[Issue 10845] std.range.Cycle broken for reference type forward ranges

d-bugmail at puremagic.com d-bugmail at puremagic.com
Sun Aug 18 05:47:50 PDT 2013


http://d.puremagic.com/issues/show_bug.cgi?id=10845


monarchdodra at gmail.com changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
                 CC|                            |monarchdodra at gmail.com


--- Comment #1 from monarchdodra at gmail.com 2013-08-18 05:47:49 PDT ---
(In reply to comment #0)
> import std.stdio;
> import std.range;
> import std.algorithm;
> 
> void main()
> {
>     auto a = inputRangeObject(iota(3).filter!"true");
>     writeln(a.cycle.take(10));
> }
> 
> Outputs
> 
> [0, 1, 2, 0, 1, 2,
> 
> then asserts: http://dpaste.dzfl.pl/84de028d
> 
> The problem is Cycle.popFront:
> 
> void popFront()
> {
>     _current.popFront();
>     if (_current.empty) _current = _original;
> }
> 
> It should be _current = _original.save, otherwise the _original range is
> consumed on the second iteration.

I have an open pull for improving cycle:
https://github.com/D-Programming-Language/phobos/pull/1149

So I integrated the fix in it. That said, given how often my pulls get reviewed
and pulled, it might be best if you created a pull for this that _I_ could
review and pull...

-- 
Configure issuemail: http://d.puremagic.com/issues/userprefs.cgi?tab=email
------- You are receiving this mail because: -------


More information about the Digitalmars-d-bugs mailing list