Merge pull request #795 from Silvertree2010/fix/h814-html-paste-drops-text-from-a-bar

This commit is contained in:
Brandon Thomas
2026-10-18 11:16:17 -04:00
+24 -14
View File
@@ -274,20 +274,22 @@ fn decode_text(raw: &str) -> String {
break;
};
let entity = &rest[1..end];
rest = &rest[end + 1..];
match entity {
"amp" => out.push('&'),
"lt" => out.push('<'),
"gt" => out.push('>'),
"quot" => out.push('"'),
"apos" => out.push('\''),
"nbsp" => out.push(' '),
_ => {
let ch = entity.strip_prefix('#').and_then(numeric_char);
if let Some(c) = ch {
out.push(c);
}
}
let decoded = match entity {
"amp" => Some('&'),
"lt" => Some('<'),
"gt" => Some('>'),
"quot" => Some('"'),
"apos" => Some('\''),
"nbsp" => Some(' '),
_ => entity.strip_prefix('#').and_then(numeric_char),
};
if let Some(c) = decoded {
out.push(c);
rest = &rest[end + 1..];
} else {
// Not a well-formed entity (e.g. a bare "R&D"): keep the '&' as typed.
out.push('&');
rest = &rest[1..];
}
}
if !rest.is_empty() && !rest.starts_with('&') {
@@ -350,6 +352,14 @@ mod tests {
assert!(last.props.underline.is_some());
}
#[test]
fn bare_ampersand_keeps_text_up_to_next_semicolon() {
let text = |html: &str| html_to_body(html).unwrap().text();
assert_eq!(text("<p>R&D costs; up</p>"), "R&D costs; up");
assert_eq!(text("<p>AT&T; Q&A</p>"), "AT&T; Q&A");
assert_eq!(text("<p>A &amp; B &#38; C &#x26; D</p>"), "A & B & C & D");
}
#[test]
fn pages_style_and_escapes() {
let html = "<html><head><style>p{}</style></head><body><p><span style=\"font-weight: bold\">A &amp; B</span></p><p><span style=\"font-style: italic\">C</span></p></body></html>";