLoader.loadImage (callback,
"http://upload.wikimedia.org/wikipedia/commons/thumb/" +
"1/19/Caerulea3_crop.jpg/800px-Caerulea3_crop.jpg", 1);
ImageLoader.loadImage (callback,
"http://upload.wikimedia.org/wikipedia/commons/thumb/" +
"8/85/Calliphora_sp_Portrait.jpg/402px-Calliphora_sp_Portrait.jpg", 2);
With the way you have your code setup, I would try and call your resize function from an onload event.
function resize() {
var img = document.getElementById('currentImage');
var maxh = 375;
var maxw = 655;
var ratio = maxh/maxw;
if (img.height/img.width > ratio){
// height is the problem
if (img.height > maxh){
img.width = Math.round(img.width*(maxh/img.height));
img.height = maxh;
}
} else {
// width is the problem
if (img.width > maxw){
img.height = Math.round(img.height*(maxw/img.width));
img.width = maxw;
}
}
};
function changeCurrentImage(conteneur)
{
var img = conteneur.getElementsByTagName("img");
img.onload = resize;
var imgUrl = img[0].src;
var imgFirstPart = imgUrl.substring(0, imgUrl.lastIndexOf('.') - 9);
var imgLastPart = imgUrl.substring(imgUrl.lastIndexOf('.'));
var currentImg = document.getElementById('currentImage');
currentImg.src = imgFirstPart + "borne" + imgLastPart;
}
I would play around with that. Maybe use global variables for your maxH/W and image ID(s);
#Comments: No, I can't do that server side since it would refresh the page everytime someone click on a new image. That would be way too bothersome and annoying for the users.
As for the thumbnails, those image are already saved in the appropriate size. Only the big image that shows is about 33% of its size. Since we already have 3 images PER uploaded images, I didn't want to upload a 4th one for each upload, that would take too much server space!
As for the "currentImage", I forgot to add it, so that might be helful lol:
<div id="currentImageContainer">
<div id="currentImageContainer1">
<div id="currentImageContainer2">
<img id="currentImage" src="#" alt="" onload="resize(this, 375, 655)" />
</div>
</div>
</div>
#rob: I'll try the ImageLoader class, that might do the trick.
I found an alternative that is working really well. Instead of changing that IMG width and height, I delete it and create a new one:
function changeCurrentImage(conteneur)
{
var thumbnailImg = conteneur.getElementsByTagName("img");
var thumbnailImgUrl = thumbnailImg[0].src;
var newImgUrl = thumbnailImgUrl.replace("thumbnail", "borne");
var currentImgDiv = document.getElementById('currentImageContainer2');
var currentImg = currentImgDiv.getElementById("currentImage");
if (currentImg != null)
{
currentImgDiv.removeChild(currentImg);
}
var newImg = document.createElement("img");
newImageDiv = document.getElementById('currentImageContainer2');
newImg.id = "currentImage";
newImg.onload = function() {
Resize(newImg, 375, 655);
newImageDiv.appendChild(newImg);
}
newImg.src = newImgUrl;
}
Also, in case people wonder, you MUST put the .onload before the .src when assigning an a new source for an image!