Re: How to handle copyright notices in testdata intended for license validation tools?

Soren Stoutner <[email protected]> Tue, 25 Nov 2025 12:52:33 -0700

Newsgroups	gmane.linux.debian.devel.legal
Organization	Debian
Message-ID	<2469726.1EsLJ0VmTh@soren-desktop>

On Sunday, November 23, 2025 8:20:15 PM Mountain Standard Time Richard Laager 
wrote:
> All of this is just my opinion, of course...
> 
> If it were my package, I'd probably do something like this (which could 
> list multiple files):
> 
> Files: testdata/valid-evidence.json
> Copyright: OWASP Foundation
> License: Apache-2.0
> Comment:
>    This file contains strings that are test copyright statements, which
>    are being used as a test vector for proper license/copyright parsing.
>    They do not correspond to any copyright of this package.
> 
> That way, you are really explicit about which files have the issue. If 
> you put the comment at the top-level and just say "Several files", it's 
> a lot less clear which files you're talking about. You said it was many, 
> and I haven't looked at the package itself, so I might do something 
> different if it was really a large number that couldn't be covered by 
> glob patterns.
> 
> I wouldn't copy the fake copyright statements into debian/copyright. 
> That feels wrong to me.

I also agree with this approach.

-- 
Soren Stoutner
[email protected]

signature.asc (application/pgp-signature, 833 B)