![[+]](./images/find.png "show/hide quicksearch"){kind=small}
Currently Jam does a transform on selectors, converting what look like tag selectors to id selectors and vice-versa. Maybe this good, becuase usually id’s are used to jam data into templates. However it is non-standard and thus less flexible. So the nominal behavior will be made to use regular selectors. An alternate method will be provided to handle the selector transform though.
Add erase attribute, jam=“erase” or erase=“true” or other? This will be used to to remove the tag from the xml/xhtml.
It would be much easier to create cherry, and then use that to provide different backend adapters.
I may limit direct support to Nokogiri and let Cherry handle the others. Hpricot adapter works pretty well though (despite the unit test bombing). So maybe support Hpricot directly too. I’d like to support LibXML-Ruby directly since I helped with that project for so long, but it’s API is not quite as easy to support. So… get Nokogiri working first at least.