* Add IANA Time Zone Support PEP
* Add section rejecting Windows native time zones
* Normalize code-block and drop python3
Apparently this is rendered without pygments support, which is causing
the build to fail.
* Add justification for a separate module
* Clean up and reorganize footnotes and citations
This adds headlines / citations for each footnote, rather than bare
links, and tries to put them in a more logical grouping, and makes
them all render without issues.
* Add / adjust proposal targets
* Add justification for set_tzpath
Basically, "people are going to do this anyway, might as well have them
use a function that we can at least deprecate if it causes major
problems."
* Add citation for rearguard format
* Fix typos and formatting in abstract
Fixes a few typos and reformats the abstract using semantic newlines.
* Update location of reference implementation
* Tweaks to motivation section
* Clarify wording about PEP organization
* Tweaks to section on the ZoneInfo class
This adjusts the wording and fixes typos in the constructors and pickle
sections.
* Tweaks 'Sources for time zone data'
These are typos, clarifications and wording adjustments in the section
on time zone data sources.
* Use semantic newlines in 'Security implications'
* Clarify what 'no special verification' means
* Add clarification to custom compiler section
* Tweaks to wording on 'tzdata' in the standard library
* Tweaks to Windows section
* Adjust the justification for rejecting pytz
* Add quotation marks around blog titles
* Abandon semantic newlines
* Rename title
* Move module name into 'Open issues'
* Comment out temporarily unused headers
* Add open issue about TZPATH structure
* Remove erroneous blockquotes
* Add discourse link and Post-History
James Bennett pointed out that "package signing" is widely misunderstood
as referring specifically to end-to-end signing with individual
publisher keys, rather than to metdata signing in general.
This updates the title and abstract to instead use the term "signed
repository metadata", and also updates a few other sections that
still gave the impression that implementing PEP 458 would be
enough to give the full end-to-end signing support that is actually
covered in PEP 480.
Also removes the exec() and eval() caveats on the reference implementation
(those have been migrated now, albeit not using a code structure that we
would genuinely want to merge).
* Rename __parameters__ to __args__, for typing.py compatibility
* Genericize, re.{Pattern,Match} and io.IO
* list != list[int], but list[int] == list[int] (and list[str] != list[int])
* Add a lazy __parameters__ that contains the unique type vars in __args__ (also for typing.py compatibility)
* make dict[str][str] fail, but dict[T, str][int] return dict[int, str]
* expose proxy type as types.GenericAlias
* Link to implementation
* Explicitly state that pickling or copying should work
* PEP 376: Use ReST code syntax for literal characters
Without markup, it is very easy to confuse code with adjoining
punctuation.
* PEP 376: Use ReST syntax for code, rather than Markdown syntax
* Made various edits and clarifications to the pep including:
* letting the PSF appoint offline key holders
* moving content from abstract to motivation to make the abstract more succinct
* updating the threat model
* resolving some minor inconsistencies
* add link to cncf blog post
* PEP 458: use "OpenPGP" instead of "GPG". The signature format is OpenPGP. Other
OpenPGP implementations exist aside from gpg, the OpenPGP tool from the GnuPG project.
* PEP 611: relax hard limits and make 1 million a soft limit. Include recursion depth in limits.
* Reorganize part of motivation to account for soft vs. hard limits. Add qualification about recursion depth.
* Add non-goals section to clarify that this PEP does not remove support for GPG signatures.
* Update pep-0458.txt
Co-Authored-By: Trishank Karthik Kuppusamy <33133073+trishankatdatadog@users.noreply.github.com>
Co-authored-by: Trishank Karthik Kuppusamy <33133073+trishankatdatadog@users.noreply.github.com>
- Remove the MRO walker, PyType_DefiningTypeFromSlotFunc, as it was
found it can give wrong results when a type's special methods are
changed (which can be done from Python code).
Solving the issue is deferred to a future PEP (and/or perhaps a
private CPython function).
- Split public C-API from CPython implementation details. It turns out
that alternate C-API implementations have grown more important
since 2016, if only for experimenting with possible optimizations.
- Add potentially confusing terms "Defining Class" and "C-API" to the
terminology section.
- Mention that ht_module is not inherited by subclasses.
- Remove modules to be ported initially: zipimport (since rewritten
in Python), _io and _cvs (need module state from slot methods).
* Mention __class__ cell, expand slot methods in Future extensions
And fix a typo.
Łukasz Langa
Guido van Rossum
Paul Ganssle
Ethan Smith
Brandt Bucher
Éric Araujo
Barry Warsaw
Nick Coghlan
Christopher Hunt
Hugo van Kemenade
Sumana Harihareswara
Petr Viktorin
Trishank Karthik Kuppusamy
Inada Naoki
Georg Brandl
Stéphane Bidoul
Shannon Zhu
Jason Fried
mnm678
dkg
Mark Shannon
Stéphane Bidoul
Shannon Zhu
Jonathan Jackson