use html_escape::decode_html_entities;
use regex::Regex;

pub fn normalize_news_content(content: Option<String>) -> Option<String> {
    content.as_ref()?;
    let content = content.unwrap();

    let re_tags = Regex::new("<[^>]+>").unwrap();
    let re_spaces = Regex::new("[\\u00A0\\s]+").unwrap();

    let content = content.replace('\n', " ");
    let content = re_tags.replace_all(&content, "");
    let content = re_spaces.replace_all(&content, " ");
    let content = decode_html_entities(&content);
    let content = content.trim();

    if content.is_empty() {
        None
    } else {
        Some(content.to_string())
    }
}