Skip to content

[Integration] main (3683457) -> swift/main #557

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Merged
merged 29 commits into from
Jul 7, 2022

Conversation

hamishknight
Copy link
Contributor

No description provided.

stephentyrone and others added 29 commits June 29, 2022 21:13
We'd like to provide this, but we haven't been able to reach agreement on what the semantics actually ought to be. The original version used wholeMatch; contains is arguably more flexible, but feels less natural when compared to how ~= is used for other types in the language.

Additionally, we'd like to investigate the possibility of using named captures to effect bindings in case statements in the future, and we don't want to wall ourselves off from that until we've had time to think about it some more.
Adjust the upper bound of the Range down when
forming the quantification, as it expects an
inclusive upper bound.
Seems like we ought to crash in a release build
instead of potentially crashing during byte code
emission or silently failing to match.
_swift_stdlib_XX symbols defined in _CUnicode module are also defined in
runtime library as hidden visibility. It works in linking as a shared
library, but it doesn't when statically linking with lld, which resolves
lib_StringProcessing.a's _swift_stdlib_getScript and loads
UnicodeScalarProps.c.o, then loads libswiftCore.a's
UnicodeScalarProps.cpp.o for other _swift_stdlib_XX symbols that are not
defined in UnicodeScalarProps.c.o.
Throw error when Output is wrong type
Use word breaking SPI for \b
Fix crash when emitting an atom of any
We have decided not to support this for now.
We have decided not to support this for now.
Update to match the proposal.
Previously we would ignore the case where the
match fails, but the test expects the match to
succeed.
This doesn't appear to be used.
Previously we would wrap a `nil` in
`optionalCount - 1` outer layers of `.some(...)`.
Change this to return an `optionalCount` nested
optional with a top-level value of `nil`.
When computing the CaptureList for AST nodes,
including converted AST -> DSL nodes, only permit
at most one level of optionality. This means that
regex literal captures are now either `Substring`
or `Substring?`.

Optional nesting is however still performed in the
DSL (due to result builder limitations). If a
regex literal is nested in the DSL, it may only
add at most one extra level of optionality to the
current nesting level.
This is no longer needed as the capture list has
the right nesting information.
@hamishknight
Copy link
Contributor Author

@swift-ci please test

@hamishknight hamishknight merged commit 0ec8b08 into swiftlang:swift/main Jul 7, 2022
@hamishknight hamishknight deleted the main-merge branch July 7, 2022 09:46
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

Successfully merging this pull request may close these issues.

4 participants