Skip to content

/rfcs

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

RFC: `..` in patterns #1492

Merged
merged 2 commits into from May 13, 2016

Conversation

Reviewers
No reviews
Assignees

@aturon aturon

Labels
A-data-types A-patterns A-slice A-syntax A-tuples T-lang final-comment-period
Projects
None yet
Milestone
No milestone
7 participants
@petrochenkov @oli-obk @nikomatsakis @brendanzab @aturon @nrc @Centril
@petrochenkov
Copy link
Contributor

petrochenkov commented Feb 6, 2016

Permit the .. pattern fragment in more contexts.

Rendered

@petrochenkov petrochenkov force-pushed the petrochenkov:patdotdot branch from cf5dc8c to 2cd24ce Feb 6, 2016

@petrochenkov

This comment has been minimized.

Sign in to view
Copy link
Contributor Author

petrochenkov commented Feb 6, 2016

cc rust-lang/rust#31443
text/0000-dotdot-in-patterns.md
[unresolved]: #unresolved-questions

Sublist binding syntax conflicts with possible exclusive range patterns
`begin .. end`/`begin..`/`..end`. This problem already exists for slice patterns and has to be

This comment has been minimized.

Sign in to view
Copy link
@oli-obk

oli-obk Feb 8, 2016

Contributor

Slice patterns are unstable (and buggy), and thus can easily be changed.

This comment has been minimized.

Sign in to view
Copy link
@petrochenkov

petrochenkov Feb 8, 2016

Author Contributor

Extensions proposed in this RFC are going to be unstable too and are supposed to be changed together with slice patterns if changed at all.
text/0000-dotdot-in-patterns.md Outdated
S(subpat1, .., subpatN)
// **NEW**: Tuple struct patterns, any position with a sublist binding.
// The binding has a tuple type.
S(subpat1, binding.., subpatN)

This comment has been minimized.

Sign in to view
Copy link
@oli-obk

oli-obk Feb 8, 2016

Contributor

these are somewhat problematic. Afaik there's no guarantee that (B, C) has the same memory layout as the last two parts of (A, B, C), so taking a slice into a tuple might not be so trivial.

This comment has been minimized.

Sign in to view
Copy link
@petrochenkov

petrochenkov Feb 8, 2016

Author Contributor

Hmm, this is an interesting observation.
It means that at least by ref bindings have to be restricted.
I'll update the RFC.
@oli-obk

This comment has been minimized.

Sign in to view
Copy link
Contributor

oli-obk commented Feb 8, 2016

Should S(a, b, ..) be allowed for struct S(A, B); ? For structs it currently is: http://is.gd/TsMz6N .
@petrochenkov

This comment has been minimized.

Sign in to view
Copy link
Contributor Author

petrochenkov commented Feb 8, 2016

Should S(a, b, ..) be allowed for struct S(A, B); ?

Yes, .. means 0 or more subpatterns.

@nrc nrc added the T-lang label Feb 8, 2016

@aturon aturon self-assigned this Feb 11, 2016

@nikomatsakis

This comment has been minimized.

Sign in to view
Copy link
Contributor

nikomatsakis commented Feb 11, 2016

This was pretty Always The Plan -- but we never did it because it involved work. I'm basically 👍 though I've not read the details of the RFC.

@petrochenkov petrochenkov referenced this pull request Feb 15, 2016

Merged

refactoring: Rename hir::Pat_ and its variants #31685

bors added a commit to rust-lang/rust that referenced this pull request Feb 17, 2016

@bors
Auto merge of #31685 - petrochenkov:patrefact2, r=eddyb 
And split `PatKind::Enum` into `PatKind::TupleStruct` and `PatKind::Path`.
This is the HIR part of #31581.
This is also kind of a preparation for rust-lang/rfcs#1492.

r? @eddyb
eb56607

bors added a commit to rust-lang/rust that referenced this pull request Feb 17, 2016

@bors
Auto merge of #31685 - petrochenkov:patrefact2, r=eddyb 
And split `PatKind::Enum` into `PatKind::TupleStruct` and `PatKind::Path`.
This is the HIR part of #31581.
This is also kind of a preparation for rust-lang/rfcs#1492.

r? @eddyb
0d1cd9b

@petrochenkov petrochenkov referenced this pull request Mar 6, 2016

Closed

[RFC] Implement `..` in tuple (struct) patterns #32079

@petrochenkov

This comment has been minimized.

Sign in to view
Copy link
Contributor Author

petrochenkov commented Mar 6, 2016

Implementation (without sub-list bindings): rust-lang/rust#32079
@nikomatsakis

This comment has been minimized.

Sign in to view
Copy link
Contributor

nikomatsakis commented Apr 12, 2016

Nominating for FCP.

@nikomatsakis nikomatsakis added the I-nominated label Apr 12, 2016

@petrochenkov

This comment has been minimized.

Sign in to view
Copy link
Contributor Author

petrochenkov commented Apr 27, 2016

I should probably strike out the sub-tuple bindings (a, b, c..) from this RFC given the discussion in #1582 and #376, at least for now.
@nikomatsakis

This comment has been minimized.

Sign in to view
Copy link
Contributor

nikomatsakis commented May 2, 2016

Hear ye, hear ye! This RFC is now entering final comment period.
@nikomatsakis

This comment has been minimized.

Sign in to view
Copy link
Contributor

nikomatsakis commented May 2, 2016

@petrochenkov very well... to be honest I've not read this closely, though I'm in favor of the spirit. We had decided to put this into FCP, but would you rather we wait? (We can keep your proposed edit in mind, I imagine.)

@nikomatsakis nikomatsakis added final-comment-period and removed I-nominated labels May 2, 2016

text/0000-dotdot-in-patterns.md
[subpat1, .., subpatN]
// Slice patterns, any of the above with a subslice binding.
// By ref bindings are allowed, slices and subslices always have compatible layouts.
[subpat1, binding.., subpatN]

This comment has been minimized.

Sign in to view
Copy link
@nikomatsakis

nikomatsakis May 12, 2016

Contributor

What does this binding mean? This feels like it overlaps with slice patterns, which (I think) still require some further RFC to "complete"? I can't remember if we postponed that work or not. :)

This comment has been minimized.

Sign in to view
Copy link
@petrochenkov

petrochenkov May 12, 2016

Author Contributor

What does this binding mean?

A subslice pattern, see above - "The binding is not actually a binding, but one more pattern ..."
@nikomatsakis

This comment has been minimized.

Sign in to view
Copy link
Contributor

nikomatsakis commented May 12, 2016

I'd be (more) in favor of a variant of this that excludes all "bindings". In other words, allowing:

(pat*, .., pat*)
Foo(pat*, .., pat*)

and I'd probably prefer to exclude [] patterns which I think ought to be handled separately.

The binding forms introduce complicated semantics.
@petrochenkov

This comment has been minimized.

Sign in to view
Copy link
Contributor Author

petrochenkov commented May 12, 2016

@nikomatsakis

I'd probably prefer to exclude [] patterns

This RFC doesn't propose any changes to slice patterns, it only uses them for analogy.

I'd be (more) in favor of a variant of this that excludes all "bindings".

With this I agree.
@nikomatsakis

This comment has been minimized.

Sign in to view
Copy link
Contributor

nikomatsakis commented May 12, 2016

@petrochenkov

This RFC doesn't propose any changes to slice patterns, it only uses them for analogy.

Huh. OK, I must have misread it. So much the better.
@brendanzab

This comment has been minimized.

Sign in to view
Copy link
Member

brendanzab commented May 12, 2016

I missed that subtlety too - maybe removing the slice examples might make it clearer?
@nikomatsakis

This comment has been minimized.

Sign in to view
Copy link
Contributor

nikomatsakis commented May 13, 2016

Huzzah! The @rust-lang/lang team has decided to accept this RFC. Specifically:

@nikomatsakis nikomatsakis referenced this pull request May 13, 2016

Closed

`..` in tuple-like patterns (tracking issue for RFC 1492) #33627

@nikomatsakis

This comment has been minimized.

Sign in to view
Copy link
Contributor

nikomatsakis commented May 13, 2016

Tracking issue: rust-lang/rust#33627

If you'd like to keep following the development of this feature, please subscribe to that issue, thanks! :)

@nikomatsakis nikomatsakis merged commit 4e9e843 into rust-lang:master May 13, 2016

@Stebalien Stebalien referenced this pull request Aug 19, 2016

Closed

Feature request: wildcards in tuple struct patterns #552

@petrochenkov petrochenkov deleted the petrochenkov:patdotdot branch Sep 21, 2016

critiqjo pushed a commit to critiqjo/rustdoc that referenced this pull request Dec 16, 2016

@bors
Auto merge of #31685 - petrochenkov:patrefact2, r=eddyb 
And split `PatKind::Enum` into `PatKind::TupleStruct` and `PatKind::Path`.
This is the HIR part of rust-lang/rust#31581.
This is also kind of a preparation for rust-lang/rfcs#1492.

r? @eddyb
13dcd0c

@Centril Centril added A-syntax A-patterns A-data-types A-tuples A-slice labels Nov 23, 2018

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
You can’t perform that action at this time.