(1) is certainly incorrect as you've got <body-content>tagdependent</body-content> and not <body-content>empty</body-content>.
(2) is correct - any tag
may be empty (but those declared as body-content empty
must be empty). All that happens is the body, which isn't even there, is never evaluated. Classic tags skip all the lifecycle methods relating to bodies if the tag is empty at translation time.
(3) is false - given that (2) is correct and it may be empty.
(4) is false - no-one said this had to be a BodyTag at all; for that matter it could equally well be a SimpleTag. The only difference is that for a simple tag the <body-content> value must not be
JSP, and it isn't here, so that could work.
(5) "It may take an attribute called 'name'" is true, but the wording in the second sentence is dodgy: "its value
must be dynamic". I don't agree - the fact that <rtexprvalue>true</rtexprvalue> is declared means that the tag attribute
can accept a dynamic runtime expression, not that it
must... In essence I suppose this last answer is correct, but I think "its value may be dynamic" would be more accurate.
[ April 28, 2006: Message edited by: Charles Lyons ]