Add support for subscript extension

src/cm.rs

@@ -396,6 +396,7 @@ impl<'a, 'o, 'c> CommonMarkFormatter<'a, 'o, 'c> {
             }
             NodeValue::Math(ref math) => self.format_math(math, allow_wrap, entering),
             NodeValue::WikiLink(ref nl) => return self.format_wikilink(nl, entering),
+            NodeValue::Subscript => self.format_subscript(),
             NodeValue::Underline => self.format_underline(),
             NodeValue::SpoileredText => self.format_spoiler(),
             NodeValue::EscapedTag(ref net) => self.format_escaped_tag(net),
@@ -711,6 +712,10 @@ impl<'a, 'o, 'c> CommonMarkFormatter<'a, 'o, 'c> {
         write!(self, "^").unwrap();
     }
 
+    fn format_subscript(&mut self) {
+        write!(self, "%").unwrap();
+    }
+
     fn format_underline(&mut self) {
         write!(self, "__").unwrap();
     }

src/html.rs

@@ -1123,6 +1123,18 @@ impl<'o, 'c: 'o> HtmlFormatter<'o, 'c> {
                     self.output.write_all(b"</a>")?;
                 }
             }
+            NodeValue::Subscript => {
+                // Unreliable sourcepos.
+                if entering {
+                    self.output.write_all(b"<sub")?;
+                    if self.options.render.experimental_inline_sourcepos {
+                        self.render_sourcepos(node)?;
+                    }
+                    self.output.write_all(b">")?;
+                } else {
+                    self.output.write_all(b"</sub>")?;
+                }
+            }
             NodeValue::Underline => {
                 // Unreliable sourcepos.
                 if entering {

src/main.rs

@@ -183,6 +183,7 @@ enum Extension {
     MathCode,
     WikilinksTitleAfterPipe,
     WikilinksTitleBeforePipe,
+    Subscript,
     Underline,
     Spoiler,
     Greentext,
@@ -266,6 +267,7 @@ fn main() -> Result<(), Box<dyn Error>> {
         .math_code(exts.contains(&Extension::MathCode))
         .wikilinks_title_after_pipe(exts.contains(&Extension::WikilinksTitleAfterPipe))
         .wikilinks_title_before_pipe(exts.contains(&Extension::WikilinksTitleBeforePipe))
+        .subscript(exts.contains(&Extension::Subscript))
         .underline(exts.contains(&Extension::Underline))
         .spoiler(exts.contains(&Extension::Spoiler))
         .greentext(exts.contains(&Extension::Greentext))

src/nodes.rs

@@ -186,6 +186,9 @@ pub enum NodeValue {
     /// **Inline**.  A wikilink to some URL.
     WikiLink(NodeWikiLink),
 
+    /// **Inline**.  Subscript. Enabled with `subscript` option.
+    Subscript,
+
     /// **Inline**.  Underline. Enabled with `underline` option.
     Underline,
 
@@ -513,6 +516,7 @@ impl NodeValue {
             NodeValue::Escaped => "escaped",
             NodeValue::Math(..) => "math",
             NodeValue::WikiLink(..) => "wikilink",
+            NodeValue::Subscript => "subscript",
             NodeValue::Underline => "underline",
             NodeValue::SpoileredText => "spoiler",
             NodeValue::EscapedTag(_) => "escaped_tag",
@@ -762,6 +766,7 @@ pub fn can_contain_type<'a>(node: &'a AstNode<'a>, child: &NodeValue) -> bool {
         | NodeValue::WikiLink(..)
         | NodeValue::Strikethrough
         | NodeValue::Superscript
+        | NodeValue::Subscript
         | NodeValue::SpoileredText
         | NodeValue::Underline
         // XXX: this is quite a hack: the EscapedTag _contains_ whatever was
@@ -789,6 +794,7 @@ pub fn can_contain_type<'a>(node: &'a AstNode<'a>, child: &NodeValue) -> bool {
                 | NodeValue::WikiLink(..)
                 | NodeValue::FootnoteReference(..)
                 | NodeValue::Superscript
+                | NodeValue::Subscript
                 | NodeValue::SpoileredText
                 | NodeValue::Underline
         ),
@@ -808,6 +814,7 @@ pub fn can_contain_type<'a>(node: &'a AstNode<'a>, child: &NodeValue) -> bool {
             | NodeValue::WikiLink(..)
             | NodeValue::FootnoteReference(..)
             | NodeValue::Superscript
+            | NodeValue::Subscript
             | NodeValue::SpoileredText
             | NodeValue::Underline
             | NodeValue::ShortCode(..)

src/parser/inlines.rs

@@ -160,6 +160,9 @@ impl<'a, 'r, 'o, 'c, 'd, 'i> Subject<'a, 'r, 'o, 'c, 'd, 'i> {
         if options.extension.shortcodes {
             s.special_chars[b':' as usize] = true;
         }
+        if options.extension.subscript {
+            s.special_chars[b'%' as usize] = true;
+        }
         if options.extension.underline {
             s.special_chars[b'_' as usize] = true;
         }
@@ -286,6 +289,7 @@ impl<'a, 'r, 'o, 'c, 'd, 'i> Subject<'a, 'r, 'o, 'c, 'd, 'i> {
                 Some(self.handle_delim(b'^'))
             }
             '$' => Some(self.handle_dollars()),
+            '%' if self.options.extension.subscript => Some(self.handle_delim(b'%')),
             '|' if self.options.extension.spoiler => Some(self.handle_delim(b'|')),
             _ => {
                 let endpos = self.find_special_char();
@@ -384,7 +388,7 @@ impl<'a, 'r, 'o, 'c, 'd, 'i> Subject<'a, 'r, 'o, 'c, 'd, 'i> {
         // This array is an important optimization that prevents searching down
         // the stack for openers we've previously searched for and know don't
         // exist, preventing exponential blowup on pathological cases.
-        let mut openers_bottom: [usize; 12] = [stack_bottom; 12];
+        let mut openers_bottom: [usize; 13] = [stack_bottom; 13];
 
         // This is traversing the stack from the top to the bottom, setting `closer` to
         // the delimiter directly above `stack_bottom`. In the case where we are processing
@@ -408,13 +412,14 @@ impl<'a, 'r, 'o, 'c, 'd, 'i> Subject<'a, 'r, 'o, 'c, 'd, 'i> {
                 let mut mod_three_rule_invoked = false;
 
                 let ix = match c.delim_char {
-                    b'|' => 0,
-                    b'~' => 1,
-                    b'^' => 2,
-                    b'"' => 3,
-                    b'\'' => 4,
-                    b'_' => 5,
-                    b'*' => 6 + (if c.can_open { 3 } else { 0 }) + (c.length % 3),
+                    b'%' => 0,
+                    b'|' => 1,
+                    b'~' => 2,
+                    b'^' => 3,
+                    b'"' => 4,
+                    b'\'' => 5,
+                    b'_' => 6,
+                    b'*' => 7 + (if c.can_open { 3 } else { 0 }) + (c.length % 3),
                     _ => unreachable!(),
                 };
 
@@ -463,6 +468,7 @@ impl<'a, 'r, 'o, 'c, 'd, 'i> Subject<'a, 'r, 'o, 'c, 'd, 'i> {
                     || c.delim_char == b'_'
                     || (self.options.extension.strikethrough && c.delim_char == b'~')
                     || (self.options.extension.superscript && c.delim_char == b'^')
+                    || (self.options.extension.subscript && c.delim_char == b'%')
                     || (self.options.extension.spoiler && c.delim_char == b'|')
                 {
                     if opener_found {
@@ -1105,6 +1111,8 @@ impl<'a, 'r, 'o, 'c, 'd, 'i> Subject<'a, 'r, 'o, 'c, 'd, 'i> {
                 NodeValue::Strikethrough
             } else if self.options.extension.superscript && opener_char == b'^' {
                 NodeValue::Superscript
+            } else if self.options.extension.subscript && opener_char == b'%' {
+                NodeValue::Subscript
             } else if self.options.extension.spoiler && opener_char == b'|' {
                 if use_delims == 2 {
                     NodeValue::SpoileredText

src/parser/mod.rs

@@ -458,6 +458,23 @@ pub struct ExtensionOptions {
     #[builder(default)]
     pub wikilinks_title_before_pipe: bool,
 
+    /// Enables subscripts using percent signs
+    ///
+    /// ```md
+    /// H%2%O
+    /// ```
+    ///
+    /// ```
+    /// # use comrak::{markdown_to_html, Options};
+    /// let mut options = Options::default();
+    /// options.extension.subscript = true;
+    ///
+    /// assert_eq!(markdown_to_html("H%2%O", &options),
+    ///            "<p>H<sub>2</sub>O</p>\n");
+    /// ```
+    #[builder(default)]
+    pub subscript: bool,
+
     /// Enables underlines using double underscores
     ///
     /// ```md

src/tests.rs

@@ -24,6 +24,7 @@ mod regressions;
 mod shortcodes;
 mod spoiler;
 mod strikethrough;
+mod subscript;
 mod superscript;
 mod table;
 mod tagfilter;

src/tests/api.rs

@@ -70,6 +70,7 @@ fn exercise_full_api() {
     let _extension = extension
         .wikilinks_title_after_pipe(true)
         .wikilinks_title_before_pipe(true)
+        .subscript(true)
         .underline(true)
         .spoiler(true)
         .greentext(true);
@@ -270,6 +271,7 @@ fn exercise_full_api() {
         nodes::NodeValue::WikiLink(nl) => {
             let _: String = nl.url;
         }
+        nodes::NodeValue::Subscript => {}
         nodes::NodeValue::Underline => {}
         nodes::NodeValue::SpoileredText => {}
         nodes::NodeValue::EscapedTag(data) => {

src/tests/subscript.rs

@@ -0,0 +1,10 @@
+use super::*;
+
+#[test]
+fn subscript() {
+    html_opts!(
+        [extension.subscript],
+        concat!("H%2%O\n"),
+        concat!("<p>H<sub>2</sub>O</p>\n"),
+    );
+}

src/xml.rs

@@ -284,6 +284,7 @@ impl<'o, 'c> XmlFormatter<'o, 'c> {
                     self.escape(nl.url.as_bytes())?;
                     self.output.write_all(b"\"")?;
                 }
+                NodeValue::Subscript => {}
                 NodeValue::Underline => {}
                 NodeValue::SpoileredText => {}
                 NodeValue::EscapedTag(ref data) => {