|
Boost : |
Subject: Re: [boost] [CMake] what to do now?
From: Paul Fultz II (pfultz2_at_[hidden])
Date: 2016-04-15 19:18:49
On Friday, April 15, 2016 at 6:12:37 PM UTC-5, Andrey Semashev wrote:
>
> On 2016-04-16 01:46, Paul Fultz II wrote:
> >
> >
> > On Friday, April 15, 2016 at 4:17:11 PM UTC-5, Glen Fernandes wrote:
> >>
> >> Paul wrote:
> >>> This would be unfortunate, as it would make it a pain to use a lot of
> >>> tools(cmake, travis, appveyor, etc) with boost libraries just because
> of
> >>> an
> >>> arbitrary rule 15 years ago.
> >>
> >> Right. It would be better if the Boost guidelines were less
> restrictive.
> >>
> >> The approach I would prefer is: Instead of restricting the contents of
> the
> >> library repository root directory, restrict the contents of the
> '.boost'
> >> (or
> >> 'boost', or 'meta') directory. There should be sufficient specification
> in
> >> there to inform the Boost build and release process about the library.
> >>
> >
> > I agree.
>
> I'll remind where those requirements come from:
>
> http://thread.gmane.org/gmane.comp.lib.boost.devel/264776
> http://lists.boost.org/Archives/boost/2015/06/223027.php
>
> The second link gives the rationale for having the restriction on the
> directory structure. Although I have my minor disagreements with the
> policy (not related to cmake), I can understand why Rene, as the
> maintainer for various Boost tools, is requesting a certain degree of
> conformance.
>
> Really, there's no problem with putting CMakeLists.txt into
> <library>/build.
>
Yes, there is, as there are many tools that rely on files being at the root
directory(not just cmake).
>
>
> _______________________________________________
> Unsubscribe & other changes:
> http://lists.boost.org/mailman/listinfo.cgi/boost
>
Boost list run by bdawes at acm.org, gregod at cs.rpi.edu, cpdaniel at pacbell.net, john at johnmaddock.co.uk