Alan Malloy <a...@malloys.org> writes:

> Stuff like this always worries me. It may happen to work now, but I
> doubt if that's a guarantee. extend-protocol's contract is that you
> give it a Symbol, which it resolves into a Class. It surely isn't
> expecting a Class as an argument, because you can't enter that as a
> source-code literal. If you give it a Class, it could conceivably do
> something dreadful like try to call (resolve s) on it to figure out
> what Class you mean, and that will fail when passed a Class.

I'm not entirely happy with it either, but I don't think the situation
is quite that dire.  The docstring for `extend-protocol' doesn't say
anything about *how* it divides up the stanzas for extended types, and
as pointed out elsewhere in this thread the implementation uses seq vs
not-seq, not symbol? vs not-symbol?.  The docstrings for `extend-type'
and `extend' explicitly say that they take a "type/class" -- I'd
interpret that to mean that they can be called with anything which
resolves to a type/class via normal resolution methods, including a
literal value generated via #=.

> For that matter, trying out your example, it doesn't seem to work for
> me. The first fix is to use java.lang.Class instead of just Class,
> since reader evaluation happens in a no-frills environment.

Apologies -- I just modified the OP's original example without thinking
it all the way through or running the code.  One does need to specify
java.lang.Class/forName, which is what I've been doing in my own source.

> But after that, it seems the extend-protocol just silently does
> nothing:

(Addressed by another poster.)

-Marshall

-- 
You received this message because you are subscribed to the Google
Groups "Clojure" group.
To post to this group, send email to clojure@googlegroups.com
Note that posts from new members are moderated - please be patient with your 
first post.
To unsubscribe from this group, send email to
clojure+unsubscr...@googlegroups.com
For more options, visit this group at
http://groups.google.com/group/clojure?hl=en

Reply via email to