We had the review of this library proposal/addition quite a while back and ever since I've been [eagerly] waiting for its official entry into Boost. https://github.com/erenon/double_ended does not show any activity and there is nothing in the Boost.Container going on either. What is the status? degski -- *“If something cannot go on forever, it will stop" - Herbert Stein*
On 24/11/2018 7:55, degski via Boost wrote:
We had the review of this library proposal/addition quite a while back and ever since I've been [eagerly] waiting for its official entry into Boost.
https://github.com/erenon/double_ended does not show any activity and there is nothing in the Boost.Container going on either.
What is the status?
There is no new activity. I haven't contacted with Benedek because I've been too busy to try to properly merge it into Boost.Container. I hope to find some time to resume some work in christmas holidays. Nice to hear there is interest. Any preference on what container would like to get first so that we can focus our efforts? Ion
On Sat, 24 Nov 2018 at 13:28, Ion Gaztañaga via Boost
Nice to hear there is interest. Any preference on what container would like to get first so that we can focus our efforts?
boost::container::devector would be what I am waiting for. degski -- *“If something cannot go on forever, it will stop" - Herbert Stein*
On Sat, Nov 24, 2018 at 7:55 AM degski via Boost
We had the review of this library proposal/addition quite a while back and ever since I've been [eagerly] waiting for its official entry into Boost.
https://github.com/erenon/double_ended does not show any activity and there is nothing in the Boost.Container going on either.
What is the status?
degski --
Hi, Thanks for the interest! As per the review verdict, the double ended classes should be part of Boost.Container, with the following major changes: - support of pre-C++11 compilers (e.g: move emulation) - integration with the Container util components and extras - integration with the Container test practices and documentation - removal of the built-in small buffer optimization (should be provided by the allocator) Judging by the review activity and results, my perception was, that there isn't a great interest in these components - therefore I decided to focus my attention on different projects. Of course, given the need, we can change that. Do you have any particular devector feature you really want? Thanks, Benedek
On Sat, 24 Nov 2018 at 22:16, Benedek Thaler
Judging by the review activity and results, my perception was, that there isn't a great interest in these components
I had a different impression. I thought it was quite a lively discussion and the interest increased as the discussion went on (and a number of decisions got changed [on paper at least]). As Boost local container guru Ion Gaztañaga was in on the discussion, I think many just left the discussion develop and observe that it was [mostly] going in the right direction. To me devector seems very interesting as it's got great performance characteristics in view of the functionality it provides. And just anecdotally, plf::colony got boldly rejected [on the ML] for entry into Boost, while it will be std::colony in C++20, so go reckon. degski -- *“If something cannot go on forever, it will stop" - Herbert Stein*
participants (4)
-
Benedek Thaler -
degski -
Glen Fernandes -
Ion Gaztañaga