Implement `?` macro repetition #47752

mark-i-m · 2018-01-25T20:53:33Z

See rust-lang/rfcs#2298 (with disposition merge)

rust-highfive · 2018-01-25T20:53:46Z

Thanks for the pull request, and welcome! The Rust team is excited to review your changes, and you should hear from @nikomatsakis (or someone else) soon.

If any changes to this PR are deemed necessary, please add them as extra commits. This ensures that the reviewer can see what has changed since they last reviewed the code. Due to the way GitHub handles out-of-date commits, this should also make it reasonably obvious what issues have or haven't been addressed. Large or tricky changes may require several passes of review and changes.

Please see the contribution instructions for more information.

kennytm · 2018-01-25T21:36:35Z

Marking as blocked by #47732, #47603, and of course rust-lang/rfcs#2298. I don't think this can be easily reviewed with all those formatting change 😂

mark-i-m · 2018-01-25T21:37:16Z

@kennytm I 100% agree. Thanks :)

kennytm · 2018-01-26T14:36:34Z

@mark-i-m Please put the new ? syntax behind a feature gate since this is an unstable feature. You may refer to :vis (search for macro_vis_matcher) to know how this works.

bors · 2018-01-26T20:35:08Z

☔ The latest upstream changes (presumably #47748) made this pull request unmergeable. Please resolve the merge conflicts.

mark-i-m · 2018-01-26T23:11:03Z

@kennytm Done.

mark-i-m · 2018-01-26T23:18:27Z

This doesn't have a tracking issue yet, so when that happens, I should update the feature gate...

mark-i-m · 2018-01-26T23:22:32Z

Also the unstable book...

nikomatsakis

Good start, nice work! Left various comments, will re-review once addressed.

nikomatsakis · 2018-01-29T19:47:44Z

src/libsyntax/ext/tt/quoted.rs

+        // #1 is `?` token, but it could be a Kleene::ZeroOrOne without a separator or it could
+        // be a `?` separator followed by any Kleene operator. We need to look ahead 1 token to
+        // find out which.
+        Ok(Ok(op)) => {


Nit: can we add assert_eq!(op, KleeneOp::ZeroOrOne); here? I find it helps catch mistakes when later somebody messes with the arms accidentally....

nikomatsakis · 2018-01-29T19:50:13Z

src/test/compile-fail/macro-at-most-once-rep-ambig.rs

+// <LICENSE-MIT or http://opensource.org/licenses/MIT>, at your
+// option. This file may not be copied, modified, or distributed
+// except according to those terms.
+


I always appreciate a comment here explaining what this test is trying to do. Example:

Test that in $(a)?+, the ? is interpreted as a separator, but not for $(a)?. This test focuses on error cases.

nikomatsakis · 2018-01-29T19:51:23Z

src/test/run-pass/macro-at-most-once-rep.rs

+
+macro_rules! bar {
+    ($(a)?+) => {}
+}


what about $(a)?*? we should test that too, no?

nikomatsakis · 2018-01-29T19:51:26Z

src/test/compile-fail/macro-at-most-once-rep-ambig.rs

+
+macro_rules! bar {
+    ($(a)?+) => {}
+}


what about $(a)?*? we should test that too, no?

nikomatsakis · 2018-01-29T19:51:34Z

src/test/run-pass/macro-at-most-once-rep.rs

+// <LICENSE-MIT or http://opensource.org/licenses/MIT>, at your
+// option. This file may not be copied, modified, or distributed
+// except according to those terms.
+


nikomatsakis · 2018-01-29T19:52:05Z

src/test/run-pass/macro-at-most-once-rep.rs

+// except according to those terms.
+
+macro_rules! foo {
+    ($(a)?) => {}


these tests would be better if they had things in the body that made use of a

nikomatsakis · 2018-01-29T19:53:31Z

src/test/compile-fail/issue-39388.rs

@@ -11,7 +11,7 @@
 #![allow(unused_macros)]

 macro_rules! assign {
-    (($($a:tt)*) = ($($b:tt))*) => { //~ ERROR expected `*` or `+`
+    (($($a:tt)*) = ($($b:tt))*) => { //~ ERROR 14:22: 14:29: expected one of: `*`, `+`, or `?`


Please move this to a ui test instead of hard-coding spans and things (i.e., the 14:22)

Oh, I intended to remove the line/col numbers but forgot... Should I still move to a ui test if I remove them?

Doesn't matter so much. I believe ui tests are generally "the future" but this test can stay a compile-fail test.

nikomatsakis · 2018-01-29T19:54:22Z

src/test/run-pass/macro-at-most-once-rep.rs

@@ -12,13 +12,19 @@ macro_rules! foo {
    ($(a)?) => {}
 }

+macro_rules! baz {
+    ($(a),?) => {} // comma separator is meaningless for `?`


wait-- so $(a),? is accepted but the , is just a no-op? seems like it should be an error

nikomatsakis · 2018-01-29T19:55:31Z

src/test/ui/feature-gate-macro_at_most_once_rep.rs

+// option. This file may not be copied, modified, or distributed
+// except according to those terms.
+
+// Test that the MSP430 interrupt ABI cannot be used when msp430_interrupt


This comment is wrong =)

mark-i-m · 2018-01-29T21:52:58Z

@nikomatsakis

wait-- so $(a),? is accepted but the , is just a no-op? seems like it should be an error

How would you feel about a lint instead? I would like to keep the separator simply for consistency with + and * even though having a separator on "at most once" is silly.

mark-i-m · 2018-01-29T22:27:46Z

@nikomatsakis I think I addressed your comments. You may want to wait on a re-review until #47732, #47603 are merged, as they will make the diffs easier to read...

bors · 2018-01-30T15:08:50Z

☔ The latest upstream changes (presumably #47870) made this pull request unmergeable. Please resolve the merge conflicts.

mark-i-m · 2018-01-30T15:11:35Z

until #47732, #47603 are merged

They have merged now 🎉 ... I will rebase shortly

nikomatsakis · 2018-01-30T17:48:54Z

How would you feel about a lint instead? I would like to keep the separator simply for consistency with + and * even though having a separator on "at most once" is silly.

Maybe we should bring this up in the RFC thread and get others' opinions. I don't have a very strong opinion, but it seems like an error would be better to me.

mark-i-m · 2018-02-09T00:40:25Z

Updated

CAD97 · 2018-02-09T03:57:05Z

Failure is due to added (see tracking issue):

[00:53:32] ---- [ui] ui/feature-gate-macro_at_most_once_rep.rs stdout ----
[00:53:32] 	diff of stderr:
[00:53:32] 
[00:53:32] -	error[E0658]: Using the `?` macro Kleene operator for "at most one" repetition is unstable
[00:53:32] +	error[E0658]: Using the `?` macro Kleene operator for "at most one" repetition is unstable (see issue #48075)

mark-i-m · 2018-02-09T04:01:09Z

Thanks! Fixed.

mark-i-m · 2018-02-09T04:04:41Z

hmm... tidy is now complaining that that line is too long 🤕

nikomatsakis · 2018-02-09T17:26:18Z

@bors r+

bors · 2018-02-09T17:26:19Z

📌 Commit b92e542 has been approved by nikomatsakis

bors · 2018-02-11T13:24:21Z

⌛ Testing commit b92e542 with merge e8b05f5bd4c89faa3f2b4718e6582acd05da08be...

bors · 2018-02-11T15:10:43Z

💔 Test failed - status-appveyor

kennytm · 2018-02-11T15:15:06Z

@bors retry #48116

bors · 2018-02-11T18:11:07Z

⌛ Testing commit b92e542 with merge b8398d9...

Implement `?` macro repetition See rust-lang/rfcs#2298 (with disposition merge)

bors · 2018-02-11T21:07:48Z

☀️ Test successful - status-appveyor, status-travis
Approved by: nikomatsakis
Pushing b8398d9 to master...

rust-highfive assigned nikomatsakis Jan 25, 2018

mark-i-m force-pushed the at-most-once-rep branch from 4652013 to abfdbdf Compare January 26, 2018 20:51

mark-i-m mentioned this pull request Jan 27, 2018

RFC: ? repetition in macro rules rust-lang/rfcs#2298

Merged

nikomatsakis requested changes Jan 29, 2018

View reviewed changes

mark-i-m changed the title ~~[WIP] Implement ? macro repetition~~ Implement ? macro repetition Jan 29, 2018

mark-i-m added 9 commits January 30, 2018 12:30

Allow ? as a KleeneOp in the macro parser

760879b

Update the macro parser to allow at most once repetitions for ? Kleene

bb8110c

Run rustfmt on quoted.rs

c33649c

Add a couple of tests

711f71c

Run rustfmt on macro_parser.rs

5ac48ec

Attempted fix for ? kleene op

f59b821

Fix typo in error message + update tests

51ef739

Fix a couple of tests

4897a05

Add feature gate + tests

3c15405

kennytm mentioned this pull request Feb 8, 2018

Comprehensively support trailing commas in std/core macros #48056

Merged

Use the right tracking issue

4cf3b65

Update feature gate test

1bd0862

Fix the test

b92e542

bors added S-waiting-on-bors Status: Waiting on bors to run and complete tests. Bors will change the label on completion. and removed S-waiting-on-author Status: This is awaiting some action (such as code changes or more information) from the author. labels Feb 9, 2018

bors added S-waiting-on-review Status: Awaiting review from the assignee but also interested parties. and removed S-waiting-on-bors Status: Waiting on bors to run and complete tests. Bors will change the label on completion. labels Feb 11, 2018

bors added S-waiting-on-bors Status: Waiting on bors to run and complete tests. Bors will change the label on completion. and removed S-waiting-on-review Status: Awaiting review from the assignee but also interested parties. labels Feb 11, 2018

kennytm mentioned this pull request Feb 11, 2018

Spurious (?) error in "compile-fail\rfc-2126-extern-in-paths\single-segment.rs" etc #48116

Closed

bors added a commit that referenced this pull request Feb 11, 2018

Auto merge of #47752 - mark-i-m:at-most-once-rep, r=nikomatsakis

b8398d9

Implement `?` macro repetition See rust-lang/rfcs#2298 (with disposition merge)

bors merged commit b92e542 into rust-lang:master Feb 11, 2018

This was referenced Feb 11, 2018

Hygiene opt-out for idents in expansion of declarative macros #47992

Closed

Implement tool_attributes feature (RFC 2103) #47773

Closed

WIP: Parallelize passes using rayon #46564

Closed

Centril mentioned this pull request Feb 27, 2018

Tracking issue for RFC 2298, ? repetition in macro rules #48591

Closed

5 tasks

mark-i-m deleted the at-most-once-rep branch November 14, 2018 16:40

Implement ? macro repetition #47752

Implement ? macro repetition #47752

Conversation

mark-i-m commented Jan 25, 2018

rust-highfive commented Jan 25, 2018

kennytm commented Jan 25, 2018

mark-i-m commented Jan 25, 2018

kennytm commented Jan 26, 2018

bors commented Jan 26, 2018

mark-i-m commented Jan 26, 2018

mark-i-m commented Jan 26, 2018

mark-i-m commented Jan 26, 2018

nikomatsakis left a comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

mark-i-m commented Jan 29, 2018

mark-i-m commented Jan 29, 2018

bors commented Jan 30, 2018

mark-i-m commented Jan 30, 2018

nikomatsakis commented Jan 30, 2018

mark-i-m commented Feb 9, 2018

CAD97 commented Feb 9, 2018

mark-i-m commented Feb 9, 2018

mark-i-m commented Feb 9, 2018

nikomatsakis commented Feb 9, 2018

bors commented Feb 9, 2018

bors commented Feb 11, 2018

bors commented Feb 11, 2018

kennytm commented Feb 11, 2018

bors commented Feb 11, 2018

bors commented Feb 11, 2018

Implement `?` macro repetition #47752

Implement `?` macro repetition #47752