Thanks I'll take a look at your libraries.
One problem I found with the zipper API is that if you have a recursive
function that takes a loc, when you call it on the clojure.zip/children of
a given loc it won't work because this function returns a seq of nodes
instead of a seq of locs. Shouldn't we add a function like
children-as-locs to the API for this purpose or is there another way?
(defn children-as-locs
"Returns a seq of locs of the children of loc."
[loc]
(when (z/branch? loc)
(let [[node path] loc
cs (z/children loc)]
(map-indexed (fn [i c]
(with-meta [c {:l (vec (take i cs))
:pnodes (if path (conj (:pnodes path)
node) [node])
:ppath path
:r (vec (drop (inc i) cs))}]
(meta loc)))
cs))))
On Wednesday, December 4, 2013 11:46:29 PM UTC, Alexander Hudek wrote:
>
> Additionally, if you need more complex access patterns you could see if
> this helps:
>
> https://github.com/akhudek/zip-visit
>
> For performance, there is a fast-zip library that is api compatible with
> clojure.zip. You can just swap the clojure.zip namespace for the fast-zip
> namespace. Note that you'd need to make the same swap in any libraries you
> use with your zippers (like zip-visit) since fast-zip is not implementation
> compatible with clojure.zip.
>
> More broadly, I've found that straight up recursive algorithms can be a
> lot faster than zippers, especially for read only operations. Of course the
> stack size limits what you can do with tree recursion. As others have said,
> it's best to see if you actually have performance problems first.
>
> On Wednesday, December 4, 2013 5:30:02 PM UTC-5, James Reeves wrote:
>>
>> On 4 December 2013 21:09, dabd <[email protected]> wrote:
>>
>>> I didn't get there because I ran into problems with the zipper API.
>>> When you call 'children' on a loc you get a seq of nodes instead of a seq
>>> of locs which causes me problems in a recursive algorithm operating on locs.
>>>
>>
>> Have you tried using next and end? for a depth-first transversal?
>>
>> - James
>>
>
--
--
You received this message because you are subscribed to the Google
Groups "Clojure" group.
To post to this group, send email to [email protected]
Note that posts from new members are moderated - please be patient with your
first post.
To unsubscribe from this group, send email to
[email protected]
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 [email protected].
For more options, visit https://groups.google.com/groups/opt_out.
