//! Public API of micromark. //! //! This module exposes [`micromark`][] (and [`micromark_with_options`][]). //! `micromark` is a safe way to transform (untrusted?) markdown into HTML. //! `micromark_with_options` allows you to configure how markdown is turned into //! HTML, such as by allowing dangerous HTML when you trust it. #![no_std] extern crate alloc; mod compiler; mod construct; mod event; mod parser; mod resolve; mod state; mod subtokenize; mod tokenizer; mod util; use crate::compiler::compile; use crate::parser::parse; use alloc::{boxed::Box, fmt, string::String}; /// Type of line endings in markdown. #[derive(Clone, Debug, Default, Eq, PartialEq)] pub enum LineEnding { /// Both a carriage return (`\r`) and a line feed (`\n`). /// /// ## Example /// /// ```markdown /// a␍␊ /// b /// ``` CarriageReturnLineFeed, /// Sole carriage return (`\r`). /// /// ## Example /// /// ```markdown /// a␍ /// b /// ``` CarriageReturn, /// Sole line feed (`\n`). /// /// ## Example /// /// ```markdown /// a␊ /// b /// ``` #[default] LineFeed, } impl LineEnding { /// Turn the line ending into a [str]. fn as_str(&self) -> &str { match self { LineEnding::CarriageReturnLineFeed => "\r\n", LineEnding::CarriageReturn => "\r", LineEnding::LineFeed => "\n", } } /// Turn a string into a line ending. /// /// ## Panics /// /// Panics if `code` is not `\r\n`, `\r`, or `\n`. fn from_str(str: &str) -> LineEnding { match str { "\r\n" => LineEnding::CarriageReturnLineFeed, "\r" => LineEnding::CarriageReturn, "\n" => LineEnding::LineFeed, _ => unreachable!("invalid str"), } } } /// Signal used as feedback when parsing MDX esm/expressions. #[derive(Clone, Debug)] pub enum MdxSignal { /// A syntax error. /// /// `micromark-rs` will crash with error message `String`, and convert the /// `usize` (byte offset into `&str` passed to `MdxExpressionParse` or /// `MdxEsmParse`) to where it happened in the whole document. /// /// ## Examples /// /// ```rust ignore /// MdxSignal::Error("Unexpected `\"`, expected identifier".to_string(), 1) /// ``` Error(String, usize), /// An error at the end of the (partial?) expression. /// /// `micromark-rs` will either crash with error message `String` if it /// doesn’t have any more text, or it will try again later when more text /// is available. /// /// ## Examples /// /// ```rust ignore /// MdxSignal::Eof("Unexpected end of file in string literal".to_string()) /// ``` Eof(String), /// Done, successfully. /// /// `micromark-rs` knows that this is the end of a valid expression/esm and /// continues with markdown. /// /// ## Examples /// /// ```rust ignore /// MdxSignal::Ok /// ``` Ok, } /// Expression kind. #[derive(Clone, Debug)] pub enum MdxExpressionKind { /// Kind of expressions in prose: `# {Math.PI}` and `{Math.PI}`. Expression, /// Kind of expressions as attributes: `` AttributeExpression, /// Kind of expressions as attribute values: ``. AttributeValueExpression, } /// Signature of a function that parses expressions. /// /// Can be passed as `mdx_expression_parse` in [`Options`][] to support /// expressions according to a certain grammar (typically, a programming /// language). pub type MdxExpressionParse = dyn Fn(&str, MdxExpressionKind) -> MdxSignal; /// Signature of a function that parses ESM. /// /// Can be passed as `mdx_esm_parse` in [`Options`][] to support /// ESM according to a certain grammar (typically, a programming /// language). pub type MdxEsmParse = dyn Fn(&str) -> MdxSignal; /// Control which constructs are enabled. /// /// Not all constructs can be configured. /// Notably, blank lines and paragraphs cannot be turned off. #[allow(clippy::struct_excessive_bools)] #[derive(Clone, Debug)] pub struct Constructs { /// Attention. /// /// ```markdown /// > | a *b* c **d**. /// ^^^ ^^^^^ /// ``` pub attention: bool, /// Autolink. /// /// ```markdown /// > | a b . /// ^^^^^^^^^^^^^^^^^^^^^ ^^^^^^^^^^^^^^^^^^ /// ``` pub autolink: bool, /// Block quote. /// /// ```markdown /// > | > a /// ^^^ /// ``` pub block_quote: bool, /// Character escape. /// /// ```markdown /// > | a \* b /// ^^ /// ``` pub character_escape: bool, /// Character reference. /// /// ```markdown /// > | a & b /// ^^^^^ /// ``` pub character_reference: bool, /// Code (indented). /// /// ```markdown /// > | a /// ^^^^^ /// ``` pub code_indented: bool, /// Code (fenced). /// /// ```markdown /// > | ~~~js /// ^^^^^ /// > | console.log(1) /// ^^^^^^^^^^^^^^ /// > | ~~~ /// ^^^ /// ``` pub code_fenced: bool, /// Code (text). /// /// ```markdown /// > | a `b` c /// ^^^ /// ``` pub code_text: bool, /// Definition. /// /// ```markdown /// > | [a]: b "c" /// ^^^^^^^^^^ /// ``` pub definition: bool, /// Frontmatter. /// /// ````markdown /// > | --- /// ^^^ /// > | title: Neptune /// ^^^^^^^^^^^^^^ /// > | --- /// ^^^ /// ```` pub frontmatter: bool, /// GFM: autolink literal. /// /// ```markdown /// > | https://example.com /// ^^^^^^^^^^^^^^^^^^^ /// ``` pub gfm_autolink_literal: bool, /// GFM: footnote definition. /// /// ```markdown /// > | [^a]: b /// ^^^^^^^ /// ``` pub gfm_footnote_definition: bool, /// GFM: footnote label start. /// /// ```markdown /// > | a[^b] /// ^^ /// ``` pub gfm_label_start_footnote: bool, /// /// ```markdown /// > | a ~b~ c. /// ^^^ /// ``` pub gfm_strikethrough: bool, /// GFM: table. /// /// ```markdown /// > | | a | /// ^^^^^ /// > | | - | /// ^^^^^ /// > | | b | /// ^^^^^ /// ``` pub gfm_table: bool, /// GFM: task list item. /// /// ```markdown /// > | * [x] y. /// ^^^ /// ``` pub gfm_task_list_item: bool, /// Hard break (escape). /// /// ```markdown /// > | a\ /// ^ /// | b /// ``` pub hard_break_escape: bool, /// Hard break (trailing). /// /// ```markdown /// > | a␠␠ /// ^^ /// | b /// ``` pub hard_break_trailing: bool, /// Heading (atx). /// /// ```markdown /// > | # a /// ^^^ /// ``` pub heading_atx: bool, /// Heading (setext). /// /// ```markdown /// > | a /// ^^ /// > | == /// ^^ /// ``` pub heading_setext: bool, /// HTML (flow). /// /// ```markdown /// > |
/// ^^^^^ /// ``` pub html_flow: bool, /// HTML (text). /// /// ```markdown /// > | a c /// ^^^ /// ``` pub html_text: bool, /// Label start (image). /// /// ```markdown /// > | a ![b](c) d /// ^^ /// ``` pub label_start_image: bool, /// Label start (link). /// /// ```markdown /// > | a [b](c) d /// ^ /// ``` pub label_start_link: bool, /// Label end. /// /// ```markdown /// > | a [b](c) d /// ^^^^ /// ``` pub label_end: bool, /// List items. /// /// ```markdown /// > | * a /// ^^^ /// ``` pub list_item: bool, /// Math (flow). /// /// ```markdown /// > | $$ /// ^^ /// > | \frac{1}{2} /// ^^^^^^^^^^^ /// > | $$ /// ^^ /// ``` pub math_flow: bool, /// Math (text). /// /// ```markdown /// > | a $b$ c /// ^^^ /// ``` pub math_text: bool, /// MDX: ESM. /// /// ```markdown /// > | import a from 'b' /// ^^^^^^^^^^^^^^^^^ /// ``` /// /// > 👉 **Note**: you *must* pass [`options.mdx_esm_parse`][MdxEsmParse] /// > too. /// > Otherwise, this option has no affect. pub mdx_esm: bool, /// MDX: expression (flow). /// /// ```markdown /// > | {Math.PI} /// ^^^^^^^^^ /// ``` /// /// > 👉 **Note**: you *can* pass /// > [`options.mdx_expression_parse`][MdxExpressionParse] /// > to parse expressions according to a certain grammar (typically, a /// > programming language). pub mdx_expression_flow: bool, /// MDX: expression (text). /// /// ```markdown /// > | a {Math.PI} c /// ^^^^^^^^^ /// ``` /// /// > 👉 **Note**: you *can* pass /// > [`options.mdx_expression_parse`][MdxExpressionParse] /// > to parse expressions according to a certain grammar (typically, a /// > programming language). pub mdx_expression_text: bool, /// MDX: JSX (flow). /// /// ```markdown /// > | /// ^^^^^^^^^^^^^ /// ``` /// /// > 👉 **Note**: you *can* pass /// > [`options.mdx_expression_parse`][MdxExpressionParse] /// > to parse expressions in JSX according to a certain grammar /// > (typically, a programming language). pub mdx_jsx_flow: bool, /// MDX: JSX (text). /// /// ```markdown /// > | a c /// ^^^^^^^^^^^^^ /// ``` /// /// > 👉 **Note**: you *can* pass /// > [`options.mdx_expression_parse`][MdxExpressionParse] /// > to parse expressions in JSX according to a certain grammar /// > (typically, a programming language). pub mdx_jsx_text: bool, /// Thematic break. /// /// ```markdown /// > | *** /// ^^^ /// ``` pub thematic_break: bool, } impl Default for Constructs { /// `CommonMark`. fn default() -> Self { Self { attention: true, autolink: true, block_quote: true, character_escape: true, character_reference: true, code_indented: true, code_fenced: true, code_text: true, definition: true, frontmatter: false, gfm_autolink_literal: false, gfm_label_start_footnote: false, gfm_footnote_definition: false, gfm_strikethrough: false, gfm_table: false, gfm_task_list_item: false, hard_break_escape: true, hard_break_trailing: true, heading_atx: true, heading_setext: true, html_flow: true, html_text: true, label_start_image: true, label_start_link: true, label_end: true, list_item: true, math_flow: false, math_text: false, mdx_esm: false, mdx_expression_flow: false, mdx_expression_text: false, mdx_jsx_flow: false, mdx_jsx_text: false, thematic_break: true, } } } impl Constructs { /// GFM. /// /// /// /// This turns on `CommonMark` + GFM. #[must_use] pub fn gfm() -> Self { Self { gfm_autolink_literal: true, gfm_footnote_definition: true, gfm_label_start_footnote: true, gfm_strikethrough: true, gfm_table: true, gfm_task_list_item: true, ..Self::default() } } /// MDX. /// /// /// /// This turns on `CommonMark`, turns off some conflicting constructs /// (autolinks, code (indented), html), and turns on MDX (JSX, /// expressions, ESM). /// /// > 👉 **Note**: you *must* pass [`options.mdx_esm_parse`][MdxEsmParse] /// > to support ESM. /// > You *can* pass /// > [`options.mdx_expression_parse`][MdxExpressionParse] /// > to parse expressions according to a certain grammar (typically, a /// > programming language). #[must_use] pub fn mdx() -> Self { Self { autolink: false, code_indented: false, html_flow: false, html_text: false, mdx_esm: true, mdx_expression_flow: true, mdx_expression_text: true, mdx_jsx_flow: true, mdx_jsx_text: true, ..Self::default() } } } /// Configuration (optional). #[allow(clippy::struct_excessive_bools)] pub struct Options { // Note: when adding fields, don’t forget to add them to `fmt::Debug` below. /// Whether to allow (dangerous) HTML. /// The default is `false`, you can turn it on to `true` for trusted /// content. /// /// ## Examples /// /// ``` /// use micromark::{micromark, micromark_with_options, Options}; /// # fn main() -> Result<(), String> { /// /// // micromark is safe by default: /// assert_eq!( /// micromark("Hi, venus!"), /// "

Hi, <i>venus</i>!

" /// ); /// /// // Turn `allow_dangerous_html` on to allow potentially dangerous HTML: /// assert_eq!( /// micromark_with_options( /// "Hi, venus!", /// &Options { /// allow_dangerous_html: true, /// ..Options::default() /// } /// )?, /// "

Hi, venus!

" /// ); /// # Ok(()) /// # } /// ``` pub allow_dangerous_html: bool, /// Whether to allow (dangerous) protocols in links and images. /// The default is `false`, you can turn it on to `true` for trusted /// content. /// /// ## Examples /// /// ``` /// use micromark::{micromark, micromark_with_options, Options}; /// # fn main() -> Result<(), String> { /// /// // micromark is safe by default: /// assert_eq!( /// micromark(""), /// "
" /// ); /// /// // Turn `allow_dangerous_protocol` on to allow potentially dangerous protocols: /// assert_eq!( /// micromark_with_options( /// "", /// &Options { /// allow_dangerous_protocol: true, /// ..Options::default() /// } /// )?, /// "

javascript:alert(1)

" /// ); /// # Ok(()) /// # } /// ``` pub allow_dangerous_protocol: bool, /// Which constructs to enable and disable. /// The default is to follow `CommonMark`. /// /// ## Examples /// /// ``` /// use micromark::{micromark, micromark_with_options, Options, Constructs}; /// # fn main() -> Result<(), String> { /// /// // micromark follows CommonMark by default: /// assert_eq!( /// micromark(" indented code?"), /// "
indented code?\n
" /// ); /// /// // Pass `constructs` to choose what to enable and disable: /// assert_eq!( /// micromark_with_options( /// " indented code?", /// &Options { /// constructs: Constructs { /// code_indented: false, /// ..Constructs::default() /// }, /// ..Options::default() /// } /// )?, /// "

indented code?

" /// ); /// # Ok(()) /// # } /// ``` pub constructs: Constructs, /// Default line ending to use, for line endings not in `value`. /// /// Generally, micromark copies line endings (`\r`, `\n`, `\r\n`) in the /// markdown document over to the compiled HTML. /// In some cases, such as `> a`, CommonMark requires that extra line /// endings are added: `
\n

a

\n
`. /// /// To create that line ending, the document is checked for the first line /// ending that is used. /// If there is no line ending, `default_line_ending` is used. /// If that isn’t configured, `\n` is used. /// /// ## Examples /// /// ``` /// use micromark::{micromark, micromark_with_options, Options, LineEnding}; /// # fn main() -> Result<(), String> { /// /// // micromark uses `\n` by default: /// assert_eq!( /// micromark("> a"), /// "
\n

a

\n
" /// ); /// /// // Define `default_line_ending` to configure the default: /// assert_eq!( /// micromark_with_options( /// "> a", /// &Options { /// default_line_ending: LineEnding::CarriageReturnLineFeed, /// ..Options::default() /// } /// )?, /// "
\r\n

a

\r\n
" /// ); /// # Ok(()) /// # } /// ``` pub default_line_ending: LineEnding, /// Label to use for the footnotes section. /// /// Change it when the markdown is not in English. /// Typically affects screen readers (change `gfm_footnote_label_attributes` /// to make it visible). /// /// ## Examples /// /// ``` /// use micromark::{micromark, micromark_with_options, Options, Constructs}; /// # fn main() -> Result<(), String> { /// /// // `"Footnotes"` is used by default: /// assert_eq!( /// micromark_with_options( /// "[^a]\n\n[^a]: b", /// &Options { /// constructs: Constructs::gfm(), /// ..Options::default() /// } /// )?, /// "

1

\n

Footnotes

\n
    \n
  1. \n

    b

    \n
  2. \n
\n
\n" /// ); /// /// // Pass `gfm_footnote_label` to use something else: /// assert_eq!( /// micromark_with_options( /// "[^a]\n\n[^a]: b", /// &Options { /// constructs: Constructs::gfm(), /// gfm_footnote_label: Some("Notes de bas de page".to_string()), /// ..Options::default() /// } /// )?, /// "

1

\n

Notes de bas de page

\n
    \n
  1. \n

    b

    \n
  2. \n
\n
\n" /// ); /// # Ok(()) /// # } /// ``` pub gfm_footnote_label: Option, /// HTML tag to use for the footnote label. /// /// Change it to match your document structure and play well with your CSS. /// /// ## Examples /// /// ``` /// use micromark::{micromark, micromark_with_options, Options, Constructs}; /// # fn main() -> Result<(), String> { /// /// // `"h2"` is used by default: /// assert_eq!( /// micromark_with_options( /// "[^a]\n\n[^a]: b", /// &Options { /// constructs: Constructs::gfm(), /// ..Options::default() /// } /// )?, /// "

1

\n

Footnotes

\n
    \n
  1. \n

    b

    \n
  2. \n
\n
\n" /// ); /// /// // Pass `gfm_footnote_label_tag_name` to use something else: /// assert_eq!( /// micromark_with_options( /// "[^a]\n\n[^a]: b", /// &Options { /// constructs: Constructs::gfm(), /// gfm_footnote_label_tag_name: Some("h1".to_string()), /// ..Options::default() /// } /// )?, /// "

1

\n

Footnotes

\n
    \n
  1. \n

    b

    \n
  2. \n
\n
\n" /// ); /// # Ok(()) /// # } /// ``` pub gfm_footnote_label_tag_name: Option, /// Attributes to use on the footnote label. /// /// > 👉 **Note**: `id="footnote-label"` is always added, because footnote /// > calls use it with `aria-describedby` to provide an accessible label. /// /// A `class="sr-only"` is added by default to hide the label from sighted /// users. /// Change it to make the label visible, or add other classes or other /// attributes. /// /// ## Examples /// /// ``` /// use micromark::{micromark, micromark_with_options, Options, Constructs}; /// # fn main() -> Result<(), String> { /// /// // `"class=\"sr-only\""` is used by default: /// assert_eq!( /// micromark_with_options( /// "[^a]\n\n[^a]: b", /// &Options { /// constructs: Constructs::gfm(), /// ..Options::default() /// } /// )?, /// "

1

\n

Footnotes

\n
    \n
  1. \n

    b

    \n
  2. \n
\n
\n" /// ); /// /// // Pass `gfm_footnote_label_attributes` to use something else: /// assert_eq!( /// micromark_with_options( /// "[^a]\n\n[^a]: b", /// &Options { /// constructs: Constructs::gfm(), /// gfm_footnote_label_attributes: Some("class=\"footnote-heading\"".to_string()), /// ..Options::default() /// } /// )?, /// "

1

\n

Footnotes

\n
    \n
  1. \n

    b

    \n
  2. \n
\n
\n" /// ); /// # Ok(()) /// # } /// ``` pub gfm_footnote_label_attributes: Option, /// Label to use from backreferences back to their footnote call. /// /// Change it when the markdown is not in English. /// Affects screen readers. /// /// ## Examples /// /// ``` /// use micromark::{micromark, micromark_with_options, Options, Constructs}; /// # fn main() -> Result<(), String> { /// /// // `"Back to content"` is used by default: /// assert_eq!( /// micromark_with_options( /// "[^a]\n\n[^a]: b", /// &Options { /// constructs: Constructs::gfm(), /// ..Options::default() /// } /// )?, /// "

1

\n

Footnotes

\n
    \n
  1. \n

    b

    \n
  2. \n
\n
\n" /// ); /// /// // Pass `gfm_footnote_back_label` to use something else: /// assert_eq!( /// micromark_with_options( /// "[^a]\n\n[^a]: b", /// &Options { /// constructs: Constructs::gfm(), /// gfm_footnote_back_label: Some("Arrière".to_string()), /// ..Options::default() /// } /// )?, /// "

1

\n

Footnotes

\n
    \n
  1. \n

    b

    \n
  2. \n
\n
\n" /// ); /// # Ok(()) /// # } /// ``` pub gfm_footnote_back_label: Option, /// Prefix to use before the `id` attribute on footnotes to prevent them /// from *clobbering*. /// /// DOM clobbering is this: /// /// ```html ///

/// /// ``` /// /// The above example shows that elements are made available by browsers, /// by their ID, on the `window` object, which is a security risk because /// you might be expecting some other variable at that place. /// Using a prefix solves this problem. /// /// ## Examples /// /// ``` /// use micromark::{micromark, micromark_with_options, Options, Constructs}; /// # fn main() -> Result<(), String> { /// /// // `"user-content-"` is used by default: /// assert_eq!( /// micromark_with_options( /// "[^a]\n\n[^a]: b", /// &Options { /// constructs: Constructs::gfm(), /// ..Options::default() /// } /// )?, /// "

1

\n

Footnotes

\n
    \n
  1. \n

    b

    \n
  2. \n
\n
\n" /// ); /// /// // Pass `gfm_footnote_clobber_prefix` to use something else: /// assert_eq!( /// micromark_with_options( /// "[^a]\n\n[^a]: b", /// &Options { /// constructs: Constructs::gfm(), /// gfm_footnote_clobber_prefix: Some("".to_string()), /// ..Options::default() /// } /// )?, /// "

1

\n

Footnotes

\n
    \n
  1. \n

    b

    \n
  2. \n
\n
\n" /// ); /// # Ok(()) /// # } /// ``` pub gfm_footnote_clobber_prefix: Option, /// Whether to support GFM strikethrough (if enabled in `constructs`) with /// a single tilde (default: `true`). /// /// Single tildes work on github.com but are technically prohibited by GFM. /// /// ## Examples /// /// ``` /// use micromark::{micromark, micromark_with_options, Options, Constructs}; /// # fn main() -> Result<(), String> { /// /// // micromark supports single tildes by default: /// assert_eq!( /// micromark_with_options( /// "~a~", /// &Options { /// constructs: Constructs::gfm(), /// ..Options::default() /// } /// )?, /// "

a

" /// ); /// /// // Pass `gfm_strikethrough_single_tilde: false` to turn that off: /// assert_eq!( /// micromark_with_options( /// "~a~", /// &Options { /// constructs: Constructs::gfm(), /// gfm_strikethrough_single_tilde: false, /// ..Options::default() /// } /// )?, /// "

~a~

" /// ); /// # Ok(()) /// # } /// ``` pub gfm_strikethrough_single_tilde: bool, /// Whether to support the GFM tagfilter, when `allow_dangerous_html` is on /// (default: `false`). /// /// The tagfilter is kinda weird and kinda useless. /// The tag filter is a naïve attempt at XSS protection. /// You should use a proper HTML sanitizing algorithm. /// /// ## Examples /// /// ``` /// use micromark::{micromark_with_options, Options, Constructs}; /// # fn main() -> Result<(), String> { /// /// // With `allow_dangerous_html`, micromark passes HTML through untouched: /// assert_eq!( /// micromark_with_options( /// "

javascript:alert(1)