CSS Counter Styles Level 3 (original) (raw)

1. Introduction

CSS 1 defined a handful of useful counter styles based on the styles that HTML traditionally allowed on ordered and unordered lists. While this was expanded slightly by CSS2.1, it doesn’t address the needs of worldwide typography.

This module introduces the @counter-style rule which allows CSS to address this in an open-ended manner, by allowing the author to define their own counter styles. These styles can then be used in the list-style-type property or in the counter() and counters() functions. It also defines some additional predefined counter styles, particularly ones which are common but complicated to represent with @counter-style.

2. Counter Styles

A counter style defines how to convert a counter value into a string. Counter styles are composed of:

• a name, to identify the style
• an algorithm, which transforms integer counter values into a basic string representation
• a negative sign, which is prepended or appended to the representation of a negative counter value.
• a prefix, to prepend to the representation
• a suffix to append to the representation
• a range, which limits the values that a counter style handles
• a spoken form, which describes how to read out the counter style in a speech synthesizer
• and a fallback style, to render the representation with when the counter value is outside the counter style’s range or the counter style otherwise can’t render the counter value

When asked to generate a counter representation using a particular counter style for a particular counter value, follow these steps:

1. If the counter style is unknown, exit this algorithm and instead generate a counter representation using the decimal style and the same counter value.
2. If the counter value is outside the range of the counter style, exit this algorithm and instead generate a counter representation using the counter style’s fallback style and the same counter value.
3. Using the counter value and the counter algorithm for the counter style, generate an initial representation for the counter value. If the counter value is negative and the counter style uses a negative sign, instead generate an initial representation using the absolute value of the counter value.
4. Prepend symbols to the representation as specified in the pad descriptor.
5. If the counter value is negative and the counter style uses a negative sign, wrap the representation in the counter style’s negative sign as specified in the negative descriptor.
6. Return the representation.

Note: prefix and suffix don’t play a part in this algorithm. This is intentional; the prefix and suffix aren’t part of the string returned by the counter() or counters() functions. Instead, the prefix and suffix are added by the algorithm that constructs the value of the content property for the ::marker pseudo-element. This also implies that the prefix and suffix always come from the specified counter-style, even if the actual representation is constructed by a fallback style.

Some values of system (symbolic, additive) and some descriptors (pad) can generate representations with size linear to an author-supplied number. This can potentially be abused to generate excessively large representations and consume undue amounts of the user’s memory or even hang their browser. User agents must support representations at least 60 Unicode codepoints long, but they may choose to instead use the fallback style for representations that would be longer than 60 codepoints.

3. Defining Custom Counter Styles: the @counter-style rule

The @counter-style rule allows authors to define a custom counter style. The components of a counter style are specified by descriptors in the @counter-style rule. The algorithm is specified implicitly by a combination of the system, symbols, and additive-symbols properties.

The general form of an @counter-style rule is:

@counter-style { }

is a that is not an ASCII case-insensitive match for "none".

Additionally, in the context of the prelude of an @counter-style rule, an ASCII case-insensitive match for "decimal" or "disc" is also an invalid and makes the @counter-style rule invalid.

Note: Note that also automatically excludes the CSS-wide keywords. In addition, some names, like inside, are valid as counter style names, but conflict with the existing values of properties like list-style, and so won’t be usable there.

Counter style names are case-sensitive. However, the names defined in this specification are ASCII lower-cased on parse wherever they are used as counter styles, e.g. in the list-style set of properties, in the @counter-style rule, and in the counter() functions.

Each @counter-style rule specifies a value for every counter-style descriptor, either implicitly or explicitly. Those not given explicit value in the rule take the initial value listed with each descriptor in this specification. These descriptors apply solely within the context of the @counter-style rule in which they are defined, and do not apply to document language elements. There is no notion of which elements the descriptors apply to or whether the values are inherited by child elements. When a given descriptor occurs multiple times in a given @counter-style rule, only the last-specified valid value is used; all prior values for that descriptor must be ignored.

Defining a @counter-style makes it available to the entire document in which it is included. If multiple @counter-style rules are defined with the same name, only one wins, according to standard cascade rules. @counter-style rules cascade "atomically": if one replaces another of the same name, it replaces it entirely, rather than just replacing the specific descriptors it specifies.

Note: Note that even the predefined counter styles can be overridden; the UA stylesheet occurs before any other stylesheets, so the predefined ones always lose in the cascade.

This at-rule conforms with the forward-compatible parsing requirement of CSS; conformant parsers that don’t understand these rules will ignore them without error. Any descriptors that are not recognized or implemented by a given user agent, or whose value does not match the grammars given here or in a future version of this specification, must be ignored in their entirety; they do not make the @counter-style rule invalid.

3.1. Counter algorithms: the system descriptor

The system descriptor specifies which algorithm will be used to construct the counter’s representation based on the counter value. For example, cyclic counter styles just cycle through their symbols repeatedly, while numeric counter styles interpret their symbols as digits and build their representation accordingly. The systems are defined in the following subsections.

3.1.1. Cycling Symbols: the cyclic system

The cyclic counter system cycles repeatedly through its provided symbols, looping back to the beginning when it reaches the end of the list. It can be used for simple bullets (just provide a single counter symbol), or for cycling through multiple symbols. The first counter symbol is used as the representation of the value 1, the second counter symbol (if it exists) is used as the representation of the value 2, etc.

If the system is cyclic, the symbols descriptor must contain at least one counter symbol, or else the @counter-style rule is invalid. This system is defined over all counter values.

A "triangle bullet" counter style can be defined as:

@counter-style triangle { system: cyclic; symbols: ‣; suffix: " "; }

It will then produce lists that look like:

‣ One ‣ Two ‣ Three

If there are N counter symbols and a representation is being constructed for the integer value, the representation is the counter symbol at index ( (value-1) mod N) of the list of counter symbols (0-indexed).

3.1.2. Exhaustible Symbols: the fixed system

The fixed counter system runs through its list of counter symbols once, then falls back. It is useful for representing counter styles that only have a finite number of representations. For example, Unicode defines several limited-length runs of special characters meant for lists, such as circled digits.

If the system is fixed, the symbols descriptor must contain at least one counter symbol, or else the @counter-style rule is invalid. This system is defined over counter values in a finite range, starting with the first symbol value and having a length equal to the length of the list of counter symbols.

When this system is specified, it may optionally have an integer provided after it, which sets the first symbol value. If it is omitted, the first symbol value is 1.

A "box-corner" counter style can be defined as:

@counter-style box-corner { system: fixed; symbols: ◰ ◳ ◲ ◱; suffix: ': '; }

It will then produce lists that look like:

◰: One ◳: Two ◲: Three ◱: Four 5: Five 6: Six

The first counter symbol is the representation for the first symbol value, and subsequent counter values are represented by subsequent counter symbols. Once the list of counter symbols is exhausted, further values cannot be represented by this counter style, and must instead be represented by the fallback counter style.

3.1.3. Repeating Symbols: the symbolic system

The symbolic counter system cycles repeatedly through its provided symbols, doubling, tripling, etc. the symbols on each successive pass through the list. For example, if the original symbols were "*" and "†", then on the second pass they would instead be "**" and "††", while on the third they would be "***"and "†††", etc. It can be used for footnote-style markers, and is also sometimes used for alphabetic-style lists for a slightly different presentation than what the alphabetic system presents.

If the system is symbolic, the symbols descriptor must contain at least one counter symbol, or else the @counter-style rule is invalid. This system is defined only over strictly positive counter values.

An "footnote" counter style can be defined as:

@counter-style footnote { system: symbolic; symbols: '*' ⁑ † ‡; suffix: " "; }

It will then produce lists that look like:

• One ⁑ Two † Three ‡ Four ** Five ⁑⁑ Six

Some style guides mandate a list numbering that looks similar to upper-alpha, but repeats differently after the first 26 values, instead going "AA", "BB", "CC", etc. This can be achieved with the symbolic system:

@counter-style upper-alpha-legal { system: symbolic; symbols: A B C D E F G H I J K L M N O P Q R S T U V W X Y Z; }

This style is identical to upper-alpha through the first 27 values, but they diverge after that, with upper-alpha going "AB", "AC", "AD", etc. Starting at the 53rd value, upper-alpha goes "BA", "BB", "BC", etc., while this style jumps into triple digits with "AAA", "BBB", "CCC", etc.

To construct the representation, run the following algorithm:

Let N be the length of the list of counter symbols, value initially be the counter value, S initially be the empty string, and symbol(n) be the nth counter symbol in the list of counter symbols (0-indexed).

1. Let the chosen symbol be symbol( (value - 1) mod N).
2. Let the representation length be ceil( value / N ).
3. Append the chosen symbol to S a number of times equal to the representation length.

Finally, return S.

3.1.4. Bijective Numerals: the alphabetic system

The alphabetic counter system interprets the list of counter symbols as digits to an alphabetic numbering system, similar to the default lower-alpha counter style, which wraps from "a", "b", "c", to "aa", "ab", "ac". Alphabetic numbering systems do not contain a digit representing 0; so the first value when a new digit is added is composed solely of the first digit. Alphabetic numbering systems are commonly used for lists, and also appear in many spreadsheet programs to number columns. The first counter symbol in the list is interpreted as the digit 1, the second as the digit 2, and so on.

If the system is alphabetic, the symbols descriptor must contain at least two counter symbols, or else the @counter-style rule is invalid. This system is defined only over strictly positive counter values.

If there are N counter symbols, the representation is a base N alphabetic number using the counter symbols as digits. To construct the representation, run the following algorithm:

Let N be the length of the list of counter symbols, value initially be the counter value, S initially be the empty string, and symbol(n) be the nth counter symbol in the list of counter symbols (0-indexed).

While value is not equal to 0:

1. Set value to value - 1.
2. Prepend symbol( value mod N ) to S.
3. Set value to floor( value / N ).

Finally, return S.

3.1.5. Positional Numerals: the numeric system

The numeric counter system interprets the list of counter symbols as digits to a "place-value" numbering system, similar to the default decimal counter style. The first counter symbol in the list is interpreted as the digit 0, the second as the digit 1, and so on.

If the system is numeric, the symbols descriptor must contain at least two counter symbols, or else the @counter-style rule is invalid. This system is defined over all counter values.

A "trinary" counter style can be defined as:

@counter-style trinary { system: numeric; symbols: '0' '1' '2'; }

It will then produce lists that look like:

1. One
2. Two
3. Three
4. Four
5. Five
6. Six

If there are N counter symbols, the representation is a base N number using the counter symbols as digits. To construct the representation, run the following algorithm:

Let N be the length of the list of counter symbols, value initially be the counter value, S initially be the empty string, and symbol(n) be the nth counter symbol in the list of counter symbols (0-indexed).

1. If value is 0, append symbol(0) to S and return S.
2. While value is not equal to 0:
   1. Prepend symbol( value mod N ) to S.
   2. Set value to floor( value / N ).
3. Return S.

3.1.6. Accumulating Numerals: the additive system

The additive counter system is used to represent "sign-value" numbering systems, which, rather than reusing digits in different positions to change their value, define additional digits with much larger values, so that the value of the number can be obtained by adding all the digits together. This is used in Roman numerals and other numbering systems around the world.

If the system is additive, the additive-symbols descriptor must contain at least one additive tuple, or else the @counter-style rule is invalid. This system is nominally defined over all counter values (see algorithm, below, for exact details).

A "dice" counter style can be defined as:

@counter-style dice { system: additive; additive-symbols: 6 ⚅, 5 ⚄, 4 ⚃, 3 ⚂, 2 ⚁, 1 ⚀; suffix: " "; }

It will then produce lists that look like:

⚀ One ⚁ Two ⚂ Three ... ⚅⚄ Eleven ⚅⚅ Twelve ⚅⚅⚀ Thirteen

To construct the representation, run this algorithm:

Let value initially be the counter value, S initially be the empty string, and symbol list initially be the list of additive tuples.

1. If value is initially 0, and there is an additive tuple with a weight of 0, append that tuple’s counter symbol to S and return S.
2. While value is greater than 0 and there are elements left in the symbol list:
   1. Pop the first additive tuple from the symbol list. This is the current tuple.
   2. Append the current tuple’s counter symbol to S floor( value / [current tuple](#current-tuple)’s weight ) times (this may be 0).
   3. Decrement value by the current tuple’s weight multiplied by the number of times the current tuple was appended to S in the previous step.
3. If the loop ended because value is 0, return S. Otherwise, the given counter value cannot be represented by this counter style, and must instead be represented by the fallback counter style.

Note: All of the predefined additive @counter-style rules in this specification produce representations for every value in their range, but it’s possible to produce values for additive-symbols that will fail to find a representation with the algorithm defined above, even though theoretically a representation could be found. For example, if a @counter-style was defined with additive-symbols: 3 "a", 2 "b";, the algorithm defined above will fail to find a representation for a counter value of 4, even though theoretically a "bb" representation would work. While unfortunate, this is required to maintain the property that the algorithm runs in linear time relative to the size of the counter value.

3.1.7. Building from Existing Counter Styles: the extends system

The extends system allows an author to use the algorithm of another counter style, but alter other aspects, such as the negative sign or the suffix. If a counter style uses the extends system, any unspecified descriptors must be taken from the extended counter style specified, rather than taking their initial values.

If a @counter-style uses the extends system, it must not contain a symbols or additive-symbols descriptor, or else the @counter-style rule is invalid.

If the specified counter style name isn’t the name of any currently-defined counter style, it must be treated as if it was extending the decimal counter style. If one or more @counter-style rules form a cycle with their extends values, all of the counter styles participating in the cycle must be treated as if they were extending the decimal counter style instead.

For example, if you wanted a counter style that was identical to decimal, but used a parenthesis rather than a period after it, like:

1. first item
2. second item
3. third item

Rather than writing up an entirely new counter style, this can be done by just extending decimal:

@counter-style decimal-paren { system: extends decimal; suffix: ") "; }

3.2. Formatting negative values: the negative descriptor

The negative descriptor defines how to alter the representation when the counter value is negative.

The first in the value is prepended to the representation when the counter value is negative. The second , if specified, is appended to the representation when the counter value is negative.

For example, specifying negative: "(" ")"; will make negative values be wrapped in parentheses, which is sometimes used in financial contexts, like "(2) (1) 0 1 2 3...".

Not all system values use a negative sign. In particular, a counter style uses a negative sign if its system value is symbolic, alphabetic, numeric, additive, or extends if the extended counter style itself uses a negative sign. If a counter style does not use a negative sign, it ignores the negative sign when generating a counter representation.

3.3. Symbols before the marker: the prefix descriptor

The prefix descriptor specifies a that is prepended to the marker representation. Prefixes come before any negative sign.

Note: Prefixes are only added by the algorithm for constructing the default contents of the ::marker pseudo-element; the prefix is not added automatically when the counter() or counters() functions are used.

3.4. Symbols after the marker: the suffix descriptor

The suffix descriptor specifies a that is appended to the marker representation. Suffixes are added to the representation after negative signs.

Note: Suffixes are only added by the algorithm for constructing the default contents of the ::marker pseudo-element; the suffix is not added automatically when the counter() or counters() functions are used.

3.5. Limiting the counter scope: the range descriptor

The range descriptor defines the ranges over which the counter style is defined. If a counter style is used to represent a counter value outside of its ranges, the counter style instead drops down to its fallback counter style.

auto

The range depends on the counter system:

• For cyclic, numeric, and fixed systems, the range is negative infinity to positive infinity.
• For alphabetic and symbolic systems, the range is 1 to positive infinity.
• For additive systems, the range is 0 to positive infinity.
• For extends systems, the range is whatever auto would produce for the extended system; if extending a complex predefined style (§7 Complex Predefined Counter Styles), the range is the style’s defined range.

[ [ | infinite ]{2} ]#

This defines a comma-separated list of ranges. For each individual range, the first value is the lower bound and the second value is the upper bound. This range is inclusive - it contains both the lower and upper bound numbers. If infinite is used as the first value in a range, it represents negative infinity; if used as the second value, it represents positive infinity. The range of the counter style is the union of all the ranges defined in the list.

If the lower bound of any range is higher than the upper bound, the entire descriptor is invalid and must be ignored.

Implementations must support ranges with a lower bound of at least -215 and an upper bound of at least 215-1 (the range of a signed 2-byte int). They may support higher ranges. If any specified bound is outside of the implementation’s supported bounds, it must be treated as the closest bound that the implementation does support.

3.6. Zero-Padding and Constant-Width Representations: the pad descriptor

The pad descriptor allows an author to specify a "fixed-width" counter style, where representations shorter than the pad value are padded with a particular . Representations larger than the specified pad value are constructed as normal.

&&

The specifies a minimum length that all counter representations must reach.

Let difference be the provided minus the number of grapheme clusters in the initial representation for the counter value. (Note that, per the algorithm to generate a counter representation, this occurs before adding prefixes/suffixes/negatives.) If the counter value is negative and the counter style uses a negative sign, further reduce difference by the number of grapheme clusters in the counter style’s negative descriptor’s (s).

If difference is greater than zero, prepend difference copies of the specified to the representation.

The must be non-negative. A negative value is a syntax error.

The most common example of "fixed-width" numbering is zero-padded decimal numbering. If an author knows that the numbers used will be less than a thousand, for example, it can be zero-padded with a simple pad: 3 "0"; descriptor, ensuring that all of the representations are 3 digits wide.

This will cause, for example, 1 to be represented as "001", 20 to be represented as "020", 300 to be represented as "300", 4000 to be represented as "4000", and -5 to be represented as "-05".

Note: The pad descriptor counts the number of grapheme clusters in the representation, but pads it with s. If the specified pad is multi-character, this will likely not have the desired effect. Unfortunately, there’s no way to use the number of grapheme clusters in the pad without violating useful constraints. It is recommended that authors only specify s of a single grapheme cluster in the pad descriptor.

3.7. Defining fallback: the fallback descriptor

The fallback descriptor specifies a fallback counter style to be used when the current counter style can’t create a representation for a given counter value. For example, if a counter style defined with a range of 1-10 is asked to represent a counter value of 11, the counter value’s representation is instead constructed with the fallback counter style (or possibly the fallback style’s fallback style, if the fallback style can’t represent that value, etc.).

If the value of the fallback descriptor isn’t the name of any currently-defined counter style, the used value of the fallback descriptor is decimal instead. Similarly, while following fallbacks to find a counter style that can render the given counter value, if a loop in the specified fallbacks is detected, the decimal style must be used instead.

Note that it is not necessarily an error to specify fallback loops. For example, if an author desires a counter style with significantly different representations for even and odd counter values, they may find it easiest to define one style that can only represent odd values and one that can only represent even values, and specify each as the fallback for the other one. Though the fallback graph is circular, at no point do you encounter a loop while following these fallbacks - every counter value is represented by one or the other counter style.

3.8. Marker characters: the symbols and additive-symbols descriptors

= | |

The symbols and additive-symbols descriptors specify the symbols used by the marker-construction algorithm specified by the system descriptor. The @counter-style rule must have a valid symbols descriptor if the counter system is cyclic, numeric, alphabetic, symbolic, or fixed, or a valid additive-symbols descriptor if the counter system is additive; otherwise, the @counter-style does not define a counter style (but is still a valid at-rule).

Some counter systems specify that the symbols descriptor must have at least two entries. If the counter style’s system is such, and the symbols descriptor has only a single entry, the descriptor is invalid and must be ignored.

Each entry in the symbols descriptor’s value defines a counter symbol, which is interpreted differently based on the counter style’s system. Each entry in the additive-symbols descriptor’s value defines an additive tuple, which consists of a counter symbol and a non-negative integer weight. Each weight must be a non-negative integer, and the additive tuples must be specified in order of strictly descending weight; otherwise, the declaration is invalid and must be ignored.

Counter symbols may be strings, images, or identifiers, and the three types can be mixed in a single descriptor. Counter representations are constructed by concatenating counter symbols together. Identifiers are rendered as strings containing the same characters. Images are rendered as inline replaced elements. The default object size of an image counter symbol is a 1em by 1em square.

Note: The syntax in is currently at-risk. No implementations have plans to implement it currently, and it complicates some usages of counter() in ways that haven’t been fully handled.

Note: If using identifiers rather than strings to define the symbols, be aware of the syntax of identifiers. In particular, ascii non-letters like "*" are not identifiers, and so must be quoted in a string. Hex escapes, used in several of the counter styles defined in this specification, "eat" the following space (to allow a digit to follow a hex escape without ambiguity), so two spaces must be put after a hex escape to separate it from the following one, or else they’ll be considered adjacent, and part of the same identifier. For example, symbols: \660 \661; only defines a single symbol, consisting of the U+0660 and U+0661 characters, rather than the two that were intended; either quote the escapes in strings, like symbols: "\660" "\661", or put two spaces between the escapes.

3.9. Speech Synthesis: the speak-as descriptor

A counter style can be constructed with a meaning that is obvious visually, but impossible to meaningfully represent via a speech synthesizer, or possible but nonsensical when naively read out. The speak-as descriptor describes how to synthesize the spoken form of a counter formatted with the given counter style. Values have the following meanings:

auto

If the counter style’s system is alphabetic, this value has the same effect as spell-out. If the system is cyclic, this value has the same effect as bullets. If the system is extends, this value has the same effect as auto would have for the extended style. Otherwise, this value has the same effect as numbers.

bullets

The UA speaks a UA-defined phrase or audio cue that represents an unordered list item being read out.

numbers

The counter’s value is spoken as a number in the content language.

words

Generate a counter representation for the value as normal, then speak it as normal text in the content language.

spell-out

Generate a counter representation for the value as normal, then spell it out letter-by-letter in the content language. If the UA does not know how to pronounce the symbols, it may handle it as numbers.

For example, lower-greek in English would be read out as "alpha", "beta", "gamma", etc. Conversely, upper-latin in French would be read out as (in phonetic notation) /a/, /be/, /se/, etc.

The counter’s value is instead spoken out in the specified style (similar to the behavior of the fallback descriptor when generating representations for a counter value). If the specified style does not exist, this value is treated as auto. If a loop is detected when following speak-as references, this value is treated as auto for the counter styles participating in the loop.

The ability to defer pronunciation to another counter style can help when the symbols being used aren’t actually letters. For example, here’s a possible definition of a circled-lower-latin counter-style, using some special Unicode characters:

@counter-style circled-lower-latin { system: alphabetic; speak-as: lower-latin; symbols: ⓐ ⓑ ⓒ ⓓ ⓔ ⓕ ⓖ ⓗ ⓘ ⓙ ⓚ ⓛ ⓜ ⓝ ⓞ ⓟ ⓠ ⓡ ⓢ ⓣ ⓤ ⓥ ⓦ ⓧ ⓨ ⓩ; suffix: " "; }

Setting its system to alphabetic would normally make the UA try to read out the names of the characters, but in this case that might be something like "Circled Letter A", which is unlikely to make sense. Instead, explicitly setting speak-as to lower-latin ensures that they get read out as their corresponding latin letters, as intended.

4. Defining Anonymous Counter Styles: the symbols() function

The symbols() function allows a counter style to be defined inline in a property value, for when a style is used only once in a stylesheet and defining a full @counter-style rule would be overkill. It does not provide the full feature-set of the @counter-style rule, but provides a sufficient subset to still be useful. The syntax of the symbols() rule is:

symbols() = symbols( ? [ | ]+ ); = cyclic | numeric | alphabetic | symbolic | fixed;

The symbols() function defines an anonymous counter style with no name, a prefix of "" (empty string) and suffix of " " (U+0020 SPACE), a range of auto, a fallback of decimal, a negative of "\2D" ("-" hyphen-minus), a pad of 0 "", and a speak-as of auto. The counter style’s algorithm is constructed by consulting the previous chapter using the provided system — or symbolic if the system was omitted — and the provided s and s as the value of the symbols property. If the system is fixed, the first symbol value is 1.

If the system is alphabetic or numeric, there must be at least two s or s, or else the function is invalid.

This code:

ol { list-style: symbols("*" "\2020" "\2021" "\A7"); }

will produce lists that look like:

• One † Two ‡ Three § Four ** Five †† Six ‡‡ Seven

On the other hand, specifying the system of counter, like so:

ol { list-style: symbols(cyclic "*" "\2020" "\2021" "\A7"); }

will produce lists that look like:

• One † Two ‡ Three § Four
• Five † Six ‡ Seven

Note: the symbols() function only allows strings and images, while the symbols descriptor of a @counter-style rule also allows identifiers.

5. Extending list-style-type, counter(), and counters()

In CSS Level 2 [CSS21] the list-style-type property and the counter() and counters() notations accept various pre-defined keywords, each identifying a counter style. This module extends these features to take instead the type, defined below:

= | symbols();

If a is used that does not refer to any existing counter style, it must act identically to the decimal counter style (but does not compute to decimal).

6. Simple Predefined Counter Styles

The following stylesheet uses the @counter-style rule to redefine all of the counter styles defined in CSS 2 and CSS 2.1. This stylesheet is normative—UAs must include it in their UA stylesheet (or at least act as if these rules were defined at that level).

6.1. Numeric: decimal, decimal-leading-zero, arabic-indic, armenian, upper-armenian, lower-armenian, bengali, cambodian, khmer, cjk-decimal, devanagari, georgian, gujarati, gurmukhi, hebrew, kannada, lao, malayalam, mongolian, myanmar, oriya, persian, lower-roman, upper-roman, tamil, telugu, thai, tibetan

decimal

Western decimal numbers (e.g., 1, 2, 3, ..., 98, 99, 100).

decimal-leading-zero

Decimal numbers padded by initial zeros (e.g., 01, 02, 03, ..., 98, 99, 100).

arabic-indic

Arabic-indic numbering (e.g., ١‎, ٢‎, ٣‎, ٤‎, ..., ٩٨‎, ٩٩‎, ١٠٠‎).

armenian

upper-armenian

Traditional uppercase Armenian numbering (e.g., Ա, Բ, Գ, ..., ՂԸ, ՂԹ, Ճ).

lower-armenian

Lowercase Armenian numbering (e.g., ա, բ, գ, ..., ղը, ղթ, ճ).

bengali

Bengali numbering (e.g., ১, ২, ৩, ..., ৯৮, ৯৯, ১০০).

cambodian

khmer

Cambodian/Khmer numbering (e.g., ១, ២, ៣, ..., ៩៨, ៩៩, ១០០).

cjk-decimal

Han decimal numbers (e.g., 一, 二, 三, ..., 九八, 九九, 一〇〇).

devanagari

devanagari numbering (e.g., १, २, ३, ..., ९८, ९९, १००).

georgian

Traditional Georgian numbering (e.g., ა, ბ, გ, ..., ჟჱ, ჟთ, რ).

gujarati

Gujarati numbering (e.g., ૧, ૨, ૩, ..., ૯૮, ૯૯, ૧૦૦).

gurmukhi

Gurmukhi numbering (e.g., ੧, ੨, ੩, ..., ੯੮, ੯੯, ੧੦੦).

hebrew

Traditional Hebrew numbering (e.g., א‎, ב‎, ג‎, ..., צח‎, צט‎, ק‎).

kannada

Kannada numbering (e.g., ೧, ೨, ೩, ..., ೯೮, ೯೯, ೧೦೦).

lao

Laotian numbering (e.g., ໑, ໒, ໓, ..., ໙໘, ໙໙, ໑໐໐).

malayalam

Malayalam numbering (e.g., ൧, ൨, ൩, ..., ൯൮, ൯൯, ൧൦൦).

mongolian

Mongolian numbering (e.g., ᠑, ᠒, ᠓, ..., ᠙᠘, ᠙᠙, ᠑᠐᠐).

myanmar

Myanmar (Burmese) numbering (e.g., ၁, ၂, ၃, ..., ၉၈, ၉၉, ၁၀၀).

oriya

Oriya numbering (e.g., ୧, ୨, ୩, ..., ୯୮, ୯୯, ୧୦୦).

persian

Persian numbering (e.g., ۱, ۲, ۳, ۴, ..., ۹۸, ۹۹, ۱۰۰).

lower-roman

Lowercase ASCII Roman numerals (e.g., i, ii, iii, ..., xcviii, xcix, c).

upper-roman

Uppercase ASCII Roman numerals (e.g., I, II, III, ..., XCVIII, XCIX, C).

tamil

Tamil numbering (e.g., ௧, ௨, ௩, ..., ௯௮, ௯௯, ௧௦௦).

telugu

Telugu numbering (e.g., ౧, ౨, ౩, ..., ౯౮, ౯౯, ౧౦౦).

thai

Thai (Siamese) numbering (e.g., ๑, ๒, ๓, ..., ๙๘, ๙๙, ๑๐๐).

tibetan

Tibetan numbering (e.g., ༡, ༢, ༣, ..., ༩༨, ༩༩, ༡༠༠).

The decimal counter-style must not be overridable with a @counter-style rule, so that it is always available as the ultimate fallback style.

The following stylesheet fragment provides the normative definition of these predefined counter styles:

@counter-style decimal { system: numeric; symbols: '0' '1' '2' '3' '4' '5' '6' '7' '8' '9'; }

@counter-style decimal-leading-zero { system: extends decimal; pad: 2 '0'; }

@counter-style arabic-indic { system: numeric; symbols: "\660" "\661" "\662" "\663" "\664" "\665" "\666" "\667" "\668" "\669"; /* ٠ ١ ٢ ٣ ٤ ٥ ٦ ٧ ٨ ٩ */ }

@counter-style armenian { system: additive; range: 1 9999; additive-symbols: 9000 \554, 8000 \553, 7000 \552, 6000 \551, 5000 \550, 4000 \54F, 3000 \54E, 2000 \54D, 1000 \54C, 900 \54B, 800 \54A, 700 \549, 600 \548, 500 \547, 400 \546, 300 \545, 200 \544, 100 \543, 90 \542, 80 \541, 70 \540, 60 \53F, 50 \53E, 40 \53D, 30 \53C, 20 \53B, 10 \53A, 9 \539, 8 \538, 7 \537, 6 \536, 5 \535, 4 \534, 3 \533, 2 \532, 1 \531; /* 9000 Ք, 8000 Փ, 7000 Ւ, 6000 Ց, 5000 Ր, 4000 Տ, 3000 Վ, 2000 Ս, 1000 Ռ, 900 Ջ, 800 Պ, 700 Չ, 600 Ո, 500 Շ, 400 Ն, 300 Յ, 200 Մ, 100 Ճ, 90 Ղ, 80 Ձ, 70 Հ, 60 Կ, 50 Ծ, 40 Խ, 30 Լ, 20 Ի, 10 Ժ, 9 Թ, 8 Ը, 7 Է, 6 Զ, 5 Ե, 4 Դ, 3 Գ, 2 Բ, 1 Ա */ }

@counter-style upper-armenian { system: extends armenian; }

@counter-style lower-armenian { system: additive; range: 1 9999; additive-symbols: 9000 "\584", 8000 "\583", 7000 "\582", 6000 "\581", 5000 "\580", 4000 "\57F", 3000 "\57E", 2000 "\57D", 1000 "\57C", 900 "\57B", 800 "\57A", 700 "\579", 600 "\578", 500 "\577", 400 "\576", 300 "\575", 200 "\574", 100 "\573", 90 "\572", 80 "\571", 70 "\570", 60 "\56F", 50 "\56E", 40 "\56D", 30 "\56C", 20 "\56B", 10 "\56A", 9 "\569", 8 "\568", 7 "\567", 6 "\566", 5 "\565", 4 "\564", 3 "\563", 2 "\562", 1 "\561"; /* 9000 ք, 8000 փ, 7000 ւ, 6000 ց, 5000 ր, 4000 տ, 3000 վ, 2000 ս, 1000 ռ, 900 ջ, 800 պ, 700 չ, 600 ո, 500 շ, 400 ն, 300 յ, 200 մ, 100 ճ, 90 ղ, 80 ձ, 70 հ, 60 կ, 50 ծ, 40 խ, 30 լ, 20 ի, 10 ժ, 9 թ, 8 ը, 7 է, 6 զ, 5 ե, 4 դ, 3 գ, 2 բ, 1 ա */ }

@counter-style bengali { system: numeric; symbols: "\9E6" "\9E7" "\9E8" "\9E9" "\9EA" "\9EB" "\9EC" "\9ED" "\9EE" "\9EF"; /* ০ ১ ২ ৩ ৪ ৫ ৬ ৭ ৮ ৯ */ }

@counter-style cambodian { system: numeric; symbols: "\17E0" "\17E1" "\17E2" "\17E3" "\17E4" "\17E5" "\17E6" "\17E7" "\17E8" "\17E9"; /* ០ ១ ២ ៣ ៤ ៥ ៦ ៧ ៨ ៩ */ }

@counter-style khmer { system: extends cambodian; }

@counter-style cjk-decimal { system: numeric; range: 0 infinite; symbols: \3007 \4E00 \4E8C \4E09 \56DB \4E94 \516D \4E03 \516B \4E5D; /* 〇 一 二 三 四 五 六 七 八 九 / suffix: "\3001"; / "、" */ }

@counter-style devanagari { system: numeric; symbols: "\966" "\967" "\968" "\969" "\96A" "\96B" "\96C" "\96D" "\96E" "\96F"; /* ० १ २ ३ ४ ५ ६ ७ ८ ९ */ }

@counter-style georgian { system: additive; range: 1 19999; additive-symbols: 10000 \10F5, 9000 \10F0, 8000 \10EF, 7000 \10F4, 6000 \10EE, 5000 \10ED, 4000 \10EC, 3000 \10EB, 2000 \10EA, 1000 \10E9, 900 \10E8, 800 \10E7, 700 \10E6, 600 \10E5, 500 \10E4, 400 \10F3, 300 \10E2, 200 \10E1, 100 \10E0, 90 \10DF, 80 \10DE, 70 \10DD, 60 \10F2, 50 \10DC, 40 \10DB, 30 \10DA, 20 \10D9, 10 \10D8, 9 \10D7, 8 \10F1, 7 \10D6, 6 \10D5, 5 \10D4, 4 \10D3, 3 \10D2, 2 \10D1, 1 \10D0; /* 10000 ჵ, 9000 ჰ, 8000 ჯ, 7000 ჴ, 6000 ხ, 5000 ჭ, 4000 წ, 3000 ძ, 2000 ც, 1000 ჩ, 900 შ, 800 ყ, 700 ღ, 600 ქ, 500 ფ, 400 ჳ, 300 ტ, 200 ს, 100 რ, 90 ჟ, 80 პ, 70 ო, 60 ჲ, 50 ნ, 40 მ, 30 ლ, 20 კ, 10 ი, 9 თ, 8 ჱ, 7 ზ, 6 ვ, 5 ე, 4 დ, 3 გ, 2 ბ, 1 ა */ }

@counter-style gujarati { system: numeric; symbols: "\AE6" "\AE7" "\AE8" "\AE9" "\AEA" "\AEB" "\AEC" "\AED" "\AEE" "\AEF"; /* ૦ ૧ ૨ ૩ ૪ ૫ ૬ ૭ ૮ ૯ */ }

@counter-style gurmukhi { system: numeric; symbols: "\A66" "\A67" "\A68" "\A69" "\A6A" "\A6B" "\A6C" "\A6D" "\A6E" "\A6F"; /* ੦ ੧ ੨ ੩ ੪ ੫ ੬ ੭ ੮ ੯ */ }

@counter-style hebrew { system: additive; range: 1 10999; additive-symbols: 10000 \5D9\5F3, 9000 \5D8\5F3, 8000 \5D7\5F3, 7000 \5D6\5F3, 6000 \5D5\5F3, 5000 \5D4\5F3, 4000 \5D3\5F3, 3000 \5D2\5F3, 2000 \5D1\5F3, 1000 \5D0\5F3, 400 \5EA, 300 \5E9, 200 \5E8, 100 \5E7, 90 \5E6, 80 \5E4, 70 \5E2, 60 \5E1, 50 \5E0, 40 \5DE, 30 \5DC, 20 \5DB, 19 \5D9\5D8, 18 \5D9\5D7, 17 \5D9\5D6, 16 \5D8\5D6, 15 \5D8\5D5, 10 \5D9, 9 \5D8, 8 \5D7, 7 \5D6, 6 \5D5, 5 \5D4, 4 \5D3, 3 \5D2, 2 \5D1, 1 \5D0; /* 10000 י׳, 9000 ט׳, 8000 ח׳, 7000 ז׳, 6000 ו׳, 5000 ה׳, 4000 ד׳, 3000 ג׳, 2000 ב׳, 1000 א׳, 400 ת, 300 ש, 200 ר, 100 ק, 90 צ, 80 פ, 70 ע, 60 ס, 50 נ, 40 מ, 30 ל, 20 כ, 19 יט, 18 יח, 17 יז, 16 טז, 15 טו, 10 י, 9 ט, 8 ח, 7 ז, 6 ו, 5 ה, 4 ד, 3 ג, 2 ב, 1 א / / This system manually specifies the values for 19-15 to force the correct display of 15 and 16, which are commonly rewritten to avoid a close resemblance to the Tetragrammaton. / / Implementations MAY choose to implement this manually to a higher range; see note below. */ }

@counter-style kannada { system: numeric; symbols: "\CE6" "\CE7" "\CE8" "\CE9" "\CEA" "\CEB" "\CEC" "\CED" "\CEE" "\CEF"; /* ೦ ೧ ೨ ೩ ೪ ೫ ೬ ೭ ೮ ೯ */ }

@counter-style lao { system: numeric; symbols: "\ED0" "\ED1" "\ED2" "\ED3" "\ED4" "\ED5" "\ED6" "\ED7" "\ED8" "\ED9"; /* ໐ ໑ ໒ ໓ ໔ ໕ ໖ ໗ ໘ ໙ */ }

@counter-style malayalam { system: numeric; symbols: "\D66" "\D67" "\D68" "\D69" "\D6A" "\D6B" "\D6C" "\D6D" "\D6E" "\D6F"; /* ൦ ൧ ൨ ൩ ൪ ൫ ൬ ൭ ൮ ൯ */ }

@counter-style mongolian { system: numeric; symbols: "\1810" "\1811" "\1812" "\1813" "\1814" "\1815" "\1816" "\1817" "\1818" "\1819"; /* ᠐ ᠑ ᠒ ᠓ ᠔ ᠕ ᠖ ᠗ ᠘ ᠙ */ }

@counter-style myanmar { system: numeric; symbols: "\1040" "\1041" "\1042" "\1043" "\1044" "\1045" "\1046" "\1047" "\1048" "\1049"; /* ၀ ၁ ၂ ၃ ၄ ၅ ၆ ၇ ၈ ၉ */ }

@counter-style oriya { system: numeric; symbols: "\B66" "\B67" "\B68" "\B69" "\B6A" "\B6B" "\B6C" "\B6D" "\B6E" "\B6F"; /* ୦ ୧ ୨ ୩ ୪ ୫ ୬ ୭ ୮ ୯ */ }

@counter-style persian { system: numeric; symbols: "\6F0" "\6F1" "\6F2" "\6F3" "\6F4" "\6F5" "\6F6" "\6F7" "\6F8" "\6F9"; /* ۰ ۱ ۲ ۳ ۴ ۵ ۶ ۷ ۸ ۹ */ }

@counter-style lower-roman { system: additive; range: 1 3999; additive-symbols: 1000 m, 900 cm, 500 d, 400 cd, 100 c, 90 xc, 50 l, 40 xl, 10 x, 9 ix, 5 v, 4 iv, 1 i; }

@counter-style upper-roman { system: additive; range: 1 3999; additive-symbols: 1000 M, 900 CM, 500 D, 400 CD, 100 C, 90 XC, 50 L, 40 XL, 10 X, 9 IX, 5 V, 4 IV, 1 I; }

@counter-style tamil { system: numeric; symbols: "\BE6" "\BE7" "\BE8" "\BE9" "\BEA" "\BEB" "\BEC" "\BED" "\BEE" "\BEF"; /* ௦ ௧ ௨ ௩ ௪ ௫ ௬ ௭ ௮ ௯ */ }

@counter-style telugu { system: numeric; symbols: "\C66" "\C67" "\C68" "\C69" "\C6A" "\C6B" "\C6C" "\C6D" "\C6E" "\C6F"; /* ౦ ౧ ౨ ౩ ౪ ౫ ౬ ౭ ౮ ౯ */ }

@counter-style thai { system: numeric; symbols: "\E50" "\E51" "\E52" "\E53" "\E54" "\E55" "\E56" "\E57" "\E58" "\E59"; /* ๐ ๑ ๒ ๓ ๔ ๕ ๖ ๗ ๘ ๙ */ }

@counter-style tibetan { system: numeric; symbols: "\F20" "\F21" "\F22" "\F23" "\F24" "\F25" "\F26" "\F27" "\F28" "\F29"; /* ༠ ༡ ༢ ༣ ༤ ༥ ༦ ༧ ༨ ༩ */ }

Implementations must implement hebrew at least to the range specified in the @counter-style rule above, but may implement it to a higher range. If they do so, the corresponding range descriptor must reflect the implemented range.

6.2. Alphabetic: lower-alpha, lower-latin, upper-alpha, upper-latin, cjk-earthly-branch, cjk-heavenly-stem, lower-greek, hiragana, hiragana-iroha, katakana, katakana-iroha

lower-alpha

lower-latin

Lowercase ASCII letters (e.g., a, b, c, ..., z, aa, ab).

upper-alpha

upper-latin

Uppercase ASCII letters (e.g., A, B, C, ..., Z, AA, AB).

cjk-earthly-branch

Han "Earthly Branch" ordinals (e.g., 子, 丑, 寅, ..., 亥, 子子, 子丑).

cjk-heavenly-stem

Han "Heavenly Stem" ordinals (e.g., 甲, 乙, 丙, ..., 癸, 甲甲, 甲乙).

lower-greek

Lowercase classical Greek (e.g., α, β, γ, ..., ω, αα, αβ).

hiragana

Dictionary-order hiragana lettering (e.g., あ, い, う, ..., ん, ああ, あい).

hiragana-iroha

Iroha-order hiragana lettering (e.g., い, ろ, は, ..., す, いい, いろ).

katakana

Dictionary-order katakana lettering (e.g., ア, イ, ウ, ..., ン, アア, アイ).

katakana-iroha

Iroha-order katakana lettering (e.g., イ, ロ, ハ, ..., ス, イイ, イロ)

The following stylesheet fragment provides the normative definition of these predefined counter styles:

@counter-style lower-alpha { system: alphabetic; symbols: a b c d e f g h i j k l m n o p q r s t u v w x y z; }

@counter-style lower-latin { system: extends lower-alpha; }

@counter-style upper-alpha { system: alphabetic; symbols: A B C D E F G H I J K L M N O P Q R S T U V W X Y Z; }

@counter-style upper-latin { system: extends upper-alpha; }

@counter-style cjk-earthly-branch { system: alphabetic; symbols: "\5B50" "\4E11" "\5BC5" "\536F" "\8FB0" "\5DF3" "\5348" "\672A" "\7533" "\9149" "\620C" "\4EA5"; /* 子 丑 寅 卯 辰 巳 午 未 申 酉 戌 亥 */ suffix: "、"; }

@counter-style cjk-heavenly-stem { system: alphabetic; symbols: "\7532" "\4E59" "\4E19" "\4E01" "\620A" "\5DF1" "\5E9A" "\8F9B" "\58EC" "\7678"; /* 甲 乙 丙 丁 戊 己 庚 辛 壬 癸 */ suffix: "、"; }

@counter-style lower-greek { system: alphabetic; symbols: "\3B1" "\3B2" "\3B3" "\3B4" "\3B5" "\3B6" "\3B7" "\3B8" "\3B9" "\3BA" "\3BB" "\3BC" "\3BD" "\3BE" "\3BF" "\3C0" "\3C1" "\3C3" "\3C4" "\3C5" "\3C6" "\3C7" "\3C8" "\3C9"; /* α β γ δ ε ζ η θ ι κ λ μ ν ξ ο π ρ σ τ υ φ χ ψ ω */ }

@counter-style hiragana { system: alphabetic; symbols: "\3042" "\3044" "\3046" "\3048" "\304A" "\304B" "\304D" "\304F" "\3051" "\3053" "\3055" "\3057" "\3059" "\305B" "\305D" "\305F" "\3061" "\3064" "\3066" "\3068" "\306A" "\306B" "\306C" "\306D" "\306E" "\306F" "\3072" "\3075" "\3078" "\307B" "\307E" "\307F" "\3080" "\3081" "\3082" "\3084" "\3086" "\3088" "\3089" "\308A" "\308B" "\308C" "\308D" "\308F" "\3090" "\3091" "\3092" "\3093"; /* あ い う え お か き く け こ さ し す せ そ た ち つ て と な に ぬ ね の は ひ ふ へ ほ ま み む め も や ゆ よ ら り る れ ろ わ ゐ ゑ を ん */ suffix: "、"; }

@counter-style hiragana-iroha { system: alphabetic; symbols: "\3044" "\308D" "\306F" "\306B" "\307B" "\3078" "\3068" "\3061" "\308A" "\306C" "\308B" "\3092" "\308F" "\304B" "\3088" "\305F" "\308C" "\305D" "\3064" "\306D" "\306A" "\3089" "\3080" "\3046" "\3090" "\306E" "\304A" "\304F" "\3084" "\307E" "\3051" "\3075" "\3053" "\3048" "\3066" "\3042" "\3055" "\304D" "\3086" "\3081" "\307F" "\3057" "\3091" "\3072" "\3082" "\305B" "\3059"; /* い ろ は に ほ へ と ち り ぬ る を わ か よ た れ そ つ ね な ら む う ゐ の お く や ま け ふ こ え て あ さ き ゆ め み し ゑ ひ も せ す */ suffix: "、"; }

@counter-style katakana { system: alphabetic; symbols: "\30A2" "\30A4" "\30A6" "\30A8" "\30AA" "\30AB" "\30AD" "\30AF" "\30B1" "\30B3" "\30B5" "\30B7" "\30B9" "\30BB" "\30BD" "\30BF" "\30C1" "\30C4" "\30C6" "\30C8" "\30CA" "\30CB" "\30CC" "\30CD" "\30CE" "\30CF" "\30D2" "\30D5" "\30D8" "\30DB" "\30DE" "\30DF" "\30E0" "\30E1" "\30E2" "\30E4" "\30E6" "\30E8" "\30E9" "\30EA" "\30EB" "\30EC" "\30ED" "\30EF" "\30F0" "\30F1" "\30F2" "\30F3"; /* ア イ ウ エ オ カ キ ク ケ コ サ シ ス セ ソ タ チ ツ テ ト ナ ニ ヌ ネ ノ ハ ヒ フ ヘ ホ マ ミ ム メ モ ヤ ユ ヨ ラ リ ル レ ロ ワ ヰ ヱ ヲ ン */ suffix: "、"; }

@counter-style katakana-iroha { system: alphabetic; symbols: "\30A4" "\30ED" "\30CF" "\30CB" "\30DB" "\30D8" "\30C8" "\30C1" "\30EA" "\30CC" "\30EB" "\30F2" "\30EF" "\30AB" "\30E8" "\30BF" "\30EC" "\30BD" "\30C4" "\30CD" "\30CA" "\30E9" "\30E0" "\30A6" "\30F0" "\30CE" "\30AA" "\30AF" "\30E4" "\30DE" "\30B1" "\30D5" "\30B3" "\30A8" "\30C6" "\30A2" "\30B5" "\30AD" "\30E6" "\30E1" "\30DF" "\30B7" "\30F1" "\30D2" "\30E2" "\30BB" "\30B9"; /* イ ロ ハ ニ ホ ヘ ト チ リ ヌ ル ヲ ワ カ ヨ タ レ ソ ツ ネ ナ ラ ム ウ ヰ ノ オ ク ヤ マ ケ フ コ エ テ ア サ キ ユ メ ミ シ ヱ ヒ モ セ ス */ suffix: "、"; }

6.3. Symbolic: disc, circle, square, disclosure-open, disclosure-closed

disc

A filled circle, similar to • U+2022 BULLET.

circle

A hollow circle, similar to ◦ U+25E6 WHITE BULLET.

square

A filled square, similar to ◾ U+25FE BLACK MEDIUM SMALL SQUARE.

disclosure-open

disclosure-closed

Symbols appropriate for indicating an open or closed disclosure widget, such as the HTML <details> element.

The disc counter-style must not be overridable with a @counter-style rule, so that the initial value of list-style-type is well-known to the User Agent.

The following stylesheet fragment provides the normative definition of these predefined counter styles:

@counter-style disc { system: cyclic; symbols: \2022; /* • */ suffix: " "; }

@counter-style circle { system: cyclic; symbols: \25E6; /* ◦ */ suffix: " "; }

@counter-style square { system: cyclic; symbols: \25FE; /* ◾ */ suffix: " "; }

@counter-style disclosure-open { system: cyclic; suffix: " "; /* for symbols, see normative text below */ }

@counter-style disclosure-closed { system: cyclic; suffix: " "; /* for symbols, see normative text below */ }

Alternately, a browser may render these styles using a browser-generated image instead of the defined character. If so, the image must look similar to the character, and must be sized to attractively fill a 1em by 1em square.

For the disclosure-open and disclosure-closed counter styles, the marker must be an image or character suitable for indicating the open and closed states of a disclosure widget, such as HTML’s <details> element. If the image is directional, it must respond to the writing mode of the element, similar to the bidi-sensitive images feature of the Images 4 module. For example, the disclosure-closed style might use the characters U+25B8 BLACK RIGHT-POINTING SMALL TRIANGLE (▸) and U+25C2 BLACK LEFT-POINTING SMALL TRIANGLE (◂), while the disclosure-open style might use the character U+25BE BLACK DOWN-POINTING SMALL TRIANGLE (▾).

7. Complex Predefined Counter Styles

While authors may define their own counter styles using the @counter-style rule or rely on the set of predefined counter styles, a few counter styles are described by rules that are too complex to be captured by the predefined algorithms. These counter styles are described in this section.

Some of the counter styles specified in this section have custom algorithms for generating counter values, but are otherwise identical to a counter style defined via the @counter-style rule. For example, an author can reference one of these styles in an extends system, reusing the algorithm but swapping out some of the other descriptors.

All of the counter styles defined in this section have a spoken form of numbers, and use a negative sign.

7.1. Longhand East Asian Counter Styles

Chinese, Japanese, and Korean have counter styles which have a “longhand” nature, similar to “thirteen thousand one hundred and twenty-three” in English. Each has both formal and informal variants. The formal styles are typically used in financial and legal documents, as their characters are more difficult to alter into each other.

The following table shows examples of these styles, particularly some ways in which they differ.

Because opinions differ on how best to represent numbers 10k or greater using the longhand CJK styles, all of the counter styles defined in this section are defined to have a range of -9999 to 9999, but implementations may support a larger range. Outside the implementation-supported range, the fallback is cjk-decimal.

Note: Implementations are encouraged to research and implement counter representations beyond 10k and report back to the CSS Working Group with data when a generally-accepted answer is discovered. Some previous research on this topic is contained in an earlier draft.

7.1.1. Japanese: japanese-informal and japanese-formal

japanese-informal

Informal Japanese Kanji numbering (e.g., 千百十一)

japanese-formal

Formal Japanese Kanji numbering (e.g. 壱阡壱百壱拾壱)

@counter-style japanese-informal { system: additive; range: -9999 9999; additive-symbols: 9000 \4E5D\5343, 8000 \516B\5343, 7000 \4E03\5343, 6000 \516D\5343, 5000 \4E94\5343, 4000 \56DB\5343, 3000 \4E09\5343, 2000 \4E8C\5343, 1000 \5343, 900 \4E5D\767E, 800 \516B\767E, 700 \4E03\767E, 600 \516D\767E, 500 \4E94\767E, 400 \56DB\767E, 300 \4E09\767E, 200 \4E8C\767E, 100 \767E, 90 \4E5D\5341, 80 \516B\5341, 70 \4E03\5341, 60 \516D\5341, 50 \4E94\5341, 40 \56DB\5341, 30 \4E09\5341, 20 \4E8C\5341, 10 \5341, 9 \4E5D, 8 \516B, 7 \4E03, 6 \516D, 5 \4E94, 4 \56DB, 3 \4E09, 2 \4E8C, 1 \4E00, 0 \3007; /* 9000 九千, 8000 八千, 7000 七千, 6000 六千, 5000 五千, 4000 四千, 3000 三千, 2000 二千, 1000 千, 900 九百, 800 八百, 700 七百, 600 六百, 500 五百, 400 四百, 300 三百, 200 二百, 100 百, 90 九十, 80 八十, 70 七十, 60 六十, 50 五十, 40 四十, 30 三十, 20 二十, 10 十, 9 九, 8 八, 7 七, 6 六, 5 五, 4 四, 3 三, 2 二, 1 一, 0 〇 / suffix: '\3001'; / 、 / negative: "\30DE\30A4\30CA\30B9"; / マイナス */ fallback: cjk-decimal; }

@counter-style japanese-formal { system: additive; range: -9999 9999; additive-symbols: 9000 \4E5D\9621, 8000 \516B\9621, 7000 \4E03\9621, 6000 \516D\9621, 5000 \4F0D\9621, 4000 \56DB\9621, 3000 \53C2\9621, 2000 \5F10\9621, 1000 \58F1\9621, 900 \4E5D\767E, 800 \516B\767E, 700 \4E03\767E, 600 \516D\767E, 500 \4F0D\767E, 400 \56DB\767E, 300 \53C2\767E, 200 \5F10\767E, 100 \58F1\767E, 90 \4E5D\62FE, 80 \516B\62FE, 70 \4E03\62FE, 60 \516D\62FE, 50 \4F0D\62FE, 40 \56DB\62FE, 30 \53C2\62FE, 20 \5F10\62FE, 10 \58F1\62FE, 9 \4E5D, 8 \516B, 7 \4E03, 6 \516D, 5 \4F0D, 4 \56DB, 3 \53C2, 2 \5F10, 1 \58F1, 0 \96F6; /* 9000 九阡, 8000 八阡, 7000 七阡, 6000 六阡, 5000 伍阡, 4000 四阡, 3000 参阡, 2000 弐阡, 1000 壱阡, 900 九百, 800 八百, 700 七百, 600 六百, 500 伍百, 400 四百, 300 参百, 200 弐百, 100 壱百, 90 九拾, 80 八拾, 70 七拾, 60 六拾, 50 伍拾, 40 四拾, 30 参拾, 20 弐拾, 10 壱拾, 9 九, 8 八, 7 七, 6 六, 5 伍, 4 四, 3 参, 2 弐, 1 壱, 0 零 / suffix: '\3001'; / 、 / negative: "\30DE\30A4\30CA\30B9"; / マイナス */ fallback: cjk-decimal; }

7.1.2. Korean: korean-hangul-formal, korean-hanja-informal, and korean-hanja-formal

korean-hangul-formal

Korean Hangul numbering (e.g., 일천일백일십일)

korean-hanja-informal

Informal Korean Hanja numbering (e.g., 千百十一)

korean-hanja-formal

Formal Korean Han (Hanja) numbering (e.g., 壹仟壹百壹拾壹)

@counter-style korean-hangul-formal { system: additive; range: -9999 9999; additive-symbols: 9000 \AD6C\CC9C, 8000 \D314\CC9C, 7000 \CE60\CC9C, 6000 \C721\CC9C, 5000 \C624\CC9C, 4000 \C0AC\CC9C, 3000 \C0BC\CC9C, 2000 \C774\CC9C, 1000 \C77C\CC9C, 900 \AD6C\BC31, 800 \D314\BC31, 700 \CE60\BC31, 600 \C721\BC31, 500 \C624\BC31, 400 \C0AC\BC31, 300 \C0BC\BC31, 200 \C774\BC31, 100 \C77C\BC31, 90 \AD6C\C2ED, 80 \D314\C2ED, 70 \CE60\C2ED, 60 \C721\C2ED, 50 \C624\C2ED, 40 \C0AC\C2ED, 30 \C0BC\C2ED, 20 \C774\C2ED, 10 \C77C\C2ED, 9 \AD6C, 8 \D314, 7 \CE60, 6 \C721, 5 \C624, 4 \C0AC, 3 \C0BC, 2 \C774, 1 \C77C, 0 \C601; /* 9000 구천, 8000 팔천, 7000 칠천, 6000 육천, 5000 오천, 4000 사천, 3000 삼천, 2000 이천, 1000 일천, 900 구백, 800 팔백, 700 칠백, 600 육백, 500 오백, 400 사백, 300 삼백, 200 이백, 100 일백, 90 구십, 80 팔십, 70 칠십, 60 육십, 50 오십, 40 사십, 30 삼십, 20 이십, 10 일십, 9 구, 8 팔, 7 칠, 6 육, 5 오, 4 사, 3 삼, 2 이, 1 일, 0 영 / suffix: ', '; negative: "\B9C8\C774\B108\C2A4 "; / 마이너스 (followed by a space) */ }

@counter-style korean-hanja-informal { system: additive; range: -9999 9999; additive-symbols: 9000 \4E5D\5343, 8000 \516B\5343, 7000 \4E03\5343, 6000 \516D\5343, 5000 \4E94\5343, 4000 \56DB\5343, 3000 \4E09\5343, 2000 \4E8C\5343, 1000 \5343, 900 \4E5D\767E, 800 \516B\767E, 700 \4E03\767E, 600 \516D\767E, 500 \4E94\767E, 400 \56DB\767E, 300 \4E09\767E, 200 \4E8C\767E, 100 \767E, 90 \4E5D\5341, 80 \516B\5341, 70 \4E03\5341, 60 \516D\5341, 50 \4E94\5341, 40 \56DB\5341, 30 \4E09\5341, 20 \4E8C\5341, 10 \5341, 9 \4E5D, 8 \516B, 7 \4E03, 6 \516D, 5 \4E94, 4 \56DB, 3 \4E09, 2 \4E8C, 1 \4E00, 0 \96F6; /* 9000 九千, 8000 八千, 7000 七千, 6000 六千, 5000 五千, 4000 四千, 3000 三千, 2000 二千, 1000 千, 900 九百, 800 八百, 700 七百, 600 六百, 500 五百, 400 四百, 300 三百, 200 二百, 100 百, 90 九十, 80 八十, 70 七十, 60 六十, 50 五十, 40 四十, 30 三十, 20 二十, 10 十, 9 九, 8 八, 7 七, 6 六, 5 五, 4 四, 3 三, 2 二, 1 一, 0 零 / suffix: ', '; negative: "\B9C8\C774\B108\C2A4 "; / 마이너스 (followed by a space) */ }

@counter-style korean-hanja-formal { system: additive; range: -9999 9999; additive-symbols: 9000 \4E5D\4EDF, 8000 \516B\4EDF, 7000 \4E03\4EDF, 6000 \516D\4EDF, 5000 \4E94\4EDF, 4000 \56DB\4EDF, 3000 \53C3\4EDF, 2000 \8CB3\4EDF, 1000 \58F9\4EDF, 900 \4E5D\767E, 800 \516B\767E, 700 \4E03\767E, 600 \516D\767E, 500 \4E94\767E, 400 \56DB\767E, 300 \53C3\767E, 200 \8CB3\767E, 100 \58F9\767E, 90 \4E5D\62FE, 80 \516B\62FE, 70 \4E03\62FE, 60 \516D\62FE, 50 \4E94\62FE, 40 \56DB\62FE, 30 \53C3\62FE, 20 \8CB3\62FE, 10 \58F9\62FE, 9 \4E5D, 8 \516B, 7 \4E03, 6 \516D, 5 \4E94, 4 \56DB, 3 \53C3, 2 \8CB3, 1 \58F9, 0 \96F6; /* 9000 九仟, 8000 八仟, 7000 七仟, 6000 六仟, 5000 五仟, 4000 四仟, 3000 參仟, 2000 貳仟, 1000 壹仟, 900 九百, 800 八百, 700 七百, 600 六百, 500 五百, 400 四百, 300 參百, 200 貳百, 100 壹百, 90 九拾, 80 八拾, 70 七拾, 60 六拾, 50 五拾, 40 四拾, 30 參拾, 20 貳拾, 10 壹拾, 9 九, 8 八, 7 七, 6 六, 5 五, 4 四, 3 參, 2 貳, 1 壹, 0 零 / suffix: ', '; negative: "\B9C8\C774\B108\C2A4 "; / 마이너스 (followed by a space) */ }

7.1.3. Chinese: simp-chinese-informal, simp-chinese-formal, trad-chinese-informal, and trad-chinese-formal

simp-chinese-informal

Simplified Chinese informal numbering (e.g., 一千一百一十一)

simp-chinese-formal

Simplified Chinese formal numbering (e.g. 壹仟壹佰壹拾壹)

trad-chinese-informal

Traditional Chinese informal numbering (e.g., 一千一百一十一)

trad-chinese-formal

Traditional Chinese formal numbering (e.g., 壹仟壹佰壹拾壹)

cjk-ideographic

This counter style is identical to trad-chinese-informal. (It exists for legacy reasons.)

The Chinese longhand styles are defined by almost identical algorithms (specified as a single algorithm here, with the differences called out when relevant), but use different sets of characters, as specified by the table following the algorithm.

1. If the counter value is 0, the representation is the character for 0 specified for the given counter style. Skip the rest of this algorithm.
2. Initially represent the counter value as a decimal number. For each digit that is not 0, append the appropriate digit marker to the digit. The ones digit has no marker.
3. For the informal styles, if the counter value is between ten and nineteen, remove the tens digit (leave the digit marker).
4. Drop any trailing zeros and collapse any remaining zeros into a single zero digit.
5. Replace the digits 0-9 with the appropriate character for the given counter style. Return the resultant string as the representation of the counter value.

For all of these counter styles, the suffix is "、" U+3001, the fallback is cjk-decimal, the range is -9999 9999, and the negative value is given in the table of symbols for each style.

The following tables define the characters used in these styles:

For reference, here are the first 120 values for the simp-chinese-informal style:

1　　　　 一 41　　 四十一 81　　 八十一 2　　　　 二 42　　 四十二 82　　 八十二 3　　　　 三 43　　 四十三 83　　 八十三 4　　　　 四 44　　 四十四 84　　 八十四 5　　　　 五 45　　 四十五 85　　 八十五 6　　　　 六 46　　 四十六 86　　 八十六 7　　　　 七 47　　 四十七 87　　 八十七 8　　　　 八 48　　 四十八 88　　 八十八 9　　　　 九 49　　 四十九 89　　 八十九 10　　　　 十 50　　　 五十 90　　　 九十 11　　　 十一 51　　 五十一 91　　 九十一 12　　　 十二 52　　 五十二 92　　 九十二 13　　　 十三 53　　 五十三 93　　 九十三 14　　　 十四 54　　 五十四 94　　 九十四 15　　　 十五 55　　 五十五 95　　 九十五 16　　　 十六 56　　 五十六 96　　 九十六 17　　　 十七 57　　 五十七 97　　 九十七 18　　　 十八 58　　 五十八 98　　 九十八 19　　　 十九 59　　 五十九 99　　 九十九 20　　　 二十 60　　　 六十 100　　　 一百 21　　 二十一 61　　 六十一 101　 一百零一 22　　 二十二 62　　 六十二 102　 一百零二 23　　 二十三 63　　 六十三 103　 一百零三 24　　 二十四 64　　 六十四 104　 一百零四 25　　 二十五 65　　 六十五 105　 一百零五 26　　 二十六 66　　 六十六 106　 一百零六 27　　 二十七 67　　 六十七 107　 一百零七 28　　 二十八 68　　 六十八 108　 一百零八 29　　 二十九 69　　 六十九 109　 一百零九 30　　　 三十 70　　　 七十 110　 一百一十 31　　 三十一 71　　 七十一 111 一百一十一 32　　 三十二 72　　 七十二 112 一百一十二 33　　 三十三 73　　 七十三 113 一百一十三 34　　 三十四 74　　 七十四 114 一百一十四 35　　 三十五 75　　 七十五 115 一百一十五 36　　 三十六 76　　 七十六 116 一百一十六 37　　 三十七 77　　 七十七 117 一百一十七 38　　 三十八 78　　 七十八 118 一百一十八 39　　 三十九 79　　 七十九 119 一百一十九 40　　　 四十 80　　　 八十 120　 一百二十

7.2. Ethiopic Numeric Counter Style: ethiopic-numeric

The ethiopic-numeric counter style is defined for all positive non-zero numbers. The following algorithm converts decimal digits to ethiopic numbers:

1. If the number is 1, return "፩" (U+1369).
2. Split the number into groups of two digits, starting with the least significant decimal digit.
3. Index each group sequentially, starting from the least significant as group number zero.
4. If the group has the value zero, or if the group is the most significant one and has the value 1, or if the group has an odd index (as given in the previous step) and has the value 1, then remove the digits (but leave the group, so it still has a separator appended below).
5. For each remaining digit, substitute the relevant ethiopic character from the list below.

   Tens	Units
   Values	Codepoints	Values	Codepoints
   10	፲	U+1372	1	፩	U+1369
   20	፳	U+1373	2	፪	U+136A
   30	፴	U+1374	3	፫	U+136B
   40	፵	U+1375	4	፬	U+136C
   50	፶	U+1376	5	፭	U+136D
   60	፷	U+1377	6	፮	U+136E
   70	፸	U+1378	7	፯	U+136F
   80	፹	U+1379	8	፰	U+1370
   90	፺	U+137A	9	፱	U+1371

6. For each group with an odd index (as given in the second step), except groups which originally had a value of zero, append ፻ U+137B.
7. For each group with an even index (as given in the second step), except the group with index 0, append ፼ U+137C.
8. Concatenate the groups into one string, and return it.

For this system, the name is "ethiopic-numeric", the range is 1 infinite, the suffix is "/ " (U+002F SOLIDUS followed by a U+0020 SPACE), and the rest of the descriptors have their initial value.

The decimal number 100, in ethiopic, is ፻ U+137B

The decimal number 78010092, in ethiopic, is ፸፰፻፩፼፺፪ U+1378 U+1370 U+137B U+1369 U+137C U+137A U+136A.

The decimal number 780100000092, in ethiopic, is ፸፰፻፩፼፼፺፪ U+1378 U+1370 U+137B U+1369 U+137C U+137C U+137A U+136A.

8. Additional Predefined Counter Styles

The Internationalization Working Group maintains a large list of predefined @counter-style rules for various world languages in their Predefined Counter Styles document. [predefined-counter-styles]

These additional counter styles are not intended to be supported by user-agents by default, but can be used by users or authors copying them directly into style sheets.

9. APIs

9.1. Extensions to the CSSRule interface

The CSSRule interface is extended as follows:

partial interface CSSRule { const unsigned short COUNTER_STYLE_RULE = 11; };

9.2. The CSSCounterStyleRule interface

The CSSCounterStyleRule interface represents a @counter-style rule.

interface CSSCounterStyleRule : CSSRule { attribute CSSOMString name; attribute CSSOMString system; attribute CSSOMString symbols; attribute CSSOMString additiveSymbols; attribute CSSOMString negative; attribute CSSOMString prefix; attribute CSSOMString suffix; attribute CSSOMString range; attribute CSSOMString pad; attribute CSSOMString speakAs; attribute CSSOMString fallback; };

name, of type CSSOMString

The name attribute on getting must return a CSSOMString object that contains the serialization of the defined for the associated rule.

On setting the name attribute, run the following steps:

1. If the value is an ASCII case-insensitive match for any of the predefined counter styles, lowercase it.
2. If the value is not "decimal", "disc", or "none", replace the associated rule’s name with an identifier equal to the value.
3. Otherwise, do nothing.

system, of type CSSOMString

symbols, of type CSSOMString

additiveSymbols, of type CSSOMString

negative, of type CSSOMString

prefix, of type CSSOMString

suffix, of type CSSOMString

range, of type CSSOMString

pad, of type CSSOMString

speakAs, of type CSSOMString

fallback, of type CSSOMString

The remaining attributes on getting must return a CSSOMString object that contains the serialization of the associated descriptor defined for the associated rule. If the descriptor was not specified in the associated rule, the attribute must return an empty string.

On setting, run the following steps:

1. Parse a list of component values from the value.
2. If the returned value is invalid according to the given descriptor’s grammar, or would cause the @counter-style rule to become invalid, do nothing and abort these steps. (For example, some systems require the symbols descriptor to contain two values.)
3. If the attribute being set is system, and the new value would change the algorithm used, do nothing and abort these steps. It’s okay to change an aspect of the algorithm, like the first symbol value of a fixed system.
4. Set the descriptor to the value.

10. Sample style sheet for HTML

This section is informative, not normative. HTML itself defines the styles that apply to its elements, and in some cases defers to the user agent’s discretion.

details > summary { display: list-item; list-style: disclosure-closed inside; }

details[open] > summary { list-style: disclosure-open inside; }

Changes

Changes since the Jun 2015 Candidate Recommendation

Significant changes since the June 11 2015 CR draft:

• Exclude "none" and "disc" from being the name of a counter style.
• When setting CSSCounterStyle.name, take the string directly; don’t parse it as an ident.
• Clarify that counter styles are read out in the element’s content language.
• Clarified that additive-symbols tuples must be of strictly decreasing weight.
• Specified that invalid values just invalidate the declaration, not the whole rule.
• @counter-style rules that are invalid due to missing descriptors just fail to create a counter style; they’re otherwise still valid rules.

A Disposition of Comments is available.

Changes since the Feb 2015 Candidate Recommendation

• Allowed UAs to extend the hebrew style past the spec-defined limits (since the current limits are mostly just an artifact of how annoying it is to go higher with the @counter-style-based definition).

Acknowledgments

The following people and documentation they wrote were very useful for defining the numbering systems: Alexander Savenkov, Arron Eicholz, Aryeh Gregor, Christopher Hoess, Daniel Yacob, Frank Tang, Jonathan Rosenne, Karl Ove Hufthammer, Musheg Arakelyan, Nariné Renard Karapetyan, Randall Bart, Richard Ishida, Simon Montagu (Mozilla, smontagu@smontagu.org)

Special thanks to Xidorn Quan for extensive reviews of all aspects of the spec.

Privacy and Security Considerations

This specification introduces no new privacy or security considerations.

Conformance requirements are expressed with a combination of descriptive assertions and RFC 2119 terminology. The key words “MUST”, “MUST NOT”, “REQUIRED”, “SHALL”, “SHALL NOT”, “SHOULD”, “SHOULD NOT”, “RECOMMENDED”, “MAY”, and “OPTIONAL” in the normative parts of this document are to be interpreted as described in RFC 2119. However, for readability, these words do not appear in all uppercase letters in this specification.

All of the text of this specification is normative except sections explicitly marked as non-normative, examples, and notes. [RFC2119]

Examples in this specification are introduced with the words “for example” or are set apart from the normative text with class="example", like this:

Informative notes begin with the word “Note” and are set apart from the normative text with class="note", like this:

Note, this is an informative note.

Advisements are normative sections styled to evoke special attention and are set apart from other normative text with <strong class="advisement">, like this: UAs MUST provide an accessible alternative.

A style sheet is conformant to this specification if all of its statements that use syntax defined in this module are valid according to the generic CSS grammar and the individual grammars of each feature defined in this module.

A renderer is conformant to this specification if, in addition to interpreting the style sheet as defined by the appropriate specifications, it supports all the features defined by this specification by parsing them correctly and rendering the document accordingly. However, the inability of a UA to correctly render a document due to limitations of the device does not make the UA non-conformant. (For example, a UA is not required to render color on a monochrome monitor.)

An authoring tool is conformant to this specification if it writes style sheets that are syntactically correct according to the generic CSS grammar and the individual grammars of each feature in this module, and meet all other conformance requirements of style sheets as described in this module.

The following sections define several conformance requirements for implementing CSS responsibly, in a way that promotes interoperability in the present and future.

So that authors can exploit the forward-compatible parsing rules to assign fallback values, CSS renderers must treat as invalid (and ignore as appropriate) any at-rules, properties, property values, keywords, and other syntactic constructs for which they have no usable level of support. In particular, user agents must not selectively ignore unsupported property values and honor supported values in a single multi-value property declaration: if any value is considered invalid (as unsupported values must be), CSS requires that the entire declaration be ignored.

Once a specification reaches the Candidate Recommendation stage, implementers should release an unprefixed implementation of any CR-level feature they can demonstrate to be correctly implemented according to spec, and should avoid exposing a prefixed variant of that feature.

To establish and maintain the interoperability of CSS across implementations, the CSS Working Group requests that non-experimental CSS renderers submit an implementation report (and, if necessary, the testcases used for that implementation report) to the W3C before releasing an unprefixed implementation of any CSS features. Testcases submitted to W3C are subject to review and correction by the CSS Working Group.

For this specification to be advanced to Proposed Recommendation, there must be at least two independent, interoperable implementations of each feature. Each feature may be implemented by a different set of products, there is no requirement that all features be implemented by a single product. For the purposes of this criterion, we define the following terms:

The specification will remain Candidate Recommendation for at least six months.

No properties defined.