Parsing inline code span with a single newline loses the newline

[mgeisler]

According to the Commonmark playground and the Commonmark spec, the string

"`\n`"

should be parsed into

[
  Start(Paragraph)
  Code(" ")
  End(Paragraph)
]

since newline is normalized as a space.

However, when parsing the string with pulldown-cmark 0.9.3, I get

[
  Start(Paragraph)
  Code("")
  End(Paragraph)
]

instead. I used this little test program:

use pulldown_cmark::{Event, Parser};

fn main() {
    let text = "`\n`";
    eprintln!("{text:?} -> [");
    for event in Parser::new(&text) {
        match event {
            Event::Code(code) => eprintln!("  Code({:?})", &*code),
            _ => eprintln!("  {event:?}"),
        }
    }
    eprintln!("]");
}

Note that "`foo\n`" and "`\nfoo`" are parsed correctly with a trailing and leading space, respectively.

I found this while trying round-trip strings through pulldown-cmark-to-cmark (Byron/pulldown-cmark-to-cmark#55).

[Martin1887]

I will check this for the next release, thanks for reporting it.

[mgeisler]

I will check this for the next release, thanks for reporting it.

Thanks for taking a look!

[Martin1887]

Check if fixed by #776.