Merge branch 'main' into test/unordered-and-ordered-list

Signed-off-by: Titus <[email protected]>

Author: wooorm

Untitled.txt

@@ -6,7 +6,6 @@ micromark.js: `atLineEnding` in html (text) should always eat arbitrary whitespa
 ```rs
 // ---------------------
 // Useful helper:
-extern crate std;
 use std::println;
 use alloc::string::String;
 

benches/bench.rs

@@ -1,6 +1,4 @@
-#[macro_use]
-extern crate criterion;
-use criterion::{BenchmarkId, Criterion};
+use criterion::{criterion_group, criterion_main, BenchmarkId, Criterion};
 use std::fs;
 
 fn readme(c: &mut Criterion) {

build.rs

@@ -1,4 +1,3 @@
-extern crate reqwest;
 use regex::Regex;
 use std::fs;
 
@@ -67,7 +66,6 @@ async fn commonmark() {
 // > 👉 **Important**: this module is generated by `build.rs`.
 // > It is generate from the latest CommonMark website.
 
-extern crate markdown;
 use markdown::{{to_html_with_options, CompileOptions, Options}};
 use pretty_assertions::assert_eq;
 

examples/lib.rs

@@ -1,5 +1,3 @@
-extern crate markdown;
-
 fn main() -> Result<(), String> {
     // Turn on debugging.
     // You can show it with `RUST_LOG=debug cargo run --example lib`

fuzz/fuzz_targets/markdown.rs

@@ -1,14 +1,10 @@
 #![no_main]
 use libfuzzer_sys::fuzz_target;
-extern crate markdown;
 
 fuzz_target!(|data: &[u8]| {
     if let Ok(s) = std::str::from_utf8(data) {
         let _ = markdown::to_html(s);
-        let _ = markdown::to_html_with_options(
-            s,
-            &markdown::Options::gfm()
-        );
+        let _ = markdown::to_html_with_options(s, &markdown::Options::gfm());
         let _ = markdown::to_mdast(s, &markdown::ParseOptions::default());
         let _ = markdown::to_mdast(s, &markdown::ParseOptions::gfm());
     }

readme.md

@@ -90,8 +90,6 @@ cargo add [email protected]
 ## Use
 
 ```rs
-extern crate markdown;
-
 fn main() {
     println!("{}", markdown::to_html("## Hello, *world*!"));
 }
@@ -106,8 +104,6 @@ Yields:
 Extensions (in this case GFM):
 
 ```rs
-extern crate markdown;
-
 fn main() -> Result<(), String> {
     println!(
         "{}",
@@ -136,8 +132,6 @@ Yields:
 Syntax tree ([mdast][]):
 
 ```rs
-extern crate markdown;
-
 fn main() -> Result<(), String> {
     println!(
         "{:?}",

src/configuration.rs

@@ -1236,7 +1236,6 @@ impl Options {
 
 #[cfg(test)]
 mod tests {
-    extern crate std;
     use super::*;
     use crate::util::mdx::Signal;
     use alloc::format;

src/construct/attention.rs

@@ -92,8 +92,10 @@ use alloc::{vec, vec::Vec};
 struct Sequence {
     /// Marker as a byte (`u8`) used in this sequence.
     marker: u8,
-    /// The depth in events where this sequence resides.
-    balance: usize,
+    /// We track whether sequences are in balanced events, and where those
+    /// events start, so that one attention doesn’t start in say, one link, and
+    /// end in another.
+    stack: Vec<usize>,
     /// The index into events where this sequence’s `Enter` currently resides.
     index: usize,
     /// The (shifted) point where this sequence starts.
@@ -172,7 +174,7 @@ pub fn resolve(tokenizer: &mut Tokenizer) -> Option<Subresult> {
                 // An opener matching our closer:
                 if sequence_open.open
                     && sequence_close.marker == sequence_open.marker
-                    && sequence_close.balance == sequence_open.balance
+                    && sequence_close.stack == sequence_open.stack
                 {
                     // If the opening can close or the closing can open,
                     // and the close size *is not* a multiple of three,
@@ -219,23 +221,20 @@ pub fn resolve(tokenizer: &mut Tokenizer) -> Option<Subresult> {
     }
 
     tokenizer.map.consume(&mut tokenizer.events);
-
     None
 }
 
 /// Get sequences.
 fn get_sequences(tokenizer: &mut Tokenizer) -> Vec<Sequence> {
     let mut index = 0;
-    let mut balance = 0;
+    let mut stack = vec![];
     let mut sequences = vec![];
 
     while index < tokenizer.events.len() {
         let enter = &tokenizer.events[index];
 
-        if enter.kind == Kind::Enter {
-            balance += 1;
-
-            if enter.name == Name::AttentionSequence {
+        if enter.name == Name::AttentionSequence {
+            if enter.kind == Kind::Enter {
                 let end = index + 1;
                 let exit = &tokenizer.events[end];
 
@@ -255,7 +254,7 @@ fn get_sequences(tokenizer: &mut Tokenizer) -> Vec<Sequence> {
 
                 sequences.push(Sequence {
                     index,
-                    balance,
+                    stack: stack.clone(),
                     start_point: enter.point.clone(),
                     end_point: exit.point.clone(),
                     size: exit.point.index - enter.point.index,
@@ -272,8 +271,10 @@ fn get_sequences(tokenizer: &mut Tokenizer) -> Vec<Sequence> {
                     marker,
                 });
             }
+        } else if enter.kind == Kind::Enter {
+            stack.push(index);
         } else {
-            balance -= 1;
+            stack.pop();
         }
 
         index += 1;

src/construct/document.rs

@@ -423,7 +423,7 @@ pub fn flow_end(tokenizer: &mut Tokenizer) -> State {
     if !document_lazy_continuation_current && !child.events.is_empty() {
         let before = skip::opt_back(&child.events, child.events.len() - 1, &[Name::LineEnding]);
         let name = &child.events[before].name;
-        if name == &Name::Content {
+        if name == &Name::Content || name == &Name::HeadingSetextUnderline {
             document_lazy_continuation_current = true;
         }
     }

src/construct/gfm_autolink_literal.rs

@@ -334,7 +334,7 @@ pub fn www_prefix_inside(tokenizer: &mut Tokenizer) -> State {
 /// ```
 pub fn www_prefix_after(tokenizer: &mut Tokenizer) -> State {
     // If there is *anything*, we can link.
-    if tokenizer.current == None {
+    if tokenizer.current.is_none() {
         State::Nok
     } else {
         State::Ok

src/construct/gfm_table.rs

@@ -883,6 +883,7 @@ pub fn resolve(tokenizer: &mut Tokenizer) -> Option<Subresult> {
         flush_table_end(tokenizer, last_table_end, last_table_has_body);
     }
 
+    tokenizer.map.consume(&mut tokenizer.events);
     None
 }
 

src/construct/gfm_task_list_item_check.rs

@@ -61,7 +61,7 @@ pub fn start(tokenizer: &mut Tokenizer) -> State {
             .tokenize_state
             .document_at_first_paragraph_of_list_item
         && tokenizer.current == Some(b'[')
-        && tokenizer.previous == None
+        && tokenizer.previous.is_none()
     {
         tokenizer.enter(Name::GfmTaskListItemCheck);
         tokenizer.enter(Name::GfmTaskListItemMarker);
@@ -149,7 +149,7 @@ pub fn after(tokenizer: &mut Tokenizer) -> State {
 /// ```
 pub fn after_space_or_tab(tokenizer: &mut Tokenizer) -> State {
     // End of paragraph, after whitespace, after check, is not okay.
-    if tokenizer.current == None {
+    if tokenizer.current.is_none() {
         State::Nok
     } else {
         State::Ok

src/construct/heading_atx.rs

@@ -280,5 +280,6 @@ pub fn resolve(tokenizer: &mut Tokenizer) -> Option<Subresult> {
         index += 1;
     }
 
+    tokenizer.map.consume(&mut tokenizer.events);
     None
 }

src/construct/heading_setext.rs

@@ -92,13 +92,12 @@ pub fn start(tokenizer: &mut Tokenizer) -> State {
         && !tokenizer.pierce
         // Require a paragraph before.
         && (!tokenizer.events.is_empty()
-            && tokenizer.events[skip::opt_back(
+            && matches!(tokenizer.events[skip::opt_back(
                 &tokenizer.events,
                 tokenizer.events.len() - 1,
                 &[Name::LineEnding, Name::SpaceOrTab],
             )]
-            .name
-                == Name::Content)
+            .name, Name::Content | Name::HeadingSetextUnderline))
     {
         tokenizer.enter(Name::HeadingSetextUnderline);
 
@@ -185,8 +184,6 @@ pub fn after(tokenizer: &mut Tokenizer) -> State {
 
 /// Resolve heading (setext).
 pub fn resolve(tokenizer: &mut Tokenizer) -> Option<Subresult> {
-    tokenizer.map.consume(&mut tokenizer.events);
-
     let mut enter = skip::to(&tokenizer.events, 0, &[Name::HeadingSetextUnderline]);
 
     while enter < tokenizer.events.len() {
@@ -280,6 +277,5 @@ pub fn resolve(tokenizer: &mut Tokenizer) -> Option<Subresult> {
     }
 
     tokenizer.map.consume(&mut tokenizer.events);
-
     None
 }

src/construct/label_end.rs

@@ -669,7 +669,6 @@ pub fn resolve(tokenizer: &mut Tokenizer) -> Option<Subresult> {
     mark_as_data(tokenizer, &starts);
 
     tokenizer.map.consume(&mut tokenizer.events);
-
     None
 }
 

src/construct/list_item.rs

@@ -469,5 +469,6 @@ pub fn resolve(tokenizer: &mut Tokenizer) -> Option<Subresult> {
         index += 1;
     }
 
+    tokenizer.map.consume(&mut tokenizer.events);
     None
 }

src/construct/mdx_esm.rs

@@ -216,7 +216,7 @@ fn parse_esm(tokenizer: &mut Tokenizer) -> State {
             State::Error(format!("{}:{}: {}", point.line, point.column, message))
         }
         MdxSignal::Eof(message) => {
-            if tokenizer.current == None {
+            if tokenizer.current.is_none() {
                 State::Error(format!(
                     "{}:{}: {}",
                     tokenizer.point.line, tokenizer.point.column, message

src/construct/partial_data.rs

@@ -74,8 +74,6 @@ pub fn inside(tokenizer: &mut Tokenizer) -> State {
 
 /// Merge adjacent data events.
 pub fn resolve(tokenizer: &mut Tokenizer) -> Option<Subresult> {
-    tokenizer.map.consume(&mut tokenizer.events);
-
     let mut index = 0;
 
     // Loop through events and merge adjacent data events.
@@ -107,5 +105,6 @@ pub fn resolve(tokenizer: &mut Tokenizer) -> Option<Subresult> {
         index += 1;
     }
 
+    tokenizer.map.consume(&mut tokenizer.events);
     None
 }

src/construct/partial_mdx_jsx.rs

@@ -1119,7 +1119,7 @@ fn crash(tokenizer: &Tokenizer, at: &str, expect: &str) -> State {
         "{}:{}: Unexpected {} {}, expected {}",
         tokenizer.point.line,
         tokenizer.point.column,
-        format_char_opt(if tokenizer.current == None {
+        format_char_opt(if tokenizer.current.is_none() {
             None
         } else {
             char_after_index(tokenizer.parse_state.bytes, tokenizer.point.index)

src/construct/partial_whitespace.rs

@@ -67,8 +67,6 @@ use alloc::vec;
 
 /// Resolve whitespace.
 pub fn resolve_whitespace(tokenizer: &mut Tokenizer, hard_break: bool, trim_whole: bool) {
-    tokenizer.map.consume(&mut tokenizer.events);
-
     let mut index = 0;
 
     while index < tokenizer.events.len() {
@@ -86,6 +84,8 @@ pub fn resolve_whitespace(tokenizer: &mut Tokenizer, hard_break: bool, trim_whol
 
         index += 1;
     }
+
+    tokenizer.map.consume(&mut tokenizer.events);
 }
 
 /// Trim a [`Data`][Name::Data] event.

src/construct/text.rs

@@ -259,5 +259,6 @@ pub fn resolve(tokenizer: &mut Tokenizer) -> Option<Subresult> {
         resolve_gfm_autolink_literal(tokenizer);
     }
 
+    tokenizer.map.consume(&mut tokenizer.events);
     None
 }

src/mdast.rs

@@ -250,7 +250,7 @@ impl ToString for Node {
             | Node::Image(_)
             | Node::ImageReference(_)
             | Node::ThematicBreak(_)
-            | Node::Definition(_) => "".into(),
+            | Node::Definition(_) => String::new(),
         }
     }
 }

src/subtokenize.rs

@@ -96,7 +96,7 @@ pub fn subtokenize(
             debug_assert_eq!(event.kind, Kind::Enter);
 
             // No need to enter linked events again.
-            if link.previous == None
+            if link.previous.is_none()
                 && (filter.is_none() || &link.content == filter.as_ref().unwrap())
             {
                 // Index into `events` pointing to a chunk.
@@ -148,7 +148,7 @@ pub fn subtokenize(
                     let link_curr = enter.link.as_ref().expect("expected link");
                     debug_assert_eq!(enter.kind, Kind::Enter);
 
-                    if link_curr.previous != None {
+                    if link_curr.previous.is_some() {
                         tokenizer.define_skip(enter.point.clone());
                     }
 

src/to_html.rs

@@ -685,7 +685,7 @@ fn on_enter_paragraph(context: &mut CompileContext) {
 /// Handle [`Enter`][Kind::Enter]:[`Resource`][Name::Resource].
 fn on_enter_resource(context: &mut CompileContext) {
     context.buffer(); // We can have line endings in the resource, ignore them.
-    context.media_stack.last_mut().unwrap().destination = Some("".into());
+    context.media_stack.last_mut().unwrap().destination = Some(String::new());
 }
 
 /// Handle [`Enter`][Kind::Enter]:[`ResourceDestinationString`][Name::ResourceDestinationString].