Hi Peter,

Sure. And I do hope that not only the compilers will be able to read the BMIs at some point when their format is stabilized. At least, this is the plan for ifcs.

But having a json near it with additional info seem to simplify many scenarios.

Olga

From: Peter Brett <pbrett@cadence.com>
Sent: Monday, February 7, 2022 01:27
To: Olga Arkhipova <olgaark@microsoft.com>
Cc: Jon Chesterfield <jonathanchesterfield@gmail.com>; sg15@lists.isocpp.org
Subject: RE: [SG15] Meeting on February 4th at 9AM Pacific

You don't often get email from pbrett@cadence.com. Learn why this is important

Hi Olga,

When packaging software we already use a variety of tools to extract information from executables and libraries. I don’t think adding another tool for inspecting a file and dumping out the info required for module compilation is an obviously unreasonable proposal.

                   Peter

From: SG15 <sg15-bounces@lists.isocpp.org> On Behalf Of Olga Arkhipova via SG15
Sent: 06 February 2022 23:45
To: Jon Chesterfield <jonathanchesterfield@gmail.com>; sg15@lists.isocpp.org
Cc: Olga Arkhipova <olgaark@microsoft.com>
Subject: Re: [SG15] Meeting on February 4th at 9AM Pacific

EXTERNAL MAIL

>> What's the point of embedding information in a second file next to the first one, when it could be written directly into the first one?

Only the compiler currently can read and write BMI, but anybody can read and write a json file. The ability to write (and not only read) is needed in package production where source/dependencies locations might be different than in the original build.

Olga

From: Jon Chesterfield <jonathanchesterfield@gmail.com>
Sent: Thursday, February 3, 2022 18:43
To: sg15@lists.isocpp.org
Cc: Olga Arkhipova <olgaark@microsoft.com>; Steve Downey <sdowney@gmail.com>
Subject: Re: [SG15] Meeting on February 4th at 9AM Pacific

You don't often get email from jonathanchesterfield@gmail.com. Learn why this is important

What's the point of embedding information in a second file next to the first one, when it could be written directly into the first one?

Jon

On Fri, 4 Feb 2022, 01:44 Steve Downey via SG15, <sg15@lists.isocpp.org> wrote:

On Thu, Feb 3, 2022 at 8:07 PM Olga Arkhipova <olgaark@microsoft.com> wrote:

The compiler will have to find all BMIs so their locations should be defined by some command line options.

My point is that the same options can be used to find the .d.json files.

Thanks,

Olga

I agree that if the build system can figure out how to do one, it can do the other, as long as there is some discernible relationship between the bmi and the .d.json file. But in a typical unixy environment, libraries and other artifacts to be consumed are not separated out. Perhaps, though the bmi and .d.json both live together in an isolated filesystem-like thing based on the module name? E.g. a directory or zip file, or some such. On the other hand, since .d.json is intended to be portable, I would expect to find it in something like /usr/share/module_${name} in an FHS style system? Or if a library provides multiple modules, underneath /usr/share/lib${name}/? 

Replace /usr with /usr/local/, ~, ${etcetera}, etc above.  

(sorry I sent this only to Olga, now replying on list, Olga if you reply, either here or add the list back?) 

_______________________________________________
SG15 mailing list
SG15@lists.isocpp.org
https://lists.isocpp.org/mailman/listinfo.cgi/sg15