background-attachment: fixed
in CSS, at best, does not work well in mobile browsers, and at worst is not even …
The Fixed Background Attachment Hack originally published on CSS-Tricks, which is part of the DigitalOcean family. You should get the newsletter.
]]>background-attachment: fixed
in CSS, at best, does not work well in mobile browsers, and at worst is not even supported by the most widely used mobile browsers. You can ditch this idea completely and let the background scroll on small screens using media queries.
Or get around it with a small fix. I suppose we could call it a “hack” since it’s a workaround in code that arguably we shouldn’t have to do at all.
Before I show you the fix, let’s examine the issue. We can see it by looking at two different approaches to CSS backgrounds:
I want to keep the background gradient in a fixed position on scroll, so let’s apply basic CSS styling to the body that does exactly that:
body {
background: linear-gradient(335deg, rgba(255,140,107,1) 0%, rgba(255,228,168,1) 100%);
background-attachment: fixed;
background-position: center;
background-repeat: no-repeat;
height: 100vh;
}
Here are the results in Chrome and Firefox, both on Android, respectively:
The gradient simply scrolls along with other content then jumps back. I don’t know exactly why that is — maybe when the URL tab goes up or disappears on scroll and the browser finds it difficult to re-render the gradient in real time? That’s my best guess since it only seems to happen in mobile browsers.
If you’re wondering about iOS Safari, I haven’t tested on iOS personally, but the issue is there too. Some have already reported the issue and it appears to behave similarly.
This issue with images is no different.
body {
background: url(../assets/test_pic.jpg);
background-repeat: no-repeat;
background-size: cover;
background-position: center;
background-attachment: fixed;
height: 100vh;
}
Another interesting thing to note is that when background-attachment: fixed
is applied, the height is ignored even if we explicitly specify it. That’s because background-attachment
calculates a fixed background position relative to the viewport.
Even if we say the body is 100vh, background-attachment: fixed
is not exactly in accordance with it. Weird! Perhaps the reason is that background-attachment: fixed
relies on the smallest possible viewport while elements rely on the largest possible viewport. David Bokan explains,
Lengths defined in viewport units (i.e.
vh
) will not resize in response to the URL bar being shown or hidden. Instead,vh
units will be sized to the viewport height as if the URL bar is always hidden. That is,vh
units will be sized to the “largest possible viewport”. This means100vh
will be larger than the visible height when the URL bar is shown.
The issues are nicely documented over at caniuse:
- Firefox does not appear to support the local value when applied on a textarea element.
- Chrome has an issue that occurs when using the
will-change
property on a selector which also hasbackground-attachment: fixed
defined. It causes the image to get cut off and gain whitespace around it.- iOS has an issue preventing
background-attachment: fixed
from being used withbackground-size: cover
.
Call it a temporary hack, if you will. Some of you may have already tried it. Whatever the case, it fixes the linear gradient and background image issues we just saw.
So, as you know, we are getting in trouble with the background-attachment: fixed
property and, as you might have guessed, we are removing it from our code. If it’s looking at the smallest possible viewport, then maybe we should be working with an element that looks for the largest possible viewport and position that instead.
So, we are creating two separate elements — one for the background-gradient
and another for the rest of the content. We are replacing background-attachment: fixed
with position: fixed
.
<div class="bg"></div>
<div class="content">
<!-- content -->
</div>
.bg {
background: linear-gradient(335deg, rgba(255,140,107,1) 0%, rgba(255,228,168,1) 100%);
background-repeat: no-repeat;
background-position: center;
height: 100vh;
width: 100vw;
position: fixed;
/* z-index usage is up to you.. although there is no need of using it because the default stack context will work. */
z-index: -1; // this is optional
}
Now, wrap up the rest of the content — except for the element containing the background image — inside a main container.
.content{
position: absolute;
margin-top: 5rem;
left: 50%;
transform: translateX(-50%);
width: 80%;
}
Success!
We can use the same trick hack with background images and it works fine. However, you do get some sort of background scrolling when the URL bar hides itself, but the white patch is no longer there.
.img {
background: url('../assets/test_pic.jpg');
background-position: center;
background-repeat: no-repeat;
background-size: cover;
position: fixed;
height: 100vh;
width: 100vw;
}
.content {
position: absolute;
left: 50%;
margin-top: 5rem;
transform: translateX(-50%);
width: 80%;
}
A fixed-position element with a height set to 100% behaves just like the element with background-attachment: fixed
property, which is clearly evident in the example below! Just observe the right-most bar (purple color) in the video.
Even though, David Bokan in his article states that:
That is, a
position: fixed
element whose containing block is the ICB will resize in response to the URL bar showing or hiding. For example, if its height is100%
it will always fill exactly the visible height, whether or not the URL bar is shown. Similarly forvh
lengths, they will also resize to match the visible height taking the URL bar position into account.
If we take into account that last sentence, that doesn’t seem to be the case here. Elements that have fixed positioning and 100vh height don’t change their height whether the URL bar is shown or not. In fact, the height is according to the height of the “largest possible viewport”. This is evident in the example below. Just observe the light blue colored bar in the video.
So, it appears that, when working with a container that is 100vh, background-attachment: fixed
considers the smallest possible viewport height while elements in general consider the largest possible viewport height.
For example, background-attachment: fixed
simply stops working when a repaint is needed, like when a mobile browser’s address bar goes away on scroll. The browser adjusts the background according to the largest possible viewport (which is now, in fact, the smallest possible viewport as URL bar is hidden) and the browser isn’t efficient enough to repaint on the fly, which results in a major lag.
Our hack addresses this by making the background an element instead of, well, an actual background. We give the element containing the content an absolute position to stack it on top of the element containing the image, then apply a fixed position on the latter. Hey, it works!
Note that the viewport height is calculated excluding the navigation bar at the bottom (if present). Here’s a comparison between the presence and absence of navigation bar at the bottom in Chrome Android.
Is there a downside? Perhaps! We’re using a general <div>
instead of an actual <img>
tag, so I wouldn’t say the markup is semantic. And that can lead to accessibility issues. If you’re working with an image that adds meaning or context to the content, then an <img>
is the correct way to go, utilizing a proper alt
description for screen readers.
But if we go the proper <img>
route, then we’re right back where we started. Also, if you have a navigation bar at the bottom which too auto hides itself, then I can’t help it. If the hack won’t cut it, then perhaps JavaScript can come to the rescue.
The Fixed Background Attachment Hack originally published on CSS-Tricks, which is part of the DigitalOcean family. You should get the newsletter.
]]>Typewriter Animation That Handles Anything You Throw at It originally published on CSS-Tricks, which is part of the DigitalOcean family. You should get the newsletter.
]]>Like Kevin, I decided to recreate the animation, but open it up to JavaScript. That way, we have a few extra tools that can make the typing feel a little more natural and even more dynamic. Many of the CSS solutions rely on magic numbers based on the length of the text, but with JavaScript, we can make something that’s capable of taking any text we throw at it.
So, let’s do that. In this tutorial, I’m going to show that we can animate multiple words just by changing the actual text. No need to modify the code every time you add a new word because JavaScript will do that for you!
Let’s start with text. We are using a monospace font to achieve the effect. Why? Because each character or letter occupies an equal amount of horizontal space in a monospaced font, which will come handy when we’ll use the concept of steps()
while animating the text. Things are much more predictable when we already know the exact width of a character and all characters share the same width.
We have three elements placed inside a container: one element for the actual text, one for hiding the text, and one for animating the cursor.
<div class="container">
<div class="text_hide"></div>
<div class="text">Typing Animation</div>
<div class="text_cursor"></div>
</div>
We could use ::before
and ::after
pseudo-elements here, but they aren’t great for JavaScript. Pseudo-elements are not part of the DOM, but instead are used as extra hooks for styling an element in CSS. It’d be better to work with real elements.
We’re completely hiding the text behind the .text_hide
element. That’s key. It’s an empty div that stretches the width of the text and blocks it out until the animation starts—that’s when we start to see the text move out from behind the element.
In order to cover the entire text element, position the .text_hide
element on top of the text element having the same height and width as that of the text element. Remember to set the background-color
of the .text_hide
element exactly same as that of the background surrounding the text so everything blends in together.
.container {
position: relative;
}
.text {
font-family: 'Roboto Mono', monospace;
font-size: 2rem;
}
.text_hide {
position: absolute;
top: 0;
bottom: 0;
left: 0;
right: 0;
background-color: white;
}
Next, let’s make that little cursor thing that blinks as the text is being typed. We’ll hold off on the blinking part for just a moment and focus just on the cursor itself.
Let’s make another element with class .text_cursor
. The properties are going to be similar to the .text_hide
element with a minor difference: instead of setting a background-color
, we will keep the background-color transparent
(since its technically unnecessary, then add a border to the left edge of the new .text_cursor
element.
.text_cursor{
position: absolute;
top: 0;
bottom: 0;
left: 0;
right: 0;
background-color: transparent;
border-left: 3px solid black;
}
Now we get something that looks like a cursor that’s ready to move as the text moves:
Now comes the super fun part—let’s animate this stuff with JavaScript! We’ll start by wrapping everything inside a function called typing_animation()
.
function typing_animation(){
// code here
}
typing_animation();
Next task is to store each and every character of text in a single array using the split()
method. This divides the string into a substring that has only one character and an array containing all the substrings is returned.
function typing_animation(){
let text_element = document.querySelector(".text");
let text_array = text_element.innerHTML.split("");
}
For example, if we take “Typing Animation” as a string, then the output is:
We can also determine the total number of characters in the string. In order to get just the words in the string, we replace split("")
with split(" ")
. Note that there is a difference between the two. Here, " "
acts as a separator. Whenever we encounter a single space, it will terminate the substring and store it as an array element. Then the process goes on for the entire string.
function typing_animation(){
let text_element = document.querySelector(".text");
let text_array = text_element.innerHTML.split("");
let all_words = text_element.innerHTML.split(" ");
}
For example, for a string ‘Typing Animation’, the output will be,
Now, let’s calculate the length of the entire string as well as the length of each and every individual word.
function typing_animation() {
let text_element = document.querySelector(".text");
let text_array = text_element.innerHTML.split("");
let all_words = text_element.innerHTML.split(" ");
let text_len = text_array.length;
const word_len = all_words.map((word) => {
return word.length;
});
}
To get the length of the entire string, we have to access the length of the array containing all the characters as individual elements. If we’re talking about the length of a single word, then we can use the map()
method, which accesses one word at a time from the all_words
array and then stores the length of the word into a new array called word_len
. Both the arrays have the same number of elements, but one contains the actual word as an element, and the other has the length of the word as an element.
Now we can animate! We’re using the Web Animation API because we’re going with pure JavaScript here—no CSS animations for us in this example.
First, let’s animate the cursor. It needs to blink on and off infinitely. We need keyframes and animation properties, both of which will be stored in their own JavaScript object. Here are the keyframes:
document.querySelector(".text_cursor").animate([
{
opacity: 0
},
{
opacity: 0, offset: 0.7
},
{
opacity: 1
}
], cursor_timings);
We have defined three keyframes as objects which are stored in an array. The term offset: 0.7
simply means that after 70% completion of the animation, the opacity will transition from 0 to 1.
Now, we have to define the animation properties. For that, let’s create a JavaScript object that holds them together:
let cursor_timings = {
duration: 700, // milliseconds (0.7 seconds)
iterations: Infinity, // number of times the animation will work
easing: 'cubic-bezier(0,.26,.44,.93)' // timing-function
}
We can give the animation a name, just like this:
let animation = document.querySelector(".text_cursor").animate([
// keyframes
], //properties);
Here’s a demo of what we have done so far:
Great! Now, let’s animate the .text_hide
element that, true to its name, hides the text. We define animation properties for this element:
let timings = {
easing: `steps(${Number(word_len[0])}, end)`,
delay: 2000, // milliseconds
duration: 2000, // milliseconds
fill: 'forwards'
}
The easing
property defines how the rate of animation will change over time. Here, we have used the steps()
timing function. This animates the element in discrete segments rather than a smooth continuous animation—you know, for a more natural typing movement. For example, the duration of the animation is two seconds, so the steps()
function animates the element in 9
steps (one step for each character in “Animation”) for two seconds, where each step has a duration of 2/9 = 0.22
seconds.
The end
argument makes the element stay in its initial state until the duration of first step is complete. This argument is optional and its default value is set to end
. If you want an in-depth insight on steps()
, then you can refer this awesome article by Joni Trythall.
The fill
property is the same as animation-fill-mode
property in CSS. By setting its value to forwards
, the element will stay at the same position as defined by the last keyframe after the animation gets completed.
Next, we will define the keyframes.
let reveal_animation_1 = document.querySelector(".text_hide").animate([
{ left: '0%' },
{ left: `${(100 / text_len) * (word_len[0])}%` }
], timings);
Right now we are animating just one word. Later, we will see how to animate multiple words.
The last keyframe is crucial. Let’s say we want to animate the word “Animation.” Its length is 9
(as there are nine characters) and we know that it’s getting stored as a variable thanks to our typing_animation()
function. The declaration 100/text_len
results to 100/9
, or 11.11%, which is the width of each and every character in the word “Animation.” That means the width of each and every character is 11.11% the width of the entire word. If we multiply this value by the length of the first word (which in our case is 9
), then we get 100%. Yes, we could have directly written 100% instead of doing all this stuff. But this logic will help us when we are animating multiple words.
The result of all of this is that the .text_hide
element animates from left: 0%
to left: 100%
. In other words, the width of this element decreases from 100% to 0% as it moves along.
We have to add the same animation to the .text_cursor
element as well because we want it to transition from left to right along with the .text_hide
element.
Yayy! We animated a single word. What if we want to animate multiple words? Let’s do that next.
Let’s say we have two words we want typed out, perhaps “Typing Animation.” We animate the first word by following the same procedure we did last time. This time, however, we are changing the easing function value in the animation properties.
let timings = {
easing: `steps(${Number(word_len[0] + 1)}, end)`,
delay: 2000,
duration: 2000,
fill: 'forwards'
}
We have increased the number by one step. Why? Well, what about a single space after a word? We must take that into consideration. But, what if there is only one word in a sentence? For that, we will write an if
condition where, if the number of words is equal to 1, then steps(${Number(word_len[0])}, end)
. If the number of words is not equal to 1, then steps(${Number(word_len[0] + 1)}, end)
.
function typing_animation() {
let text_element = document.querySelector(".text");
let text_array = text_element.innerHTML.split("");
let all_words = text_element.innerHTML.split(" ");
let text_len = text_array.length;
const word_len = all_words.map((word) => {
return word.length;
})
let timings = {
easing: `steps(${Number(word_len[0])}, end)`,
delay: 2000,
duration: 2000,
fill: 'forwards'
}
let cursor_timings = {
duration: 700,
iterations: Infinity,
easing: 'cubic-bezier(0,.26,.44,.93)'
}
document.querySelector(".text_cursor").animate([
{
opacity: 0
},
{
opacity: 0, offset: 0.7
},
{
opacity: 1
}
], cursor_timings);
if (all_words.length == 1) {
timings.easing = `steps(${Number(word_len[0])}, end)`;
let reveal_animation_1 = document.querySelector(".text_hide").animate([
{ left: '0%' },
{ left: `${(100 / text_len) * (word_len[0])}%` }
], timings);
document.querySelector(".text_cursor").animate([
{ left: '0%' },
{ left: `${(100 / text_len) * (word_len[0])}%` }
], timings);
} else {
document.querySelector(".text_hide").animate([
{ left: '0%' },
{ left: `${(100 / text_len) * (word_len[0] + 1)}%` }
], timings);
document.querySelector(".text_cursor").animate([
{ left: '0%' },
{ left: `${(100 / text_len) * (word_len[0] + 1)}%` }
], timings);
}
}
typing_animation();
For more than one word, we use a for
loop to iterate and animate every word that follows the first word.
for(let i = 1; i < all_words.length; i++){
// code
}
Why did we take i = 1
? Because by the time this for
loop is executed, the first word has already been animated.
Next, we will access the length of the respective word:
for(let i = 1; i < all_words.length; i++){
const single_word_len = word_len[i];
}
Let’s also define the animation properties for all words that come after the first one.
// the following code goes inside the for loop
let timings_2 = {
easing: `steps(${Number(single_word_len + 1)}, end)`,
delay: (2 * (i + 1) + (2 * i)) * (1000),
duration: 2000,
fill: 'forwards'
}
The most important thing here is the delay
property. As you know, for the first word, we simply had the delay
property set to two seconds; but now we have to increase the delay for the words following the first word in a dynamic way.
The first word has a delay of two seconds. The duration of its animation is also two seconds which, together, makes four total seconds. But there should be some interval between animating the first and the second word to make the animation more realistic. What we can do is add a two-second delay between each word instead of one. That makes the second word’s overall delay 2 + 2 + 2
, or six seconds. Similarly, the total delay to animate the third word is 10 seconds, and so on.
The function for this pattern goes something like this:
(2 * (i + 1) + (2 * i)) * (1000)
…where we’re multiplying by 1000 to convert seconds to milliseconds.
Length of the word | Duration taken by one character to animate |
6 | 2/6 = 0.33 seconds |
8 | 2/8 = 0.25 seconds |
9 | 2/9 = 0.22 seconds |
12 | 2/12 = 0.17 seconds |
The longer the word, the faster it is revealed. Why? Because the duration remains the same no matter how lengthy the word is. Play around with the duration and delay properties to get things just right.
Remember when we changed the steps()
value by taking into consideration a single space after a word? In the same way, the last word in the sentence doesn’t have a space after it, and thus, we should take that into consideration in another if
statement.
// the following code goes inside the for loop
if (i == (all_words.length - 1)) {
timings_2.easing = `steps(${Number(single_word_len)}, end)`;
let reveal_animation_2 = document.querySelector(".text_hide").animate([
{ left: `${left_instance}%` },
{ left: `${left_instance + ((100 / text_len) * (word_len[i]))}%` }
], timings_2);
document.querySelector(".text_cursor").animate([
{ left: `${left_instance}%` },
{ left: `${left_instance + ((100 / text_len) * (word_len[i]))}%` }
], timings_2);
} else {
document.querySelector(".text_hide").animate([
{ left: `${left_instance}%` },
{ left: `${left_instance + ((100 / text_len) * (word_len[i] + 1))}%` }
], timings_2);
document.querySelector(".text_cursor").animate([
{ left: `${left_instance}%` },
{ left: `${left_instance + ((100 / text_len) * (word_len[i] + 1))}%` }
], timings_2);
}
What’s that left_instance
variable? We haven’t discussed it, yet it is the most crucial part of what we’re doing. Let me explain it.
0%
is the initial value of the first word’s left
property. But, the second word’s initial value should equal the first word’s final left
property value.
if (i == 1) {
var left_instance = (100 / text_len) * (word_len[i - 1] + 1);
}
word_len[i - 1] + 1
refers to the length of the previous word (including a white space).
We have two words, “Typing Animation.” That makes text_len
equal 16
meaning that each character is 6.25% of the full width (100/text_len = 100/16
) which is multiplied by the length of the first word, 7
. All that math gives us 43.75
which is, in fact, the width of the first word. In other words, the width of the first word is 43.75%
the width of the entire string. This means that the second word starts animating from where the first word left off.
Last, let’s update the left_instance
variable at the end of the for
loop:
left_instance = left_instance + ((100 / text_len) * (word_len[i] + 1));
You can now enter as many words as you want in HTML and the animation just works!
Have you noticed that the animation only runs once? What if we want to loop it infinitely? It’s possible:
There we go: a more robust JavaScript version of a typewriting animation. It’s super cool that CSS also has an approach (or even multiple approaches) to do the same sort of thing. CSS might even be the better approach in a given situation. But when we need enhancements that push beyond what CSS can handle, sprinkling in some JavaScript does the trick quite nicely. In this case, we added support for all words, regardless of how many characters they contain, and the ability to animate multiple words. And, with a small extra delay between words, we get a super natural-looking animation.
That’s it, hope you found this interesting! Signing off.
Typewriter Animation That Handles Anything You Throw at It originally published on CSS-Tricks, which is part of the DigitalOcean family. You should get the newsletter.
]]>