All of lore.kernel.org
 help / color / mirror / Atom feed
From: Quentin Schulz <quentin.schulz@cherry.de>
To: Tom Rini <trini@konsulko.com>
Cc: Simon Glass <sjg@chromium.org>,
	U-Boot Mailing List <u-boot@lists.denx.de>,
	Alper Nebi Yasak <alpernebiyasak@gmail.com>
Subject: Re: [PATCH 2/2] tools: Update the license specifiers
Date: Tue, 29 Apr 2025 16:04:32 +0200	[thread overview]
Message-ID: <290f70ad-bc06-41cc-910f-cecb89bdd6ff@cherry.de> (raw)
In-Reply-To: <20250429135241.GC1261075@bill-the-cat>

Hi Tom,

On 4/29/25 3:52 PM, Tom Rini wrote:
> On Tue, Apr 29, 2025 at 03:28:35PM +0200, Quentin Schulz wrote:
>> Hi Simon,
>>
>> On 4/29/25 3:15 PM, Simon Glass wrote:
>>> Recent versions of Python complain about the license being in the
>>
>> I believe this isn't related to Python but rather setuptools.
>>
>> setuptools 77.0.3 and later support PEP-639 which recommends to ditch the
>> License :: classifier for a license property.
>>
>> The issue is that this license property also changed in PEP-639, with
>> something that isn't compatible with pre-PEP-639.
>>
>> Therefore we need to bump the minimum requirement for the setuptools
>> dependency in the various pyproject.toml to 77.0.3 or later.
>>
>> I'm also wondering if we don't need to add a license-files property (also
>> PEP-639) to comply with the GPL-2.0-or-later which requires to provide a
>> copy of the license.
> 
> Thanks for digging in to this more. Is there not some way to both meet
> the PEP and utilize SPDX?
> 

AFAICT, the license field is for the SPDX identifier.

Not all licenses require you to provide the license to comply with it I 
guess? But my reading of GPL-2.0-or-later makes me believe that we need 
to. You would then have the license file(s) included when license-files 
contains a file that is found at the root of the python project? I 
believe that the default value for license-files is ['LICEN[CS]E*', 
'COPYING*', 'NOTICE*', 'AUTHORS*'] so it could be automatically included 
(and thus the license-files not be explicit) if the license file(s) is 
named like that.

It isn't necessarily a big deal (IANAL) to not have the license shipped 
if we don't plan on pushing those to pypi or share sdist/wheels through 
another mean (e.g. GitLab/GitHub releases generated with `python3 -m 
build`)?

I am not sure exactly what you meant by "both meet the PEP and utilize 
SPDX" as I believe they aren't related here? I mean, license property is 
the way to define the SPDX identifier in PEP-639, but the missing 
license-files doesn't have anything to do with SPDX I believe? Just that 
it wouldn't be complying (I believe, still haven't become a lawyer in 
the last few paragraphs) with the license itself if the license file 
isn't shipped.

What I meant is that in order to use the license property, we need to 
bump the setuptools version dependency in pyproject.toml so we're sure 
we're building with a recent enough setuptools.

Let me know if something isn't clear, I just happened to have had a 
glance at the PEP this morning for another project, so I am as confident 
as someone who spent 5min reading a spec :)

Cheers,
Quentin

  reply	other threads:[~2025-04-29 14:04 UTC|newest]

Thread overview: 7+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2025-04-29 13:15 [PATCH 1/2] tools/make_pip.sh: Use debian packages instead of pip Simon Glass
2025-04-29 13:15 ` [PATCH 2/2] tools: Update the license specifiers Simon Glass
2025-04-29 13:28   ` Quentin Schulz
2025-04-29 13:52     ` Tom Rini
2025-04-29 14:04       ` Quentin Schulz [this message]
2025-04-29 14:27         ` Tom Rini
2025-04-29 13:43 ` [PATCH 1/2] tools/make_pip.sh: Use debian packages instead of pip Quentin Schulz

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=290f70ad-bc06-41cc-910f-cecb89bdd6ff@cherry.de \
    --to=quentin.schulz@cherry.de \
    --cc=alpernebiyasak@gmail.com \
    --cc=sjg@chromium.org \
    --cc=trini@konsulko.com \
    --cc=u-boot@lists.denx.de \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.