C++ Logo


Advanced search

Re: [isocpp-ext] Can we expect that all C++ source files can have the same suffix?

From: Daniela Engert <dani_at_[hidden]>
Date: Wed, 18 May 2022 18:34:55 +0200
Am 18.05.2022 um 13:26 schrieb Stephen Kelly:
> I was able to build fmt with

You should -DFMT_TEST=ON to engage building both the module and the
related unit test that imports the module.

> The reason I got interested in your work there was to see what
> artifacts you install and where. It seems that in that mode, headers
> are installed, but I don't see anything in the install prefix that
> would make it possible for downstreams to use it (would that be the
> PMUI in theory?). I am guessing that was out of scope of your efforts.
Right. My efforts were only about proper unit testing the modular build
mode and the consumption of the built BMI, checking the presence of all
of {fmt}'s API and successful instantiation of all exported templates.
This alone was difficult enough because at that time CMake had no notion
of a compilation of a TU depending on the BMI created by another TU.

>> I need to check once more if CMake is in better shape these days, and
>> also if the visibility issue is still present in msvc 17.2. TBH I'm
>> following {fmt} development (and my modules implementation for it)
>> much less recently because afaik neither clang nor gcc can compile
>> fmt.cc (the PMIU) and modules support in CMake is still a mystery to
>> me (as a Windows-only person).
> Do you or anyone else know of any library with a similar effort of
> modules support, with also some installed artifacts relating to
> modules? (possibly also installing headers for compatibility).

Matusz Pusz units library is also striving for modules but the folks
over there are pretty lost because of the state of gcc and clang.

> Thanks,
> Stephen.

PGP/GPG: 2CCB 3ECB 0954 5CD3 B0DB 6AA0 BA03 56A1 2C4638C5

Received on 2022-05-18 16:34:55