I doubt you can extend it to all n-dimensional arrays of foo.
What you can do, though, is (defmacro add-support-for-dimension [n]
`(extend-protocol ...)) and document that users of your library should
invoke this macro once for each number of dimensions that will need
supporting. If that set's open-ended, they can construct and eval calls to
the macro at run time, though that's a bit icky, every time a new dimension
is first encountered.
If it's true that if you have a protocol extended to a class C and its
subclass S, and give it an S, it will always use the S specialization
rather than the C version, and more generally will use the most specific
extension that applies to the run-time type of an object, then you can
extend the protocol to Object and have the Object version test for an
array, handle that case generically, and throw an exception otherwise.
You could also petition for clojure to be amended to add dummy types for
x-dimensional arrays, arrays of foo, and x-dimensional arrays of foo
generally, so one could put something in extend-protocol to extend it to
all arrays of (ultimately) doubles or all three-dimensional arrays; but
clojure itself changes relatively slowly, so you'd likely have to wait a
year or two, if it ever did get done.
On Wed, Jun 19, 2013 at 1:59 PM, Jim - FooBar(); jimpil1...@gmail.comwrote:
Hi again,
quick question:
Is it at all possible to extend a protocol to a particular primitive array
type but also handle nesting of arrays? A double[] is not the same class as
double[][] which in turn is not the same Class as double[][][] etc etc and
there is no interface tying them together So this means I can't do the
regular trick that I can do with any other Collection type.
example:
(defprotocol Normalisable
(normalise [this transformer]))
(extend-protocol Normalisable
Number
(normalise [this transform]
(transform this))
String
(normalise [this stem]
(stem this))
java.util.List ;;if this fires, we're dealing with a Java list-like
collection - return a vector
(normalise [this transform]
(if (instance? java.util.Collection (first this)) ;;handle nesting at the
top interface and pray that an extension-point for that type exists
(mapv #(normalise % transform) this)
(let [top(delay (apply max this))
bottom (delay (apply min this))]
(mapv (fn [x] (normalise x #(transform % [top bottom]))) this))) )
and so forth for Collections... for arrays it's the same story (but with
amap) but it will only work for 1d arrays. Extending to object-arrays,
thinking that the outer array is an Object[], fails miserably. How on earth
am I supposed to handle nesting in a general way?
thanks in advance...
Jim
--
--
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+unsubscribe@**googlegroups.comclojure%2bunsubscr...@googlegroups.com
For more options, visit this group at
http://groups.google.com/**group/clojure?hl=enhttp://groups.google.com/group/clojure?hl=en
--- You received this message because you are subscribed to the Google
Groups Clojure group.
To unsubscribe from this group and stop receiving emails from it, send an
email to
clojure+unsubscribe@**googlegroups.comclojure%2bunsubscr...@googlegroups.com
.
For more options, visit
https://groups.google.com/**groups/opt_outhttps://groups.google.com/groups/opt_out
.
--
--
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
---
You received this message because you are subscribed to the Google Groups
Clojure group.
To unsubscribe from this group and stop receiving emails from it, send an email
to clojure+unsubscr...@googlegroups.com.
For more options, visit https://groups.google.com/groups/opt_out.
