- switch from the new `TargetScopeError` to a plain `SyntaxError`
as per referenced python-dev discussion
- clarify rationale for those extra SyntaxError cases (we don't
want current CPython implementation details to implicitly leak into
the "do what CPython does" de facto language specification)
- broaden one of the error cases to handle the fact that CPython's
symbol table analysis for a comprehension involves two different
scopes and hence makes it difficult to detect when the target of
a named expression in the iterable expression gets re-used as an
iteration variable in the comprehension
- note that even dead code affects the symbol analysis pass
See bpo-36798.
This update specifies the interplay between f-strings and
the assignment operator. Specifically, it attempts to
clarify how to use assignment operators inside of f-strings.
- Delete 'Pending Acceptance' section
- Change reference implementation to proof of concept
- Add the ^L back before the Emacs Local Variables section
- Add Resolution link
* PEP 572: Tweaks suggested on python-dev by Glenn Linderman.
* Use Tim's rewording of the exceptional cases.
* Clarify that TargetScopeError is a *new* subclass of SyntaxError.
- Dropped "TODO: Include Guido's evidence, and do a more systematic search."
I think the current text is good enough.
- Relented on assignment expressions in default values other than top
level. (Mainly because the syntactic/semantic check would be
awkward, and it was already called out as a possible style
recommendation instead.)
- Clarified that a lambda counts as "the current scope" (containing an
assignment expression).
- Added prohibition on [... for i in i := ...].
- Specified that scope-related prohibitions should raise
TargetScopeError, a subclass of SyntaxError.
- Remove changes to comprehension scope
- Make := in comprehensions assign to containing non-comprehension scope
- Clarify binding precedence (tighter than comma, not at same level as =)
- Remove mention of more complex targets in the future
- Explicitly disallow toplevel :=
- Rewrite section on differences with =, enumerating all of them
- Remove "here's how this could be written without :=" from examples
- Tweak first paragraph of "Syntax and semantics"
- Add "Exception cases" (various explicit prohibitions)
- Clarify that lambda is a containing scope
- Clarify that := and = just don't mix
- Added "Open questions" section
- Added two new rejected alternatives: "Allowing commas to the right"
and "Always requiring parentheses"
- Minor edits
* Start a section on real code
* Correct/clarify "commas to the right" section
* Add Tim and Guido as authors
* Update abstract to mention :=
* Rule out targets conflicting with comprehension loop control
* Add timcode.txt as Appendix A
* Add os.fork() example
* Add TODOs about evaluation order
Nick Coghlan
Logan Jones
Pradyun Gedam
Guido van Rossum
Harmon
Chris Angelico
Giampaolo Rodola
Tim Peters
Dmitry Alimov
Brett Cannon
JimJJewett
Georg Brandl
Mariatta
Jonathan Goble
Neil Schemenauer