Status of ranges for C++1z? [closed]

2019-02-08 22:39发布

问题:

There is a study group on ranges in the C++ committee:

but I have not followed the history of this study group and I am not sure of what kind of delivery is expected for C++1z (furthermore I do not use boost.range so I do not have a clear view of existing practices). Will we have:

  1. ranges as a pair of first/last iterators?

  2. union and other set operations on ranges(for example [v.begin()+5, v.begin()+7[ U [v.begin()+10, v.begin()+15[ U [v.begin()+21, v.begin()+42[), namely: union, intersection, disjoint union, complement?

  3. iterator filters (in order to execute a for_each where a condition is satisfied)?

  4. other things?

回答1:

Eric Niebler initiated discussion of his Iterable work (four-part blog posting: Part 1, Part 2, Part 3, Part 4) on the SG9 mailing list on Feb 27. There's been quite a bit of back and forth since then, especially considering that the list had been dead for about a year.

As near as I can tell as an outsider, the SG is in the "arguing about what they're trying to do" phase which is quite far from the "developing wording for a Standard/Technical Specification" phase. This is, however, a substantial move forward from the "dormant" phase. The meeting minutes from the last few committee meetings are illuminating:

  • 2013-05-03 (Bristol):

    Marshall Clow [the study group chair] reports group is currently suffering from an excess of vision. There are no papers. SG9 is trying to constrain scope. Nothing expected to go into C++14

  • 2013-10-16 (Chicago):

    Clow reports no activity in past few months. One paper to discuss [I believe N3752 "Index Based Ranges"]. Will have evening session.

  • 2014-03-03 (Issaquah):

    Clow reports no papers and no meeting. [Stephan] Lavavej reports a ranged for paper in EWG that the range people may care about [I believe referring to N3853 "Range-Based For-Loops: The Next Generation"]. Dawes asks if work is stalled. Clow says his work is stalled and he has not seen work others are doing. [Jeffrey] Yasskin says that he is unsure of where the papers are going which may be stalling SG9 consideration of papers.

It hasn't exactly been going like gangbusters, and there is no set deliverable goal AFAIK.