PEP440 examples missing operator?
the first 3 examples here are confusing. https://www.python.org/dev/peps/pep-0440/#examples
there's no comparison operator?
oh, "A compatible release clause consists of either a version identifier without any comparison operator or else the compatible release operator ~="
On Mon, Dec 22, 2014 at 9:57 PM, Marcus Smith qwcode@gmail.com wrote:
the first 3 examples here are confusing. https://www.python.org/dev/peps/pep-0440/#examples
there's no comparison operator?
On Dec 23, 2014, at 1:03 AM, Marcus Smith qwcode@gmail.com wrote:
oh, "A compatible release clause consists of either a version identifier without any comparison operator or else the compatible release operator ~="
On Mon, Dec 22, 2014 at 9:57 PM, Marcus Smith <qwcode@gmail.com mailto:qwcode@gmail.com> wrote: the first 3 examples here are confusing. https://www.python.org/dev/peps/pep-0440/#examples https://www.python.org/dev/peps/pep-0440/#examples
there's no comparison operator?
Distutils-SIG maillist - Distutils-SIG@python.org https://mail.python.org/mailman/listinfo/distutils-sig
Heh, that’s actually a hold over from before we made specifiers mandatory. That needs updated.
--- Donald Stufft PGP: 7C6B 7C5D 5E2B 6356 A926 F04F 6E3C BCE9 3372 DCFA
On 23 Dec 2014 19:02, "Donald Stufft" donald@stufft.io wrote:
On Dec 23, 2014, at 1:03 AM, Marcus Smith qwcode@gmail.com wrote:
oh, "A compatible release clause consists of either a version identifier
without any comparison operator or else the compatible release operator ~="
On Mon, Dec 22, 2014 at 9:57 PM, Marcus Smith qwcode@gmail.com wrote:
the first 3 examples here are confusing. https://www.python.org/dev/peps/pep-0440/#examples
there's no comparison operator?
Distutils-SIG maillist - Distutils-SIG@python.org https://mail.python.org/mailman/listinfo/distutils-sig
Heh, that’s actually a hold over from before we made specifiers
mandatory. That needs updated.
Should we add a reference to the metadata issue tracker to the PEPs themselves? That's a better home for bug reports than the mailing list (Note I'm also happy to move the repo for metadata drafts to GitHub if folks would prefer that - it's a pretty simple repo, so git-hg bridges should handle the conversion easily. The issue migration would involve a bit more work, though).
Cheers, Nick.
Donald Stufft PGP: 7C6B 7C5D 5E2B 6356 A926 F04F 6E3C BCE9 3372 DCFA
Distutils-SIG maillist - Distutils-SIG@python.org https://mail.python.org/mailman/listinfo/distutils-sig
Well you know my preferences is for github ;). I don't mind much either way though since its a simple repo I rarely need to interact with. The main benefit would be consistency with most of the other pypa things.
I'm +1 on adding a bug tracker link. Maybe peps should support a header for it similar to discussions-to
On Dec 23, 2014, at 6:42 AM, Nick Coghlan ncoghlan@gmail.com wrote:
On 23 Dec 2014 19:02, "Donald Stufft" donald@stufft.io wrote:
On Dec 23, 2014, at 1:03 AM, Marcus Smith qwcode@gmail.com wrote:
oh, "A compatible release clause consists of either a version identifier without any comparison operator or else the compatible release operator ~="
On Mon, Dec 22, 2014 at 9:57 PM, Marcus Smith qwcode@gmail.com wrote:
the first 3 examples here are confusing. https://www.python.org/dev/peps/pep-0440/#examples
there's no comparison operator?
Distutils-SIG maillist - Distutils-SIG@python.org https://mail.python.org/mailman/listinfo/distutils-sig
Heh, that’s actually a hold over from before we made specifiers mandatory. That needs updated.
Should we add a reference to the metadata issue tracker to the PEPs themselves? That's a better home for bug reports than the mailing list (Note I'm also happy to move the repo for metadata drafts to GitHub if folks would prefer that - it's a pretty simple repo, so git-hg bridges should handle the conversion easily. The issue migration would involve a bit more work, though).
Cheers, Nick.
Donald Stufft PGP: 7C6B 7C5D 5E2B 6356 A926 F04F 6E3C BCE9 3372 DCFA
Distutils-SIG maillist - Distutils-SIG@python.org https://mail.python.org/mailman/listinfo/distutils-sig
Heh, that’s actually a hold over from before we made specifiers mandatory. That needs updated.
you mean made operators mandatory? so the bit about "version identifier without any comparison operator" needs to be removed?
Donald Stufft PGP: 7C6B 7C5D 5E2B 6356 A926 F04F 6E3C BCE9 3372 DCFA
On Dec 23, 2014, at 2:36 PM, Marcus Smith qwcode@gmail.com wrote:
Heh, that’s actually a hold over from before we made specifiers mandatory. That needs updated.
you mean made operators mandatory? so the bit about "version identifier without any comparison operator" needs to be removed?
Yea.
Originally PEP 426 changed the syntax to be ``foo (==2.0)`` and such, and in that regards it made sense to indicate a default operation for ``foo (2.0)``. However we removed them because:
* That syntax is backwards incompatible with the syntax that setuptools uses, and it’s sort of a pain to enter on the CLI. * A default isn’t that useful, no matter what we pick it’s going to be wrong, originally PEP 386 had it equal to == (but a weird == that doesn’t make sense to me), then PEP 440 originally changed it to ~= but that doesn’t make sense for date based releases which would be better implemented as
= or >.
--- Donald Stufft PGP: 7C6B 7C5D 5E2B 6356 A926 F04F 6E3C BCE9 3372 DCFA
participants (3)
-
Donald Stufft -
Marcus Smith -
Nick Coghlan