Data::Dumper::Simple is already on the CPAN and it's my third upload of the module. The idea was trivial to implement, but it's surprising how many weird little issues popped up with it. It works quite well, though, despite source filtering issues. I've had several people thank me for the module (and it's only been out two days!) and at least one request to port the work to another module of similar intent.
In other news, I'm thinking about a serious reworking of the HTML::TokeParser::Simple interface. I'm rather embarrassed about how I put it together. I need to drop the return_ portion of accessor names. What was I thinking? Every time I use the module I try and type $token->attr('foo') and moan about it not working. If the author keeps misusing a module, there's a serious problem. I'll probably deprecate these methods and leave them in as wrappers to the new methods (and I might refactor the code base to get rid of the awful &_synch_arrays method.) Still, I hate to do this as this is my most popular module and it's been out there for quite some time. Yuck.
I'm also thinking about (gasp!) changing the default behavior of the set_ and get methods to automatically encode and decode HTML entities. Or should that be an option? I think default is a much better choice, but that also involves changing the interface. Maybe optional at first and then have it be default in the 3.0 version. Suggestions welcome.