0001-make-interleave-handle-odd-arugments-in-the-same-man.patch
clj-863-make-interleave-handle-odd-args-like-concat-patch-v1.txt
[CLJ-863] interleave should accept 1 or 0 arguments Created: 24/Oct/11 Updated: 24/May/13 Resolved: 24/May/13 |
|
| Status: | Closed |
| Project: | Clojure |
| Component/s: | None |
| Affects Version/s: | Release 1.3 |
| Fix Version/s: | Release 1.6 |
| Type: | Enhancement | Priority: | Trivial |
| Reporter: | Joe Gallo | Assignee: | Unassigned |
| Resolution: | Completed | Votes: | 5 |
| Labels: | None | ||
| Attachments: |
0001-make-interleave-handle-odd-arugments-in-the-same-man.patch
clj-863-make-interleave-handle-odd-args-like-concat-patch-v1.txt
|
| Patch: | Code and Test |
| Approval: | Ok |
| Description |
|
interleave should handle 0 and 1 arguments in the same way that concat does (i.e., 0 args --> empty seq, 1 args --> identity). |
| Comments |
| Comment by Rich Hickey [ 15/Jun/12 9:31 AM ] |
|
(lazy-seq nil) should be () |
| Comment by Joe Gallo [ 15/Jun/12 10:13 AM ] |
|
Hey Rich, if you're talking about the first line of the diff: + ([] (lazy-seq nil)) Then that's the implementation, not the tests – given an empty vector of arguments, return (lazy-seq nil), which I just copied from the existing definition of concat. Cheers, |
| Comment by Marc Dzaebel [ 03/Oct/12 1:19 PM ] |
|
(defn interleave [& s] (apply mapcat list s)) |
| Comment by Matthew O. Smith [ 03/Oct/12 8:47 PM ] |
|
Marc's definition doesn't work for no arguments. Maybe: (defn interleave |
| Comment by Marc Dzaebel [ 05/Oct/12 1:07 PM ] |
|
Yes, but my solution is too slow, as it uses "apply". |
| Comment by Andy Fingerhut [ 01/Jan/13 11:54 AM ] |
|
Patch clj-863-make-interleave-handle-odd-args-like-concat-patch-v1.txt dated Jan 1 2013 is identical to Joe Gallo's 0001-make-interleave-handle-odd-arugments-in-the-same-man.patch patch dated Oct 24 2011, except it returns () instead of (lazy-seq nil) as per Rich's comment. If concat should also return () instead of (lazy-seq nil), perhaps another ticket should be created to fix that. Also presumptuously changing ticket state from Incomplete back to Vetted, since the reason it was marked Incomplete should now be addressed, and it was Screened before. |