Screenreaders will read whatever string is set to the "alt" attribute. The use of this attribute is specifically for image tags.
If I have a div like so:
<div id=myCoolDiv tabindex="0"> 2 <div>
Is there a way to have a screen reader pickup an attribute to read a string the same way an alt tag is used?
So for the div listed below, the screen reader will say ie: "shopping cart items 2"?
I tried using aria-label but the screenreader won't pick it up:
<div id=myCoolDiv tabindex="0" aria-label="shopping cart items"> 2 <div>
You can just put a title tag in the div which will do the same as an alt tag like so:
<div title="I AM HELLO WORLD">HELLO WORLD</div>
"I AM HELLO WORLD" will be printed once you move your cursor around it on a browser
There are two ways (which can be combined) to have screen reader to read alternative text:
Anything with ARIA role img can (MUST) have alt attribute. See WAI-ARIA img role.
<div role="img" alt="heart">
♥︎
</div>
UPDATE: In 2017 the WAI-ARIA document was changed and the following text does not apply anymore. See comments below.
However this should be used only in case the element really represent an image (e.g. the heart unicode character).
If an element contain actual text, that just need different reading, you should set ARIA role to text and add aria-label with whatever you want to be read by the screen reader. See WAI-ARIA text role.
<div role="text" aria-label="Rating: 60%">
Rating: ★★★☆☆︎
</div>
Do not mismatch it with aria-labeledby which should contain ID of an related element.
You can combine the previous two cases into one using two ARIA roles and adding both alt and aria-label:
<div role="img text" alt="heart" aria-label="heart">
♥︎
</div>
When more ARIA roles are defined, browser should use the first one that is supported and process the element with that role.
One last important thing is that you must set page type to HTML5 (which support ARIA by design).
<!DOCTYPE html>
Using HTML4 or XHTML requires special DTD to enable ARIA support.
<!DOCTYPE html PUBLIC "-//W3C//DTD XHTML+ARIA 1.0//EN"
"http://www.w3.org/WAI/ARIA/schemata/xhtml-aria-1.dtd">
Try role="listitem" or role="group" and aria-labelledby="shopping cart items". See Example 1. The 2 is text content which should be read by screen reader already with the attribute read as context to the content. Refer to this section.
UPDATE 2
Add aria-readonly=true role=textbox if you use an input. If there are doubts whether to use aria-label or aria-labelledby, read this article. In the documentation for JAWS and testing it myself supports the fact that aria-label is ignored. Furthermore, semantics are very important when accessibility is your concern. Using a div when you could use an input is not semantically sound and like I said before, JAWS would accept a form element more readily than a div. I assume that this "shopping cart" is a form or part of a form, and if you don't like it's borders, input {border: 0 none transparent} or use <output>* which would be A+ as far as semantics are concerned.
Sorry, #RadekPech reminded me; I forgot to add that using aria-labelledby needs visible text and that the text needs an id which is also listed as the value(s) of aria-labelledby. If you don't want text because of aesthetics, use color: transparent, line-height: 0, or color:<same as background>. That should satisfy visibility as far as the DOM is concerned* and still be invisible to the naked eye. Keep in mind these measures are because JAWS ignores aria-label.
*untested
EXAMPLE 3
<span id="shopping">Shopping</span>
<span id="cart">Cart</span>
<span id="items">Items</span>
<input id='cart' tabindex="0" aria-readonly=true readonly role="textbox" aria-labelledby="shopping cart items" value='2'>
UPDATE 1
For JAWS, you probably have to configure it a little:
Click the Utilities menu item.
Then Settings Center.
Speech and Sounds Schemes
Modiy Scheme...
HTML Tab
In this particular dialog box, you can add specific attributes and what is said when an element is tabbed to. JAWS will respond to form elements easier because they can trigger the focus event. You'll have an easier time doing Example 2 instead:
EXAMPLE 1
<div id=myCoolDiv tabindex="0" role="listitem" aria-labelledby="shopping cart items"> 2 <div>
EXAMPLE 2
<input id='semantic' tabindex="0" role="listitem" aria-labelledby="shopping cart items" value='2' readonly>
In case you use Bootstrap Framework there is a quick and easy solution. You should use sr-only or sr-only sr-only-focusable Bootstrap's CSS classes in a span element where your screen-reader-only text will be written.
Check the following example, a span element with class glyphicon glyphicon-shopping-cart is also used as cart icon.
<div id="myCoolDiv">
<h5>
<span class="glyphicon glyphicon-shopping-cart"></span> 2
<span class="sr-only sr-only-focusable" tabindex="0">shopping cart items</span>
</h5>
<div>
Screen Reader Output: "two shopping cart items"
provided by Fangs Screen Reader Emulator Addon for Firefox
You can find the above working example in this: Fiddle
As suggested by Oriol, in case you don't use Bootstrap Framework then simply add the following in your CSS file.
.sr-only {
position: absolute;
width: 1px;
height: 1px;
padding: 0;
margin: -1px;
overflow: hidden;
clip: rect(0, 0, 0, 0);
border: 0;
}
.sr-only-focusable:active,
.sr-only-focusable:focus {
position: static;
width: auto;
height: auto;
margin: 0;
overflow: visible;
clip: auto;
}
According to the text alternative computation algorithm of the W3C and the
Accessible Name and Description: Computation and API Mappings 1.1 you definitely should use aria-label.
That being said, it does not work with Jaws. Text alternative is only computed for elements having an ARIA role.
The remaining option is to use a link that will go to your cart page, using both title and aria-label to satisfy anyone:
2
You can also use a transparent 1 pixel option:
2 <img src="pixel.png" height="1" width="1" alt="shopping cart items" />
No, there is no equivalent to an alt attribute for <div> elements.
For what you are trying to do, an ARIA-based solution is overkill. Not only are you bumping into screen reader compatibility problems, you are applying ARIA attributes where they are not needed (and arguably do not belong if on something like a <div>).
Instead, consider using an off-screen technique (such as this one from The Paciello Group or this one from WebAIM). Content hidden using this technique will still be read by screen readers but will be visually hidden.
From reading your question, I think this is what you are after.
I made a pen demonstrating this technique. It may be easier to test in the full-page version.
Edit: Added HTML and CSS from the example, but please note that both the specs and browser / assistive technology support change over time, so if you are reading this in a year you should continue to use the links above to verify this CSS is still the current best practice.
HTML
<div tabindex="0">
<span class="offscreen">Items in shopping cart: </span>2
</div>
CSS
.offscreen {
position: absolute;
clip: rect(1px 1px 1px 1px);
/* for Internet Explorer */
clip: rect(1px, 1px, 1px, 1px);
padding: 0;
border: 0;
height: 1px;
width: 1px;
overflow: hidden;
}
Accessibility (Screen readers) can be achieved through role and aria-label tags on div. This can be very useful while using svg.
<div role="img" aria-label="I can speak the text">
<svg>...</svg>
</div>
Try:
HTML
<div id=myCoolDiv tabindex="0"><span class="aria-hidden">shopping cart items</span>2<div>
CSS
.aria-hidden {
position: absolute;
left: -100000px;
}
This will announce the text inside the span. And the Parent div will not lose visual focus. Aria-hidden class will hide the span from the visible screen area but will read it as its inside the div that has focus.
You can create a class such as screen-reader-text with the following css:
.screen-reader-text {
clip: rect(1px, 1px, 1px, 1px);
height: 1px;
width: 1px;
overflow: hidden;
position: absolute !important;
}
Then, in your code, you can just add a <span> with the screenreader text as so:
<div>
I am a div!
<span class="screen-reader-text">This is my screen reader text</span>
</div>
See an example over here: https://jsfiddle.net/zj1zuk9y/
(Source: http://www.coolfields.co.uk/2016/05/text-for-screen-readers-only-updated/)
Use an image inside the div that has the label as its alt attribute. That way, those without screen readers just see the number and an image, whereas those with readers will hear the whole sentence:
<div>
<img src="http://tny.im/57j" alt="Shopping cart items" />
2
</div>
Seen as:
2
Read as: "Shopping cart items: 2"
The alt attribute exists for images because there is no way to "read aloud" the content of the image, so the provided text is used instead. But for the div, it already contains text and images. Therefore, if you want it to be read by a screen-reader, you need to include the text and alt text in the content of the div.
Related
I'm working on a project where you can insert your working hours, in which you have to insert start- & end times. But after you put in the start time and tab to go on, it does focus first on the icon, which comes from the
<input type="time">
I want that the tabbing only hits the input area, not the icon next to it.
This is the current state:
example input field:
<div class="text-section">
<label for="startTime"
class="text label without-margin">#((MarkupString)loc["StartTime"].ToString())</label>
<div class="property-container-three">
<div class="property-icon">
<div class="icon solid start" tabIndex="-2">#Icons.Startzeit</div>
</div>
<input class="input-text nomargin centerInputContentVertical"
type="time"
id="startTime"
#ref="startTimeElement"
#bind-value="blazorStartTime"
#onblur="HandleStartTimeSet">
</div>
</div>
I already tried everything with tabindex:"-1", it just makes no difference. Also I'm just able to modify this icon due css, which goes like:
input[type="text"], input[type="date"], input[type="number"], input[type="search"], input[type="time"], input[type="datetime"], input[type="password"], [type="email"], select, textarea {
padding: 8px 12px;
height: 38px;
width: 100%;
}
I do not have any more ideas or approaches...
After some googling I found, it is a known issue with Edge... see this answer, it states that Microsoft do not plan to fix it; but the link they mention is dead.
I can only replicate this bug on Edge. And it seems MS won't solve it...
You can target it with CSS: idea from here
input[type="time"]::-webkit-calendar-picker-indicator` {
display: none:
}
Perhaps setting display: none will be enough and maybe adjusting padding/margin for it too?
Unfortunately, there is currently no stable CSS way to change the tab-index; and currently no way to change the HTML attributes.
The current CSS equivalent for tab-index is -moz-user-focus but this is non-standard and the documentation stresses that you "should not use this".
Similar things exist for grabbing the pseudo element with JavaScript like this question, but again this is for computedStyles which is back to the CSS issue again.
Maybe in future this sort of feature will be introduced and there will be a working answer for it....
I'm trying to figure out out to create a event to appear in front of my home without it opening a new page. It would, for lack of a better word, expand to fill the browser. I know I'll have to do some work with z-index and javascript. The month would hover and then the user would click to see the event.
Home and event
My HTML
<div class= "month sep_box">
<h1 class= "sep">SEP</h1>
<div class= "year">2016</div>
</div>
CSS
.sep_box{
background-image: url("images/design_disrupt.svg");
background-repeat: no-repeat;
background-size: cover;
background-clip: content-box;
background-position: center;
float: left;
width: 25%;
height: 25vh;
transition:.25s ease;
}
EDIT: Screen-shoted HTML now copied
<article>
<div><h1 id="design_disruptors">
DESIGN <br />DISRUPTORS</h1></div>
<div><p class="child_day">THURSDAY</p></div>
<div><p class="child_day_number">15</p></div>
<div><p class="child_event_about">JCM 2121<br />7:00pm</p></div>
<div><p class="child_rsvp">RSVP</p></div>
<div><p class="child_desc">Design Disruptors reveals<br />
a never-before-seen<br />
perspective on the design approaches of these<br />
companies and how they<br />
are overtaking billion dollar industries though design.</p>
</div>
</article>
https://jsfiddle.net/es60r7cv/
The comments aren't going to work at this point because of the character limit, so I'm going to try my best to give you some hints here. I am a little unsure as to how far along you are in your development to this point, and the intent of the design, but let's give it a shot.
Firstly, if I understand your design image correctly, you want almost the entire screen to look different except for the square that was clicked. This is going to be difficult, as you'll need to position a lot of elements in just such a way that you can have a transparency in exactly the right spot. Given your design, and how important pixel-perfection is going to be to making it work, and where you are in your development, I'm wondering if it would be ideal to simply fix the width of the whole design (no growing or shrinking with the screen).
I would also recommend you use jQuery for this project, as it will be easier for you.
To add an event listener to all your month boxes using jQuery, you would write it:
$(document).on('click', '.month', function (evt) {
// your event handling code here
}
I would give each month element an id for the month it represented, then create your overlays with a similar id. So, for example, the December month box would be <div class="month" id="december"><!--your_content--></div> and the overlay for the month could be <div class="overlay" id="decemberOverlay"><!--your_overlay_content--></div>. That way you could target it by getting the clicked month boxes id, and getting the overlay by doing that id + "Overlay".
You could fetch your overlay content on the fly using AJAX, but to reduce complexity for yourself you may just always load all overlays to the page and hide them with CSS, but also include the positioning code:
.overlay {
display: none;
z-index: 10;
position: absolute; /* this will position it to the document, or the first parent that is relatively or absolutely positioned */
top: 0;
left: 0
}
We are using absolute positioning because:
we want to be able to position the overlay directly over the original image, and not influence the flow of the rest of the document, and
z-index requires some non-static position value to be applied
Then, in your script, you would update it do be this:
$(document).on('click', '.month', function (evt) {
var clickedMonth = this.id;
var correspondingOverlay = $(clickedMonth+"Overlay");
correspondingOverlay.show();
}
Based on your fiddle and code, I think perhaps you are not very far along yet. Hopefully this gives you a bit of a head start on how to achieve your desired result.
Edit:
One last thing-- this is a cleaner way to style your markup:
<article>
<div>
<h1 id="design_disruptors">DESIGN <br />DISRUPTORS</h1>
</div>
<div>
<p class="child_day">THURSDAY</p>
</div>
<div>
<p class="child_day_number">15</p>
</div>
<div>
<p class="child_event_about">JCM 2121<br />7:00pm</p>
</div>
<div>
<p class="child_rsvp">RSVP</p>
</div>
<div>
<p class="child_desc">
Design Disruptors reveals<br />
a never-before-seen<br />
perspective on the design approaches of these<br />
companies and how they<br />
are overtaking billion dollar industries though design.
</p>
</div>
</article>
Clean HTML will be easier to read and easier to spot errors.
I have a page of text and it is formatted similar to this
<div class="container">
<p style="text-align: center;">
<span style="text-decoration: underline;">
<strong>
<img src="//cdn.shopify.com/s/files/1/0652/9219/files/Horizontal.jpg?13817493546805475501" alt="">
</strong>
</span>
</p>
<p style="text-align: center;">
<span style="text-decoration: underline;">
<strong>The Hosting</strong>
</span>
</p>
<p>
The in-laws are arriving, friends are in town, and everyone is heading to your abode for a night filled with holiday cheer. As stress levels tend to rise during these events, expenses do as well. Here are a few tips to nail your hostess game, without breaking the bank and <em>still</em> shopping consciously.
</p>
</div>
I am looking to keep the images which fit the entire content width of the class container the same but only change the text within the paragraph tags to either be a smaller width (so it looks indented on both sides) or have margins but not affect the images at all. I cannot change how the code is outputted so the images will always be wrapped in paragraph tags.
This code is a small sample on the page of content and there are several images and text throughout.
So basically I am looking for a way in css to style only the actual text within the paragraph tags and leave the images unchanged. Any help would be great.
Here is a fiddle example: https://jsfiddle.net/jpautt8v/
If your html is static or if you know which child you want to modify then you could simply use the nth child css selector to apply css like below 3rd in your sample code case. You could play around margins and see what works best for your solution.
p:nth-child(3)
{
margin: 0 50px;
}
Without changing any of the existing markup, you can accomplish what you want using negative margin.
Something like this will work:
img {
width: 120%; max-width: 120%;
margin: 0 -10%;
}
.content > div, .content > p {
margin: 0 10%;
}
You can see it working in this fiddle: https://jsfiddle.net/igor_9000/jpautt8v/1/
Sorry if this is a dumb question but I'm very new to HTML and javascript.
I'm trying to click a button/link on web page not having a unique identifier (at least that's what I'm seeing).
Below is the snippet of the HTML Code.
Is it possible if I could reference the span value/text immediately below the 'a' element (e.g. "Search Incident", "New Incident")?
Or are there any other ways I could go about this?
"<div class="item EP lvl1 " navmode="1" artype="NavBarItem" arid="app1598" lvl="1" arwindowid="0" style="height: 25px; overflow: visible; border-top: medium none; border-bottom: medium none;">
<a class="btn" style="z-index: 1;" onclick="javascript:CallARGHPD_58INC_58AppListEntryPointamrrm6100EPFunc(false, this);">
<span class="navLabel lvl1 ">
New Incident
</span>
</a>
</div>
<div class="item EP lvl1 " navmode="1" artype="NavBarItem" arid="app1599" lvl="1" arwindowid="0" style="height: 23px; overflow: visible; border-left: 1px solid rgb(…, 233); border-top: medium none; border-bottom: medium none;">
<a class="btn" style="z-index: 1;" onclick="javascript:CallARGHPD_58INC_58AppListEntryPoint_95Searchamrrm6100EPFunc(false, this);">
<span class="navLabel lvl1 ">
Search Incident
</span>
</a>
</div>
</div>
</div>"
EDIT
This is what I've done so far:
Store the whole form into a variable by referencing the ID and filtering by tag name, as below;
var links = IE.Document.getElementById("WIN_0_80077").getElementsByTagName("a");
Basically the form is outlined like below, clicking through each like a folder structure, to access the inner link.
Form
| Incident Management
| New Incident (links to another page)
| Search Incident (links to another page)
| Problem Management
| New Problem (links to another page)
| Search Problem (links to another page)
Each has span attribute with their corresponding text as described on the outline.
Navigate to the inner structure by clicking each button/link (e.g. Incident Mgt > Search Incident)
- Im trying to achieve this by looping on the links using span text as reference (or any viable reference)
for (link in links) {
for (span in link.getElementsByTagName("span")) {
if (condition) //validation for span text {
span.click;
}
}
}
Does this even make sense at all? Again, sorry if it did not. Appreciate you help.
UPDATE:
Thanks for the suggestions.
It might not be the best solution but at the least, I was able to make a way out by implementing below.
// Reference the intended element/class/item
var link = objIE.Document.querySelectorAll('a[onclick="javascript:CallARGHPD_58INC_58AppListEntryPoint_95Searchamrrm6100EPFunc(false, this);"]');
// Do something with it (e.g. click the link)
link[0].click();
You could do this a number of ways in JavaScript but one way would be to reference the span elements by class using
var labels = document.querySelectorAll('.navLabel');
This will return all the elements with the class 'navLabel'
Then to reference the individual label, you could do:
var firstLabel = label[0].innerHTML;
var secondLabel = label[1].innerHTML;
Again, there are multiple ways to do this, it just depends what you want to do with the references after you get them.
Simply use:
var label=document.getElementByClassName('navLabel lvl1');
label[0].value; //New Incident
label[1].value; //Search Incident
I'd like to implement this situation: user hovers link and on his screen appears div with additional information.
There is no problem, to generate div with absolute position, populate data and display it with jQuery, but the problem is with maintainability. I want to separate logic and view. What if I'd like to change page layout in the future? How programmer will know, that some part of page exist in JavaScript file?
Is it some elegant way to separate view (in my case HTML structure) and logic (data obtained from server in JSON using JS script) and combine them? Is it any ... templating engine or something like that in JavaScript?
You can generate your HTML entries like this
<a>
My Cool Link
<div class="tooltip">
This link is awesome!
</div>
</a>
And use CSS to style the tooltip:
(The important part is a:hover .tooltip and everything except background on .tooltip
a {
display: inline-block;
}
a:hover .tooltip {
visibility: visible;
}
.tooltip {
position: relative;
top: 20px;
left: -50%;
display: inline;
visibility: hidden;
background: #eee;
}
Example: http://jsfiddle.net/4K7sB/
Then you just need to populate the elements with the correct text using JS.
Bootstrap does all this and more: http://getbootstrap.com/javascript/#tooltips
To put it in:
$('#example').tooltip(options)
And the markup being:
<div class="tooltip">
<div class="tooltip-inner">
Tooltip text here!
</div>
<div class="tooltip-arrow"></div>
</div>
Go to the link for a more detailed explanation on how to implement it
You can alter the Title="information" dynamicaly by some program or script and use the following code to display it as a tool tip,found it on net
http://www.webdesignerdepot.com/2012/11/how-to-create-a-simple-css3-tooltip/
Just add title attribute in div tag like below code....
<div title="First Div">StackOverFlow StackOverFlow StackOverFlow
StackOverFlow StackOverFlow StackOverFlow StackOverFlow StackOverFlow
StackOverFlow </div>
</div>
<br /><br />
<div title="Second Div">StackOverFlow StackOverFlow StackOverFlow
StackOverFlow StackOverFlow StackOverFlow StackOverFlow StackOverFlow
StackOverFlow </div>
</div>
just copy paste and check.