CSS Syntax Module Level 3

Abstract

CSS is a language for describing the rendering of structured documents (such as HTML and XML) on screen, on paper, in speech, etc. This module describes, in general terms, the basic structure and syntax of CSS stylesheets. It defines, in detail, the syntax and parsing of CSS - how to turn a stream of bytes into a meaningful stylesheet.

Status of this document

The following features are at risk: …

Table of contents

Introduction

This section is not normative.

This module defines the abstract syntax and parsing of CSS stylesheets and other things which use CSS syntax (such as the HTML style attribute).

It defines algorithms for converting a stream of codepoints (in other words, text) into a stream of CSS tokens, and then further into CSS objects such as stylesheets, rules, and declarations.

Module interactions

This module defines the syntax and parsing of CSS stylesheets. It supersedes the lexical scanner and grammar defined in CSS 2.1.

Description of CSS's Syntax

This section is not normative.

A CSS document is a series of style rules, which apply CSS properties to elements in the source document, and at-rules, which define special processing rules or values for the CSS document.

A style rule starts with a selector (defined by the Selectors specification), then has a {}-wrapped block containing a sequence of declarations. The selector specifies which elements the declarations will apply to. Each declaration has a property name, followed by a colon and the property value, and finished with a semicolon.

p > a {
	color: blue;
	text-decoration: underline;
}

At-rules are all different, but they have a basic structure in common. They start with an "@" character followed by their name. Some at-rules are simple statements, with their name followed by more CSS values to specify their behavior, and finally ended by a semicolon. Others are blocks; they can have CSS values following their name, but they end with a {}-wrapped block, similar to a rule. Even the contents of these blocks are specific to the given at-rule: sometimes they contain a sequence of declarations, like a rule; other times, they may contain additional blocks, or at-rules, or other structures altogether.

@import "my-styles.css";

@page :left {
	margin-left: 4cm;
	margin-right: 3cm;
}

@media print {
	body { font-size: 10pt }
}

Property names and at-rule names are always idents, which have to start with a letter or a hyphen followed by a letter, and then can contain letters, numbers, hyphens, or underscores. You can include any character at all, even ones that CSS uses in its syntax, by escaping it with a backslash (\) or by using a hexadecimal escape.

The syntax of selectors is defined in the Selectors spec. Similarly, the syntax of the wide variety of CSS values is defined in the Values & Units spec. The special syntaxes of individual at-rules can be found in the specs that define them.

Tokenizing and Parsing CSS

User agents must use the parsing rules described in this specifiction to generate the CSSOM trees from text/css resources. Together, these rules define what is referred to as the CSS parser.

This specification defines the parsing rules for CSS documents, whether they are syntactically correct or not. Certain points in the parsing algorithm are said to be parse errors. The error handling for parse errors is well-defined: user agents must either act as described below when encountering such problems, or must abort processing at the first error that they encounter for which they do not wish to apply the rules described below.

Conformance checkers must report at least one parse error condition to the user if one or more parse error conditions exist in the document and must not report parse error conditions if none exist in the document. Conformance checkers may report more than one parse error condition if more than one parse error condition exists in the document. Conformance checkers are not required to recover from parse errors.

Overview of the Parsing Model

The input to the CSS parsing process consists of a stream of Unicode code points, which is passed through a tokenization stage followed by a tree construction stage. The output is a CSSStyleSheet object.

Implementations that do not support scripting do not have to actually create a CSSOM CSSStyleSheet object, but the CSSOM tree in such cases is still used as the model for the rest of the specification.

The input byte stream

The stream of Unicode code points that comprises the input to the tokenization stage will be initially seen by the user agent as a stream of bytes (typically coming over the network or from the local file system). The bytes encode the actual characters according to a particular character encoding, which the user agent must use to decode the bytes into characters.

To decode the stream of bytes into a stream of characters, UAs must follow these steps.

The algorithms to get an encoding and decode are defined in the Encoding Standard.

1. Let encoding be utf-8.
2. If HTTP or equivalent protocol defines an encoding (e.g. via the charset parameter of the Content-Type header), get an encoding for the specified value. If that does not return failure, set encoding to the return value and jump to the last step of this algorithm.
3. Check the byte stream. If the first several bytes match the hex sequence

   40 63 68 61 72 73 65 74 20 22 (XX)* 22 3B

   then get an encoding for the sequence of XX bytes, decoded per windows-1252, and let temp be the return value.

   Note: Anything ASCII-compatible will do, so using windows-1252 is fine.

   Note: The byte sequence above, when decoded as ASCII, is the string "@charset "…";", where the "…" is the sequence of bytes corresponding to the encoding's name.

   If temp is utf-16 or utf-16be, set temp to utf-8. If temp is not failure, set encoding to it and jump to the last step.

   This mimics HTML <meta> behavior.

4. Get an encoding for the value of the charset attribute on the <link> element or <?xml-stylesheet?> processing instruction that caused the style sheet to be included, if any. If that does not return failure, set encoding to the return value and jump to the last step.
5. Set encoding to the encoding of the referring style sheet or document, if any.
6. Decode the byte stream using fallback encoding encoding.

   Note: the decode algorithm lets the byte order mark (BOM) take precedence, hence the usage of the term "fallback" above.

Anne says that steps 4/5 should be an input to this algorithm from the specs that define importing stylesheet, to make the algorithm as a whole cleaner. Perhaps abstract it into the concept of an "environment charset" or something?

Preprocessing the input stream

The input stream consists of the characters pushed into it as the input byte stream is decoded.

Before sending the input stream to the tokenizer, implementations must make the following character substitutions:

• Replace any U+000D CARRIAGE RETURN (CR) characters or pairs of U+000D CARRIAGE RETURN (CR) followed by U+000A LINE FEED (LF) by a single U+000A LINE FEED (LF) character.
• Replace any U+0000 NULL characters with U+FFFD REPLACEMENT CHARACTER (�).

The next input character is the first character in the input stream that has not yet been consumed or explicitly ignored by the requirements in this section. Initially, the next input character is the first character in the input. The current input character is the last character to have been consumed.

The "EOF" character in the tables below is a conceptual character representing the end of the input stream.

Tokenization

Implementations must act as if they used the following state machine to tokenize CSS. The state machine must start in the data state. Most states consume a single character, which may have various side-effects, and either switches the state machine to a new state to reconsume the same character, or switches it to a new state to consume the next character, or stays in the same state to consume the next character. Some states have more complicated behavior and can consume several characters before switching to another state.

The output of the tokenization step is a series of zero or more of the following tokens: ident, function, at-keyword, hash, string, bad-string, url, bad-url, delim, number, percentage, dimension, unicode-range, include-match, dash-match, prefix-match, suffix-match, substring-match, whitespace, cdo, cdc, colon, semicolon, [, ], (, ), {, }.

ident, function, at-keyword, hash, string, and url tokens have a value composed of zero or more characters. Delim tokens have a value composed of a single character. Number, percentage, and dimension tokens have a representation composed of 1 or more character, a numeric value, and a type flag set to either "integer" or "number". The type flag defaults to "integer" if not otherwise set. Dimension tokens additionally have a unit composed of one or more characters. Unicode-range tokens have a range of characters.

Several types of tokens that are guaranteed to be language-defined, like at-keywords, functions, or dimension units, are always lowercased when serialized. Should we eagerly lowercase them in the parser, or leave that to serialization to deal with? (I slightly lean toward letting serialization deal with it, so we don't mess with the capitalization of variable values.)

The tokenizer state machine consists of the states defined in the following subsections.

Token Railroad Diagrams

This section is non-normative.

This section presents an informative view of the token parser, in the form of railroad diagrams. Railroad diagrams are more compact than a state-machine, but often easier to read than a regular expression.

These diagrams are informative and incomplete; they describe the grammar of "correct" tokens, but do not describe error-handling at all. They are provided solely to make it easier to get an intuitive grasp of the syntax of each token.

escape

IDENT

FUNCTION

AT-KEYWORD

HASH

STRING

URL

NUMBER

DIMENSION

PERCENTAGE

UNICODE-RANGE

COMMENT

CDO

CDC

Tokenizer Flags

The tokenizer can be run with any of several flags that alter its behavior.

the transform function whitespace flag

This flag is set when parsing SVG's transform attribute. When this is set, whitespace is allowed between the name of a transform function and its opening parenthesis.

Definitions

This section defines several terms used during the tokenization phase.

digit

A character between U+0030 DIGIT ZERO (0) and U+0039 DIGIT NINE (9).

hex digit

A digit, or a character between U+0041 LATIN CAPITAL LETTER A (A) and U+0046 LATIN CAPITAL LETTER F (F), or a character between U+0061 LATIN SMALL LETTER A (a) and U+0066 LATIN SMALL LETTER F (f).

uppercase letter

A character between U+0041 LATIN CAPITAL LETTER A (A) and U+005A LATIN CAPITAL LETTER Z (Z).

lowercase letter

A character between U+0061 LATIN SMALL LETTER A (a) and U+007A LATIN SMALL LETTER Z (z).

letter

An uppercase letter or a lowercase letter.

non-ASCII character

A character with a codepoint equal to or greater than U+0080 <control>.

name-start character

A letter, a non-ASCII character, or U+005F LOW LINE (_).

name character

A name-start character, A digit, or U+002D HYPHEN-MINUS (-).

non-printable character

A character between U+0000 NULL and U+0008 BACKSPACE or a character between U+000E SHIFT OUT and U+001F INFORMATION SEPARATOR ONE or a character between U+007F DELETE and U+009F APPLICATION PROGRAM COMMAND.

newline

U+000A LINE FEED or U+000C FORM FEED. Note that U+000D CARRIAGE RETURN is not included in this definition, as it is removed from the stream during preprocessing.

whitespace

A newline, U+0009 CHARACTER TABULATION, or U+0020 SPACE.

maximum allowed codepoint

The greatest codepoint defined by Unicode. This is currently U+10FFFF.

Tokenizer State Machine

Data state

Consume the next input character.

whitespace

Consume as much whitespace as possible. Emit a whitespace token. Remain in this state.

U+0022 QUOTATION MARK (")

Switch to the double-quote-string state.

U+0023 NUMBER SIGN (#)

Switch to the hash state.

U+0024 DOLLAR SIGN ($)

If the next input character is U+003D EQUALS SIGN (=), consume it and emit a suffix-match token. Remain in this state.

Otherwise, emit a delim token with its value set to the current input character. Remain in this state.

U+0027 APOSTROPHE (')

Switch to the single-quote-string state.

U+0028 LEFT PARENTHESIS (()

Emit a ( token. Remain in this state.

U+0029 RIGHT PARENTHESIS ())

Emit a ) token. Remain in this state.

U+002A ASTERISK (*)

If the next input character is U+003D EQUALS SIGN (=), consume it and emit a substring-match token. Remain in this state.

Otherwise, emit a delim token with its value set to the current input character. Remain in this state.

U+002B PLUS SIGN (+)

If the next input character is a digit, or the next 2 input characters are U+002E FULL STOP (.) followed by a digit, switch to the number state. Reconsume the current input character.

Otherwise, emit a delim token with its value set to U+002B PLUS SIGN (+). Remain in this state.

U+002D HYPHEN-MINUS (-)

If the next 2 input characters are U+002D HYPHEN-MINUS U+003E GREATER-THAN SIGN (->), consume them, emit a CDC token, and remain in this state.

Otherwise, if the next input character is a digit, or the next 2 input characters are U+002E FULL STOP (.) followed by a digit, switch to the number state. Reconsume the current input character.

Otherwise, switch to the ident state. Reconsume the current input character.

U+002E FULL STOP (.)

If the next input character is a digit, switch to the number state. Reconsume the current input character.

Otherwise, emit a delim token with its value set to U+002E FULL STOP (.). Remain in this state.

U+002F SOLIDUS (/)

If the next input character is U+002A ASTERISK (*), consume it and switch to the comment state.

Otherwise, emit a delim token with its value set to U+002F SOLIDUS (/). Remain in this state.

U+003A COLON (:)

Emit a colon token. Remain in this state.

U+003B SEMICOLON (;)

Emit a semicolon token. Remain in this state.

U+003C LESS-THAN SIGN (<)

If the next 3 input characters are U+0021 EXCLAMATION MARK U+002D HYPHEN-MINUS U+002D HYPHEN-MINUS (!--), consume them and emit a cdo token. Remain in this state.

Otherwise, emit a delim token with its value set to U+003C LESS-THAN SIGN (<). Remain in this state.

U+0040 COMMERCIAL AT (@)

Switch to the at-keyword state.

U+005B LEFT SQUARE BRACKET ([)

Emit a [ token. Remain in this state.

U+005C REVERSE SOLIDUS (\)

If the next input character is a newline or EOF, this is a parse error. Emit a delim token with its value set to U+005C REVERSE SOLIDUS (\). Remain in this state.

Otherwise, switch to the ident state. Reconsume the current input character.

U+005D RIGHT SQUARE BRACKET (])

Emit a ] token. Remain in this state.

U+005E CIRCUMFLEX ACCENT (^)

If the next input character is U+003D EQUALS SIGN (=), consume it and emit a prefix-match token. Remain in this state.

Otherwise, emit a delim token with its value set to the current input character. Remain in this state.

U+007B LEFT CURLY BRACKET ({)

Emit a { token. Remain in this state.

U+007D RIGHT CURLY BRACKET (})

Emit a } token. Remain in this state.

digit

Switch to the number state. Reconsume the current input character.

U+0055 LATIN CAPITAL LETTER U (U)

U+0075 LATIN SMALL LETTER U (u)

If the next 2 input character are U+002B PLUS SIGN (+) followed by a hex digit or U+003F QUESTION MARK (?), consume the next input character. Note: don't consume both of them. Switch to the unicode-range state.

Otherwise, switch to the ident state. Reconsume the current input character.

name-start character

Switch to the ident state. Reconsume the current input character.

U+007C VERTICAL LINE (|)

If the next input character is U+003D EQUALS SIGN (=), consume it and emit a dash-match token. Remain in this state.

Otherwise, emit a delim token with its value set to the current input character. Remain in this state.

U+007E TILDE (~)

If the next input character is U+003D EQUALS SIGN (=), consume it and emit an includes-match token. Remain in this state.

Otherwise, emit a delim token with its value set to the current input character. Remain in this state.

EOF

Emit an end-of-file token.

anything else

Emit a delim token with its value set to the current input character. Remain in this state.

Double-quote-string state

If a string token has not yet been created since entering this state, create a string token with its value initially set to the empty string.

Consume the next input character.

U+0022 QUOTATION MARK (")

Emit the string token. Return to the data state.

EOF

This is a parse error. Emit the string token. Switch to the data state. Reconsume the current input character.

newline

This is a parse error. Emit a bad-string token. Switch to the data state. Reconsume the current input character.

U+005C REVERSE SOLIDUS (\)

If the next input character is an EOF, this is a parse error. Emit a bad-string token, then switch to the data state.

Otherwise, if the next input character is a newline, consume it. Remain in this state.

Otherwise, consume an escaped character. Append the returned character to the string token's value. Remain in this state.

anything else

Append the current input character to the string token's value. Remain in this state.

Single-quote-string state

If a string token has not yet been created since entering this state, create a string token with its value initially set to the empty string.

Consume the next input character.

U+0027 APOSTROPHE (')

Emit the string token. Return to the data state.

EOF

This is a parse error. Emit the string token. Switch to the data state. Reconsume the current input character.

newline

This is a parse error. Emit a bad-string token. Switch to the data state. Reconsume the current input character.

U+005C REVERSE SOLIDUS (\)

If the next input character is an EOF, this is a parse error. Emit a bad-string token, then switch to the data state.

Otherwise, if the next input character is a newline, consume it. Remain in this state.

Otherwise, consume an escaped character. Append the returned character to the string token's value. Remain in this state.

anything else

Append the current input character to the string token's value. Remain in this state.

Hash state

Consume the next input character.

name character

Create a hash token with its value set to the current input character. Switch to the hash-rest state.

U+005C REVERSE SOLIDUS (\)

If the next input character is a newline or EOF, emit a delim token with its value set to U+0023 NUMBER SIGN (#). Switch to the data state. Reconsume the current input character.

Otherwise, consume an escaped character. Create a hash token with its value set to the returned character. Switch to the hash-rest state.

anything else

Emit a delim token with its value set to U+0023 NUMBER SIGN (#). Switch to the data state. Reconsume the current input character.

Hash-rest state

Consume the next input character.

name character

Append the current input character to the hash token's value. Remain in this state.

U+005C REVERSE SOLIDUS (\)

If the next input character is a newline or EOF, emit the hash token. Switch to the data state. Reconsume the current input character.

Otherwise, consume an escaped character. Append the returned character to the hash token's value. Remain in this state.

anything else

Emit the hash token. Switch to the data state. Reconsume the current input character.

Comment state

Consume the next input character.

U+002A ASTERISK (*)

If the next input character is U+002F SOLIDUS (/), consume it, and switch to the data state.

Otherwise, do nothing and remain in this state.

EOF

This is a parse error. Switch to the data state. Reconsume the current input character.

anything else

Do nothing and remain in this state.

At-keyword state

Consume the next input character.

U+002D HYPHEN-MINUS (-)

If the next input character is a name-start character, Create an at-keyword token with its value initially set to U+002D HYPHEN-MINUS. Switch to the at-keyword-rest state.

Otherwise, if the next two characters are U+005C REVERSE SOLIDUS (\) followed by anything but a newline or EOF, create an at-keyword token with its value initially set to U+002D HYPHEN-MINUS. Switch to the at-keyword-rest state.

Otherwise, emit a delim token with its value set to U+0040 COMMERCIAL AT (@). Switch to the data state. Reconsume the current input character.

name-start character

Create an at-keyword token with its value set to the current input character. Switch to the at-keyword-rest state.

U+005C REVERSE SOLIDUS (\)

If the next input character is a newline or EOF, emit a delim token with its value set to U+0040 COMMERCIAL AT (@). Switch to the data state. Reconsume the current input character.

Otherwise, consume an escaped character. Create an at-keyword token with its value set to the returned character. Switch to the at-keyword-rest state.

anything else

Emit a delim token with its value set to U+0040 COMMERCIAL AT (@). Switch to the data state. Reconsume the current input character.

At-keyword-rest state

Consume the next input character.

name character

Append the current input character to the at-keyword token's value. Remain in this state.

U+005C REVERSE SOLIDUS (\)

If the next input character is a newline or EOF, emit the at-keyword token. Switch to the data state. Reconsume the current input character.

Otherwise, consume an escaped character. Append the returned character to the at-keyword token's value. Remain in this state.

anything else

Emit the at-keyword token. Switch to the data state. Reconsume the current input character.

Ident state

Consume the next input character.

U+002D HYPHEN-MINUS (-)

If the next input character is a name-start character, create an identifer token with its value initially set to U+002D HYPHEN-MINUS. Switch to the ident-rest state.

Otherwise, if the next two characters are U+005C REVERSE SOLIDUS (\) followed by anything but a newline or EOF, create an identifer token with its value initially set to U+002D HYPHEN-MINUS. Switch to the ident-rest state.

Otherwise, emit a delim token with its value set to U+002D HYPHEN-MINUS (-). Switch to the data state.

name-start character

Create an ident token with its value set to the current input character. Switch to the ident-rest state.

U+005C REVERSE SOLIDUS (\)

If the next input character is a newline or EOF, switch to the data state. Reconsume the current input character.

Otherwise, consume an escaped character. Create an ident token with its value set to the returned character. Switch to the ident-rest state.

Ident-rest state

Consume the next input character.

name character

Append the current input character to the ident token's value. Remain in this state.

U+005C REVERSE SOLIDUS (\)

If the next input character is a newline or EOF, emit the ident token. Switch to the data state. Reconsume the current input character.

Otherwise, consume an escaped character. Append the returned character to the ident token's value. Remain in this state.

U+0028 LEFT PARENTHESIS (()

If the identifier token's value is an ASCII case-insensitive match for "url", switch to the url state.

Otherwise, Emit a function token with its value set to the identifier token's value. Switch to the data state.

whitespace

If the transform function whitespace flag is set, reconsume the current input character and switch to the transform-function-whitespace state.

Otherwise, emit the ident token. Switch to the data state. Reconsume the current input character.

anything else

Emit the ident token. Switch to the data state. Reconsume the current input character.

Transform-function-whitespace state

Consume the next input character.

whitespace

If the next input character is whitespace, remain in this state.

Otherwise, if the next input character is U+0028 LEFT PARENTHESES ((), emit a function token with its value set to the identifer token's value. Switch to the data state.

Otherwise, emit the identifer token. Switch to the data state. Reconsume the current input character.

Number state

Create a number token with its representation initially set to the empty string.

Consume the next input character.

U+002D HYPHEN-MINUS (-)

If the next input character is a digit, consume it. Append U+002D HYPHEN-MINUS (-) to the number token's representation. Append the digit to the number token's representation. Switch to the number-rest state.

Otherwise, if the next 2 input characters are U+002E FULL STOP followed by a digit, consume them. Append U+002D HYPHEN-MINUS (-) to the number token's representation. Append U+002E FULL STOP (.) to the number token's representation. Append the digit to the number token's representation. Switch to the number fraction state.

Otherwise, switch to the data state. Reconsume the current input character.

U+002B PLUS SIGN (+)

If the next input character is a digit, consume it. Append U+002D PLUS SIGN (+) to the number token's representation. Append the digit to the number token's representation. Switch to the number-rest state.

Otherwise, if the next 2 input characters are U+002E FULL STOP followed by a digit, consume them. Append U+002D PLUS SIGN (+) to the number token's representation. Append U+002E FULL STOP (.) to the number token's representation. Append the digit to the number token's representation. Switch to the number fraction state.

Otherwise, switch to the data state. Reconsume the current input character.

digit

Append the current input character to the number token's representation. Switch to the number-rest state.

U+002E FULL STOP (.)

If the next input character is a digit, consume it. Append U+002E FULL STOP (.) to the number token's representation. Append the digit to the number token's representation. Switch to the number-fraction state.

Otherwise, switch to the data state. Reconsume the current input character.

anything else

Switch to the data state. Reconsume the current input character.

Number-rest state

Consume the next input character.

digit

Append the current input character to the number token's representation. Remain in this state.

U+002E FULL STOP (.)

If the next input character is a digit, consume it. Append U+002E FULL STOP (.) followed by the digit to the number token's representation. Switch to the number-fraction state.

Otherwise, set the number token's value to the number produced by interpreting the number token's representation as a base-10 number and emit it. Switch to the data state. Reconsume the current input character.

U+0025 PERCENT SIGN (%)

Emit a percentage token with its value set to the number produced by interpreting the number token's representation as a base-10 number. Switch to the data state.

U+0045 LATIN CAPITAL LETTER E (E)

U+0065 LATIN SMALL LETTER E (e)

If the next input character is a digit, or the next 2 input characters are U+002B PLUS SIGN (+) or U+002D HYPHEN-MINUS (-) followed by a digit, consume them and append them to the number token's representation. Switch to the sci-notation state.

Otherwise, create a dimension token with its representation set to the number token's representation, its value set to the number produced by interpreting the number token's representation as a base-10 number, and a unit initially set to the current input character. Switch to the dimension state.

U+002D HYPHEN-MINUS (-)

If the next input character is a name-start character, consume it. Create a dimension token with its representation set to the number token's representation, its value set to the number produced by interpreting the number token's representation as a base-10 number, and a unit initially set to U+002D HYPHEN-MINUS followed by the name-start character. Switch to the dimension state.

Otherwise, if the next 2 input characters are U+005C REVERSE SOLIDUS (\) followed by a newline, or U+005C REVERSE SOLIDUS (\) followed by EOF, set the number token's value to the number produced by interpreting the number token's representation as a base-10 number and emit it. Switch to the data state. Reconsume the current input character.

Otherwise, if the next input character is U+005C REVERSE SOLIDUS (\), consume it, then consume an escaped character. Create a dimension token with its representation set to the number token's representation, its value set to the number produced by interpreting the number token's representation as a base-10 number, and a unit initially set to U+002D HYPHEN-MINUS followed by the returned character. Switch to the dimension state.

Otherwise, set the number token's value to the number produced by interpreting the number token's representation as a base-10 number and emit it. Switch to the data state. Reconsume the current input character.

name-start character

Create a dimension token with its representation set to the number token's representation, its value set to the number produced by interpreting the number token's representation as a base-10 number, and a unit initially set to the current input character. Switch to the dimension state.

U+005C REVERSE SOLIDUS (\)

If the next input character is a newline or EOF, set the number token's value to the number produced by interpreting the number token's representation as a base-10 number and emit it. Switch to the data state. Reconsume the current input character.

Otherwise, consume an escaped character. Create a dimension token with its representation set to the number token's representation, its value set to the number produced by interpreting the number token's representation as a base-10 number, and a unit initially set to the returned character. Switch to the dimension state.

anything else

Emit a number token with its value set to the number produced by interpreting the string token's value as a base-10 number. Switch to the data state. Reconsume the current input character.

Number-fraction state

Set the number token's type flag to "number".

Consume the next input character.

digit

Append the current input character to the string token's value. Remain in this state.

U+0025 PERCENT SIGN (%)

Emit a percentage token with its value set to the number produced by interpreting the number token's representation as a base-10 number. Switch to the data state.

U+0045 LATIN CAPITAL LETTER E (E)

U+0065 LATIN SMALL LETTER E (e)

If the next input character is a digit, or the next 2 input characters are U+002B PLUS SIGN (+) or U+002D HYPHEN-MINUS (-) followed by a digit, consume them and append them to the number token's representation. Switch to the sci-notation state.

Otherwise, create a dimension token with its representation set to the number token's representation, its value set to the number produced by interpreting the number token's representation as a base-10 number, and a unit initially set to the current input character. Switch to the dimension state.

U+002D HYPHEN-MINUS (-)

If the next input character is a name-start character, consume it. Create a dimension token with its representation set to the number token's representation, its value set to the number produced by interpreting the number token's representation as a base-10 number, and a unit initially set to U+002D HYPHEN-MINUS followed by the name-start character. Switch to the dimension state.

Otherwise, if the next 2 input characters are U+005C REVERSE SOLIDUS (\) followed by a newline, or U+005C REVERSE SOLIDUS (\) followed by EOF, set the number token's value to the number produced by interpreting the number token's representation as a base-10 number and emit it. Switch to the data state. Reconsume the current input character.

Otherwise, if the next input character is U+005C REVERSE SOLIDUS (\), consume it, then consume an escaped character. Create a dimension token with its representation set to the number token's representation, its value set to the number produced by interpreting the number token's representation as a base-10 number, and a unit initially set to U+002D HYPHEN-MINUS followed by the returned character. Switch to the dimension state.

Otherwise, set the number token's value to the number produced by interpreting the number token's representation as a base-10 number and emit it. Switch to the data state. Reconsume the current input character.

name-start character

Create a dimension token with its representation set to the number token's representation, its value set to the number produced by interpreting the number token's representation as a base-10 number, and a unit initially set to the current input character. Switch to the dimension state.

U+005C REVERSE SOLIDUS (\)

If the next input character is a newline or EOF, set the number token's value to the number produced by interpreting the number token's representation as a base-10 number and emit it. Switch to the data state. Reconsume the current input character.

Otherwise, consume an escaped character. Create a dimension token with its representation set to the number token's representation, its value set to the number produced by interpreting the number token's representation as a base-10 number, and a unit initially set to the returned character. Switch to the dimension state.

anything else

Emit a number token with its value set to the number produced by interpreting the string token's value as a base-10 number. Switch to the data state. Reconsume the current input character.

Dimension state

Consume the next input character.

name character

Append the current input character to the dimension token's unit. Remain in this state.

U+005C REVERSE SOLIDUS (\)

If the next input character is a newline or EOF, emit the dimension token. Switch to the data state. Reconsume the current input character.

Otherwise, consume an escaped character. Append the returned character to the dimension token's unit.

anything else

Emit the dimension token. Switch to the data state. Reconsume the current input character.

Sci-notation state

Set the number token's type flag to "number".

Consume the next input character.

digit

Append the current input character to the number token's representation. Remain in this state.

anything else

Let base be the result of interpreting the portion of the number token's representation preceding the U+0045 LATIN CAPITAL LETTER E (E) or U+0065 LATIN SMALL LETTER E (e) as a base-10 number.

Let power be the result of interpreting the portion of the number token's representation following the U+0045 LATIN CAPITAL LETTER E (E) or U+0065 LATIN SMALL LETTER E (e) as a base-10 number.

Set the number token's value to base * 10power. If the number token's value is not an integer, set the number token's type flag to "number". Emit the number token. Switch to the data state. Reconsume the current input character.

URL state

Consume the next input character.

EOF

This is a parse error. Emit a bad-url token. Switch to the data state.

U+0022 QUOTATION MARK (")

Switch to the url-double-quote state.

U+0027 APOSTROPHE (')

Switch to the url-single-quote state.

U+0029 RIGHT PARENTHESIS ())

Emit a url token with its value set to the empty string. Switch to the data state.

whitespace

Remain in this state.

anything else

Switch to the url-unquoted state. Reconsume the current input character.

URL-double-quote state

If a url token has not yet been created since entering this state, create a url token with its value initially set to the empty string.

Consume the next input character.

EOF

This is a parse error. Emit a bad-url token. Switch to the data state.

U+0022 QUOTATION MARK (")

Switch to the url-end state.

newline

This is a parse error. Switch to the bad-url state.

U+005C REVERSE SOLIDUS (\)

If the next input character is EOF, this is a parse error. Emit a bad-url token. Switch to the data state. Reconsume the current input character.

Otherwise, if the next input character is a newline, consume it and remain in this state.

Otherwise, consume an escaped character. Append the returned character to the url token's value. Remain in this state.

anything else

Append the current input character to the url token's value. Remain in this state.

URL-single-quote state

If a url token has not yet been created since entering this state, create a url token with its value initially set to the empty string.

Consume the next input character.

EOF

This is a parse error. Emit a bad-url token. Switch to the data state.

U+0027 APOSTROPHE (')

Switch to the url-end state.

newline

This is a parse error. Switch to the bad-url state.

U+005C REVERSE SOLIDUS (\)

If the next input character is EOF, this is a parse error. Emit a bad-url token. Switch to the data state. Reconsume the current input character.

Otherwise, if the next input character is a newline, consume it and remain in this state.

Otherwise, consume an escaped character. Append the returned character to the url token's value. Remain in this state.

anything else

Append the current input character to the url token's value. Remain in this state.

URL-end state

Consume the next input character.

EOF

This is a parse error. Emit a bad-url token. Switch to the data state.

whitespace

Remain in this state.

U+0029 RIGHT PARENTHESIS ())

Emit the url token. Switch to the data state.

anything else

This is a parse error. Switch to the bad-url state. Reconsume the current input character.

URL-unquoted state

If a url token has not yet been created since entering this state, create a url token with its value initially set to the empty string.

Consume the next input character.

EOF

This is a parse error. Emit a bad-url token. Switch to the data state.

whitespace

Switch to the url-end state.

U+0029 RIGHT PARENTHESIS ())

Emit the url token. Switch to the data state.

U+0022 QUOTATION MARK (")

U+0027 APOSTROPHE (')

U+0028 LEFT PARENTHESIS (()

non-printable character

This is a parse error. Switch to the bad-url state.

U+005C REVERSE SOLIDUS

If the next input character is a newline or EOF, this is a parse error. Switch to the bad-url state.

Otherwise, consume an escaped character. Append the returned character to the url token's value. Remain in this state.

anything else

Append the current input character to the url token's value. Remain in this state.

Bad-URL state

Consume the next input character.

EOF

This is a parse error. Emit a bad-url token. Switch to the data state.

U+0029 RIGHT PARENTHESIS ())

Emit a bad-url token. Switch to the data state.

U+005C REVERSE SOLIDUS

If the next input character is a newline or EOF, do nothing and remain in this state.

Otherwise, consume an escaped character. Remain in this state.

anything else

Do nothing. Remain in this state.

Unicode-range state

Create a new unicode-range token with an empty range.

Consume as many hex digits as possible, but no more than 6. If less than 6 hex digits were consumed, consume as many U+003F QUESTION MARK (?) characters as possible, but no more than enough to make the total of hex digits and U+003F QUESTION MARK (?) characters equal to 6.

If any U+003F QUESTION MARK (?) characters were consumed, first interpret the consumed characters as a hexadecimal number, with the U+003F QUESTION MARK (?) characters replaced by U+0030 DIGIT ZERO (0) characters. This is the start of the range. Then interpret the consumed characters as a hexadecimal number again, with the U+003F QUESTION MARK (?) character replaced by U+0046 LATIN CAPITAL LETTER F (F) characters. This is the end of the range. Set the unicode-range token's range, then emit it. Switch to the data state.

Otherwise, interpret the digits as a hexadecimal number. This is the start of the range.

Consume the next input character.

U+002D HYPHEN-MINUS (-)

If the next input character is a hex digit, consume as many hex digits as possible, but no more than 6. Interpret the digits as a hexadecimal number. This is the end of the range. Set the unicode-range token's range, then emit it. Switch to the data state.

Otherwise, set the unicode-range token's range and emit it. Switch to the data state. Reconsume the current input character.

anything else

Set the unicode-range token's range and emit it. Switch to the data state. Reconsume the current input character.

Consume an escaped character

This section describes how to consume an escaped character. It assumes that the U+005C REVERSE SOLIDUS (\) has already been consumed and that the next input character has already been verified to not be a newline or EOF. It will return a character.

Consume the next input character.

hex digit

Consume as many hex digits as possible, but no more than 5. Note that this means 1-6 hex digits have been consumed in total. If the next input character is whitespace, consume it as well. Interpret the hex digits as a hexadecimal number. If this number is zero, or is greater than the maximum allowed codepoint, return U+FFFD REPLACEMENT CHARACTER (�). Otherwise, return the character with that codepoint.

anything else

Return the current input character.

Set the unicode-range token's range

This section describes how to set a unicode-range token's range so that the range it describes is within the supported range of unicode characters.

It assumes that the start of the range has been defined, the end of the range might be defined, and both are non-negative integers.

If the start of the range is greater than the maximum allowed codepoint, the unicode-range token's range is empty.

If the end of the range is defined, and it is less than the start of the range, the unicode-range token's range is empty.

If the end of the range is not defined, the unicode-range token's range is the single character whose codepoint is the start of the range.

Otherwise, if the end of the range is greater than the maximum allowed codepoint, change it to the maximum allowed codepoint. The unicode-range token's range is all characters between the character whose codepoint is the start of the range and the character whose codepoint is the end of the range.

Changes from CSS 2.1 Tokenizer

This section is non-normative.

Here are the changes that I know that I've made to the tokenizer of 2.1. If there are changes not present here, please let me know as they may be unintentional.

Note that the point of this spec is to match reality; changes from CSS2.1's tokenizer are nearly always because the tokenizer specified something that doesn't match actual browser behavior, or left something unspecified. If some detail doesn't match browsers, please let me know as it's almost certainly unintentional.

1. The DASHMATCH and INCLUDES tokens have been removed. They can instead be handled simply by having them parse as DELIM tokens. It was weird to privilege just those two types of attribute equality operators, when Selectors 3 adds several more.
2. The BAD-URI token (now bad-url) is "self-contained". In other words, once the tokenizer realizes it's in a bad-url rather than a url token, it just seeks forward to look for the closing ), ignoring everything else. This behavior is simpler than treating it like a FUNCTION token and paying attention to opened blocks and such. Only WebKit exhibits this behavior, but it doesn't appear that we've gotten any compat bugs from it.
3. The number, percentage, and dimension tokens have been changed to include the preceding +/- sign as part of their value (rather than as a separate DELIM token that needs to be manually handled every time the token is mentioned in other specs). The only consequence of this is that comments can no longer be inserted between the sign and the number.
4. Some flags have been added for SVG-compatible tokenizing, so that a single state machine can be used for both "vanilla" and SVG CSS parsing.
5. Scientific notation is supported for numbers, per WG resolution.

Parsing

The input to the tree construction stage is a sequence of tokens from the tokenization stage. The output is a tree of items with a stylesheet at the root and all other nodes being at-rules, style rules, or declarations.

The construction of this stylesheet does not take into account unrecognized items, like unknown properties or at-rules. It simply produces a tree of rules according to the fundamental syntax of CSS. Unrecognized or misplaced rules are handled in the CSSOM Construction stage.

The items that can appear in the tree are a mixture of basic tokens and new objects:

stylesheet

A stylesheet has a value consisting of a list of at-rules and style rules.

at-rule

An at-rule has a name, a prelude consisting of a list of component values, and a value consisting of a list of at-rules, style rules, or declarations.

style rule

A style rule has a selector consisting of a list of component values, and a value consisting of a list of at-rules or declarations.

declaration

A declaration has a name, a value consisting of a list of component values, and an important flag which is initially unset.

component value

A component value is one of the preserved tokens, a function, or a simple block.

preserved tokens

The tokens that can appear in the tree are: ident, hash, string, url, delim, number, percentage, dimension, unicode-range, whitespace, colon, semicolon, at-keyword, include-match, dash-match, prefix-match, suffix-match, and substring-match.

This means that the following tokens emitted by the tokenizer stage will not appear in the stylesheet object: function, bad-string, bad-url, cdo, cdc, [, ], (, ), {, and }.

function

A function has a name, and a list of arguments. Each argument is a list of component values.

simple block

A simple block has an associated token (either a [, (, or { token) and a value consisting of a list of component values.

When the parser is invoked, it must be provided with a list of recognized at-rule names, where each name is additionally associated with whether the at-rule is rule-filled, declaration-filled, or neither.

Parser Flags

the quirks mode flag

This flag is set when the host document is in quirks mode. When this is set, the parser allows hash colors without the # character and pixel lengths without the 'px' unit in some circumstances.

the only declarations flag

This flag is set when the parser is invoked to consume just a set of declarations, not any rules. It causes the } token to be a syntax error, rather than have special rule-closing behavior.

Definitions

current input token

The token currently being operated on, from the list of tokens produced by the tokenizer.

next input token

The token following the current input token in the list of tokens produced by the tokenizer. If there isn't a token following the current input token, the next input token is an EOF token.

reconsume the current input token

Push the current input token back onto the list of tokens produced by the tokenizer, so that the next time a mode instructs you to consume the next input token, it will instead reconsume the current input token.

stack of open rules

A stack of rules. This stack always contains the stylesheet object at the bottom of the stack, gains additional rules as block are opened, and loses them when blocks are closed.

top-level rule

A rule is a top-level rule if the stack of open rules contains only it and the stylesheet object. Otherwise, it is not a top-level rule.

current rule

The topmost rule in the stack of open rules.

current declaration

The declaration currently being built. There may or may not be a current declaration at any given time, but there is never more than one.

rule-filled

An at-rule is rule-filled if it is in the list of recognized at-rules and marked as such. This type of at-rule may have a prelude, and finishes with a block which contains at-rules and/or style rules. Examples of rule-filled at-rules are ''@media'' and ''@keyframes''.

declaration-filled

An at-rule is declaration-filled if it is in the list of recognized at-rules and marked as such. This type of at-rule may have a prelude, and finishes with a block which contains at-rules and/or declarations. Examples of declaration-filled at-rules are ''@font-face'' and ''@page''.

ASCII case-insensitive

When two strings are to be matched ASCII case-insensitively, temporarily convert both of them to ASCII lower-case form by adding 32 (0x20) to the value of each codepoint between U+0041 LATIN CAPITAL LETTER A (A) and U+005A LATIN CAPITAL LETTER Z (Z), inclusive, and then compare them on a codepoint-by-codepoint basis.

Parser State Machine

Top-level mode

Consume the next input token.

cdo token

cdc token

whitespace token

Do nothing. Remain in this mode.

at-keyword token

Create an at-rule with its name set to the value of the current input token and push it onto the stack of open rules. Switch to the at-rule-prelude mode.

EOF token

Finish parsing.

anything else

Create a style rule and push it onto the stack of open rules. Switch to the selector mode. Reconsume the current input token.

At-rule-prelude mode

Consume the next input token.

semicolon token

If the current rule is declaration-filled or rule-filled, this is a parse error. Discard the current rule. Switch to the current rule's content mode.

Otherwise, pop the current rule. Switch to the current rule's content mode.

{ token

If the current rule is rule-filled, switch to the rule-block mode.

Otherwise, if the current rule is declaration-filled, switch to the declaration-block mode.

Otherwise, this is a parse error. Discard the current rule. Consume a component value and ignore the return value. Switch to the current rule's content mode.

} token

If the current rule is a top-level rule, this is a parse error. Discard the current rule. Switch to the next-block-or-statement error mode. Reconsume the current input token.

Otherwise, if the current rule is declaration-filled or rule-filled, this is a parse error. Discard the current rule. Switch to the current rule's content mode. Reconsume the current input token.

Otherwise, pop the current rule. Switch to the current rule's content mode. Reconsume the current input token.

EOF token

If the current rule is declaration-filled or rule-filled, this is a parse error. Discard the current rule. Finish parsing.

Otherwise, finish parsing.

anything else

Consume a component value. If anything was returned, append the returned value to the prelude of the current rule. Remain in this mode.

Otherwise, this is a parse error. Discard the current rule. If the current rule is declaration-filled or rule-filled, switch to the next-block error mode; otherwise, switch to the next-block-or-statement error mode. Reconsume the current input token.

Rule-block mode

Consume the next input token.

whitespace token

Do nothing. Remain in this mode.

} token

Pop the current rule. Switch to the current rule's content mode.

at-keyword token

Create an at-rule with its name set to the value of the current input token and push it onto the stack of open rules. Switch to the at-rule-prelude mode.

EOF token

Finish parsing.

anything else

Create a style rule and push it onto the stack of open rules. Switch to the selector mode. Reconsume the current input token.

Selector mode

Consume the next input token.

{ token

Switch to the declaration-block mode.

} token

If the current rule is a top-level rule, this is a parse error. Discard the current rule. Switch to the next-block error mode. Reconsume the current input token.

Otherwise, this is a parse error. Discard the current rule. Switch to the current rule's content mode. Reconsume the current input token.

EOF token

This is a parse error. Discard the current rule. Finish parsing.

anything else

Consume a component value. If anything was returned, append the returned value to the selector of the current rule. Remain in this mode.

Otherwise, this is a parse error. Discard the current rule. Switch to the next-block error mode. Reconsume the current input token.

Declaration-block mode

Consume the next input token.

whitespace token

semicolon token

Do nothing. Remain in this mode.

} token

If the only declarations flag is set and the current rule is a top-level rule, this is a parse error. Unset the current declaration. Switch to the next-declaration error mode. Reconsume the current input token.

Otherwise, pop the current rule. Switch to the current rule's content mode.

at-keyword token

Create an at-rule with its name set to the value of the current input token and push it onto the stack of open rules. Switch to the at-rule-prelude mode.

ident token

Create a declaration with its name set to the value of the current input token. Set it as the current declaration. Switch to the after-declaration-name mode.

EOF token

Finish parsing.

anything else

This is a parse error. Unset the current declaration. Switch to the next-declaration error mode. Reconsume the current input token.

After-declaration-name mode

Consume the next input token.

whitespace

Do nothing. Remain in this mode.

colon token

Do nothing. Switch to the declaration-value mode.

semicolon token

This is a parse error. Unset the current declaration. Switch to the current rule's content mode.

EOF token

Unset the current declaration. Finish parsing.

anything else

This is a parse error. Unset the current declaration. Switch to the next-declaration error mode. Reconsume the current input token.

Declaration-value mode

Consume the next input token.

delim token with a value of "!"

Switch to the declaration-important mode.

semicolon token

If the current declaration is grammatically valid, append it to the value of the current rule. Unset the current declaration. Switch to the current rule's content mode.

} token

If the only declarations flag is set and the current rule is a top-level rule, this is a parse error. Unset the current declaration. Switch to the next-declaration error mode. Reconsume the current input token.

Otherwise, if the current declaration is grammatically valid, append it to the value of the current rule. Unset the current declaration. Pop the current rule. Switch to the current rule's content mode.

EOF token

Finish parsing.

anything else

If the quirks mode flag is set, and the current declaration's name is an ASCII case-insensitive match for one of the property names on the hashless color quirk list, consume a component value with the hashless color quirk. If anything was returned, append the returned value to the value of the current declaration and remain in this mode; otherwise, unset the current declaration and switch to the next-declaration error mode. Reconsume the current input token.

Otherwise, if the quirks mode flag is set, and the current declaration's name is an ASCII case-insensitive match for one of the property names on the unitless length quirk list, consume a component value with the unitless length quirk. If anything was returned, append the returned value to the value of the current declaration and remain in this mode; otherwise, unset the current declaration and switch to the next-declaration error mode. Reconsume the current input token.

Otherwise, consume a component value. If anything was returned, append the returned value to the value of the current declaration and remain in this mode; otherwise, unset the current declaration and switch to the next-declaration error mode. Reconsume the current input token.

1. Initialize the stack of open rules by pushing a new stylesheet onto it.
2. Invoke the parser.
3. Return the stylesheet it returns.

1. Initialize the stack of open rules by pushing a new stylesheet onto it.
2. Invoke the parser.
3. If the returned stylesheet contains more than one rule in its value, return a syntax error.

   Otherwise, return the first rule in the value of the stylesheet it returns.

1. Initialize the stack of open rules by pushing a new stylesheet onto it, and then pushing a new style rule onto it with a selector set to the empty string and an initially empty value.
2. Invoke the parser, with the only declarations flag set, starting in the declaration-block mode.
3. Return the value of the style rule in the value of the stylesheet it returns.

Parse a component value

To parse a component value:

1. Discard whitespace tokens from the token stream until a non-whitespace token is reached. If the token stream is exhausted without finding a non-whitespace token, return a syntax error.
2. Consume a component value. If nothing is returned, return a syntax error.
3. Discard whitespace tokens from the token stream until a non-whitespace token is reached. If the token stream is exhausted without finding a non-whitespace token, return the value found in the previous step. Otherwise, return a syntax error.

This algorithm is token/component-value agnostic, which means it can be called with either a stream of tokens or a stream of component values with identical results.

Parse a list of component values

To parse a list of component values:

1. Repeatedly consume a component value until the token stream is exhausted, appending the returned values into a list. If at any point nothing is returned, return a syntax error.
2. Return the list.

This algorithm is token/component-value agnostic.

Parse a comma-separated list of component values

To parse a comma-separated list of component values:

1. Initialize val to an empty list, and temp to an empty list.
2. Repeatedly consume a component value, appending the returned values to temp, until either a comma token is returned, or the token stream is exhausted. If at any point nothing is returned, return a syntax error.
3. If a comma token is encountered, do not append it to temp. Instead, append temp to val. Create a new temp, and return to step 2.
4. If the token stream is exhausted, append temp to val, and return val.

This algorithm is token/component-value agnostic.

Parse an+b notation

To parse an+b notation:

Initialize an integer called step to 0, an integer called offset to 0, and a string called repr to the empty string.

Repeatedly consume the next input token and process it as follows:

ident token

number token with its type flag set to "integer"

dimension token

Append the token's value, representation, or representation and unit, respectively, to repr.

delim token with the value "+"

Append U+002B PLUS SIGN (+) to repr.

delim token with the value "-"

Append U+002D HYPHEN-MINUS (-) to repr.

whitespace token

Append U+0020 SPACE ( ) to repr.

EOF

Continue to the next step of the algorithm.

anything else

This is a parse error. Return a syntax error.

Trim U+0020 SPACE ( ) characters from the front and back of repr, then process it as follows:

1. If repr is an ASCII case-insensitive match for the string "odd", set step to 2 and offset to 1.
2. Otherwise, if repr is an ASCII case-insensitive match for the string "even", set step to 2 and offset to 0.
3. Otherwise, if repr consists solely of digits, optionally prefixed with a single U+002B PLUS SIGN (+) or U+002D HYPHEN-MINUS (-), interpret repr as a base-10 number, and set step to the result.
4. Otherwise, if repr contains U+004E LATIN CAPITAL LETTER N (N), or U+006E LATIN SMALL LETTER N (n), split repr into two substrings composed respectively of the characters preceding and following the first such letter.

   Interpret the first string as follows:

   • If the first string is empty, set step to 1.
   • Otherwise, if the first string consists solely of a single U+002B PLUS SIGN (+) or U+002D HYPHEN-MINUS (-), set step to 1 or -1, respectively.
   • Otherwise, if the first string consists solely of digits, optionally prefixed with a single U+002B PLUS SIGN (+) or U+002D HYPHEN-MINUS (-), interpret the first string as a base-10 number, and set step to the result.
   • Otherwise, this is a parse error; return a syntax error.

   Interpret the second string as follows:

   • If the second string is empty, set offset to 0.
   • Otherwise, if the second string consists solely of 0 or more U+0020 SPACE characters, optionally followed by a single U+002B PLUS SIGN (+) or U+002D HYPHEN-MINUS (-) character, followed by 0 or more U+0020 SPACE characters, followed by 1 or more digits, interpret the digits as a base-10 number. If there was a U+002D HYPHEN-MINUS (-) character, negate the result. Set offset to the result.
   • Otherwise, this is a parse error; return a syntax error.
5. Otherwise, this is a parse error; return a syntax error.

Return step and offset.

This algorithm is token/component-value agnostic.

Serialization

This specification does not define how to serialize CSS in general, leaving that task to the CSSOM and individual feature specifications. However, there is one important facet that must be specified here regarding comments, to ensure accurate "round-tripping" of data from text to CSS objects and back.

The tokenizer described in this specification does not produce tokens for comments, or otherwise preserve them in any way. Implementations may preserve the contents of comments and their location in the token stream. If they do, this preserved information must have no effect on the parsing step, but must be serialized in its position as "/*" followed by its contents followed by "*/".

If the implementation does not preserve comments, it must insert the text "/**/" between the serialization of adjacent tokens when the two tokens are of the following pairs:

• hash or at-keyword token followed by a number, percentage, ident, dimension, unicode-range, url, or a function token;
• number, ident, and dimension tokens in any combination;
• number, ident, or dimension token followed by a percentage, unicode-range, url, or function token;
• ident token followed by a ( token;
• a delim token containing "#" or "@" followed by any token except whitespace;
• a delim token containing "-", "+", ".", "<", ">", or "!" following or followed by any token except whitespace;
• a delim token containing "/" following or followed by a delim token containing "*".

The preceding pairs of tokens can only be adjacent due to comments in the original text, so the above rule reinserts the minimum number of comments into the serialized text to ensure an accurate round-trip. (Roughly. The delim token rules are slightly too powerful, for simplicity.)

Conformance

Document conventions

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.

Conformance classes

Conformance to CSS Syntax Module Level 3 is defined for three conformance classes:

style sheet

A CSS style sheet.

renderer

A UA that interprets the semantics of a style sheet and renders documents that use them.

authoring tool

A UA that writes a style sheet.

A style sheet is conformant to CSS Syntax Module Level 3 if it is syntactically valid according to this module.

A renderer is conformant to CSS Syntax Module Level 3 if it parses a stylesheet according to this module.

An authoring tool is conformant to CSS Syntax Module Level 3 if it writes style sheets that are syntactically valid according to this module.

Acknowledgments

Thanks for feedback and contributions from David Baron, 呂康豪 (Kang-Hao Lu), and Simon Sapin.

References

Normative references

Other references

Index

Property index