diff options
Diffstat (limited to 'src/construct')
| -rw-r--r-- | src/construct/code_fenced.rs | 4 | ||||
| -rw-r--r-- | src/construct/code_indented.rs | 4 | ||||
| -rw-r--r-- | src/construct/code_text.rs | 213 | ||||
| -rw-r--r-- | src/construct/mod.rs | 6 | ||||
| -rw-r--r-- | src/construct/paragraph.rs | 4 | ||||
| -rw-r--r-- | src/construct/partial_label.rs | 4 | ||||
| -rw-r--r-- | src/construct/raw_text.rs | 270 | ||||
| -rw-r--r-- | src/construct/text.rs | 35 |
8 files changed, 301 insertions, 239 deletions
diff --git a/src/construct/code_fenced.rs b/src/construct/code_fenced.rs index d117006..edb2d93 100644 --- a/src/construct/code_fenced.rs +++ b/src/construct/code_fenced.rs @@ -48,7 +48,7 @@ //! Which value it holds depends on what your syntax highlighter supports, if //! one is used. //! -//! In markdown, it is also possible to use [code (text)][code_text] in the +//! In markdown, it is also possible to use [code (text)][raw_text] in the //! [text][] content type. //! It is also possible to create code with the //! [code (indented)][code_indented] construct. @@ -106,7 +106,7 @@ //! [character_escape]: crate::construct::character_escape //! [character_reference]: crate::construct::character_reference //! [code_indented]: crate::construct::code_indented -//! [code_text]: crate::construct::code_text +//! [raw_text]: crate::construct::raw_text //! [html_code]: https://html.spec.whatwg.org/multipage/text-level-semantics.html#the-code-element //! [html_pre]: https://html.spec.whatwg.org/multipage/grouping-content.html#the-pre-element diff --git a/src/construct/code_indented.rs b/src/construct/code_indented.rs index 7d279c1..5fc9446 100644 --- a/src/construct/code_indented.rs +++ b/src/construct/code_indented.rs @@ -16,7 +16,7 @@ //! As this construct occurs in flow, like all flow constructs, it must be //! followed by an eol (line ending) or eof (end of file). //! -//! In markdown, it is also possible to use [code (text)][code_text] in the +//! In markdown, it is also possible to use [code (text)][raw_text] in the //! [text][] content type. //! It is also possible to create code with the [code (fenced)][code_fenced] //! construct. @@ -49,7 +49,7 @@ //! [flow]: crate::construct::flow //! [text]: crate::construct::text //! [code_fenced]: crate::construct::code_fenced -//! [code_text]: crate::construct::code_text +//! [raw_text]: crate::construct::raw_text //! [html_code]: https://html.spec.whatwg.org/multipage/text-level-semantics.html#the-code-element //! [html_pre]: https://html.spec.whatwg.org/multipage/grouping-content.html#the-pre-element diff --git a/src/construct/code_text.rs b/src/construct/code_text.rs deleted file mode 100644 index b2cfd17..0000000 --- a/src/construct/code_text.rs +++ /dev/null @@ -1,213 +0,0 @@ -//! Code (text) occurs in the [text][] content type. -//! -//! ## Grammar -//! -//! Code (text) forms with the following BNF -//! (<small>see [construct][crate::construct] for character groups</small>): -//! -//! ```bnf -//! ; Restriction: the number of markers in the closing sequence must be equal -//! ; to the number of markers in the opening sequence. -//! code_text ::= sequence 1*byte sequence -//! -//! ; Restriction: not preceded or followed by `` ` ``. -//! sequence ::= 1*'`' -//! ``` -//! -//! The above grammar shows that it is not possible to create empty code. -//! It is possible to include grave accents (ticks) in code, by wrapping it -//! in bigger or smaller sequences: -//! -//! ```markdown -//! Include more: `a``b` or include less: ``a`b``. -//! ``` -//! -//! It is also possible to include just one grave accent (tick): -//! -//! ```markdown -//! Include just one: `` ` ``. -//! ``` -//! -//! Sequences are “gready”, in that they cannot be preceded or followed by -//! more grave accents (ticks). -//! To illustrate: -//! -//! ```markdown -//! Not code: ``x`. -//! -//! Not code: `x``. -//! -//! Escapes work, this is code: \``x`. -//! -//! Escapes work, this is code: `x`\`. -//! ``` -//! -//! Yields: -//! -//! ```html -//! <p>Not code: ``x`.</p> -//! <p>Not code: `x``.</p> -//! <p>Escapes work, this is code: `<code>x</code>.</p> -//! <p>Escapes work, this is code: <code>x</code>`.</p> -//! ``` -//! -//! That is because, when turning markdown into HTML, the first and last space, -//! if both exist and there is also a non-space in the code, are removed. -//! Line endings, at that stage, are considered as spaces. -//! -//! In markdown, it is possible to create code with the -//! [code (fenced)][code_fenced] or [code (indented)][code_indented] constructs -//! in the [flow][] content type. -//! -//! ## HTML -//! -//! Code (text) relates to the `<code>` element in HTML. -//! See [*§ 4.5.15 The `code` element*][html_code] in the HTML spec for more -//! info. -//! -//! When turning markdown into HTML, each line ending is turned into a space. -//! -//! ## Tokens -//! -//! * [`CodeText`][Name::CodeText] -//! * [`CodeTextData`][Name::CodeTextData] -//! * [`CodeTextSequence`][Name::CodeTextSequence] -//! * [`LineEnding`][Name::LineEnding] -//! -//! ## References -//! -//! * [`code-text.js` in `micromark`](https://github.com/micromark/micromark/blob/main/packages/micromark-core-commonmark/dev/lib/code-text.js) -//! * [*§ 6.1 Code spans* in `CommonMark`](https://spec.commonmark.org/0.30/#code-spans) -//! -//! [flow]: crate::construct::flow -//! [text]: crate::construct::text -//! [code_indented]: crate::construct::code_indented -//! [code_fenced]: crate::construct::code_fenced -//! [html_code]: https://html.spec.whatwg.org/multipage/text-level-semantics.html#the-code-element - -use crate::event::Name; -use crate::state::{Name as StateName, State}; -use crate::tokenizer::Tokenizer; - -/// Start of code (text). -/// -/// ```markdown -/// > | `a` -/// ^ -/// > | \`a` -/// ^ -/// ``` -pub fn start(tokenizer: &mut Tokenizer) -> State { - match tokenizer.current { - Some(b'`') - if tokenizer.parse_state.options.constructs.code_text - && (tokenizer.previous != Some(b'`') - || (!tokenizer.events.is_empty() - && tokenizer.events[tokenizer.events.len() - 1].name - == Name::CharacterEscape)) => - { - tokenizer.enter(Name::CodeText); - tokenizer.enter(Name::CodeTextSequence); - State::Retry(StateName::CodeTextSequenceOpen) - } - _ => State::Nok, - } -} - -/// In opening sequence. -/// -/// ```markdown -/// > | `a` -/// ^ -/// ``` -pub fn sequence_open(tokenizer: &mut Tokenizer) -> State { - if let Some(b'`') = tokenizer.current { - tokenizer.tokenize_state.size += 1; - tokenizer.consume(); - State::Next(StateName::CodeTextSequenceOpen) - } else { - tokenizer.exit(Name::CodeTextSequence); - State::Retry(StateName::CodeTextBetween) - } -} - -/// Between something and something else -/// -/// ```markdown -/// > | `a` -/// ^^ -/// ``` -pub fn between(tokenizer: &mut Tokenizer) -> State { - match tokenizer.current { - None => { - tokenizer.tokenize_state.size = 0; - State::Nok - } - Some(b'\n') => { - tokenizer.enter(Name::LineEnding); - tokenizer.consume(); - tokenizer.exit(Name::LineEnding); - State::Next(StateName::CodeTextBetween) - } - Some(b'`') => { - tokenizer.enter(Name::CodeTextSequence); - State::Retry(StateName::CodeTextSequenceClose) - } - _ => { - tokenizer.enter(Name::CodeTextData); - State::Retry(StateName::CodeTextData) - } - } -} - -/// In data. -/// -/// ```markdown -/// > | `a` -/// ^ -/// ``` -pub fn data(tokenizer: &mut Tokenizer) -> State { - match tokenizer.current { - None | Some(b'\n' | b'`') => { - tokenizer.exit(Name::CodeTextData); - State::Retry(StateName::CodeTextBetween) - } - _ => { - tokenizer.consume(); - State::Next(StateName::CodeTextData) - } - } -} - -/// In closing sequence. -/// -/// ```markdown -/// > | `a` -/// ^ -/// ``` -pub fn sequence_close(tokenizer: &mut Tokenizer) -> State { - match tokenizer.current { - Some(b'`') => { - tokenizer.tokenize_state.size_b += 1; - tokenizer.consume(); - State::Next(StateName::CodeTextSequenceClose) - } - _ => { - if tokenizer.tokenize_state.size == tokenizer.tokenize_state.size_b { - tokenizer.exit(Name::CodeTextSequence); - tokenizer.exit(Name::CodeText); - tokenizer.tokenize_state.size = 0; - tokenizer.tokenize_state.size_b = 0; - State::Ok - } else { - let index = tokenizer.events.len(); - tokenizer.exit(Name::CodeTextSequence); - // More or less accents: mark as data. - tokenizer.events[index - 1].name = Name::CodeTextData; - tokenizer.events[index].name = Name::CodeTextData; - tokenizer.tokenize_state.size_b = 0; - State::Retry(StateName::CodeTextBetween) - } - } - } -} diff --git a/src/construct/mod.rs b/src/construct/mod.rs index c5002bb..a0065fa 100644 --- a/src/construct/mod.rs +++ b/src/construct/mod.rs @@ -38,7 +38,7 @@ //! * [character reference][character_reference] //! * [code (fenced)][code_fenced] //! * [code (indented)][code_indented] -//! * [code (text)][code_text] +//! * [code (text)][raw_text] //! * [definition][] //! * [hard break (escape)][hard_break_escape] //! * [heading (atx)][heading_atx] @@ -62,6 +62,8 @@ //! * [gfm footnote definition][gfm_footnote_definition] //! * [gfm task list item check][gfm_task_list_item_check] //! * [gfm label start footnote][gfm_label_start_footnote] +//! * math (text) (in `raw_text`) +//! * gfm strikethrough (in attention) //! //! There are also several small subroutines typically used in different places: //! @@ -143,7 +145,6 @@ pub mod character_escape; pub mod character_reference; pub mod code_fenced; pub mod code_indented; -pub mod code_text; pub mod definition; pub mod document; pub mod flow; @@ -171,6 +172,7 @@ pub mod partial_space_or_tab; pub mod partial_space_or_tab_eol; pub mod partial_title; pub mod partial_whitespace; +pub mod raw_text; pub mod string; pub mod text; pub mod thematic_break; diff --git a/src/construct/paragraph.rs b/src/construct/paragraph.rs index 9e20643..c1e7311 100644 --- a/src/construct/paragraph.rs +++ b/src/construct/paragraph.rs @@ -18,7 +18,7 @@ //! allowed to contain blank lines, or to be blank themselves. //! //! The paragraph is interpreted as the [text][] content type. -//! That means that [autolinks][autolink], [code (text)][code_text], etc are allowed. +//! That means that [autolinks][autolink], [code (text)][raw_text], etc are allowed. //! //! ## HTML //! @@ -37,7 +37,7 @@ //! [flow]: crate::construct::flow //! [text]: crate::construct::text //! [autolink]: crate::construct::autolink -//! [code_text]: crate::construct::code_text +//! [raw_text]: crate::construct::raw_text //! [html]: https://html.spec.whatwg.org/multipage/grouping-content.html#the-p-element use crate::event::{Content, Kind, Link, Name}; diff --git a/src/construct/partial_label.rs b/src/construct/partial_label.rs index ab436b2..255fde1 100644 --- a/src/construct/partial_label.rs +++ b/src/construct/partial_label.rs @@ -44,7 +44,7 @@ //! > ([label start (image)][label_start_image] or //! > [label start (link)][label_start_link]) and a closing //! > ([label end][label_end]), so as to allow further phrasing such as -//! > [code (text)][code_text] or [attention][]. +//! > [code (text)][raw_text] or [attention][]. //! //! ## References //! @@ -58,7 +58,7 @@ //! [label_start_image]: crate::construct::label_start_image //! [label_start_link]: crate::construct::label_start_link //! [label_end]: crate::construct::label_end -//! [code_text]: crate::construct::code_text +//! [raw_text]: crate::construct::raw_text //! [link_reference_size_max]: crate::util::constant::LINK_REFERENCE_SIZE_MAX use crate::construct::partial_space_or_tab_eol::{space_or_tab_eol_with_options, Options}; diff --git a/src/construct/raw_text.rs b/src/construct/raw_text.rs new file mode 100644 index 0000000..7f3990d --- /dev/null +++ b/src/construct/raw_text.rs @@ -0,0 +1,270 @@ +//! Raw (text) occurs in the [text][] content type. +//! It forms code (text) and math (text). +//! +//! ## Grammar +//! +//! Raw (text) forms with the following BNF +//! (<small>see [construct][crate::construct] for character groups</small>): +//! +//! ```bnf +//! ; Restriction: the number of markers in the closing sequence must be equal +//! ; to the number of markers in the opening sequence. +//! raw_text ::= sequence 1*byte sequence +//! +//! ; Restriction: not preceded or followed by the same marker. +//! sequence ::= 1*'`' | 1*'$' +//! ``` +//! +//! The above grammar shows that it is not possible to create empty raw (text). +//! It is possible to include the sequence marker (grave accent for code, +//! dollar for math) in raw (text), by wrapping it in bigger or smaller +//! sequences: +//! +//! ```markdown +//! Include more: `a``b` or include less: ``a`b``. +//! ``` +//! +//! It is also possible to include just one marker: +//! +//! ```markdown +//! Include just one: `` ` ``. +//! ``` +//! +//! Sequences are “gready”, in that they cannot be preceded or followed by +//! more markers. +//! To illustrate: +//! +//! ```markdown +//! Not code: ``x`. +//! +//! Not code: `x``. +//! +//! Escapes work, this is code: \``x`. +//! +//! Escapes work, this is code: `x`\`. +//! ``` +//! +//! Yields: +//! +//! ```html +//! <p>Not code: ``x`.</p> +//! <p>Not code: `x``.</p> +//! <p>Escapes work, this is code: `<code>x</code>.</p> +//! <p>Escapes work, this is code: <code>x</code>`.</p> +//! ``` +//! +//! That is because, when turning markdown into HTML, the first and last space, +//! if both exist and there is also a non-space in the code, are removed. +//! Line endings, at that stage, are considered as spaces. +//! +//! In markdown, it is possible to create code with the +//! [code (fenced)][code_fenced] or [code (indented)][code_indented], +//! and math with the [math (flow)][math_flow] constructs in the [flow][] +//! content type. +//! +//! ## HTML +//! +//! Code (text) relates to the `<code>` element in HTML. +//! See [*§ 4.5.15 The `code` element*][html_code] in the HTML spec for more +//! info. +//! +//! Math (text) does not relate to HTML elements. +//! `MathML`, which is sort of like SVG but for math, exists but it doesn’t work +//! well and isn’t widely supported. +//! Instead, it is recommended to use client side JavaScript with something like +//! `KaTeX` or `MathJax` to process the math +//! For that, the math is compiled as a `<code>` element with two classes: +//! `lang-math` and `math-inline`. +//! Client side JavaScript can look for these classes to process them further. +//! +//! When turning markdown into HTML, each line ending in raw (text) is turned +//! into a space. +//! +//! ## Recommendations +//! +//! When authoring markdown with math, keep in mind that math doesn’t work in +//! most places. +//! Notably, GitHub currently has a really weird crappy client-side regex-based +//! thing. +//! But on your own (math-heavy?) site it can be great! +//! Alternatively, set `options.math_text_single_dollar: false`, which prevents +//! single dollars from being seen as math, and thus prevents normal dollars in +//! text from being seen as math. +//! +//! ## Tokens +//! +//! * [`CodeText`][Name::CodeText] +//! * [`CodeTextData`][Name::CodeTextData] +//! * [`CodeTextSequence`][Name::CodeTextSequence] +//! * [`MathText`][Name::MathText] +//! * [`MathTextData`][Name::MathTextData] +//! * [`MathTextSequence`][Name::MathTextSequence] +//! * [`LineEnding`][Name::LineEnding] +//! +//! ## References +//! +//! * [`code-text.js` in `micromark`](https://github.com/micromark/micromark/blob/main/packages/micromark-core-commonmark/dev/lib/code-text.js) +//! * [`micromark-extension-math`](https://github.com/micromark/micromark-extension-math) +//! * [*§ 6.1 Code spans* in `CommonMark`](https://spec.commonmark.org/0.30/#code-spans) +//! +//! [flow]: crate::construct::flow +//! [text]: crate::construct::text +//! [code_indented]: crate::construct::code_indented +//! [code_fenced]: crate::construct::code_fenced +//! [math_flow]: # "to do" +//! [html_code]: https://html.spec.whatwg.org/multipage/text-level-semantics.html#the-code-element + +use crate::event::Name; +use crate::state::{Name as StateName, State}; +use crate::tokenizer::Tokenizer; + +/// Start of raw (text). +/// +/// ```markdown +/// > | `a` +/// ^ +/// > | \`a` +/// ^ +/// ``` +pub fn start(tokenizer: &mut Tokenizer) -> State { + // Code (text): + if ((tokenizer.parse_state.options.constructs.code_text && tokenizer.current == Some(b'`')) + // Math (text): + || (tokenizer.parse_state.options.constructs.math_text && tokenizer.current == Some(b'$'))) + // Not the same marker (except when escaped). + && (tokenizer.previous != tokenizer.current + || (!tokenizer.events.is_empty() + && tokenizer.events[tokenizer.events.len() - 1].name == Name::CharacterEscape)) + { + let marker = tokenizer.current.unwrap(); + if marker == b'`' { + tokenizer.tokenize_state.token_1 = Name::CodeText; + tokenizer.tokenize_state.token_2 = Name::CodeTextSequence; + tokenizer.tokenize_state.token_3 = Name::CodeTextData; + } else { + tokenizer.tokenize_state.token_1 = Name::MathText; + tokenizer.tokenize_state.token_2 = Name::MathTextSequence; + tokenizer.tokenize_state.token_3 = Name::MathTextData; + } + tokenizer.tokenize_state.marker = marker; + tokenizer.enter(tokenizer.tokenize_state.token_1.clone()); + tokenizer.enter(tokenizer.tokenize_state.token_2.clone()); + State::Retry(StateName::RawTextSequenceOpen) + } else { + State::Nok + } +} + +/// In opening sequence. +/// +/// ```markdown +/// > | `a` +/// ^ +/// ``` +pub fn sequence_open(tokenizer: &mut Tokenizer) -> State { + if tokenizer.current == Some(tokenizer.tokenize_state.marker) { + tokenizer.tokenize_state.size += 1; + tokenizer.consume(); + State::Next(StateName::RawTextSequenceOpen) + } + // Not enough markers in the sequence. + else if tokenizer.tokenize_state.marker == b'$' + && tokenizer.tokenize_state.size == 1 + && !tokenizer.parse_state.options.math_text_single_dollar + { + tokenizer.tokenize_state.marker = 0; + tokenizer.tokenize_state.size = 0; + tokenizer.tokenize_state.token_1 = Name::Data; + tokenizer.tokenize_state.token_2 = Name::Data; + tokenizer.tokenize_state.token_3 = Name::Data; + State::Nok + } else { + tokenizer.exit(tokenizer.tokenize_state.token_2.clone()); + State::Retry(StateName::RawTextBetween) + } +} + +/// Between something and something else +/// +/// ```markdown +/// > | `a` +/// ^^ +/// ``` +pub fn between(tokenizer: &mut Tokenizer) -> State { + match tokenizer.current { + None => { + tokenizer.tokenize_state.marker = 0; + tokenizer.tokenize_state.size = 0; + tokenizer.tokenize_state.token_1 = Name::Data; + tokenizer.tokenize_state.token_2 = Name::Data; + tokenizer.tokenize_state.token_3 = Name::Data; + State::Nok + } + Some(b'\n') => { + tokenizer.enter(Name::LineEnding); + tokenizer.consume(); + tokenizer.exit(Name::LineEnding); + State::Next(StateName::RawTextBetween) + } + _ => { + if tokenizer.current == Some(tokenizer.tokenize_state.marker) { + tokenizer.enter(tokenizer.tokenize_state.token_2.clone()); + State::Retry(StateName::RawTextSequenceClose) + } else { + tokenizer.enter(tokenizer.tokenize_state.token_3.clone()); + State::Retry(StateName::RawTextData) + } + } + } +} + +/// In data. +/// +/// ```markdown +/// > | `a` +/// ^ +/// ``` +pub fn data(tokenizer: &mut Tokenizer) -> State { + if matches!(tokenizer.current, None | Some(b'\n')) + || tokenizer.current == Some(tokenizer.tokenize_state.marker) + { + tokenizer.exit(tokenizer.tokenize_state.token_3.clone()); + State::Retry(StateName::RawTextBetween) + } else { + tokenizer.consume(); + State::Next(StateName::RawTextData) + } +} + +/// In closing sequence. +/// +/// ```markdown +/// > | `a` +/// ^ +/// ``` +pub fn sequence_close(tokenizer: &mut Tokenizer) -> State { + if tokenizer.current == Some(tokenizer.tokenize_state.marker) { + tokenizer.tokenize_state.size_b += 1; + tokenizer.consume(); + State::Next(StateName::RawTextSequenceClose) + } else { + tokenizer.exit(tokenizer.tokenize_state.token_2.clone()); + if tokenizer.tokenize_state.size == tokenizer.tokenize_state.size_b { + tokenizer.exit(tokenizer.tokenize_state.token_1.clone()); + tokenizer.tokenize_state.marker = 0; + tokenizer.tokenize_state.size = 0; + tokenizer.tokenize_state.size_b = 0; + tokenizer.tokenize_state.token_1 = Name::Data; + tokenizer.tokenize_state.token_2 = Name::Data; + tokenizer.tokenize_state.token_3 = Name::Data; + State::Ok + } else { + // More or less accents: mark as data. + let len = tokenizer.events.len(); + tokenizer.events[len - 2].name = tokenizer.tokenize_state.token_3.clone(); + tokenizer.events[len - 1].name = tokenizer.tokenize_state.token_3.clone(); + tokenizer.tokenize_state.size_b = 0; + State::Retry(StateName::RawTextBetween) + } + } +} diff --git a/src/construct/text.rs b/src/construct/text.rs index 5535e3f..3cb0f10 100644 --- a/src/construct/text.rs +++ b/src/construct/text.rs @@ -1,16 +1,16 @@ //! The text content type. //! //! **Text** contains phrasing content such as -//! [attention][crate::construct::attention] (emphasis, strong), -//! [code (text)][crate::construct::code_text], and actual text. +//! [attention][crate::construct::attention] (emphasis, gfm strikethrough, strong), +//! [raw (text)][crate::construct::raw_text] (code (text), math (text)), and actual text. //! //! The constructs found in text are: //! -//! * [Attention][crate::construct::attention] +//! * [Attention][crate::construct::attention] (emphasis, gfm strikethrough, strong) //! * [Autolink][crate::construct::autolink] //! * [Character escape][crate::construct::character_escape] //! * [Character reference][crate::construct::character_reference] -//! * [Code (text)][crate::construct::code_text] +//! * [Raw (text)][crate::construct::raw_text] (code (text), math (text)) //! * [GFM: Label start (footnote)][crate::construct::gfm_label_start_footnote] //! * [GFM: Task list item check][crate::construct::gfm_task_list_item_check] //! * [Hard break (escape)][crate::construct::hard_break_escape] @@ -29,17 +29,18 @@ use crate::state::{Name as StateName, State}; use crate::tokenizer::Tokenizer; /// Characters that can start something in text. -const MARKERS: [u8; 10] = [ +const MARKERS: [u8; 11] = [ b'!', // `label_start_image` + b'$', // `raw_text` (math (text)) b'&', // `character_reference` - b'*', // `attention` + b'*', // `attention` (emphasis, strong) b'<', // `autolink`, `html_text` b'[', // `label_start_link` b'\\', // `character_escape`, `hard_break_escape` b']', // `label_end`, `gfm_label_start_footnote` - b'_', // `attention` - b'`', // `code_text` - b'~', // `attention` (w/ `gfm_strikethrough`) + b'_', // `attention` (emphasis, strong) + b'`', // `raw_text` (code (text)) + b'~', // `attention` (gfm strikethrough) ]; /// Start of text. @@ -81,6 +82,14 @@ pub fn before(tokenizer: &mut Tokenizer) -> State { ); State::Retry(StateName::LabelStartImageStart) } + // raw (text) (code (text), math (text)) + Some(b'$' | b'`') => { + tokenizer.attempt( + State::Next(StateName::TextBefore), + State::Next(StateName::TextBeforeData), + ); + State::Retry(StateName::RawTextStart) + } Some(b'&') => { tokenizer.attempt( State::Next(StateName::TextBefore), @@ -88,6 +97,7 @@ pub fn before(tokenizer: &mut Tokenizer) -> State { ); State::Retry(StateName::CharacterReferenceStart) } + // attention (emphasis, gfm strikethrough, strong) Some(b'*' | b'_' | b'~') => { tokenizer.attempt( State::Next(StateName::TextBefore), @@ -124,13 +134,6 @@ pub fn before(tokenizer: &mut Tokenizer) -> State { ); State::Retry(StateName::LabelEndStart) } - Some(b'`') => { - tokenizer.attempt( - State::Next(StateName::TextBefore), - State::Next(StateName::TextBeforeData), - ); - State::Retry(StateName::CodeTextStart) - } _ => State::Retry(StateName::TextBeforeData), } } |