[Tom]
> Some things we really need to keep at least as simple as they are today
> and introducing a build system requirement doesn't do that

I don't have problem with this -- I agree with it. In fact, I have not seen objection to having some form of driver doing that.

-- Gaby

-----Original Message-----
From: Tom Honermann <tom_at_honermann.net>
Sent: Wednesday, April 20, 2022 11:59 AM
To: Daniel Ruoso <daniel_at_[hidden]>; nico_at_[hidden]
Cc: René Ferdinand Rivera Morell via Ext <ext_at_[hidden]cpp.org>; ISO C++ Tooling Study Group <sg15_at_lists.isocpp.org>; Gabriel Dos Reis <gdr_at_microsoft.com>; Nathan Sidwell <nathan_at_acm.org>; Peter Dimov <pdimov_at_gmail.com>
Subject: Re: [isocpp-ext] [SG15] Can we expect that all C++ source files can have the same suffix?

On 4/20/22 2:32 PM, Daniel Ruoso wrote:
>
> IMHO, it makes more sense to just accept that C++ with modules is not
> really viable without a proper build system.

I'm having a hard time with the notion that the answer to "How do I
build C++ Hello World using a modularized standard library?" would be
"First, choose and install a build system that works with your compiler".

The use cases I want to preserve involve ad-hoc interactive compiler
invocations that produce a few object files or a single executable or a
single shared library (along with any adjacent files like import
libraries, dependency files, symbol files, etc...); things that I do
every day. Performance isn't significant because, however slow it might
be, it will still be faster than me having to 1) figure out what build
tool happens to be available on the host I'm working on (if any), 2)
figure out how to use it with the toolchain I happen to be using at the
moment, and 3) author a build script for it.

Some things we really need to keep at least as simple as they are today
and introducing a build system requirement doesn't do that.

Tom.

>
> daniel