I'm aware that there are plenty of methods to use SVG sprites in HTML. My preference to this date has been to use something like
<div class="container">
<svg class="icon">
<title>Icon Title</title>
<use xlink:href="/svg/sprite.svg#icon"/>
</svg>
</div>
However, now I wanted to load the sprite from a subdomain, like this:
<div class="container">
<svg class="icon">
<title>Icon Title</title>
<use xlink:href="https://sub.domain.com/svg/sprite-home.svg#icon"/>
</svg>
</div>
Unfortunately, this doesn't work as the file is not fetched. I've tried with <object> as well, but that doesn't seem to work either.
So far, I'm only able to use
<?php include_once("https://sub.domain.com/svg/sprite.svg"); ?>
It's ok as a quick fix, as this doesn't involve much refactoring. However, this also means the HTML gets bloated.
With the <use> method the sprite gets cached. But with the include method the sprite isn't cached, only gets embedded, and so it is far from ideal.
Does anybody use any alternative to the php include method that is compatible with cross domain requests and browser caching?
Thanks to this post at css-tricks I've been able to work out how to do this. The idea is to AJAX to bring the SVG sprite with jQuery like this (see post for vanilla version):
$j.get("https://sub.domain.com/svg/sprite-home.svg", function(data) {
var div = document.createElement("div");
div.className = 'no-display';
div.innerHTML = new XMLSerializer().serializeToString(data.documentElement);
document.body.insertBefore(div, document.body.childNodes[0]);
});
What this does is insert the SVG at the beginning of the document. Unlike the original post, I've added a class to make it hidden, as otherwise you get a blank big space at the top in Chrome. The result is great (it works with local files too) and now you can reference icons by just their id.
<div class="container">
<svg class="icon">
<title>Icon Title</title>
<use xlink:href="#icon"/>
</svg>
</div>
There are many advantages to this technique:
SVG sprite is cached
Really simple to use as you only reference the icon
You can request several SVG sprites and they all work the same
The only thing to bear in mind is that, this requires CORS AJAX to be set up. For those using nginx, it would be simple enough:
location ~* \.svg$ {
...
add_header 'Access-Control-Allow-Methods' 'GET';
add_header 'Access-Control-Allow-Origin' 'https://your.domain.com';
}
Related
I'm playing around with a basic Dot-Net web assembly application. In the application I'm displaying two images using two different image tags image and img. The size of the image is bound to a private variable Size. I've noticed a problem where images do not render in a specific scenario using the image tag.
Replication:
dotnet new blazorwasm
I downloaded the SVG from: Bootstrap icons, then I placed the SVG file in "wwwroot/Media/".
In index.razor I've updated the code as follows:
#page "/"
<PageTitle>Index</PageTitle>
<h1>Hello, world!</h1>
Welcome to your new app.
<img src="Media/search.svg" alt="Bootstrap" width="#Size" height="#Size">
<image src="Media/search.svg" alt="Bootstrap" width="#Size" height="#Size"/>
#code
{
private static string Size => "75";
}
The result of running the above code shows only one image displaying
Through process of elimination, the image defined using the image tag is the problem here. If you tweak the code to use hardcoded values i.e.
<image src="Media/search.svg" alt="Bootstrap" width="75" height="75"/>
then the code works again as expected.
I'm aware that <image> is deprecated, but I'd like to understand if the reason the binding breaks the image displaying is due to the deprecation or something else?
Update
The generated HTML using the template is
<!--!--><div class="top-row px-4" b-vv8m6rf2se="">About</div>
<article class="content px-4" b-vv8m6rf2se=""><!--!--><!--!--><!--!--><!--!-->
<!--!--><h1 tabindex="-1">Hello, world!</h1>
Welcome to your new app.
<img src="Media/search.svg" alt="Bootstrap" width="75" height="75">
<image src="Media/search.svg" alt="Bootstrap" width="75" height="75"></image></article>
An interesting find, although of course not of any practical value, just use <img>.
I could easily reproduce this with a jpg image so it's not about svg.
Now for a speculative answer:
Blazor treats <image> like any other tag and the generated HTML looks like expected. But according to this answer,
The HTML5 parsing spec requires that the tag is mapped to the img element at the tree construction stage
This makes me think that when the complete tag is rendered just once it works fine, handling is up to the browser.
But after Blazor has filled in the #Size it will try to update the HTML it generated earlier. If the Browser really changed <image> to <img> internally then the JS Bridge will have trouble finding the element again and the updates fail.
In our project we are calling all the icons like this:
<svg focusable="false" aria-hidden="true" class="icon icon--extra-small ">
<use xlink:href="#icon-search" href="#icon-search"></use>
</svg>
Right now we are loading all the icons in a hidden div at the footer of the document.
Is there a way to use an external source for the svg without having to rewrite all the icons?
I assume JavaScript is acceptable for your project. Here is an option that you could choose to do.
Create another file called svg-icons.js or anything really.
Add <script type="text/javascript" src="svg-icons.js"></script> below the closing </body> tag.
In that file, add something along the lines of:
document.querySelector('.my-icon').innerHTML = '<svg focusable="false" aria-hidden="true" class="icon icon--extra-small "><use xlink:href="#icon-search" href="#icon-search"></use></svg>';
Then whenever you want to use that icon in your html, add a div or span (according to your needs) with the same class:
<div class="my-icon"></div>
The component is fairly small, so I don't suggest doing this. Additionally, if you are using JQuery, you could look into $.load(). That will help a lot when starting to use the same components across the entire website.
I have large, wide images within a portfolio page. The images are saved "progressive" and they load fine.
I was wondering if there's a way though to kind of preload those images to make them appearing faster and smoother. Or maybe there's a way to preload all the other pages and images into the cache so that at least all the following pages after the first appear smooth and fast? Whatever helps to make the pages load faster and smoother.
Any suggestions?
Each image consists of a variety of images, all of them within one wide image (prepared in PSD) and the visitor can shift left and right to call for the respective image to appear in the center.
Unfortunately sacrificing on the image quality or make them smaller is not an option here.
I know there are posts here on preloading images ad stuff but I can't find any that work with the image embedded in the HTML code.
Please have merci, I'm a CSS and Javascript novice, so the simpler the more likely I'll understand it. I'm afraid breaking up the images in single instances (make it a row of images instead of one whole image), place them in a floated div and change the respective Javascript code could be too challenging for me, right...? How else could I do that?
Appreciated!
Here's what I have (I guess it would be overkill to post all my HTML, Javascript and CSS here, I'll post some). The large images are placed within the HTML page and called via Javascript.
see here
<div class="ShiftGroup">
<div class="ShiftGroupC">
<div class="ShiftGroupI"><div id="ShiftGalleryFive"><img src="images/gallery_anzic1.png" width="3348" height="372" alt="imagegallery1" /></div></div>
<div class="ShiftGroupP" style="margin-left: -990px;"><div id="ShiftLeft" class="ShiftGroupD"><span class="pointer"><img src="images/arrowleft.png" width="78" height="50" alt="arrowsleft" /></span></div></div>
<div class="ShiftGroupP" style="margin-left: 341px;"><div id="ShiftRight" class="ShiftGroupD"><span class="pointer"><img src="images/arrowright.png" width="78" height="50" alt="arrowright" /></span></div></div>
and
gallery = document.getElementById('ShiftGalleryFour').style;
This is how we preloaded images in one of our projects:
preloadImages = function(imageIndex) {
// Finds div element containing index (1..n) and url of image
// (both stored as data attributes)
var image = $(".data .image[data-index=" + imageIndex + "]");
// Creates an image (just virtually) that is not embedded to HTML.
// Its source is the url of the image that is to be preloaded
var preloadedImage = $(new Image()).attr("src", image.attr("data-full-url"));
// Bind an event to the "virtual" image to start preloading next image when
// this one is done
preloadedImage.load(function() {
// Start preloading the next one
preloadImages(imageIndex + 1);
});
}
// Start preloading the first image
preloadImages(1)
In your case this solves only one part of the problem - preloading.
I see you include all images in html as img tags. If you want to achieve better performance, do not place any img tags in your html of the gallery. Just div tags that will become the future containers of your images. These divs may have indexes and contain data attributes with image urls (as seen in my example). When your page gets loaded, start preloading procedure. When an "virtual image" gets loaded. Create new image tag inside its container and start preloading the next image.
This will definitely cut off the download time of your page.
My example uses jQuery which simplifies the script. Pure javascript would be more complicated.
UPDATE:
This is how preloading example may work like.
HTML
Let's say you have 4 images and all of them has its container - a div in which individual image is to be placed.
<div class="images">
<div class="image" data-index="1" data-src="image_1_source_path"></div>
<div class="image" data-index="2" data-src="image_2_source_path"></div>
<div class="image" data-index="3" data-src="image_3_source_path"></div>
<div class="image" data-index="4" data-src="image_4_source_path"></div>
</div>
JavaScript
After the the document is loaded, preloading procedure may start. You start by preloading the first image. After this one is loaded, you append it to its container and trigger preloading of the next image. There is also return called if all images are loaded and no container is found.
preloadImages = function(imageIndex) {
var imageContainer = $(".images .image[data-index=" + imageIndex + "]");
return if imageContainer.length === 0
var preloadedImage = $(new Image()).attr("src", image.attr("data-full-url"));
preloadedImage.load(function() {
imageContainer.append(preloadedImage);
preloadImages(imageIndex + 1);
});
}
$(document).ready(function(){
preloadImages(1);
});
Hopefully you get the idea.
Is there a way to tell the browser to look down a list of image URLs until it finds one that works? Pure HTML would be preferred, but I'm guessing JavaScript is probably necessary here (I'm already using JQuery, so it's not an issue).
EDIT: Thanks for your answers! I'll add a few clarifications:
By "works" I mean the image can be displayed.
I specifically want to do this on the client side.
This seems like a bad idea to me. What is the purpose of this feature? It sounds like you want something equivalent to this:
<img src="/images/file1.jpg" src2="/images/file2.jpg" src3="/images/file3.jpg">
Where the browser would try each file in succession. The problem with this approach is that it significantly increases the http traffic required and the latency. The best approach is to dynamically construct the page using the correct image tags ahead of time. Using a server-side approach you can try to load the image from the disk (or database or wherever the images are) and dynamically include the best url in the page's image tag.
If you insist on doing it client-side, you can try loading multiple image tags:
<img src="file1.jpg" alt="" onerror="this.style.display='none'">
<img src="file2.jpg" alt="" onerror="this.style.display='none'">
<img src="file3.jpg" alt="" onerror="this.style.display='none'">
<img src="file4.jpg" alt="" onerror="this.style.display='none'">
<img src="file5.jpg" alt="" onerror="this.style.display='none'">
<img src="file6.jpg" alt="" onerror="this.style.display='none'">
This will result in a page that appears to have lots of images but they disappear as the page loads. The alt="" is required to make Opera not show the broken image placeholder; the onerror is required for Chrome and IE.
If that's not spiffy enough, and if all your images are the same size, and that size is known, you could stack a bunch of images one on top of the other, so that the first image that loads hides all the others. This worked for me in Opera, FF, and IE8. It loads the last image in the list that exists. Note that this wastes bandwidth and memory because every image is loaded.
<div style="width: 50px; height:38px; background-image: url(file1.jpg);">
<div style="width: 50px; height:38px; background-image: url(file2.jpg);">
<div style="width: 50px; height:38px; background-image: url(file3.jpg);">
<div style="width: 50px; height:38px; background-image: url(file4.jpg);">
<div style="width: 50px; height:38px; background-image: url(file5.jpg);">
<div style="width: 50px; height:38px; background-image: url(file6.jpg);">
<div style="width: 50px; height:38px; background-image: url(file7.jpg);">
</div></div></div></div></div></div>
Finally, there is the JavaScript approach:
<!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 4.01 Transitional//EN" "http://www.w3.org/TR/html4/loose.dtd">
<html>
<head>
<title></title>
<script type="text/javascript">
var image_array = ['file1.jpg', 'file2.jpg', 'file3.jpg', 'file4.jpg', 'file5.jpg','file6.jpg' ];
function load_img(imgId, image, images, index) {
image.onerror = function() {
load_img(imgId, this, images, index+1);
};
image.onload = function() {
var element = document.getElementById(imgId);
if (element) {
element.src = this.src;
element.style.display = 'block';
}
};
image.src = images[index];
}
</script>
</head>
<body>
<img id="id_1" alt="" style="display: none;">
</body>
<script>
load_img('id_1', new Image(), image_array, 0);
</script>
</html>
If you're trying setting multiple sources to the image tag depending on the resolution, srcset is the paramenter you're looking for.
<img src="images/space-needle.jpg"
srcset="images/space-needle.jpg 1x, images/space-needle-2x.jpg 2x,
images/space-needle-hd.jpg 3x">
If I am reading the specification correctly, you should be able to do this with the HTML object element. <object> tags can be nested and thereby provide a chain of resources that are tried each in turn to be rendered and upon failure the user agent continues with the next one.
Note, though, that this behaviour is/was buggy for several browsers and versions.
Assuming you mean the browser being able to retrieve some content with an HTTP response code 200 for a specific URL, then the answer is : NO from the client side using only HTML.
In other words, you can't have an element (e.g. img) and specify multiple URLs to "try".
Of course you can craft something on the server side: a request comes in for resource X and the server has a list of URLs that "work".
INAJNBAM (I'm not a Javascript Ninja by any means), but in pseudo code, maybe try something like this after the page has loaded: (OR, now that I think about it, this would work well with PHP too)
$images = array('img1.jpg', 'img2.jpg', 'img3.png'....)
foreach $images as $img
{if $img.height > 0px
{print "<img src="$img" />"
end}
};;;;
In fact PHP would be even better because I presume in JS this would result in images flashing up on the screen at the end of the pageload. Try it out in PHP and see if something like this fits your bill.
NOTE: I added 4 semi colons at the end. I know Javascript always wants 'em, I just didn't know where to stick them.
If by saying "works" you mean the image can be loaded, you can use the "load" function on an image( in your case a bunch of images) of jQuery and inside of it declare the functionality that will be fire once the loading of the an image is completed.
If by saying "works" you mean that the HTTP status code is ok then use an ajax call using jquery.
function getUrlStatus(url) {
$.ajax({
url: url,
complete: function(xhr) {
return xhr.status;
}
});
}
You could enter the URL of some server-side application/script that serves up the image from whatever image source it can find.
You could do this in ASP.Net with an HTTPHandler that sends a response of content-type=image/jpg.
Other than ASP.Net there are amny other server-side options such as Perl, PHP...
I have tried to change the images on my site from img to svg, changing img tags to embed and object tags. But, implementing the onclick function, which previously was contained in the img tag, is proving most difficult.
I found onclick had no effect when placed inside the object or embed tag.
So, I made a div exclusively for the svg, and placed onclick in this div tag. But, no effect unless visitor clicks on the edges/padding of the image.
I have read about overlaying a div, but am trying to avoid using absolute positioning, or specifying position at all.
Is there another way to apply onclick to a svg?
Has anyone encountered this problem? Questions and suggestions are welcome.
You can have an onclick event in the svg itself, I do this all the time in my work. make a rect over the space of your svg, (so define it last, remember svg uses the painters model)
rect.btn {
stroke:#fff;
fill:#fff;
fill-opacity:0;
stroke-opacity:0;
}
then as an attribute to the rect add the onclick (this can be done with js or jquery as well).
<div>
<svg xmlns="http://www.w3.org/2000/svg" version="1.1">
<g>
<circle ... //your img svg
<rect class="btn" x="0" y="0" width="10" height="10" onclick="alert('click!')" />
</g>
</svg>
</div>
this will work in almost all browsers: http://caniuse.com/svg
If you just use inline svg there is no problem.
<svg id="svg1" xmlns="http://www.w3.org/2000/svg" style="width: 3.5in; height: 1in">
<circle id="circle1" r="30" cx="34" cy="34" onclick="circle1.style.fill='yellow';"
style="fill: red; stroke: blue; stroke-width: 2"/>
</svg>
This started as a comment on RGB's solution but I could not fit it in so have converted it to an answer. The inspiration for which is entirely RGB's.
RGB's solution worked for me. However, I wished to note a couple of points which may help others arriving at this post (like me) who are not that familiar which SVG and who may very well have generated their SVG file from a graphics package (as I had).
So to apply RGB's solutions I used:
The CSS
<style>
rect.btn {
stroke:#fff;
fill:#fff;
fill-opacity:0;
stroke-opacity:0;
}
</style>
The jquery script
<script type="text/javascript" src="../_public/_jquery/jquery-1.7.1.js"></script>
<script type="text/javascript">
$("document").ready(function(){
$(".btn").bind("click", function(event){alert("clicked svg")});
});
</script>
The HTML to code the inclusion of your pre-existing SVG file in the group tag inside the SVG code.
<div>
<svg xmlns="http://www.w3.org/2000/svg" version="1.1">
<g>
<image x="0" y="0" width="10" height="10"
xlink:href="../_public/_icons/booked.svg" width="10px"/>
<rect class="btn" x="0" y="0" width="10" height="10"/>
</g>
</svg>
</div>
However, in my case I have several SVG icons which I wish to be clickable and incorporating each of these into the SVG tag was starting to become cumbersome.
So as an alternative approach where I could employ Classes I used jquery.svg. This is probably a shameful application of this plugin which can do all sorts of stuff with SVG's. But it worked using the following code:
<script type="text/javascript" src="../_public/_jquery/jquery-1.7.1.js"></script>
<script type="text/javascript" src="jquery.svg.min.js"></script>
<script type="text/javascript">
$("document").ready(function(){
$(".svgload").bind("click", function(event){alert("clicked svg")});
for (var i=0; i < 99; i++) {
$(".svgload:eq(" + i + ")").svg({
onLoad: function(){
var svg = $(".svgload:eq(" + i + ")").svg('get');
svg.load("../_public/_icons/booked.svg", {addTo: true, changeSize: false});
},
settings: {}}
);
}
});
</script>
where HTML
<div class="svgload" style="width: 10px; height: 10px;"></div>
The advantage to my thinking is that I can use the appropriate class where ever the icons are needed and avoid quite a lot of code in the body of the HTML which aids readability. And I only need to incorporate the pre-existing SVG file once.
Edit: Here is a neater version of the script courtesy of Keith Wood: using .svg's load URL setting.
<script type="text/javascript" src="../_public/_jquery/jquery-1.7.1.js"></script>
<script type="text/javascript" src="jquery.svg.min.js"></script>
<script type="text/javascript">
$("document").ready(function(){
$('.svgload').on('click', function() {
alert('clicked svg new');
}).svg({loadURL: '../_public/_icons/booked.svg'});
});
</script>
I got this working accross the latest versions of Firefox, Chrome, Safari and Opera.
It relies on a transparent div before the object that has absolute position and set width and height so it covers the object tag below.
Here it is, I've been a bit lazy and used inline styes:
<div id="toolbar" style="width: 600px; height: 100px; position: absolute; z-index: 1;"></div>
<object data="interface.svg" width="600" height="100" type="image/svg+xml">
</object>
I used the following JavaScript to hook up an event to it:
<script type="text/javascript">
var toolbar = document.getElementById("toolbar");
toolbar.onclick = function (e) {
alert("Hello");
};
</script>
In case you're fine with wrapping the svg in another element (a for example) and putting onclick on the wrapper, svg {pointer-events: none;} CSS will do the trick.
It worked by simply replacing the <embed/> tag with <img/> and deleting the type attribute.
For instance, in my code, instead of:
<embed src=\"./images/info_09c.svg\" type=\"image/svg+xml\" width=\"45\" onClick='afiseaza_indicatie($i, \"$indicatii[$i]\")'>
which does not answer the clicking, I wrote:
<img src=\"./images/info_09c.svg\" height=\"25\" width=\"25\" onClick='afiseaza_indicatie($i, \"$indicatii[$i]\")'>
It works in Internet Explorer and Google Chrome, and I hope that in the other browsers too.
You could use following code:
<style>
.svgwrapper {
position: relative;
}
.svgwrapper {
position: absolute;
z-index: -1;
}
</style>
<div class="svgwrapper" onClick="function();">
<object src="blah" />
</div>
b3ng0 wrote similar code but it does not work. z-index of parent must be auto.
When embedding same-origin SVGs using <object>, you can access the internal contents using objectElement.contentDocument.rootElement. From there, you can easily attach event handlers (e.g. via onclick, addEventListener(), etc.)
For example:
var object = /* get DOM node for <object> */;
var svg = object.contentDocument.rootElement;
svg.addEventListener('click', function() {
console.log('hooray!');
});
Note that this is not possible for cross-origin <object> elements unless you also control the <object> origin server and can set CORS headers there. For cross-origin cases without CORS headers, access to contentDocument is blocked.
Have you looked into using the CSS z-index property to make the container dev be "on top" of the svg? Because the div is (presumably) transparent, you will still see the image exactly as before.
This, I believe, is the best-practice, non-hack, intended way of solving your problem. z-index is only useful for elements that have a position property of fixed, relative, or, as you've heard, absolute. However, you don't actually have to move the object.
For example:
<style>
.svgwrapper {
position: relative;
z-index: 1;
}
</style>
<div class="svgwrapper" onClick="function();">
<object src="blah" />
</div>
For what it's worth, it would also be a little more elegant and safe to not use onClick at all, but instead to bind the click event using javascript. That's another issue altogether, though.
Assuming you don't need cross browser support (which is impossible without a plugin for IE), have you tried using svg as a background image?
Experimental stuff for sure, but thought I would mention it.
Perhaps what you're looking for is the SVG element's pointer-events property, which you can read about at the SVG w3C working group docs.
You can use CSS to set what happens to the SVG element when it is clicked, etc.
Click on SVG's <g> element in <object> with click event. Works 100%. Take a look on the nested javascript in <svg>. Don't forget to insert window.parent.location.href= if you want to redirect the parent page.
https://www.tutorialspoint.com/svg/svg_interactivity.htm
I wrapped the 'svg' tag in 'a' tag and put the onClick event in the 'a' tag
I had a similar issue: it only seems that the onclick event is not occurring, but it is firing twice
the svg tag had an onclick option, like: <svg ... onclick="someJsFunction() ...> it opens a dropdown menu, and had a class option named f.e. class-for-svg
the path (included in the svg tag) had no any class option,
in the other hand, I had a window.onclick function to close the dropdown when the user clicks on the somthing else, here the dropdown was clossing - when the object's class option matched to the svg's class option
so when I clicked on the path portion inside the svg tag - the window.onclick event accurred too, and because of the does not matching(to the class name), the dropdown was clossed imediatelly, and it seems that the onclick event is not happening )
in reality it happens twise, one opens, an the second closses the drop down
solution: add the same class option to the path tag too )
enjoy please )