text-transform

The text-transform property takes into account language-specific case mapping rules such as the following:

• In Turkic languages, like Turkish (tr), Azerbaijani (az), Crimean Tatar (crh), Volga Tatar (tt), and Bashkir (ba), there are two kinds of i, with and without the dot, and two case pairings: i/İ and ı/I.
• In German (de), the ß becomes SS in uppercase.
• In Dutch (nl), the ij digraph becomes IJ, even with text-transform: capitalize, which only puts the first letter of a word in uppercase.
• In Greek (el), vowels lose their accent when the whole word is in uppercase (ά/Α), except for the disjunctive eta (ή/Ή). Also, diphthongs with an accent on the first vowel lose the accent and gain a diaeresis on the second vowel (άι/ΑΪ).
• In Greek (el), the lowercase sigma character has two forms: σ and ς. ς is used only when sigma terminates a word. When applying text-transform: lowercase to an uppercase sigma (Σ), the browser needs to choose the right lowercase form based on context.
• in Irish (ga), certain prefixed letters remain in lowercase when the base initial is capitalized, so for example text-transform: uppercase will change ar aon tslí to AR AON tSLÍ and not, as one might expect, AR AON TSLÍ (Firefox only). In some cases, a hyphen is also removed upon uppercasing: an t-uisce transforms to AN tUISCE (and the hyphen is correctly reinserted by text-transform: lowercase).

The language is defined by the lang HTML attribute or the xml:lang XML attribute.

/* Keyword values */
text-transform: none;
text-transform: capitalize;
text-transform: uppercase;
text-transform: lowercase;
text-transform: full-width;
text-transform: full-size-kana;
text-transform: math-auto;

/* Global values */
text-transform: inherit;
text-transform: initial;
text-transform: revert;
text-transform: revert-layer;
text-transform: unset;

capitalize

Is a keyword that converts the first letter of each word to uppercase. Other characters remain unchanged (they retain their original case as written in the element's text). A letter is defined as a character that is part of Unicode's Letter or Number general categories; thus, any punctuation marks or symbols at the beginning of a word are ignored.

Note: Authors should not expect capitalize to follow language-specific title casing conventions (such as skipping articles in English).

Note: The capitalize keyword was under-specified in CSS 1 and CSS 2.1. This resulted in differences between browsers in the way the first letter was calculated (Firefox considered - and _ as letters, but other browsers did not. Both Webkit and Gecko incorrectly considered letter-based symbols like ⓐ to be real letters.) By precisely defining the correct behavior, CSS Text Level 3 cleans this mess up. The capitalize line in the browser compatibility table contains the version the different engines started to support this now precisely-defined behavior.

uppercase

Is a keyword that converts all characters to uppercase.

lowercase

Is a keyword that converts all characters to lowercase.

none

Is a keyword that prevents the case of all characters from being changed.

full-width

Is a keyword that forces the writing of a character — mainly ideograms and Latin scripts — inside a square, allowing them to be aligned in the usual East Asian scripts (like Chinese or Japanese).

full-size-kana

Generally used for <ruby> annotation text, the keyword converts all small Kana characters to the equivalent full-size Kana, to compensate for legibility issues at the small font sizes typically used in ruby.

math-auto

Used to automatically render text in math italic where appropriate. It transforms Latin and Greek letters, and a few other math-related symbols, to italic mathematical symbols but only if it's applied on a text node containing a single character. For example, "x" will become "𝑥" (U+1D465), but "exp" will stay as "exp". It is primarily used to specify the behavior of <mi> elements in MathML. You should generally use MathML markup which automatically applies the right styling.

Large sections of text set with a text-transform value of uppercase may be difficult for people with cognitive concerns such as Dyslexia to read.

• MDN Understanding WCAG, Guideline 1.4 explanations
• W3C Understanding WCAG 2.1

text-transform = 
  none                                                |
  [ capitalize | uppercase | lowercase ] || full-width || full-size-kana  |
  math-auto                                           

<p>
  Initial String
  <strong>Lorem ipsum dolor sit amet, consectetur adipisicing elit…</strong>
</p>
<p>
  text-transform: none
  <strong
    ><span
      >Lorem ipsum dolor sit amet, consectetur adipisicing elit…</span
    ></strong
  >
</p>

span {
  text-transform: none;
}
strong {
  float: right;
}

This demonstrates no text transformation.

<p>
  Initial String
  <strong>Lorem ipsum dolor sit amet, consectetur adipisicing elit…</strong>
</p>
<p>
  text-transform: capitalize
  <strong
    ><span
      >Lorem ipsum dolor sit amet, consectetur adipisicing elit…</span
    ></strong
  >
</p>

span {
  text-transform: capitalize;
}
strong {
  float: right;
}

This demonstrates text capitalization.

<p>
  Initial String
  <strong
    >(this) "is" [a] –short– -test- «for» *the* _css_ ¿capitalize?
    ?¡transform!</strong
  >
</p>
<p>
  text-transform: capitalize
  <strong
    ><span
      >(this) "is" [a] –short– -test- «for» *the* _css_ ¿capitalize?
      ?¡transform!</span
    ></strong
  >
</p>

span {
  text-transform: capitalize;
}
strong {
  float: right;
}

This demonstrates how initial punctuations of a word are ignored. The keyword target the first letter, that is the first Unicode character part of the Letter or Number general category.

<p>
  Initial String
  <strong>ⓐⓑⓒ (ⓓⓔⓕ) —ⓖⓗⓘ— ⓙkl</strong>
</p>
<p>
  text-transform: capitalize
  <strong><span>ⓐⓑⓒ (ⓓⓔⓕ) —ⓖⓗⓘ— ⓙkl</span></strong>
</p>

span {
  text-transform: capitalize;
}
strong {
  float: right;
}

This demonstrates how initial symbols are ignored. The keyword target the first letter, that is the first Unicode character part of the Letter or Number general category.

<p>
  Initial String
  <strong lang="nl">The Dutch word: "ijsland" starts with a digraph.</strong>
</p>
<p>
  text-transform: capitalize
  <strong
    ><span lang="nl"
      >The Dutch word: "ijsland" starts with a digraph.</span
    ></strong
  >
</p>

span {
  text-transform: capitalize;
}
strong {
  float: right;
}

This demonstrates how the Dutch ij digraph must be handled like one single letter.

<p>
  Initial String
  <strong>Lorem ipsum dolor sit amet, consectetur adipisicing elit…</strong>
</p>
<p>
  text-transform: uppercase
  <strong
    ><span
      >Lorem ipsum dolor sit amet, consectetur adipisicing elit…</span
    ></strong
  >
</p>

span {
  text-transform: uppercase;
}
strong {
  float: right;
}

This demonstrates transforming the text to uppercase.

<p>
  Initial String
  <strong>Θα πάμε στο "Θεϊκό φαΐ" ή στη "Νεράιδα"</strong>
</p>
<p>
  text-transform: uppercase
  <strong
    ><span lang="el">Θα πάμε στο "Θεϊκό φαΐ" ή στη "Νεράιδα"</span></strong
  >
</p>

span {
  text-transform: uppercase;
}
strong {
  float: right;
}

This demonstrates how Greek vowels except disjunctive eta should have no accent, and the accent on the first vowel of a vowel pair becomes a diaeresis on the second vowel.

<p>
  Initial String
  <strong>Lorem ipsum dolor sit amet, consectetur adipisicing elit…</strong>
</p>
<p>
  text-transform: lowercase
  <strong
    ><span
      >Lorem ipsum dolor sit amet, consectetur adipisicing elit…</span
    ></strong
  >
</p>

span {
  text-transform: lowercase;
}
strong {
  float: right;
}

This demonstrates transforming the text to lowercase.

<p>
  Initial String
  <strong>Σ IS A greek LETTER that appears SEVERAL TIMES IN ΟΔΥΣΣΕΥΣ.</strong>
</p>
<p>
  text-transform: lowercase
  <strong
    ><span
      >Σ IS A greek LETTER that appears SEVERAL TIMES IN ΟΔΥΣΣΕΥΣ.</span
    ></strong
  >
</p>

span {
  text-transform: lowercase;
}
strong {
  float: right;
}

This demonstrates how the Greek character sigma (Σ) is transformed into the regular lowercase sigma (σ) or the word-final variant (ς), according the context.

<p>
  Initial String
  <strong>Ĩ is a Lithuanian LETTER as is J́</strong>
</p>
<p>
  text-transform: lowercase
  <strong><span lang="lt">Ĩ is a Lithuanian LETTER as is J́</span></strong>
</p>

span {
  text-transform: lowercase;
}
strong {
  float: right;
}

This demonstrates how the Lithuanian letters Ĩ and J́ retain their dot when transformed to lowercase.

<p>
  Initial String
  <strong
    >0123456789abcdefghijklmnopqrstuvwxyzABCDEFGHIJKLMNOPQRSTUVWXYZ!"#$%&()*+,-./:;<=>?@{|}~</strong
  >
</p>
<p>
  text-transform: full-width
  <strong
    ><span
      >0123456789abcdefghijklmnopqrstuvwxyzABCDEFGHIJKLMNOPQRSTUVWXYZ!"#$%&()*+,-./:;<=>?@{|}~</span
    ></strong
  >
</p>

span {
  text-transform: full-width;
}
strong {
  width: 100%;
  float: right;
}

Some characters exist in two formats: normal width and a full-width, with different Unicode code points. The full-width version is used to mix them smoothly with Asian ideographic characters.

<p>
  Initial String
  <strong>ｳｪﾌﾞﾌﾟﾛｸﾞﾗﾐﾝｸﾞの勉強</strong>
</p>
<p>
  text-transform: full-width
  <strong><span>ｳｪﾌﾞﾌﾟﾛｸﾞﾗﾐﾝｸﾞの勉強</span></strong>
</p>

span {
  text-transform: full-width;
}
strong {
  width: 100%;
  float: right;
}

The Japanese half-width katakana was used to represent katakana in 8-bit character codes. Unlike regular (full-width) katakana characters, a letter with dakuten (voiced sound mark) is represented as two code points, the body of letter and dakuten. The full-width combines these into a single code point when converting these characters into full-width.

In this example, we use pure HTML markup to create a math formula:

<div>
  (<span class="math-id">sin</span>&#8198;<span class="math-id">x</span>)<sup
    >2</sup
  >
  + (<span class="math-id">cos</span>&#8198;<span class="math-id">x</span>)<sup
    >2</sup
  >
  = 1
</div>

We give every .math-id element text-transform: math-auto. However, note how only the x characters become italic, while the sin and cos remain unchanged.

.math-id {
  text-transform: math-auto;
}

However, you are encouraged to use MathML for mathematical formulas, as it provides a more robust and accessible way to represent mathematical content. Here's the same formula using MathML:

<math xmlns="http://www.w3.org/1998/Math/MathML" display="block">
  <semantics>
    <mrow>
      <mo stretchy="false">(</mo>
      <mo lspace="0em" rspace="0em">sin</mo>
      <mspace width="0.16666666666666666em"></mspace>
      <mi>x</mi>
      <msup>
        <mo stretchy="false">)</mo>
        <mn>2</mn>
      </msup>
      <mo>+</mo>
      <mo stretchy="false">(</mo>
      <mo lspace="0em" rspace="0em">cos</mo>
      <mspace width="0.16666666666666666em"></mspace>
      <mi>x</mi>
      <msup>
        <mo stretchy="false">)</mo>
        <mn>2</mn>
      </msup>
      <mo>=</mo>
      <mn>1</mn>
    </mrow>
    <annotation encoding="TeX">(\sin\,x)^2+(\cos\,x)^2=1</annotation>
  </semantics>
</math>

• font-variant