[re-sending after subscribing to list]

On 9/5/19 12:12 AM, Thiago Macieira wrote:
> > SG16 was asked to comment on P1689 and how file names can be encoded in a file
> > format for exchange of information between different tools in a buildsystem.
> > This is not SG16's official reply, it is my own opinion that other members of
> > SG16 asked to write after a discussion in our Slack channel.

Thanks!

FYI, Ben is the primary author and is on vacation for a few weeks.
When he returns we'll review this discussion thread in more detail
and consider updates to the paper. Meanwhile I'll reply to some
points elsewhere in the thread.

> > Please revise this paper and try to answer some of these questions:

Yes, we can add these to the paper. I'll answer below for purposes
of this discussion.

> > * what tools produce the file?

A C++20 module dependency scanning tool, typically expected to be
provided as part of the compiler. Eventually Fortran compilers may
do this too.

> > * what tools consume the file?

The build system will use it to infer dependencies among sources
implied by module structure.

> > * how long is the file supposed to last? Is the file supposed to be committed
> > to version control?
> > * how far is the file supposed to be spread? That is, is networking in scope?

It is short-lived as part of buildsystem operation in a single build
tree and is not meant for distribution outside of that.

The assumptions in your original post are correct.

-Brad