CVE-2009-4214: Moderate severity vulnerability that affects rails

The vulnerability stems from HTML::Tokenizer's handling of non-printable characters in tag names. The critical change in node.rb modifies the regular expression that identifies HTML tag names during parsing. The original regex [\w:-]+ didn't account for non-printing characters (\x00-\x1f range), allowing attackers to craft tags with these characters that would bypass sanitization. This parse method is directly responsible for tokenizing HTML input for strip_tags, making it the core vulnerable function. The test case addition in sanitizer_test.rb demonstrates how non-printable characters could be used to create valid tags that should be stripped.