Medium uses a strict subset of LESS for style generation. This subset includes variables and mixins, but nothing else (no nesting, etc.).
Medium's naming conventions are adapted from the work being done in the SUIT CSS framework. Which is to say, it relies on structured class names and meaningful hyphens (i.e., not using hyphens merely to separate words). This is to help work around the current limits of applying CSS to the DOM (i.e., the lack of style encapsulation) and to better communicate the relationships between classes.
Table of contents
syntax: js-<targetName>
JavaScript-specific classes reduce the risk that changing the structure or theme of components will inadvertently affect any required JavaScript behaviour and complex functionality. It is not neccesarry to use them in every case, just think of them as a tool in your utility belt. If you are creating a class, which you dont intend to use for styling, but instead only as a selector in JavaScript, you should probably be adding the js- prefix. In practice this looks like this:
<a href="/login" class="btn btn-primary js-login"></a>Again, JavaScript-specific classes should not, under any circumstances, be styled.
Medium's utility classes are low-level structural and positional traits. Utilities can be applied directly to any element; multiple utilities can be used together; and utilities can be used alongside component classes.
Utilities exist because certain CSS properties and patterns are used frequently. For example: floats, containing floats, vertical alignment, text truncation. Relying on utilities can help to reduce repetition and provide consistent implementations. They also act as a philosophical alternative to functional (i.e. non-polyfill) mixins.
<div class="u-clearfix">
<p class="u-textTruncate">{$text}</p>
<img class="u-pullLeft" src="{$src}" alt="">
<img class="u-pullLeft" src="{$src}" alt="">
<img class="u-pullLeft" src="{$src}" alt="">
</div>Syntax: u-<utilityName>
Utilities must use a camel case name, prefixed with a u namespace. What follows is an example of how various utilities can be used to create a simple structure within a component.
<div class="u-clearfix">
<a class="u-pullLeft" href="{$url}">
<img class="u-block" src="{$src}" alt="">
</a>
<p class="u-sizeFill u-textBreak">
…
</p>
</div>Syntax: <componentName>[--modifierName|-descendantName]
Component driven development offers several benefits when reading and writing HTML and CSS:
- It helps to distinguish between the classes for the root of the component, descendant elements, and modifications.
- It keeps the specificity of selectors low.
- It helps to decouple presentation semantics from document semantics.
You can think of components as custom elements that enclose specific semantics, styling, and behaviour.
The component's name must be written in camel case.
.myComponent { /* … */ }<article class="myComponent">
…
</article>A component modifier is a class that modifies the presentation of the base component in some form. Modifier names must be written in camel case and be separated from the component name by two hyphens. The class should be included in the HTML in addition to the base component class.
/* Core button */
.btn { /* … */ }
/* Default button style */
.btn--default { /* … */ }<button class="btn btn--primary">…</button>A component descendant is a class that is attached to a descendant node of a component. It's responsible for applying presentation directly to the descendant on behalf of a particular component. Descendant names must be written in camel case.
<article class="tweet">
<header class="tweet-header">
<img class="tweet-avatar" src="{$src}" alt="{$alt}">
…
</header>
<div class="tweet-body">
…
</div>
</article>Use is-stateName for state-based modifications of components. The state name must be Camel case. Never style these classes directly; they should always be used as an adjoining class.
JS can add/remove these classes. This means that the same state names can be used in multiple contexts, but every component must define its own styles for the state (as they are scoped to the component).
.tweet { /* … */ }
.tweet.is-expanded { /* … */ }<article class="tweet is-expanded">
…
</article>Syntax: <property>-<value>[--componentName]
Variable names in our CSS are also strictly structured. This syntax provides strong associations between property, use, and component.
The following variable defintion is a color property, with the value grayLight, for use with the highlightMenu component.
@color-grayLight--highlightMenu: rgb(51, 51, 50);When implementing feature styles, you should only be using color variables provided by colors.less.
When adding a color variable to colors.less, using RGB and RGBA color units are preferred over hex, named, HSL, or HSLA values.
Right:
rgb(50, 50, 50);
rgba(50, 50, 50, 0.2);Wrong:
#FFF;
#FFFFFF;
white;
hsl(120, 100%, 50%);
hsla(120, 100%, 50%, 1);Please use the z-index scale defined in z-index.less.
@zIndex-1 - @zIndex-9 are provided. Nothing should be higher then @zIndex-9.
With the additional support of web fonts font-weight plays a more important role than it once did. Different font weights will render typefaces specifically created for that weight, unlike the old days where bold could be just an algorithm to fatten a typeface. Obvious uses the numerical value of font-weight to enable the best representation of a typeface. The following table is a guide:
Raw font weights should be avoided. Instead, use the appropriate font mixin: .font-sansI7, .font-sansN7, etc.
The suffix defines the weight and style:
N = normal
I = italic
4 = normal font-weight
7 = bold font-weightRefer to type.less for type size, letter-spacing, and line height. Raw sizes, spaces, and line heights should be avoided outside of type.less.
ex:
@fontSize-micro
@fontSize-smallest
@fontSize-smaller
@fontSize-small
@fontSize-base
@fontSize-large
@fontSize-larger
@fontSize-largest
@fontSize-jumboSee Mozilla Developer Network — font-weight for further reading.
Type.less also provides a line height scale. This should be used for blocks of text.
ex:
@lineHeight-tightest
@lineHeight-tighter
@lineHeight-tight
@lineHeight-baseSans
@lineHeight-base
@lineHeight-loose
@lineHeight-looserAlternatively, when using line height to vertically center a single line of text, be sure to set the line height to the height of the container - 1.
.btn {
height: 50px;
line-height: 49px;
}Letter spacing should also be controlled with the following var scale.
@letterSpacing-tightest
@letterSpacing-tighter
@letterSpacing-tight
@letterSpacing-normal
@letterSpacing-loose
@letterSpacing-loosermixin syntax: m-<propertyName>
At Medium we only use mixins to generate polyfills for browser prefixed properties.
An example of a border radius mixin:
.m-borderRadius(@radius) {
-webkit-border-radius: @radius;
-moz-border-radius: @radius;
border-radius: @radius;
}The following are some high level page formatting style rules.
CSS rules should be comma seperated but live on new lines:
Right:
.content,
.content-edit {
…
}Wrong:
.content, .content-edit {
…
}CSS blocks should be seperated by a single new line. not two. not 0.
Right:
.content {
…
}
.content-edit {
…
}Wrong:
.content {
…
}
.content-edit {
…
}Quotes are optional in CSS and LESS. We use double quotes as it is visually clearer that the string is not a selector or a style property.
Right:
background-image: url("/img/you.jpg");
font-family: "Helvetica Neue Light", "Helvetica Neue", Helvetica, Arial;Wrong:
background-image: url(/img/you.jpg);
font-family: Helvetica Neue Light, Helvetica Neue, Helvetica, Arial;Although in the name (cascading style sheets) cascading can introduce unnecessary performance overhead for applying styles. Take the following example:
ul.user-list li span a:hover { color: red; }Styles are resolved during the renderer's layout pass. The selectors are resolved right to left, exiting when it has been detected the selector does not match. Therefore, in this example every a tag has to be inspected to see if it resides inside a span and a list. As you can imagine this requires a lot of DOM walking and and for large documents can cause a significant increase in the layout time. For further reading checkout: https://developers.google.com/speed/docs/best-practices/rendering#UseEfficientCSSSelectors
If we know we want to give all a elements inside the .user-list red on hover we can simplify this style to:
.user-list > a:hover {
color: red;
}If we want to only style specific a elements inside .user-list we can give them a specific class:
.user-list > .link-primary:hover {
color: red;
}
@chendrix and @waltfy
The thing with CSS and HTML is that a modifier class like
btn--primarydoesn't really work as a modifier ofbtnunless both classes are specified in the markup, that's just the way CSS is applied by the browser.Think about something like this:
Say all your buttons have rounded corners, but you want any button in the sidebar to have square corners.
Both of your buttons have no common class to target. They aren't actually
btncomponents unless the class attribute says they are; the mixin doesn't make a difference here.If you used a multi-class pattern, you could just use
.sidebar .btnand set the corners to square.The other option is to use
&:extendbut then you end up generating a ton of CSS bloat for no real reason other than to reduce the number of classes you are using in your HTML. You end up fighting how the browser works for no appreciable gain as far as I can tell. Why is it bad to have more classes in the HTML? What problem are you solving by making your class attributes shorter?I used to be firmly on the other side, in favor of deep nesting and almost class-less HTML, to keep the markup "pure". What I found in doing that was that my styles were completely coupled to my markup, and nothing was reusable at all anywhere. You end up with completely unmaintainable Less/Sass, fighting specificity wars everywhere, and a massive compiled CSS file, all in the name "clean markup", for which I haven't actually found a justification for.