https://issues.dlang.org/show_bug.cgi?id=12188
Iain Buclaw changed:
What|Removed |Added
Priority|P2 |P3
--
https://issues.dlang.org/show_bug.cgi?id=12188
--- Comment #6 from hst...@quickfur.ath.cx ---
Well, I'm out of ideas. Seems the only way around this is to make
nextPermutation require a random access range. :-(
--
https://issues.dlang.org/show_bug.cgi?id=12188
--- Comment #5 from Andrei Alexandrescu ---
Oh, yah, never mind.
--
https://issues.dlang.org/show_bug.cgi?id=12188
--- Comment #4 from Peter Alexander ---
(In reply to Andrei Alexandrescu from comment #3)
> It's easy to implement allBefore to return a new range type distinct (e.g.
> Voldemort) from R. That range would keep the two ranges as state and would
> say
https://issues.dlang.org/show_bug.cgi?id=12188
Andrei Alexandrescu changed:
What|Removed |Added
CC||and...@erdani.com
--- Comment #3 from
https://d.puremagic.com/issues/show_bug.cgi?id=12188
--- Comment #2 from hst...@quickfur.ath.cx 2014-02-18 13:14:25 PST ---
Hmm this sucks. It didn't occur to me that takeN on a bidirectional range
doesn't return a bidirectional range. :-( In retrospect, it makes sense (takeN
doesn't know the i
https://d.puremagic.com/issues/show_bug.cgi?id=12188
--- Comment #1 from Peter Alexander 2014-02-18
12:51:15 PST ---
BTW, you may already know this, but it's impossible to solve this with the
current range primitives. See the discussion here:
http://www.digitalmars.com/d/archives/digitalmars/
