Very clever!
On Fri, Jan 10, 2014 at 8:10 PM, Håkan Råberg hakan.rab...@gmail.comwrote:
Another style, using channels for local state, but could been plain old
iterators, slight golf warning:
(require '[clojure.core.async :refer [to-chan !!]])
(defn uniquify [s formatter]
(let [g
I've been finding uses for Brandon Bloom's transduce mini-library left and
right lately. There is a class of problems where you want to track some
state as you process a seq, and transduce.lazy/map-state enables you to do
that (https://github.com/brandonbloom/transduce).
Here's my solution using
On Sat, January 11, 2014 9:22 pm, Alex Baranosky wrote:
There is a class of problems where you want to track some state as you
process a seq
This is an interesting exercise. I find said class of problem comes up
fairly frequently and I usually end up with something scary-looking using
reduce or
Hi Colin,
Clojure has a distinct function that does this. I may be missing some
context on what you what out of your new method that 'distinct' does not
provide. The distinct functions source code is a good reference.
Thanks
On Fri, Jan 10, 2014 at 6:59 AM, Colin Yates colin.ya...@gmail.com
I would not use an atom. Think about it as doing a reduce while passing
along a set of the names you've seen so far. You might also look at the
implementation of distinct in clojure.core which is similar (you want to
detect duplicates in the same way, but emit new names instead of omitting
07:10:22 -0800
Subject: Re: How can I improve this?
From: grd...@gmail.com
To: clojure@googlegroups.com
Hi Colin,
Clojure has a distinct function that does this. I may be missing some context
on what you what out of your new method that 'distinct' does not provide. The
distinct functions source
Good call.
I keep discounting reduce as I am not 'reducing' anything, only
transforming (i.e. map) it - my mistake.
Thanks.
Col
On Friday, 10 January 2014 15:12:27 UTC, Alex Miller wrote:
I would not use an atom. Think about it as doing a reduce while passing
along a set of the names
On 10 January 2014 14:59, Colin Yates colin.ya...@gmail.com wrote:
I have a sequence of file names and I want to make them unique. (uniquify
[a b c a]) = [a b c a_1])
This is what I have come up with, but surely there is a better way?
I would do something like:
(defn uniquify
([xs]
Hi,
Use frequencies to get a map of path = nb of occurrences, then for each
entry of the map, create unique names.
Cannot provide an impl on the uPhine, sorry
Le vendredi 10 janvier 2014, Colin Yates a écrit :
I have a sequence of file names and I want to make them unique. (uniquify
[a b c
Love it. Much more readable without any nasty persistent state.
On Friday, 10 January 2014 15:19:03 UTC, Ray Miller wrote:
On 10 January 2014 14:59, Colin Yates colin...@gmail.com javascript:wrote:
I have a sequence of file names and I want to make them unique.
(uniquify [a b c a]) = [a b
I did consider that but I want to preserve the order of the incoming
sequence.
On Friday, 10 January 2014 15:22:29 UTC, Laurent PETIT wrote:
Hi,
Use frequencies to get a map of path = nb of occurrences, then for each
entry of the map, create unique names.
Cannot provide an impl on the
...
--
Date: Fri, 10 Jan 2014 07:10:22 -0800
Subject: Re: How can I improve this?
From: grd...@gmail.com
To: clojure@googlegroups.com
Hi Colin,
Clojure has a distinct function that does this. I may be missing some
context on what you what out of your new method
I quickly put together this which seems to preserver the orderof the
original seq:
(defn uniquify [coll]
(let [post-fn #(group-by first (- % meta :encountered))]
(loop [unique (with-meta [] {:encountered []})
[f more] coll]
(if (nil? f) (flatten (concat unique (reduce #(conj %
Le vendredi 10 janvier 2014, Colin Yates a écrit :
I did consider that but I want to preserve the order of the incoming
sequence.
On Friday, 10 January 2014 15:22:29 UTC, Laurent PETIT wrote:
Hi,
Use frequencies to get a map of path = nb of occurrences, then for each
entry of the map,
actually `post-fn` should be #(group-by identity (- % meta :encountered))
Jim
On 10/01/14 15:28, Jim - FooBar(); wrote:
I quickly put together this which seems to preserver the orderof the
original seq:
(defn uniquify [coll]
(let [post-fn #(group-by first (- % meta :encountered))]
(loop
oops! my fn will not keep the original ordering...sorry Colin
Jim
On 10/01/14 15:34, Jim - FooBar(); wrote:
actually `post-fn` should be #(group-by identity (- % meta :encountered))
Jim
On 10/01/14 15:28, Jim - FooBar(); wrote:
I quickly put together this which seems to preserver the
On 10/01/14, Colin Yates wrote:
I have a sequence of file names and I want to make them unique. (uniquify
[a b c a]) = [a b c a_1])
This is what I have come up with, but surely there is a better way?
What would you all do? Feedback welcome (including the word 'muppet' as I
am sure I
2014/1/10 Stefan Kanev stefan.ka...@gmail.com
On 10/01/14, Colin Yates wrote:
I have a sequence of file names and I want to make them unique.
(uniquify
[a b c a]) = [a b c a_1])
This is what I have come up with, but surely there is a better way?
What would you all do? Feedback
Here's a version using reduce:
(defn uniquify [items]
(first
(reduce (fn [[result count-map] item]
(let [n (inc (count-map item 0))]
[(conj result (str item _ n))
(assoc count-map item n)]))
[[] {}]
Somehow I totally forgot I could use destructuring. Here's a slightly
shorter version:
(defn uniquify [words]
(loop [encountered {}
result []
[word remaining] words]
(if (seq remaining)
(let [occurences (get encountered word)
no you have a bug in this last version, it now skips the last result
2014/1/10 Stefan Kanev stefan.ka...@gmail.com
Somehow I totally forgot I could use destructuring. Here's a slightly
shorter version:
(defn uniquify [words]
(loop [encountered {}
result []
What about this one?
Inspired by Stefan's, with more destructuring in loop, format-fn as a
function, initial call to (seq) then (next) instead of (rest), placing the
exit argument first so that it's not lost at the end of the function,
renamed word as item since this function does not depend on
On 10/01/14, Laurent PETIT wrote:
What about this one?
Inspired by Stefan's, with more destructuring in loop, format-fn as a
function, initial call to (seq) then (next) instead of (rest), placing the
exit argument first so that it's not lost at the end of the function,
renamed word as item
If all you need is unqiueness, why not just number *all* the filenames in
sequential order, something like:
(defn uniqueify [filenames]
(map (fn [filename number] (str filename \_ number)) filenames (iterate
inc 1)))
--
--
You received this message because you are subscribed to the Google
java.jdbc does this for column names (in joins):
https://github.com/clojure/java.jdbc/blob/master/src/main/clojure/clojure/java/jdbc.clj#L257
Sean
On Jan 10, 2014, at 6:59 AM, Colin Yates colin.ya...@gmail.com wrote:
I have a sequence of file names and I want to make them unique. (uniquify
But, for the given problem this may not be directly helpful. This method
only returns the next unique name for the given list. The logic would have
to traverse the list n times for n elements (and some factor of no of
duplicates) to get the desired result, correct?
Thanks
Guru
On Fri, Jan 10,
This and Jonas' are my current favourites, for what that's worth.
Keep the suggestions coming!
On Friday, 10 January 2014 17:29:20 UTC, Laurent PETIT wrote:
What about this one?
Inspired by Stefan's, with more destructuring in loop, format-fn as a
function, initial call to (seq) then
Actually, you might have meant line 267?
On Fri, Jan 10, 2014 at 11:03 AM, Sean Corfield s...@corfield.org wrote:
java.jdbc does this for column names (in joins):
https://github.com/clojure/java.jdbc/blob/master/src/main/clojure/clojure/java/jdbc.clj#L257
Sean
On Jan 10, 2014, at 6:59
Technically, all these solutions are flawed.
With the input
[a a a_1]
you'll get back
[a a_1 a_1]
To truly address this, you need to also add the newly formatted filename
into the seen map, which none of the suggested solutions do.
--
--
You received this message because you are subscribed to
On Fri, Jan 10, 2014 at 9:39 PM, Mark Engelberg mark.engelb...@gmail.comwrote:
Technically, all these solutions are flawed.
With the input
[a a a_1]
you'll get back
[a a_1 a_1]
To truly address this, you need to also add the newly formatted filename
into the seen map, which none of the
On Fri, Jan 10, 2014 at 11:43 AM, Jonas Enlund jonas.enl...@gmail.comwrote:
That's why I wrote my solution like I did, i.e., concatenate _1 when a
new string is found. This would result in the vector [a_1 a_2 a_1_1]
Right, I agree that works, as does my tack unique numbers onto the end of
way to take the wind out of our sails! Well spotted :).
On Friday, 10 January 2014 19:39:45 UTC, puzzler wrote:
Technically, all these solutions are flawed.
With the input
[a a a_1]
you'll get back
[a a_1 a_1]
To truly address this, you need to also add the newly formatted filename
On Fri, Jan 10, 2014 at 10:22 AM, Laurent PETIT laurent.pe...@gmail.comwrote:
Hi,
Use frequencies to get a map of path = nb of occurrences, then for each
entry of the map, create unique names.
Cannot provide an impl on the uPhine, sorry
uPhine? :)
--
--
You received this message because
On Fri, Jan 10, 2014 at 11:52 AM, Colin Yates colin.ya...@gmail.com wrote:
way to take the wind out of our sails! Well spotted :).
It's not too hard to fix. Here's an adapted version of Jonas' solution
that should do the trick:
(defn uniqueify [items]
(first
(reduce (fn [[results
I thought I would have a go myself without copying (although having read
them earlier) the other functions and this is what I came up with:
(first (reduce (fn [[results seen] item]
(let [occurrences ((fnil identity 0) (get seen item))
seen (assoc
On Fri, Jan 10, 2014 at 12:55 PM, Colin Yates colin.ya...@gmail.com wrote:
Being really anal I could claim the original a_2 should remain a_2 and the
third instance of a jump to being a_3.
Sure, but that would require two passes. Otherwise, there's no way when
you encounter the third a to
Gosh - my public humiliation continues. Here is one that actually works:
(first (reduce (fn [[results seen] item]
(let [cnt (get seen item 0)]
[(conj results (if ( cnt 0) (format-fn item cnt)
item))
(assoc seen item (inc
okay, new take solving the issue raised by Mark:
(defn uniquify [in format-fn]
(loop [[item :as in] (seq in)
{n item :as item-nbrs} {}
out []]
(if-not in
out
(let [format-fn (if n format-fn (constantly item))
new-item (format-fn item n)]
Laurent, your approach doesn't quite work:
= (uniquify [a_1 a a] (fn [s n] (str s \_ n)))
[a_1 a a_1]
On Fri, Jan 10, 2014 at 1:34 PM, Laurent PETIT laurent.pe...@gmail.comwrote:
okay, new take solving the issue raised by Mark:
(defn uniquify [in format-fn]
(loop [[item :as in] (seq in)
Indeed, I should definitely recur as you do
Le vendredi 10 janvier 2014, Mark Engelberg a écrit :
Laurent, your approach doesn't quite work:
= (uniquify [a_1 a a] (fn [s n] (str s \_ n)))
[a_1 a a_1]
On Fri, Jan 10, 2014 at 1:34 PM, Laurent PETIT laurent.pe...@gmail.comwrote:
okay, new
Colin Yates colin.ya...@gmail.com writes:
This and Jonas' are my current favourites, for what that's worth.
Keep the suggestions coming!
here's another one that uses reductions. formatter would need to be
adapted a bit, since it currently also adds _0 for the first name
seen:
(defn
I meant the code that starts at line 257 (and continues to line 274): two
functions, the second one calls the first one.
Luckily, java.jdbc's code seems to pass all the test cases posted to this
thread so far (arguably more intuitively, the second occurrence gets _2
appended, the third _3
On 10 January 2014 21:06, Colin Yates colin.ya...@gmail.com wrote:
Gosh - my public humiliation continues. Here is one that actually works:
(first (reduce (fn [[results seen] item]
(let [cnt (get seen item 0)]
[(conj results (if ( cnt 0)
On 11 January 2014 01:03, Alan Forrester
alanmichaelforres...@googlemail.com wrote:
On 10 January 2014 21:06, Colin Yates colin.ya...@gmail.com wrote:
Gosh - my public humiliation continues. Here is one that actually works:
(first (reduce (fn [[results seen] item]
(let
On 11 January 2014 01:14, Alan Forrester
alanmichaelforres...@googlemail.com wrote:
On 11 January 2014 01:03, Alan Forrester
alanmichaelforres...@googlemail.com wrote:
On 10 January 2014 21:06, Colin Yates colin.ya...@gmail.com wrote:
Gosh - my public humiliation continues. Here is one that
Another style, using channels for local state, but could been plain old
iterators, slight golf warning:
(require '[clojure.core.async :refer [to-chan !!]])
(defn uniquify [s formatter]
(let [g (memoize #(to-chan (cons % (map (partial formatter %) (next
(range))]
(map (fn f [x]
46 matches
Mail list logo
