Re: [Boost-users] [boost] Interest in a Unicode library for Boost?
On Wed, Oct 30, 2019 at 9:47 AM Peter Koch Larsen < peter.koch.larsen@gmail.com> wrote:
I saw that talk (on video) and was quite impressed but never got to try your library out. Would be happy to use a library like this! Can you tell me how it relates to the proposed std?
There is no relationship yet. Boost.Text is an experiment meant to gather feedback, which will in turn drive the standardization effort. I believe all large libraries should have Boost-equivalent visibility in the C++ community before being standardized. That being said, some of my fellow SG-16 members (that's the committee's Unicode Study Group) have started proposing some actual APIs for standardization. The first of these is the most low-level: transcoding (see http://www.open-std.org/jtc1/sc22/wg21/docs/papers/2019/p1629r0.html ). JeanHeyd's approach is more ambitious than mine -- I really only care about Unicode. His proposal covers conversions among all possible encodings, including an API that lets users add implementations for the encodings they care about that are not covered by the standard. There are no other in-flight Unicode API proposals that I'm aware of. Zach
This proposed Text library has facilities that perform dynamic allocation, but do not support custom allocators? i.e. Not even the polymorphic allocators? (even if not the C++ allocator model which would require facilities become templates) I recall this was the case a while ago, not sure if intentions have changed since. Glen
On Wed, Oct 30, 2019 at 11:20 AM Glen Fernandes
This proposed Text library has facilities that perform dynamic allocation, but do not support custom allocators?
i.e. Not even the polymorphic allocators? (even if not the C++ allocator model which would require facilities become templates)
I recall this was the case a while ago, not sure if intentions have changed since.
Yes, that's right. There are no allocators. Zach
participants (2)
-
Glen Fernandes -
Zach Laine