Add support for hard break (trailing)

4 files changed, 101 insertions, 7 deletions

diff --git a/src/construct/character_escape.rs b/src/construct/character_escape.rs
index baedd4b..743cbf8 100644
--- a/src/construct/character_escape.rs
+++ b/src/construct/character_escape.rs
@@ -14,7 +14,7 @@
 //! [character reference][character_reference] instead
 //! (as in, `&`, `{`, or say `	`).
 //! It is also possible to escape a line ending in text with a similar
-//! construct: a [hard break escape][hard_break_escape] is a backslash followed
+//! construct: a [hard break (escape)][hard_break_escape] is a backslash followed
 //! by a line ending (that is part of the construct instead of ending it).
 //!
 //! ## References
diff --git a/src/construct/hard_break_escape.rs b/src/construct/hard_break_escape.rs
index a7712d6..51da953 100644
--- a/src/construct/hard_break_escape.rs
+++ b/src/construct/hard_break_escape.rs
@@ -1,4 +1,4 @@
-//! Hard break escapes are a construct that occurs in the  [text][] content
+//! Hard break (escape) is a construct that occurs in the  [text][] content
 //! type.
 //!
 //! They’re formed with the following BNF:
@@ -8,6 +8,15 @@
 //! ; instead of ending it).
 //! hard_break_escape ::= '\\'
 //! ```
+//!
+//! Hard breaks in markdown relate to the HTML element `<br>`.
+//! See [*§ 4.5.27 The `br` element* in the HTML spec][html] for more info.
+//!
+//! It is also possible to create a hard break with a
+//! [hard break (trailing)][hard_break_trailing].
+//! That construct is not recommended because trailing spaces are typically
+//! invisible in editors, or even automatically removed, making them to use.
+//!
 //! It is also possible to escape punctuation characters with a similar
 //! construct: a [character escape][character_escape] is a backslash followed
 //! by an ASCII punctuation character.
@@ -22,12 +31,12 @@
 //! [text]: crate::content::text
 //! [character_escape]: crate::construct::character_escape
 //! [character_reference]: crate::construct::character_reference
-//!
-//! <!-- To do: link `hard_break_escape` -->
+//! [hard_break_trailing]: crate::construct::hard_break_trailing
+//! [html]: https://html.spec.whatwg.org/multipage/text-level-semantics.html#the-br-element
 
 use crate::tokenizer::{Code, State, StateFnResult, TokenType, Tokenizer};
 
-/// Start of a hard break escape.
+/// Start of a hard break (escape).
 ///
 /// ```markdown
 /// a|\
@@ -45,7 +54,7 @@ pub fn start(tokenizer: &mut Tokenizer, code: Code) -> StateFnResult {
     }
 }
 
-/// At the end of a hard break escape, after `\`.
+/// At the end of a hard break (escape), after `\`.
 ///
 /// ```markdown
 /// a\|
diff --git a/src/construct/hard_break_trailing.rs b/src/construct/hard_break_trailing.rs
new file mode 100644
index 0000000..46337c5
--- /dev/null
+++ b/src/construct/hard_break_trailing.rs
@@ -0,0 +1,83 @@
+//! Hard break (trailing) is a construct that occurs in the  [text][] content
+//! type.
+//!
+//! They’re formed with the following BNF:
+//!
+//! ```bnf
+//! ; Restriction: followed by a line ending  (that is part of the construct
+//! ; instead of ending it).
+//! hard_break_trailing ::= 2*' '
+//! ```
+//!
+//! The minimum number of the spaces is defined in
+//! [`HARD_BREAK_PREFIX_SIZE_MIN`][hard_break_prefix_size_min].
+//!
+//! Hard breaks in markdown relate to the HTML element `<br>`.
+//! See [*§ 4.5.27 The `br` element* in the HTML spec][html] for more info.
+//!
+//! It is also possible to create a hard break with a similar construct: a
+//! [hard break (escape)][hard_break_escape] is a backslash followed
+//! by a line ending.
+//! That construct is recommended because it is similar to a
+//! [character escape][character_escape] and similar to how line endings can be
+//! “escaped” in other languages.
+//! Trailing spaces are typically invisible in editors, or even automatically
+//! removed, making hard break (trailing) hard to use.
+//!
+//! ## References
+//!
+//! *   [`lib/initialize/text.js` in `micromark`](https://github.com/micromark/micromark/blob/main/packages/micromark/dev/lib/initialize/text.js)
+//! *   [*§ 6.7 Hard line breaks* in `CommonMark`](https://spec.commonmark.org/0.30/#hard-line-breaks)
+//!
+//! [text]: crate::content::text
+//! [hard_break_escape]: crate::construct::hard_break_escape
+//! [character_escape]: crate::construct::character_escape
+//! [hard_break_prefix_size_min]: crate::constant::HARD_BREAK_PREFIX_SIZE_MIN
+//! [html]: https://html.spec.whatwg.org/multipage/text-level-semantics.html#the-br-element
+
+use crate::constant::HARD_BREAK_PREFIX_SIZE_MIN;
+use crate::tokenizer::{Code, State, StateFnResult, TokenType, Tokenizer};
+
+/// Start of a hard break (trailing).
+///
+/// ```markdown
+/// a|  ␊
+/// b
+/// ```
+pub fn start(tokenizer: &mut Tokenizer, code: Code) -> StateFnResult {
+    match code {
+        Code::Char(' ') => {
+            tokenizer.enter(TokenType::HardBreakTrailing);
+            tokenizer.enter(TokenType::HardBreakTrailingSpace);
+            tokenizer.consume(code);
+            (State::Fn(Box::new(|t, c| inside(t, c, 1))), None)
+        }
+        _ => (State::Nok, None),
+    }
+}
+
+/// Inside the hard break (trailing).
+///
+/// ```markdown
+/// a  |␊
+/// b
+/// ```
+fn inside(tokenizer: &mut Tokenizer, code: Code, size: usize) -> StateFnResult {
+    match code {
+        Code::Char(' ') => {
+            tokenizer.consume(code);
+            (
+                State::Fn(Box::new(move |t, c| inside(t, c, size + 1))),
+                None,
+            )
+        }
+        Code::CarriageReturnLineFeed | Code::Char('\r' | '\n')
+            if size >= HARD_BREAK_PREFIX_SIZE_MIN =>
+        {
+            tokenizer.exit(TokenType::HardBreakTrailingSpace);
+            tokenizer.exit(TokenType::HardBreakTrailing);
+            (State::Ok, Some(vec![code]))
+        }
+        _ => (State::Nok, None),
+    }
+}
diff --git a/src/construct/mod.rs b/src/construct/mod.rs
index 27f4308..880d055 100644
--- a/src/construct/mod.rs
+++ b/src/construct/mod.rs
@@ -26,7 +26,8 @@
 //! *   [code (text)][code_text]
 //! *   content
 //! *   definition
-//! *   [hard break escape][hard_break_escape]
+//! *   [hard break (escape)][hard_break_escape]
+//! *   [hard break (trailing)][hard_break_trailing]
 //! *   [heading (atx)][heading_atx]
 //! *   heading (setext)
 //! *   [html (flow)][html_flow]
@@ -61,6 +62,7 @@ pub mod code_fenced;
 pub mod code_indented;
 pub mod code_text;
 pub mod hard_break_escape;
+pub mod hard_break_trailing;
 pub mod heading_atx;
 pub mod html_flow;
 pub mod html_text;