Hello Laurent, Thank you so much for your detailed and helpful reply.
> You could use the new as-> macro in clojure 1.5 Is it possible to get as-> to work with thread-last? > depends :-) since your 2 fns are helper fns rather than > generically-reusable fns, I would make them private to enforce that : > (defn- ...) Good point, and I'll definitely try to remember that in the future. > May I suggest alternative implementations ? > <snip> I really like both of your alternative solutions, and have learned several new ideas from them. Thank you. By the way, I'm using the Clojure Cheat Sheet (http://clojure.org/cheatsheet) as my guide to the core functions, but it doesn't seem to be up-to-date with all these fancy 1.5 functions that you're using. Is there a new website that has superseded CCS? Do you know why CCS isn't kept up-to-date? Regards, James On Thursday, July 11, 2013 2:29:38 PM UTC+2, Laurent PETIT wrote: > > Hello James, > > 2013/7/11 James Trunk <james...@gmail.com <javascript:>>: > > Hi everyone, I'm new to Clojure and trying to learn my way around the > > language. > > > > I've written a function that sums the frequencies of words starting with > the > > same letter, however I'm not fully satisfied with the result. I'm > striving > > for readability and idomaticity, but I fear that my rather limited grasp > of > > Clojure's core functions and lack of experience with functional > programming > > are letting me down on both counts. > > > > (def text (slurp "http://www.ccel.org/ccel/bible/kjv.txt";)) > > > > (defn sum-last-elements [coll] > > (reduce + (map last coll))) > > > > (defn group-first-elements [coll] > > (map first coll)) > > > > (->> text > > (re-seq > > #"(?:(?:\bm[a|e]n\b)|(?:\bwom[a|e]n\b)|(?:\bchild\b)|(?:\bchildren\b))") > > frequencies > > (group-by ffirst) > > vals > > (#(zipmap (map group-first-elements %) (map sum-last-elements %)))) > > > > Which on that file gives the result: > > > > {("woman" "women") 663, ("children" "child") 2274, ("men" "man") 5314} > > > > My questions: > > > > Is there a way to avoid using ?: everywhere in the regex? > > (group-by ffirst) vals - is there a more readable/declarative/idiomatic > way > > to group by first letter? > > you could first (group-by first), then create the final map by calling > (distinct) and count > > > Is there a trick to avoid the ugly looking anonymous function (which I'm > > only using to reorder the thread-last argument)? > > You could use the new as-> macro in clojure 1.5 > > > Is it idiomatic to extract small functions to give them names (with an > eye > > to aiding readability) or would most Clojure programmers prefer these to > be > > anonymous and in-line? > > depends :-) since your 2 fns are helper fns rather than > generically-reusable fns, I would make them private to enforce that : > (defn- ...) > > > Is thread-last the right approach when dealing with nested > data-structures > > or is list comprehension, or some other approach preferred? > > The thread-last as you employed it is fine by me. > > May I suggest alternative implementations ? > > (def text (slurp "http://www.ccel.org/ccel/bible/kjv.txt";)) > (let [words (re-seq > #"(?:(?:\bm[a|e]n\b)|(?:\bwom[a|e]n\b)|(?:\bchild\b)|(?:\bchildren\b))" > text) > words-per-initial (vals (group-by first words))] > (zipmap > (map distinct words-per-initial) > (map count words-per-initial))) > > ;; => {("woman" "women") 663, ("children" "child") 2274, ("men" "man") > 5314} > > > > or to avoid the intermediary creation of seqs of distincts and seqs of > counts: > > (defn- reduce-initial [m initial initial-words] > (assoc m (distinct initial-words) (count initial-words))) > > (let [words (re-seq > #"(?:(?:\bm[a|e]n\b)|(?:\bwom[a|e]n\b)|(?:\bchild\b)|(?:\bchildren\b))" > text) > words-per-initial (group-by first words)] > (reduce-kv reduce-initial {} words-per-initial)) > > ;; => {("woman" "women") 663, ("children" "child") 2274, ("men" "man") > 5314} > > > HTH, > > -- > Laurent > -- -- 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.
