feat: add HTML stripping utility for Owncast emoji and markup

Made-with: Cursor

src/html.rs

@@ -0,0 +1,87 @@
+use std::borrow::Cow;
+
+/// Extracts alt text from <img> tags (for Owncast emoji) and strips all other HTML.
+pub fn strip_html(input: &str) -> String {
+    let mut result = String::with_capacity(input.len());
+    let mut chars = input.chars().peekable();
+
+    while let Some(&ch) = chars.peek() {
+        if ch == '<' {
+            chars.next(); // consume '<'
+            let tag: String = chars.by_ref().take_while(|&c| c != '>').collect();
+            if tag.starts_with("img ") || tag.starts_with("img\t") {
+                if let Some(alt) = extract_attr(&tag, "alt") {
+                    result.push_str(&alt);
+                }
+            }
+        } else if ch == '&' {
+            chars.next(); // consume '&'
+            let entity_name: String = chars.by_ref().take_while(|&c| c != ';').collect();
+            let full_entity = format!("&{}", entity_name);
+            result.push_str(&decode_entity(&full_entity));
+        } else {
+            result.push(ch);
+            chars.next();
+        }
+    }
+
+    result
+}
+
+fn extract_attr(tag: &str, attr_name: &str) -> Option<String> {
+    let pattern = format!("{}=\"", attr_name);
+    let start = tag.find(&pattern)? + pattern.len();
+    let rest = &tag[start..];
+    let end = rest.find('"')?;
+    Some(rest[..end].to_string())
+}
+
+fn decode_entity(entity: &str) -> Cow<'static, str> {
+    match entity {
+        "&amp" => Cow::Borrowed("&"),
+        "&lt" => Cow::Borrowed("<"),
+        "&gt" => Cow::Borrowed(">"),
+        "&quot" => Cow::Borrowed("\""),
+        "&#39" | "&apos" => Cow::Borrowed("'"),
+        "&nbsp" => Cow::Borrowed(" "),
+        other => Cow::Owned(other.to_string()),
+    }
+}
+
+#[cfg(test)]
+mod tests {
+    use super::*;
+
+    #[test]
+    fn test_plain_text_unchanged() {
+        assert_eq!(strip_html("hello world"), "hello world");
+    }
+
+    #[test]
+    fn test_strips_basic_tags() {
+        assert_eq!(strip_html("<b>bold</b> text"), "bold text");
+    }
+
+    #[test]
+    fn test_emoji_img_to_alt_text() {
+        let input = r#"hello <img class="emoji" alt=":beerparrot:" title=":beerparrot:" src="/img/emoji/beerparrot.gif"> world"#;
+        assert_eq!(strip_html(input), "hello :beerparrot: world");
+    }
+
+    #[test]
+    fn test_multiple_emoji() {
+        let input = r#"<img class="emoji" alt=":a:" src="/a.gif"><img class="emoji" alt=":b:" src="/b.gif">"#;
+        assert_eq!(strip_html(input), ":a::b:");
+    }
+
+    #[test]
+    fn test_strips_links() {
+        let input = r#"check <a href="https://example.com">this link</a>"#;
+        assert_eq!(strip_html(input), "check this link");
+    }
+
+    #[test]
+    fn test_decodes_html_entities() {
+        assert_eq!(strip_html("a &amp; b &lt; c"), "a & b < c");
+    }
+}

src/main.rs

@@ -1,5 +1,6 @@
 mod config;
 mod events;
+mod html;
 
 fn main() {
     println!("owncast-irc-bridge");