Alternative stripping HTML with regexp to https://github.com/6ftDan/regex-is-evil/blob/master/strip_html.rb

strip-html-2.rb

HTML_TAG_REPLACEMENTS = {
  'br' => "\n",
}
HTML_QUOTE_REPLACEMENTS = {
  'quot' => '"',
  'amp' => '&',
}

def strip_html(str, tag = HTML_TAG_REPLACEMENTS, quot = HTML_QUOTE_REPLACEMENTS)
  str.gsub %r{
    # first alternative: remove tags
    <
      (?:
        (?:(\w+) # tag name
	  # alternative: attributes
          (?: 
	    \s+ 
	    \w+  # attr name
	    =
	    (?:"[^"]*"|'[^']*') # attr value
	  )*
	  /? # optional tag closes also
        )
        |
        # alternative: closing tag
        (/\w+)
       )
    >
    |
    # second alternative: replace HTML entities
    &
    (\w+)
    ;
  }x do |m|
    tg = $1 || $2

    if tg
      tag[tg]
    else
      quot[$3] 
    end
  end
end

ARGV.each do |arg|
  printf "\nTest %p\n", arg
  p strip_html(arg)
end