Rust Programming Language Language Design Team, 24 Oct 2019

Previous Meeting Next Meeting

⏯

youtube image

►

From YouTube: Lang Team Triage Meeting 2019-10-24

Description

Main topics discussed:

* `!` stabilization

A

That's a good question. Okay, maybe we could go through the license. First, yeah.

B

A

Jumping in random name, the stabilized rebalance commands thingy seems to be ready. There is possibly a bunk that someone found I think that Nikko can review that.

A

C

Yeah take a look but.

A

No, no! No! No! No.

C

Yeah I think we're ready to go unless anyone has any like last-minute concerns.

C

C

Many updates on constitute valuation.

D

There's nothing that needs like urgent attention. Okay, what's.

C

He saying this is sort of the same as last time. Do we just know I can't thank.

D

You know it's different I opened a few bugs about promotion, doing sort of the weirder forms of promotion within a cons body. So thank you look at it.

A

Seems like we needs to be redesigning the marci about conference and maybe there may be repeats expressions.

D

I, don't agree with that. I think we should I think we should be able to implement it roughly as originally envisioned it. Just we didn't have rules around promote ability, so.

A

So so I think that is important is that we speak a future can possible like wrong time. The wrong segmentation.

C

So that I I feel a bit lost with this conversation, I don't know, shall we do? We want to talk for a detail here, or do we want to just people can read these links and will like how house how settled is it? How much do you think we should all be following along versus take this offline.

A

I think we can discuss it on the issue. The people were like kidding me not me yet so that was funny we already discussed yet. So me continue.

A

D

Is good for me, okay,.

C

Well, this is an interesting one. Yeah I haven't fully caught up on this conversation, but last week so are in and around the unwind and FFI stuff. There was a sort of confusing to me turn where there was this PR opened, which allows us plus exceptions to safely unwind through rest code or claims to and there's some, which is great, did a lot of important implementation work.

C

But one thing that came up around there was a debate about whether it makes sense to we had been assuming that X, R and C would mean no unwinding, but there was a case to be made that that's the wrong default, because, basically it's the opposite of what she does. For example, we accept at least and that it would make more sense to support unwinding, which isn't that hard to do and allow people to opt out and perhaps like if you're using panic, equals abort I, don't know.

C

Maybe that has some other semantics, but there's been a lot of back and forth. What was surprising to me was that kind of mmm most people like guns of look and.

C

And on you, and other people were arguing more in this direction, they've seen the opposite of what we had been saying for quite a long time, so I haven't really caught up with it, but the one thing I like to float out here is like I, think the main objection to permitting unwinding throughout the five boundaries is around dead code elimination. Perhaps there's a correctness question whether people expect unwinding, although I don't know I'm kind of curious to take a quick temperature or.

E

C

Have people give me their thoughts on this would be a horrible change that that against it, if I have their concerns, that I don't understand.

A

About doing a 180, I.

A

Like the zoo discussion.

F

Yeah, it seems surprising, alright, I feel like that's a strange default, because I have a hard time, imagining rust code that correctly handle exceptions coming from across in a bi boundary I.

C

Don't know yeah, so there is the question of light.

C

There's a correctness argument around there's, tour of to safety arguments and I think you're, making one of them, which is especially for FF I, which tends to have unsafe code in its surroundings. It's really important to be sort of aware of online safety, and a lot of people may not expect on lining across this boundary. Is that kind of the argument.

F

Yes, but also, just in general, like arbitrary, unwinding across rust code, isn't something that we've well-defined right. Well, that would, what do you mean I mean we would be different like I would be the point defining that to be like you, you would expect drop hooks to still run everywhere. Yes, PPR.

E

F

C

E

Books run, but the chatroom wine does not work like it. Just ignores cash unwind, so I don't know. If that's what you mean like what's we're defining here, but that's definitely like the imagine code that they wouldn't be very surprised by that.

A

Norris means yes, I think so it's.

E

Opposed to a board, yeah I think it's a death, yeah I think I.

C

Think catch unwind has been. There is some debate about. This is sort of an orthogonal question, but if it affects people's opinion, I think it'll be important like either way. We have to say what kitchen line will do when the panic problem, when an exception goes I, think he can be set up either way. I mean.

F

We we don't have to say what catch unwind does in these cases. If we say these cases are just undefined behavior degree, yeah.

C

But we're not like we're gonna support unwinding at some point, whether we get a windfall say something you opt into, and then we have to find anybody right.

A

Okay, look when they're saying that it's unspecified behavior right like at present, yes, but I'm, actually want to specify it but like yeah, but me so we, but like the priestess in the previous discussions. That was important, that we didn't want to find.

A

C

Is not there's nothing to hear that.

F

This is innings rate. This is right.

C

F

We're just emo exceptions, getting thrown or unwinding triggered by non rest.

C

Code, ready I mean it doesn't define how we implement it, but it says that if an exception, propagates destructors run and.

C

Catch on wine does not catch it. That's what the current PR does. We could implement that by as we do today by integrating tightly with the native mechanism or you could integrate it by catching exception, translating it to some other thing propagating it in your own way, but they sort of not use your visible.

F

Has any money done any measurements on how much worse this makes cogent I, don't think.

A

Thousands like builds it okay and we don't really have a good wait. So we have a good way to test. Are we don't have a good way to test.

A

Unlike large just have to pick some project yeah, presumably yeah,.

F

I mean because so and and still for well I, guess that's a question so on panic equals the Bohr code panics from exceptions from rust code are clearly not well-defined right. Does this mean that we would still need to generate unwind hooks for exceptions coming from non rust code, because that seems I.

C

Think what I think what we would do in this case is release. What we had been saying is that we would abort if an exception came across the boundary and if an exception makes it into Russell Ebert aboard.

F

Essentially so so every F of five it has to have a harness around it catching unwinding and aborting right. That seems expensive to me.

C

F

C

Make it a guarantee good, is it I, don't know it's not a lot of.

A

Code in C, so one back of it, it seems like the new suggestion, essentially is what they don't likely. It's like unwind, the boards being unwind allow being the default, and then we have like unwind authority that is just reading through no.11 seems like you wanted to get away from that.

C

No I think the difference would be that, for one thing we don't behave right now. Our behavior, with respect to foreign exceptions, is not very it's not very consistent across platforms and for another we say it's undefined behavior right now, whereas we would be saying these we would. We would not be adding.

F

White attributes.

C

And so on, right.

F

Yeah well, when there's sort of a funny thing that you start wanting here, which this is the like: Z equals a former and F no except right right.

C

I think there there's a question of like I'm, not exactly sure whether, for example, I don't that we necessarily have to say this G panic people support like whether it actually aborts or mates at UB to unwind. This interesting question.

F

C

F

C

F

C

Question that would it work. I would certainly address here. What.

F

It does it addresses my concern, but it does mean that you could now write rest code that would unsafe rest code. That would be sound when compiling for the default panic mode, but for panic people support the code. Now will be unsound. Yes, which makes me very uncomfortable.

F

Because I mean there are, there are cases where the reverse is true, but I think that would be the first case of that being true right today. I think that's really missing. Go to me. I was like something.

B

That we should not make like a except like a normal part of an expectation like I, think that this code should be. If it exposes the saying of API. It should be safe under every yeah.

A

B

With a like in the F of Ike code, would now potentially become you be under late or I. Guess didn't know that this, if I, could never yeah.

C

No I mean it's not that any a cycle it becomes you be, but it's if you don't have a way, if you either don't say explicitly that I can't panic like. Presumably there would also be some way to say that it can't or you could imagine that being as in see like a no except um but yeah.

C

Okay, so I want to just write this down so I understand so I prefer, like the cookie so you're saying the concern is basically that you could have code that uses extern C, which throws exceptions under Z panic, equal support.

C

If we made that you B, then this code would be unsound by definition, sort of should an exception, propagate yeah.

F

That's queer so soundcheck we're dancing, whereas.

B

F

Default that would be a safe interface right, which are so like, presumably as.

A

Soon, as it goes, if it's expecting and handling the exception correctly I'm a bit surprised, that's let's see policy, but it's not really. As you can see. That's just me poking.

C

Wrong: okay, listen thanks! All right! Well, I will go. I will propagate these concerns around into that document and try to understand better what makes any change, but I think that it seems like the consensus in this room is still that, if I, something like the most vast majority of regardless of sees president the vast majority of effort find roles do not unwind, and thus we would prefer to have it be that you explicitly indicate when you do expect unwinding, then be a reverse. Oh.

F

A

Certain there, my bully I would also I think like having it in the air I straighten my makes the type system notice that the distinction well.

C

But that's true either way, it's just a question: if you read you write, see unwind before see no, except for it or something but yeah.

A

Because you can all no I, don't think so, because the proposal isn't that we would have a different API string for the opposite.

C

That's an important.

A

Order, that's orthogonal part.

C

Of the proposal right I agree with you that the a guys doing is better than random attributes.

F

Actually, so that's that's an interesting clarification, though I want to make. The argument here is over what today's default should be, not whether we should still have those options, there would still be a way to specify no like this function specifically is no except you would just have to do it everywhere. Oh yeah, that's right! Okay,.

F

I mean I, don't know about other people, but I would personally just write that ever everywhere, like I, would.

C

Yeah so there are, you may sort of think si.

F

Got it backwards? Basically, I am I, have almost never written bindings to a see library that had like well-defined unwinding and in Xavier I would say. That's not true for C++ libraries, which throw exceptions, let C++ libraries. You generally already have to wrap and see ABI boundary great right.

C

I think I find this argument kind of persuasive that, regardless of I, don't first of all I, don't think C programmers have a really strong expectation around this or like I, don't know at least in other words, the argument for being corollary with C is kind of weak, but that indeed unwinding does require extra auditing and it's unusual. So it makes sense to call it out.

C

All right we'll take it back, didn't rate I'm gonna move on, so we have time didn't rate no major updates, except that I've made some progress but for some reason, I'm getting realizin stammers um civilized.

A

Don't care generic items, it seems to be done so you can remove that yeah I can probably remove the stabilized yeah.

A

They sent yeah. Let's see we're sorry, that's really just.

C

Thinking that I better add it to my own to-do list or to see it through yeah.

C

Well, let's leave it up to it's actually stabilized. Okay,.

A

G

Saying we don't have any big of things, just progress as usual.

C

Okay, we'll all stop editing the same place, central editing and we've done so nothing too much here are we had I, don't know? What's going, I was six five. Three three add implicit named arguments actually I.

C

Don't want to skip over at this one I want to talk a little bit about banks. It felt like we didn't finalize this last time. I mean I'm worried well. If we leave it to the end, we'll get stuck with it.

C

So does that seem reasonable.

B

C

All recall were probably there's a PR here to stabilize paying. We talked about it last time. I, don't really know at this point. You know, I, don't know how we're going forward with us and I'd like to know I like to work out like. Are we driving this what's happening so so I? So.

A

So one thing we said last week was that it wouldn't be too difficult throughout fellate.

A

Things from my perspective, the best things do.

A

C

G

E

I was imagining.

C

Was a lint that triggered on fall back would be easy. That would be way to probably wait to course, and we'd have to iterate a little bit. That's a sort of development side thing so I said well. Do you have time like if I were to mentor you if you're doing that? Would you be interested in driving debt or said else? Who is yes.

A

Well, I would need to learn the new part of the Potter, but yes I think so they can use ice. The work I'm doing but I would need to see about some baby. Okay.

C

I mean that's one thing we can do. What are we going to get out of this lint? We expect we're gonna.

A

People one can run it on their own okay, so.

H

A

H

It's mitigation for the constant, the.

B

Fall back potentially causing undefined behavior so for the didn't change the running right.

A

Is doing it with crater-like isn't really possible because, like Apple has weird policies around licenses, Max and C I just didn't.

B

But I mean the main thing is like Simone had like some pretty reasonable, like we need to have like if we're going to make a breaking change, so we know cusses undefined behavior in like real code that exists, we should have tools to help. People find that and the are you find suddenly.

C

How how yeah, okay I think we have to figure out how exactly how precise it needs to be, but we can work with that. Do we want to consider?

C

We just want to wait on.

G

C

We want to consider stabilize and bang and work on the lint and fall back set before I. Think.

B

G

I think that that's.

B

A key fly was more productive talk about in this meeting, because I know that we don't agree on that.

A

So the first point I would make is like that's like how do I not make this pure I think you know just sat for two months and no one. Oh, we need to stabilize this, but.

B

That's like it's because the main reason to stabilize bang with, like we've already had to work around the fact that we haven't stabilized thing: I introduced the unpalatable type like sure. So it's not a years now or like blocking a change. It has like really material who, like benefits, to like the API, for like a fallback change.

B

That is a breaking change that causes you B, which, like material benefits, which, based on my conversation, the last minute and I, didn't really find that persuasive I'm like I'm, not against making the fallback change and I think that if.

B

Like like, basically, if someone is happy with the tooling that I'm definitely happy with it and happy to make the change, but the situation with the rest of the like things we like to stabilize just stabilizing Bank, is that Simone already tried to work around the problems with the fallback change by proposing to stabilize the part that doesn't introduce any like breaking changes.

B

But because, like the link team blocked that online commit the fallback change, he then went and found a different workaround of making the infallible type and stabilizing that in the center library and like we've just been like, were working around not able to have this API for years. And it's been blocked on like no one really being willing to implement this fallback thing and I. Think it's not reasonable to keep like blocking on this. One I think I. Think.

A

It's only stable since February and it's like.

B

A

Saying that I changed my mind and I I mean I I.

F

Also think it's unreasonable to keep walking.

F

Great, let's do the I'd really like to see the fallback, but otherwise I'm fine, I.

B

Definitely think that, like you, should drive implementing. If you want to fall back change, you should diplomat the like tools necessary, but holding like that. We can't move forward the things that are not actually blocked on the far back change doesn't seem like, like seems like inappropriate and, like you said in the last meeting, that you thought that the introducing bang was not worthwhile had the father change, and so I'd like to talk more about like that few, because I don't agree with and I would like me.

A

Like so I think I think and the current state, where bang is not it's like sort of like if bang is a proper type and the father is not I, think that's very confusing. It's like the opposite, I.

F

Think central I sympathize with that and I agree that, like that's sort of semantics Lee, that's what I think is like I think that's wrong. I think that's not the behavior that I bought in the language right, but I also don't think that it's gonna ruin too many people's day and certainly is gonna. Be far less can be like. Let's have a problem for those people to have a sort of not well that you would necessarily want fallback to them.

F

It would be to not have access to the never type at all which is currently the status quo. I mean I, don't know how how how strongly you feel about this, but I.

A

Think I think that, like most pressing reasons to the note are holes with track from them and stuff, the pressure is making lawyers. Bang is much lower so because palpable yeah.

F

Okay, I mean they don't use or convert infallible for things that aren't about fowl ability is the like I I use a separate, never type right.

B

And that's what everyone's doing, and that means that they are not like interconvertible right. It would be we'll just use bang great. So everybody just has like this some other type. It would be like if you had to create a new unit type. Every time you wanted to use unit, because every unit type had a name that give it like some semantic, meaning.

C

So it's the argument that introducing infallible already got us a lot of the benefits of thing.

B

Like having on one.

C

B

Unblock library on things that needed paying tapes, yeah.

F

You know not try from, but there are lots of places where I want to use Bing that have nothing to do with try from and I don't use Coar convert infallible there I.

B

Also think that, for me, one of the really compelling argument system like it's sort of interesting actually there, because I would think that this would persuade you more central from you.

B

It's the consistency thing that we already allow thing in return position. You.

A

B

A

Right about that, I do feel that it's an consistency, improvement, the same time I feel like if Bang is a bumper type and- and the fallback is like you know, it makes it even weirder. So it's like yes you're right Oh,.

C

Brittany but I just want to make one minor color area, which is that I'm not saying we shouldn't do the fallback, but I, don't think the fallback is a clear win because of how of how a fallback integrates I. Think the classic example from my dog was this: was this one?

C

They came to do with D serialization, where, if you run this code like and you don't constrain what type this is here today, it falls back to unit and the it basically gets into the fallback path, and the reason is because try winds up D sugaring to a match and one of those arms.

C

Isn't it is an infallible one or is a visit, falls back to bein, basically, and that takes a question might as well. It does yeah and, and the reason I'm bringing this up only is that my point is that it's not like this fall back only ever affects dead code. This is a case where it affects life code granted, weird and not very great life code, and that's why I think it's kind of okay, but there's.

D

C

Yeah it's like yeah. Maybe it would be better, but it's not an entirely update something falling by unit.

C

Really, no fallback, no fallback could be my preference. Unfortunately, no.

F

One like here makes sense, yeah.

B

Interesting that it's using like return type, possibly if we had like full flexibility, we could have designed better follicles. That would ever hear, unfortunately,.

C

B

This is this is the case. I tried.

C

For a while change, it I don't think we can, but basically what I'm saying is yes, you're right central, but as first of all, I think we wouldn't necessarily land there if you want to keep pushing on it, but secondly like if we did let in there when someone were to say but shouldn't it be paying I, think they're gonna be well a history, be it's sort of fall back, is weird, and not that great anyway,.

C

Ass, don't know how much you'll even notice.

A

This feels like weird.

C

But it's just like I, don't I mean there are a bunch of like there was a handful of examples like this right. This a lot of these have already been changed, but the objective I don't know what the Objective C case is. But I bet it's something similar where fall back is leaking out.

B

Okay, thanks to Derek, but is being yeah yeah.

B

So yeah I mean but like Central I, think they'll like it's like great for you to drive like making it the tooling necessary to make the fall back change if you like really believe in it, but it just doesn't seem like I. Never has likely practical use cases that we like would like immediately and like blocking on making an breaking change that, like the breaking change, is blocked on never existing, but ever is not blocked on the breaking change happening. So instead.

A

So if I, if I, if I, do implement the links, are you fine with like switching 12.

B

My opinion is basically like definitely if someone's happy, I'm happy and if someone's still not happy I would like evaluate the situation like with more thought and possibly like Simone is like my limit. Basically where I'm at neither.

F

Keith and I can't bring up that I feel like whether never type is stabilized is an inappropriate tool for sort of pressure. Pointing yeah.

B

Like what I see is sort of like.

F

The large a procedural issue with this is being done, like I'm, not I'm, not really interested in holding the future hostage. I I I'm so alright.

F

I'm, raising my hand, is like the person who put the concern on the original proposal of stabilizing just being a fallback, because I thought that, like in in everybody's head that that was tied, you the the fallback issue and that the fallback wouldn't sort of get dropped on the floor, but I think central step up and say that you know you're interested in implementing the lint and that's something that's I'm. Assuming is you're like you seem very passionate about the the fallback thing.

F

Far more than I ever was do I have faith that you'll actually go through and, like you know, implement something there and like try and have C motion on this, and so I feel comfortable. Saying that, like like I, don't think we're gonna forget about it and I think it would be. I would rather have a stable, never type than not have a stable, never type, and so whether or not Mike yeah. So.

A

Isn't so I mean I mean I do have concerns about like having sleeping there. It's like no, but I, don't see what concerns about like I mean if I, if I, if I do they actually work, I don't want to see like concern on something that is not realistic to implement one. So you wanna. Let me go change my way.

F

Yeah so unfortunately, I can't but I mean I. Guess that's that's a point where you have to trust your fellow line team members to be reasonable humans, okay and try and work with you like and I. Think everybody here is trying so yeah.

A

F

B

A

Okay, I'm going to change the PR experience, yeah.

B

But like be more like concrete settle, it seems to me like maybe like.

B

If someone posted a counter-argument, I could be persuaded differently, but, like like the lint does seem to me like it would be enough, like objective-c has made a non-breaking change to the latest version, to make it impossible like to require you to like to avoid these errors like having tooling that we can tell people if you are getting a weird like bug, it's possible that it was this breaking change, and you should run this one in May and it will tell you and then I think the only other thing that I would like to see is but I think this is what people are already doing.

B

I actually objective-c crate. Is that like making PRS didn't know, crates that are.

F

Do you mind, would you mind writing up those those comments like in a you know, comment on just yeah.

B

I can drive Central will uh like change the P R to do the thing that I'll just remove the canary with their common.

F

Would be it'll be helpful to have those steps outlined so that we know like when we fulfilled them and can't be like there's not later and arguing about whether or.

E

Not certain yeah I.

F

E

Like that can I can I just confirm I understand the conversation that just happened. We have basically collectively agreed that we are willing to decouple the stabilization of bang from a change to the fallback semantics and the central can independent work on this Lintz they'll be attached to a hypothetical change to fall back, but that can happen at its own pace, independently of landing. Never stabilization, never ambassador.

A

E

What we disagreed on the invite or not okay, thank you.

C

Does this my write this down for while we're in the moment, yeah.

B

C

Need a lint so yeah that you see with our and.

I

C

Find out or to diagnose if their behavior is related to this, and we need continued progress. This is sort of unknown trouble crates, but we have already gonna got that or we need, like you, have this already more or less yeah.

B

Like the root of the problem has been solved and then it's more like people who are using it bit, I'm making PR secreted.

A

F

Let's move it Oh actually can we can we finally clarify what the deal was with whether or not we would make be making corporate infallible an alias for bang.

H

Yes, I know that.

F

Voice but I actually feel pretty strongly about that. So I don't know I thought.

C

It was I, didn't think there was any country.

F

E

Then then that's the party, that's what Simone mentioned is that did I? Do that immunity, good lipstick, that's fine!.

F

C

F

Opposition that I heard was related to fall back so yeah.

B

C

F

Procedural all just tops.

C

It on but I, don't think it'll be an issue. Okay, okay, see: where are we so we could juggle? We can jump think that was good use of time. There's two: we can either this we already discussed, discuss things like.

G

C

What's happening with this I Rick you talking about this well.

A

F

This I feel like the new version of this, has a lot of rias. Warnings was supposedly linting on every instance of drain his under great. It feels very great.

A

To me, what what can someone is? What changed? Well, you you.

A

Well, the new creature on was.

B

Nowhere else, literally sacred.

A

Sacred segregates, the new case is probably one of the old faces, the from from the existing instance I think what changed.

B

In the analysis, oh I see like thousands of crates because it was like producing other warnings that were not like the.

C

Amount of Gator almost.

B

Basically, well.

C

I know but I'm trying to figure out what we changed before we did. The crater run right. Yes, we had, we made it opt-in. Is that what happened or.

A

Something no I think it's just special case being since a fateful morning like.

B

The original creator report was just wrong, so I think it's what happened so.

C

Ok, so it's still the original version. It's just any structure that contains a must use is now infected, as must use yeah exactly yeah, okay, okay, so.

A

Let's talk about the drain thingy well personally, if I would read some drain done and I wouldn't understand that dropping it on the floor remains everything for me, I mean.

F

It says not in the docks it's whatever drain. Does no that's yeah! That's like that's. Most of my calls to drain or, like you know, drop everything after some length sure it says.

A

E

Anything that that's that diagnostic, you see there is actually misleading, then right, because reading that I think okay, this is Noah right.

F

The the warning here is actually more confusing than helpful, like the wardens wrong like.

B

F

C

You might expect would be something like if we had a more idiomatic way or something.

B

Like I kind of steer, but that, because it's weird that, when he's an iterator that does things when you without being run through, like that yeah yeah,.

F

I mean that's sort of an argument: the Drina sort of a weird function, but, like it's impossible, there's a special.

A

Cage isn't it's possible to special case them like diagnostic, for it just not not, must use.

I

G

B

Types by adding must use to them, like my god, are.

A

These true, that's more.

B

Well, I mean if they only found 23 crates, that even in half cut and maybe not like most of the crates they found. Could they are using drain, which is what we'd be accepted from this.

B

Like I mean I, think I'm not like super against, like never like I'm, not like hard against this change, but I think that it definitely it's an RFC because to mediate seems like we're changing the semantics of the must use attribute like right now.

B

I must use like it's sort of like making less use into an auto trait right, where it sort of like leaks into, but like anything that contains the most used is also must use, should definitely be an RFC where we have like a larger discussion about it, not a PR and I'm kind of like not convinced that it's still worthwhile change based on these results. Yeah.

F

I mean I I, think I caused this by starting the conversation around how making tuples must use Reyes I feel like that. That is a case where, like you, would not be able to put, must use than a couple of a certain specific set of types right, yeah.

A

F

Don't know I I find I'm somewhat unconvinced by this.

A

Chain, maybe we could do it on these little plays or something if we do it on unique that it would affect box and friends. Maybe that's enough.

B

I'm gonna make like a box of a result into a must use right. I mean I'd, be fine with that, but, like seems.

C

Like we did special case or have a way to have a must use this like a parrot that.

B

Means like if my oh, my writers, must use done right.

F

Or an option of a must use or yeah, something like that would be fine with those yeah.

B

Yeah I think it's more when you're, like writing a struck and you have a field. That's a result right now, like just like an easy way to like have like right. Otherwise, like it's. An internal aspect is implanted like the.

A

Fact that we're discussing like interests or something on see territory, yeah.

B

Right, so that's where it like artsy would be like right. I would like give us an opportunity to these ideas like giggles.

F

Anyone inclined a lady.

B

F

To lend this PR, as is I, thinks I, think I think the answer is pretty much. No, so shall we say that we should close this, and if we want to push on this, we can ask for an RFC or some sort of design proposal seems.

F

Like maybe recommend some of these things that have come up here like.

C

F

Don't think it necessarily has to be a stable feature of the language either really having this for, like hey like do, we want, must use for like box of, must use types yeah.

A

A

Is anyone like up is most say they're both from you.

C

And one of you or I will but I'd rather semesters.

F

Sure I can leave a comment cute.

A

Let's see so when we're done with this, the fully integrated drugs are the first thing. This is.

C

A

There's a small amount of follows that came from this times executed last week: yeah I need to what did we land on him soon or not? No, okay! No! Yes I! It was actually feet, but I think we I didn't feel very sure about our rationale for EXCI being it was more like okay. Why not so I think that we'll need to read it in detail, but I can be done. Amazing, okay,.

C

A

A

C

48 after, under what conditions can you implement copy for a union seemed like there was rough consensus here they saying that you could implement copy of all variants at copy types.

A

That's the current behavior as well. So if we decide that we're fine with this, that it's basically here CP learnt or at closing or something I guess we can just that basically legit, yes for the, so that mean community can object or something if they on the other hands. It's also a conservative pack. Yes,.

C

Let's do it submerge it, someone want to write it. I can do.

C

C

A

See where are we then?

A

Can we talk yeah suddenly.

C

My Chuck something else so overlapping ranges, lint.

C

Currently, we warn, if the end point of one and the end point of another overlap. This was narrowed down from broader the thing where it weren't over any overlap. And now the question is: do we want to learn from partial overload? We know we don't want to work for subsets because that's considered useful I don't really care it's my opinion.

A

C

Think the case that I guess the main use case that Josh was bringing forward which sounded pretty convincing, was specializing and having a general purpose, fall back and then specializing various subsets of it um yeah, which doesn't really apply here. I guess not very clearly. I.

A

Guess it's maybe if you have like far there's something that's more likely, but it could be, you can imagine you.

C

Got some constants you want, if you like, using them and they don't, they happen to partially overlap.

B

Kind of more, like maybe Clippy,.

A

Territory like because it's fairly involved in the master game any-anything, maybe.

B

Maybe the interesting noise allowed by Depot loons I.

C

I

Wondering about that could be, you could turn it on big fall when.

C

I

I

Yeah but what's the point- and this is.

B

Totally got to.

I

Maintain that nobody uses yeah.

B

But that's that might have to do more to do with our, like fact that we don't advertise them like people, don't use Clippy, because they're written planet and Clippy they use it, be as quippy pushes the people to use it.

A

Oh would be nice to have site.

A

Yeah, like I, think.

B

That, like more like awareness around things that you could add, if you want to, would be yeah.

A

Anyway, I feel like we could accept this and change our minds later if we wanted to its limbs. So you can also like decide to do this when we brought you in what you asked for it. I also.

C

Think I would be somewhat persuaded by a crater run. That showed me bugs or not bugs.

A

A

C

A

C

I'm not gonna, write it right now, I'll take them all right. So no clear content conclusion, but those crater run might be helpful feels like Clippy territory,.

C

Quit imagine use cases, but not that strong eg two ranges that happen to overlap to test when they get one pair eg, exactly the thing or nothing against.

C

Okay, I, don't.

D

Think we're gonna do our drop today.

C

Because, except this is not my generator homework like.

A

Going back to the nominative yards there is as to oh yeah, Halima tested named arguments and return. The NT not importance something interrelate.

B

What the first.

H

B

I think that like could rouses right, I think this is a great idea, but it should be narsing yeah.

C

H

That's our conclusion.

B

I'll, post, okay,.

C

I also agree: it's plausibly lives, yeah.

H

I think I think it's actually probably true I think.

C

Sort of, like mildly built into the compiler but um I'd be happy to have live scan period, but I'm, okay. It seems like.

A

Just just a quick, quick like wrong table: do we think this good idea, person yeah, do.

C

You mean this so I think it's great yeah I've won nine times out of ten I'm, doing X equals x or something yeah.

B

C

Basically, I what I really want to Steph, you could put little mini expressions in there. Maybe simple expressions.

C

H

We're doing this more if it's a single, if it's.

C

A single identifier- and there are no arguments that use it eyes and default to identify our equals identifiers. That would be awesome.

C

That's what I actually want.

A

It seems like the hiding question is dealt with as well seem sensible, yeah. It doesn't seem problematic.

C

Alright, so we are generally positive but felt it should be. An arson sounds good. I posted my crazy idea resolve turned the non-empty. Glob must import something there. Doolin I think this broke some things.

A

So you can look like one things, but he's not you.

C

Can John soon we made it in error, I guess to have a glob that doesn't actually import anything.

A

I think this is fallout from interaction with something else. This thing it's not to blame directly.

C

A

Meaning intelligent was rushing out word for the existence of the error. It's feels to me like this is good reason why it exists.

C

Doesn't feel very real easy to me, but I think I, like dis, appropriate personally.

A

But well, why do we not have a little story for the same case them? Because.

C

There you said exactly what you wanted: I, don't know to me: star, expands to I realize it's not exact, that works in the mechanism but liking because of precedents, but it to ative Lee. It feels to me the star expands to some number of names, which could be none, and if it is, maybe you know, maybe it's not that's how it works. So you so basically.

A

The regular expression.

C

Right or kind of bash star right, I'm.

A

Interpreting yeah, I suppose but I feel that's very mutual like something like the interpretation will.

E

Start turn into a little star or if you actually man that same thing,.

C

E

C

But like I, don't know, I, don't expect an error and I think the reason is I I. Think of this as shorthand for putting. However, many individual imports I needed to make it work, but.

B

It's not it's like this. It's it's not like you're, using plus as synthetics write.

I

A

F

A

B

F

Okay, can can somebody convinced me that this should be error like why that's what why.

A

F

Care who rather.

A

Than a warning: well, there's one! Well! Unless that's if you read it's this commenter.

G

A

C

Basically saying you could think of prefix something as the I feel like there's a you could think of it as a filter, and so foo would be like I select the item from prefix that matches foo and star as I select the items that match I mean any items and either way for all filters. It's an error if it comes to zero, but I feel like what that saying.

C

Is this an error, because, under this way of thinking, it should be in your but not because it needed to be in your like there's some other part of the compiler that gets unblocked buying this being here or not beginner, which I would find more persuasive.

F

So, but like as a user like what experience do I want that's kind of around country.

C

F

Really good I mean.

B

I think you definitely want a warning, at least because you want to yeah.

F

I mean a warning but like like you, expect them to be something but yeah.

B

Yeah it could be prototyping, and so it'd be great if this would be like like if you're prototyping, this you wouldn't want.

F

And that's only a problem with my code like great, let it compile right.

C

So I think we want to merge this PR, which turns it into a warning to somehow I became so.

E

The only question would be: is it.

A

A future compatibility.

C

Warning I think it should just be no it's it's.

A

One one face in the.

B

Past compatibility warning.

A

So he'll get like: let's see.

H

A

H

Should just make it tiny, tiny.

A

Limbs for that, yes,.

A

So the the second bullet from from the bottom I'm his yes, so they make comments about something being harmful for the input resolution versions.

A

And the that doesn't apply to jobs or not.

C

That's what I'm interested in finding out I? Don't they don't say that it does I.

A

Don't know how I would like to know more about before.

C

Apparently so the consensus is, we feel like intuitively, it would be a warning or I would do what users want, but if there's some other argument like that, it's helpful for name resolution and some other way that we'd like to know yeah I also think the backwards compatibility rule applies here like there's, not a strong. You know to change it yeah. It's just white like this. Okay, so.

A

Wait now I didn't get away from said there is what white and black words compulsive a if you think currently.

C

We don't do an error like currently in stable I. Don't think we do I. Think no, no I I, don't think so, because I thought that was the whole point. There was a regression and it changed. That's why we made it a warning to not make your regression.

B

No I think that a Petro chenkov just felt like it was inconsistent that we have an error right now. Yeah.

A

B

Only reason your question going on at all this is.

I

I'm correct, there's.

H

I

A

B

Some things then wait a private item in there.

C

B

I don't know it's wait.

C

Right, ok, I see this error is on nightly. No.

B

No pub make it a pub use. It also to be a pub use. I think it's a circle specific.

B

Wait no just college this users, literal example.

B

C

I have no idea what we're talking about.

C

Pub use superstar.

C

B

The example be in six two three three four.

C

C

Okay, this is an error, and now we're saying to prefer it to be a warning. I.

B

Mean it seems really weird that halle.

C

Colored examples.

B

Don't do anything at all I, don't care about this area. Yes, okay,.

C

I, don't feel like I've been talking about something completely different than what is actually.

A

Super zoom: it should be like buku meeting on the All Hands for better to explain name resolution to all of us sometime, maybe all right, I gotta go, could.

B

We uh real quick next meeting. Could we talk about the turbot fish RC, specifically how to unlock it.

A

B

Want to talk about the substance of it. I want to talk about it. In January. I said it was locking it for a few days because it was to eat it, and it's so locked, but I don't believe. I was linking to a market. Oh yeah,.

C

Okay sounds good: let's note that for next time all three of the top so actually see.