[Buildroot] [PATCH] legal-info: Add site to legal info manifest

Clayton Shotwell clshotwe at rockwellcollins.com
Mon Apr 21 15:43:04 UTC 2014


Luca,

Luca Ceresoli <luca at lucaceresoli.net> wrote on 04/17/2014 04:47:42 PM:

> From: Luca Ceresoli <luca at lucaceresoli.net>
> To: Clayton Shotwell <clshotwe at rockwellcollins.com>, 
buildroot at buildroot.org
> Date: 04/17/2014 04:47 PM
> Subject: Re: [Buildroot] [PATCH] legal-info: Add site to legal info 
manifest
> 
> Hi Clayton,
> 
> Clayton Shotwell wrote:
> > I would like to propose adding the site to the legal-info manifest
> > files. This gives a little more information on where the sources came
> > from without adding much overhead. Please note that is is only for
> > packages where the source is not local or set with OVERRIDE_SRCDIR.
> >
> > This patch works for the most part. The only issue I see with it 
occurs
> > when the SITE for a package uses one of the common url macros such as
> > BR2_GNU_MIRROR. The legal info manifest ends up having an extra set of
> > double quotes in the site string.
> >
> > Signed-off-by: Clayton Shotwell <clshotwe at rockwellcollins.com>
> 
> Thanks for the proposal.
> 
> The idea is generally good to me: it provides a useful info with a
> little effort.
> 
> The project website URL would be nice as well, but that's not easily
> extracted.

Agreed

> Of course the double-quote issue with some packages needs to be fixed.
> But this is just "implementation details". Note that, besides packages
> from GNU_MIRROR it also affects the kernel from kernel.org, probably
> U-Boot and Busybox, maybe a few others.

I think this could be solved with a creative strip command but I was not
able to come up with anything that worked.  Maybe you or someone else will
have more luck?

> Instead, the case of packages downloaded from custom locations deserves
> a little more thought on what we want to do.
> 
> When using e.g. BR2_LINUX_KERNEL_CUSTOM_GIT=y, the REPO_URL may point
> either to a public server (github, gitorious, silicon vendors...) or to
> an enterprise server. In the former case the repository URL is
> informative. In the latter case it is not, and some companies may not
> like exposing their server names to the public. You know how netadmins
> are made... :)
> 
> So I'm not sure of what is the best policy when using custom (non-wget)
> download methods.

I was assuming the information could be scrubbed if needed from the 
resulting csv files.

Thanks,
Clayton

Clayton Shotwell
Software Engineer
clshotwe at rockwellcollins.com
www.rockwellcollins.com 




More information about the buildroot mailing list