On 08/17/2015 09:19 AM, Vladimir Prus wrote:
On 07-Aug-15 10:08 AM, Bjorn Reese wrote:
The source code is build using CMake, but Boost.Build is in the pipeline (already done for documentation.)
I think that per http://www.boost.org/community/reviews.html the above should have blocked formal review. If a library is reviewed, it normally means the author believes it's to be ready for Boost, which certainly is not the case if one cannot run the test using the current tools.
That is a fair point to raise. I did not see the lack of Boost.Build as an impediment to the review for various reasons: 1. This is not the first library without Boost.Build support during review (e.g. Boost.Hana.) 2. The current CMake support is easier to use for the reviewers who want to take the library for a test-drive. Using Boost.Build would require that they copy Boost.Http into their Boost source tree. I find it unfortunate to force reviewers to "pollute" their Boost source tree with candidate libraries. 3. The Boost.Http documentation is already build using Boost.Build, so the transition of the remining source code should be minimal.