[Distutils] RFC: PEP 566 - Metadata for Python Software Packages 1.3
di at di.codes
Mon Dec 11 18:19:47 EST 2017
After working a bit on an implementation of the "JSON-compatible
Metadata" section in this PEP, I'm noticing that it might be more
useful if it had the following step instead for canonicalizing the
> #. All transformed keys should be reduced to lower case. Hyphens should be
> replaced with underscores, but otherwise should retain all other characters;
Thus a field like `Description-Content-Type` would become
`description_content_type` and the resulting data structure would be
slightly more useful (as it could be passed as `**kwargs` to a
function with PEP 8 style parameter names)
If no one has any objections to this, I'll update the draft accordingly.
Also, if there is no other feedback on the current draft, I'l be
formally submitting it for Daniel's review this week.
On Sat, Dec 9, 2017 at 8:27 AM, Daniel Holth <dholth at gmail.com> wrote:
> I don't know why the parentheses were included in the older pep. They are
> widely deployed. We probably can get rid of them or make them optional in a
> practical parser.
> On Sat, Dec 9, 2017, 02:03 Nick Coghlan <ncoghlan at gmail.com> wrote:
>> On 9 December 2017 at 02:42, Thomas Kluyver <thomas at kluyver.me.uk> wrote:
>> > Dustin asked me to bring this issue to this thread:
>> > Metadata version 1.2 (PEP 345) says that version specifiers within a
>> > Requires-Dist field should go in parentheses: "zope.interface (>3.5.0)".
>> > The metadata spec on PyPUG repeats this.
>> > However, PEP 508 says that the parentheses are not needed, and tools
>> > writing dependency specifications should not create them. Its
>> > recommended format is therefore "zope.interface >3.5.0".
>> > Should the metadata 1.3 PEP note that this has changed? Or do we only
>> > need to update the metadata spec on PyPUG?
>> The new PEP already delegates to PEP 508 for the version specifier
>> format, so I think that can just be updated in PyPUG.
>> Nick Coghlan | ncoghlan at gmail.com | Brisbane, Australia
>> Distutils-SIG maillist - Distutils-SIG at python.org
> Distutils-SIG maillist - Distutils-SIG at python.org
More information about the Distutils-SIG