Rust Programming Language Language Design Team, 5 Sep 2019

Previous Meeting Next Meeting

⏯

youtube image

►

From YouTube: Lang Team Triage Meeting 2019-09-05

Description

Triage meeting

Major topics covered:

- nounwind attribute and RFC 2753
- misc nominated PRs and regressions

Paper document

https://paper.dropbox.com/doc/T-Lang-Meeting-2019.09.05--AkmYatur2MDikhHZuCgppWpxAg-qA0w3GsE7WJMX3v05orXV

A

Let me share this.

B

Do you do you want to think.

A

So we did some pre triage go through some things, one of the first ones we circle back to dishes the circle back. To is that no online attribute logic, sixty nine or nine that we talked about last time. Josh did you want to share.

C

Yeah, a quick review on that based on last time, we have a an RFC posted and that would define unlined attribute in the simplest possible way that would allow us to move forward. But, given that you said it was worth the 26 that the beta would be branching for the next release was that the day 28 for maybe.

D

C

Centons 20 something given that the twenty-something is when the beta would branch for the next release: I'm really skeptical about us. You know accepting this RFC adjusting the current implementation to match that RFC have been stabilizing the resolve all by the twenty-something, so that we have a solution for people that doesn't invoke undefined behavior I think it might make sense for us to go ahead and merge this BR, which just says hey. If we're gonna abort, we should not let tho unwind on things and then we can add no unwind back when.

D

C

The unwind attribute I would like to run that by you know, get the general consensus here, but do we think we should try to ram through unwind in the next couple of weeks or do you think we should merge this PR.

B

So the actual branching date is the 20th. Both 20 states, that's when they actually, these lands, people who use it, which.

D

Means technically we can't get through to FCPS, even if everything was checked off right now. Okay,.

B

I think that's unhealthy.

A

B

The point yeah.

E

I know I, don't I didn't realize that that was the thing we were even aiming for. You know. Thank you.

D

I think if the RFC is posted, we I have no problem with doing the thing saying: hey we're losing.

F

D

This the fix is this RFC, which is open six hours.

E

These open AG work in progress. It's not really much of an RFC in text form at the moment and there's a great opposition and already 80 comments in the last like couple days. So.

D

That was different. There.

E

Is a new ones, it's different.

E

There was the old, there was the older one and then there's.

C

2750 is the new one.

D

Okay, I saw 78, commenting, hoping that was the old one.

C

B

Attempts to kind of derailed.

C

This one again into being the always encompassing RF secret schools. You know world cookies, but that would seem to have been managed to get under control and I. Think we're all attracted just treat this as the direct focused RFC.

A

E

I saw comments from like gnz LBG I think it was saying that they were opposed to the like unwind only attribute what was I I I haven't been following this close enough because it already got somebody it comments. In the last like two days, somebody who's been following the thread more recently like it'll, like they can summarize what what the opposition is.

B

C

Rflp identity, sorry go ahead, go ahead, Josh fine I was just gonna say as far as I can tell. The main objection is whether we are just providing a minimal solution for what people are doing today, with the understanding that people will still have to, for instance, build their C code with F exceptions, and that this only works at the rust line mechanism aligns with the C++ or C unwind mechanism, but that's what people have today. The objection is basically, oh, no, we shouldn't do an interim solution.

C

We should solve everything before we can do anything as far as I can tell I'm, not I'm, trying not to be unfair to the objection, but as far as I can tell, the statement is basically, if it's not perfect, we can't proceed.

A

That's the main objection. You think there isn't a thread yeah.

C

It appears to be the objection of know who we want to fully specify how you handle cross language, unwinding and specify the restaurant line mechanism and avoid any possibility that people will have to actually know what they're doing. In order to use this feature and to unwind from rust to see rust. There is a.

D

C

Some additional complaints of oh, but if you just completely rewrote the extern function, call mechanism of things like most JPEG to use C++. Instead, then you wouldn't have to change the language.

A

B

It seems like somebody needs to poach emissions. I have different concerns, which is more around like the RFC is not specified touching like type systems, and on dates is that external things are not specified like unwind. What does that do to versions with function, pointers and so on.

C

Right and I think the current thought there is. We don't currently put no unwind on calls through function, pointers at all which may.

B

C

Was just gonna say we don't currently put no unwind on function, pointer calls at all so that doesn't create any undefined behavior the we do need a good solution going forward for function, pointer calls, but the theory is right now the people who are using this and need this don't need to do it through function, pointers, which means we could proceed with what we have and then have a second round of a let's specify a solution for function, pointers that is compatible with that the there is.

B

So some people from I think thus the loose it specifically to run box to deal with function, plunges the explicitly sense of so.

C

They do need support for doing those calls. My point is right now there is no no unwind on function, pointers, so what they're doing does correctly work and we can introduce support or just you.

B

Thoughts, the fact that we don't omits those knots to the end doesn't mean thoughts in like the rust semantics. It's not undefined behavior from volumes. There is no way to know this. Why.

E

Are we interested in introducing no unwind function? Pointers. Is that the thing that someone isn't asking for I agree that why can't we just not solve that problem when you.

A

E

A

E

Fact, like the vast majority of function, pointers that people will be using are not know unwind right, because they can panic.

C

Extern see books and quarters extern, rust, function, pointers I believe are.

E

No unwind right yeah, not no.11, they they unwind. Yes, so.

A

That we're saying is why don't we just say that that's we don't capture that in the text when you call a function via direct call, we know we're able to respect the unwind attribute and otherwise.

A

It should be in the or I agree seems fine to put in C. We.

C

Could say currently we assume that function pointer calls through function. Pointers may not unwind yeah, you know. If we change that in the future, we will provide a mechanism for tagging. The message we.

E

Don't assume that they don't unwind, we assume that they they may unwind, which is, is always a correct conservative assumption all right.

C

The fact that the at redeem attribute is phrase no unwind is producing a lot of double and triple negatives. Here, sorry, we don't currently tag function, pointers as no unwind, so.

B

We assume that they pre unwind, there's.

C

No way to see in.

B

C

Think there may be in hell of the year. It's not currently.

D

C

Regardless we're saying are.

E

No reason to do this.

A

But basically we're saying in our rust undefined behavior semantics we do not permit unwinding from calls do.

B

We, or do we just say we're.

A

Saying we're saying that now we want to allow, because we don't care to complicate the type system to Dragon and that's what we know that we want to accommodate functions like the ad ID alone. Doesn't tell us that so.

B

I think I think that's inconsistent defaults to have function, pointers not have no align but to have functions.

E

Do not functions do not default to no one wined, nothing, nothing defaults to no online natural means. Extra and C functions, protect students, new functions, I'm only talking about currency.

C

Since do default, I like.

E

Extra in C functions are special specifically because unwinding across them is a problem. I mean I am.

B

In about extern seat, bunk shinbone, just in this case, not any other function, pointers everything I've said is about exiting. Yes,.

E

Extra and c function pointers would make sense to be no unwind. If that's a thing that exists right.

B

So so I think that we need to have any like static semantics. No.

E

One needs no one needs that, though, right like we're, like the complexity in the type system that you're trying to trade off like this is like a new effect mechanism right again and like I, don't think it's worth introducing an additional effect would be like.

C

If we were even gonna go there, we would need to show that it was worth it in optimization possibilities and until then we should just treat it as I think we do so.

E

And if you care much about having things tagged, no unwind, you should be using F, no exceptions and panic equals support anyway, like it.

B

Sounds like I'm more concerned about like surprises in semantics, that then speak, but okay, so so we would need to ensure about the coercion. Behavior is such that we don't create any future into Hauser's. If that's, we do that that I'm fine with it. What what future.

A

Hazards like with what future I think it's I'm trying, because I think if we adopt the future that Taylor and I were proposing, then there are no hazards, because the type system doesn't capture that information.

A

If we're going to separate it, if we're going to in the future, consider changing the type of extra and C functions to some other type, then of course, there isn't.

B

Well, if so, so, if.

B

The function just by default, our knowledge and then you can, of course, online one that is like a function that is unwind allow.

B

That is those that version makes sense.

B

No, the new our teachers and husbands Lazard. Only if you would things.

A

That yeah, if you, if you believe that the that's why we said it, that's why we said we would not capture the information in in the function pointer of whether the target function may unwind or not right.

B

Cos means that you can never do.

A

That's what it means unless yeah pretty much. It means that I mean we could do it by introducing some new forms that you want into at some point, or if you really want it, I mean you can just make the.

B

Version on error for now.

A

But we need Dennis type, okay, as if but I, then we would need.

A

Then you would not be able to call unwind allowed functions indirectly. So we write that that I think I think the dispute is whether we would want such a type right.

C

Well, whether we want such a thing or not, I think I just made a comment on RFC 27:53. If people could take a look at that, that effectively says, let's specify what we currently assume and then it defer to some future RFC if any any mechanism that would change that assumption. I, don't want to change that assumption right now. I just want to specify the current assumption. I.

A

Yeah I well I think the challenge so well I think we should do this at the moment. I in fairness to central I think there's a point that if we we are definitely narrowing our future options, if we ever want to type I just happen to think we don't and I'm okay with it or that if we do it's, okay, that the type of opt in so anyway, I agree with your interesting. It's not like that's a free choice. It sounds like I think we need to move on from this, but it's like.

A

Maybe we need to have depending how maybe somebody should try to write out all the situation, I guess at least and.

A

It's a it's a conflict worthy of written out, I'm happy, but at the moment I'm only happy with one of the answers, but.

A

But I think it's worth writing it out. I would volunteer.

A

C

Sorry, some of that just cut.

A

Out I'm, just saying it seems like this thread is also so hey I think we need to like write out the dimensions at this conflict, but be that this thread needs an active Shepherd, making sure that there's like able to summarize the the comments and move it along there do.

C

You feel like that's you Josh I will see about doing that. Yes, please feel free to assign it to me. Okay,.

A

So there's two action items: there's sort of one of them is not a discrete action, Shepherd the thread. So are we gonna land, this PR? In the meantime, we.

C

Are I think we may want to? Should we go ahead? Let's see be it I think nothing we've discussed here indicates that the RC will not land. We.

A

Will resolve this I.

C

Don't think the RFC will land in the next. You know week.

A

Agreed I'm just saying I think Scott. Another concern that they wanted to know that the RFC was leaning open, is back and.

D

Not indefinitely postponed I think I'm fine, what they open I. It would be nice if we could get it like ps2, a PSD, P state and then merge the thing or something, but that's not a I, wouldn't certain I certainly wouldn't block on that. That's fine I get.

C

That would you say: I don't want us to release it in a state where we stopped a boarding and we still tag things yeah. No one wanted, and we have no unwind at no attribute online to to fix that. It would be at the.

B

House, in the case, for a lot of years, it's not exactly new, it's true, but.

A

It still seems like the PRS there. The premise is correct: there we are engines, it is inconsistent.

B

As long as we everyone agrees that eventually, we will change. In fact, yeah.

D

But that's why I sort of like the a if we could have the are SCP SCP that says we are going to change it back. Here's the RFC then merging that I think having that statement out there helps the this is why we're merging this for now, I. Don't.

B

I, don't it's like not the best idea to come to the quality of an RFC by like how many lines it is, but I think it needs to like 10 times as long.

C

I'm, sorry did you say 10 times as long yeah, no.

B

No, no I'm, not saying that it needs to do more and just say, needs to like be more detailed about the stuff that it does and send the alternatives, like the other part they eat they ate. All that you do is yes,.

A

Okay, let's move on it's more minutes. In so P height issues, I mean I. Think we're reasonably good. You agree at this okay, P ayisha's. So there's not much to say about this one. You can read my update if you like, fYI there's, we been chosen considered an unstable feature, but we used to accept the odd compute since they're doing anything really productive with the bench part using an unstable library.

A

We kind of fix this, but there's some code out in the wild compiling so patrician common diminutive soft future gated issues, a warning with the hope to transition rate wassup wait. This is mostly indefinitely unless anybody has strong comments.

D

No I support making unstable things unstable.

A

It was not a lot of stuff affected by enough, so there's some nominated PRS six 402 one redefines the see start type to be a new type of an extra tight, thus making it a shallow or a thin pointer, rather so that it could actually be used in a bi functions, but central points out that that sort of exposes, in some sense external types, enjoy stable surface. So the question at hand is like: are we committing to some implementation detail? We might regret by doing so.

A

I I don't know I, don't it seems, like the main thing were committing to. Is that there's the dampers and easter as a thin point.

C

Pair pointer, yes,.

A

Go BNC stir that is compatible with the see car star, correct, yeah.

C

A

Maybe we can approach it like this people, any other things think, let's write it down. Anyone object.

D

This, because does open. Does anyone care about using raw pointers? To this thing,.

E

This is he, sir.

C

Yeah has been a raw pointer to a see store would be equivalent to of a pointer pointer care.

A

In terms of its representation, I'm just.

D

Wondering if this opens the whole I expected to be able to use pointer now on this thing and makes that exposed on stable, that we don't have a solution to that there's right now. If anyone is using a pointer, Kant's Easter, they can use pointer null with it. You're.

E

Saying because pointer null fails, if the type is unsigned.

D

Doesn't it fail for an extern type.

E

Well, it also fails for anything. That's uncie's. Doesn't it? Yes, oh and.

D

It's right now it's on sized.

E

Already, on sized rate.

D

E

D

E

Yeah, so I don't think that I don't think this changes that at all this is a stable thing.

A

D

C Sturges yeah one.

A

Point, oh, okay! So it's just stable as a fat.

B

It has the implement the current implementation. Is it's a start as an inner field which is guitar bracket prize and fun spot like an ngon sized slice of see, yeah.

A

I mean it does, it seems like this is a good change. It seems like technique, probably a breaking change, but.

A

D

Knows the only thing I know of the care is really about extern. So that's not a problem. I think it seems okay, but the.

B

But the wrong point you to see, sir, is that that's.

A

Not what do you mean over a pointer, sighs?

A

Well, I mean every pointer is size, but it's not the question so.

E

What is the reference science winter null? The constructor, the the function only works for size types today. Yes, it has a generic bound great. So this you, you already cannot construct a a null pointer to a sister, and you continue after this to not be able to construct that pointer. Why is he going.

A

E

That you don't work I, don't.

D

Know if you do I'm just trying to avoid conscious.

E

Existing uses of pointer null by doing this cancer is.

C

Something you'll still be able to convert a point. Your null into an option: C stir right and you'll, get a nun, a nun. You could I, don't know you. Could you.

A

Could do this, for example, which is nullable.

A

Like which is represented as a pointer, is compatible as I.

C

Was asking yes so we're saying that ampersand sister will be compatible with care pointer if you assume that you'll never get a null and option ampersands Easter will also be a care pointer that you can pass null to and you'll get none we're guaranteeing both right. What's this I believe.

D

C

What's the second option, you'll be able to accept option, ampersand, see, stir and get back and dull. You.

A

Know what was the first that.

C

In FFI, you can accept just ampersand sister as long as you just know that you'll have brokenness. If somebody passes you and all.

E

C

E

Asked about this to constructing a box, see stir unsafely, is still unsound or at least unspecified I.

C

Guess that's what seafaring is for yeah.

A

I mean it seems like it's, even more so than ever since extras: okay, cool.

D

So I don't see how box could possibly be allocated well.

E

They can because they can look for the noble yeah. That's it! No. You can't.

B

So um I think so, if we use excellent type in this way- and it like turns out that we don't want extreme type or that we want something else, how do we support this? Like our.

E

Existing RFC that you've Isan wrote that's been in the RFC PR list for forever that I've been trying to get around to both there and Simone's uncited types RFC and those would both allow us to do the same thing. Your honor so.

D

We have, however,.

E

Options here this is not like settling us into Eckstrom type I think we should.

A

Go forward but document that these are the main things it's about to be committing to, and the first ones out to be some other side effect.

A

It's unintentional, not that, like just one big difference, people's lives, but maybe it will at least we've written it down somewhere.

B

All right who is doing this and can write the comment.

E

And I'm excited to get to go change some code after this yeah nonsense code that I get to get rid of this.

C

Is actually that's an interesting point? Should we suggest that there should be a went on extern c functions that are accepting a care and immediately converting it to a see, stir I.

E

A

E

Be happy to take it.

A

B

C

Could be shooting in central for.

A

33 minutes in like to see so not that far, nothing.

A

So, okay, there's a PR this one. Maybe we can defer but I'll cover it quickly. So currently you've got a warning if you have a range which is a subset of another range, but you don't get a warning of any kind if you have a range, that is a partial overlap. This PR changes that go no unreachable patterns, length, which is sort of not strictly correct. The pattern is only partially unreachable. The question is: do we want this at all? If so, should it be its own lint?

A

There are some legit use cases and I think the one that the Josh was raising earlier is something like this, where you have some random exceptions and then a big range and it's kind of annoying to write the range without the exceptions and I think it is a little potentially confusing.

E

Clarify this wouldn't happen if you had like a range and then an underscore or pattern after it right, but it would happen if you had like a range and then a bigger like an open range after it. It.

A

Would not work I, don't know, that's a good point.

A

D

So is there a reason we should do this for integers, but not for enums. We.

E

Definitely well so the only enum equivalent I can think of would be I guess, either/or pattern that was duplicated right or an underscore, and we definitely want to do this on underscore right- is I. Think that I believe.

D

E

Sort of is supposed to be a catch. All the attention individually, I.

B

Think if we would like introduce pattern aliases like Patton Simmons in Moscow, then it will even get worse because you might want to work together to patent aliases, which ours has some overlap.

B

Which are just using convenience.

A

So ok, so we all sort of agreed it presently. There are good reasons to write, partially overlapping patterns. I think this is all saying it should not be the same good like this is a greatly different thing. If it's a little I think.

E

B

It seems like a.

E

Clubby thing: first to bake there there is: one party exist: okay,.

B

I think there would be too many possible space to move it.

B

C

Could find ways to avoid the false positive so as an as a note here, I think that it might make sense to lint the partial overlap case. That is in the second line here. If you have 0 to 10 and 5 to 15 linting on that makes more sense than if you have say, 5 to 10 and then 0 to 20, if you're completely subsuming the previous pattern. I, don't think you want to lint on that, because that is the potential base of and now I'm handling it all gates. That's.

E

Very heuristic II, though, because you also get into like oh what, if they're constants, instead of like literals and what, if they came from the expansion of a macro, rather than being something that the user wrote in line like you're sure I saw.

E

With a set of rules that make this like apply to the like minimal specific case, but I feel like it's not worth it. I'd like to move on I feel like we should is.

D

The heuristic that it belongs in Clippy, yes,.

C

B

Sorry, that's the moment to democracy and who's gonna write the comment: I propose tail out or Josh I'll.

C

Take care of it.

B

So after chanko has proposed moving on to the next issue 693 month, fishing has proposed that we stabilize not persons and more physicians provided a report on. There is some discretion at cetera and.

B

In the neck is clear as well.

B

A

B

Would like to say something a few words.

F

I'll start by saying that the D cells of the hygiene stuff is way paged out. So don't ask me any hard questions. I.

F

Think I had kind of objected previously it's a kind of having macros and any more positions. Basically, because we weren't sure about the hygiene situation and as far as I know was still not like, we've been using it and it works and people have been using it. No one's complained, so you know there's it's probably ok, but what I was saying to central the the other day was like realistically like to get the kind of understanding that I want and I think others want.

F

You know we're probably looking at like an entire PhD thesis or something and there's, and even then only with, like you know, Sam th or another couple of people would like supervised that so like I, think realistically, when we should admit that we're never going to get the deep understanding of macro hygiene that we actually want, and maybe that's: okay, like it's macros that they all disappear by the time we've finished parsing. So, theoretically, we can't cause any soundness issues but maybe like best effort.

F

Hygiene is good enough and we should move forward because there are people who, like know enough to implement this stuff right. So we're.

E

Still on spam, def site right, we're still only Spade stabilizing spam call site. This is just introducing the new positions.

A

F

I'm still not a hundred percent convinced that even just call site or even the infrastructure that we have for using spans as hygiene things is.

F

Correct like it feels like it is like I, don't think, there's a problem, but this is novel. This isn't like borrowing something from somebody else has already done. The work like what we have mostly tried to do with with rust. This is kind of completely new and in an area where nobody here has, you know really deep expertise, yeah so.

A

I forget why, but for what is for? If somebody from racket came up to me and told me that the concept of call site was totally unfounded, it gave me a convincing example and I've been searching for that example.

A

So Tim I'll be back up you're concerned, but I also, nonetheless, well we're going in no matter what we do. I think we've stabilized.

E

Span call site and I, don't know that it's worse, like obviously not backing that out. So.

A

Yeah I'm somewhere around I I, think I'm, like I share your concerns, Nick I, guess, I. Think there's a lot of spin there's a big gap between the current state of understanding, at least amongst the link team members.

B

A

Authors, it would be nice to occupy when it's like at least half like that fake heuristic description of what they expected behaviors um do.

D

We have a reference page at all about this stuff, no.

B

Not to my knowledge, actually, we we have the sort of documentation about macaroons markers, but that's mostly arrives at transcription and.

B

A

And have good documentation does this? Does this particularly for any of these proposed sites like seem likely to have more hygiene.

A

Like we already have.

E

Micro-Expressions I mean we, we actually don't or we all my crack certain extent. The ship has sailed, yeah.

A

I mean I I, think I.

E

Pervasive uses of Brock, Mac or hack around so yeah.

A

B

Personally feel very under educated about spy team very much yeah.

A

Yeah take some work to really clarify yeah.

B

There has been some movements with PR as Matthew and that I think about having working on on some stuff there. So perhaps the understanding is better well. Alright,.

A

How much things you know do we have to say about this.

D

Feels like all the macro stuff is just sort of people go and make PRS, and it goes in and we sort of shrug our shoulders and hope for the best. Yes,.

B

I feel that way about name resolution as well: yeah.

A

We should have.

F

Macro stuff yeah anyway, did I miss anything important. Just now,.

B

I, don't know just what you missed I think Nico made the point that the ship has already sailed.

F

Yeah I kind of think that it may well of them.

F

Like we've, we have certainly stabilized enough.

F

That we're kind of like stuck somewhat on this course at this point.

A

Yeah, the one thing I will say that gives me a certain measure of good feeling. Instead, we transitioned our name resolution algorithm through several improvements, including across 2018, a vaguely unified, 2015, 2018 model sort of successfully and I suspect that hygiene will ultimately understand, of course, where we will kind of come to something better understanding, but it will not be a hundred percent compatible around all the edges, but it will mostly work we'll.

B

Be able to there has been some hygiene ization, the four of these done library, markers and whatnot.

F

Yeah and I think like it, we should like fighting the we have. The mone is mostly correct, so I think like even if we discover there's something fundamentally wrong about it, and we have to come up with like a new system, then like it's still going to be just the edge cases. That change like most macro code will continue to work. I think.

A

Really cases where it's kind of ambiguous with you, men.

F

Yeah or I mean there might be at like weird edge cases where, like you've, got a macro importing. Another macro which uses is used in a macro expansion in some way, and then that's used by a procedural macro and that.

E

Sounds exactly the sort of thing that people are doing today.

E

A

E

Like a macro that is defined by a proc macro that defines a macro rules, macro that defines another buck all.

A

Right I would sort of like to move to the next videos, because I have ten minutes. Yeah I feel like we're not going to reach a firm solution past what we've said so far.

B

If I'm reading room correctly seems like we are sort of feeling that we're not having anything fundamentally more risky with this yeah.

E

I mean I was actually gonna, say, I would be more aggressive than this, and I would also asked for it to be stable as an expression position. So we don't have to use product macro hack anymore, but it's probably more of a conversation.

A

All right, I think that's right. Maybe.

B

Maybe you would like to rightfully be in that case yep what I want to.

E

Know for expression position, yeah I, don't.

B

Know I mean the one the.

E

B

We're discussing.

E

B

Sure, okay, well, I, can still read you. The are I mean I.

E

Don't feel like I have a ton of context on it, but I'm happy to go. Look around it. What tests exist and what you know what surface area is being expand, the PR itself? Oh, so you just want me to leave a comment and there's already a stabilization report. Although I've done yeah I guess I can do it. Yeah I mean I. Read your I read to report I thought you were asking for more of something for me. So okay I will do it.

B

Okay, cool: let's thanks Nick.

B

Stay if you like,.

A

Let's jump down to this, everybody go check, boxes or John destinations, but on this stabilizing storyline is constant Thanks, so so yeah bring up here. No no stuff. We were discussing the parallel migration and we actually said what did you want? You can like, like we were discussing it. The release team had some discussions.

A

Bring it up here, we would like it would be really nice to have only one bar a compiler one block. Let's simplify things, we could watch sound souls and we would unblock further improvements, we're kind of targeting the idea we had a vague target at the end of the year, which is one for. Oh that's, coming up turns out when you cycle there's currently six 1622 known crates.

A

That would be broken by enabling hard errors for all of the soundest bugs and all fixes, however, of those only around 100 or regressions, meaning that the rest are just depending on those hundred crates. Basically,.

B

That's creates like yeah, it's it's worth, noting that we used to have fewer regressions and we even closed like three of the issues. But since more more, like github repos were out its equator.

E

This is testing github repos to not just things that are on crates dot, io, yes, right, whoa I mean.

A

Okay, so the point is there will definitely there would definitely be some impact. Just did it today. We're currently planning something like this. First of all, we're gonna target the most common root regressions. Most of them already have new versions for the major new minor versions available, but I think it looks like there's a few left to go, maybe and then essentially try to draw attention to what's happening.

A

Publishing the blog post, reiterating the timeline fixes, giving links to Felix had a great blog post that and kind of helping people transition and do it I, don't know how other ways we might do more. Such a.

B

Big key I think of one action we could be right away. Is you make it make like the migration things are there's on 2018? For now it shouldn't be very unlikely for like if you have switched to Edition 2008, and then you haven't seen the moorings to do that. Being like given fair warning, don't shoot really clear things some stuff, but it's pretty safe. You do that just right or.

E

They not heard her son 19, no.

A

We wanted to make it easy for people to move to 2018, that's right, so yeah, that's right and there's a few options, basically of what we can do sides these two things, which is the one central just said we might. You could also do as we have done in some of fine specific cases like right now we just sort of dip the ears, but we can find cases that do not you kind of make some of the errors Partiers. We can do both of those right. We can do both of those and yeah plausibly.

A

We could go too hard errors, but exempt crates playing if your crate happens to be named URL, you get a little more time which feels weird. But yes, it's kind of I.

A

E

If it means we get to delete old, borrow check, I'm fine, with there being a special case in rusty that says, if the current crate being compiled is called URL. How this error.

A

How you are, and.

B

We're in caplets mode so I'm looking at the rest crates. Almost all of them are like it. Up.Once probably have no findings of, and very nor actually wait. There's not that many there are and gets up just just just throw the question out there: how? How would you feel about like making it all the harder on 2015 right now, I.

A

Would like to make you announcement tonight but and I think this cycle is pretty ok, I mean I. Guess possibly you want to move it up to one you're, basically saying move it up to 139.

A

B

Yeah one person.

A

Just for 2018 edition, I think doing 139 2018 140 2015 feels pretty ok um and it lets us make an announcement give people a little time. It does make a difference, though, that that these regressions are primarily like. It would be good to note the number of things there aren't crazy over society blows. Give me those because I care a lot more about I. Think people will fix their own gear, because, if they're maintained but create IO can be opinion, just lingers.

B

Okay, I can actually shake that I.

A

Think the t of the artists yeah.

E

I guess I guess. The thing that is interesting to me about github is that it is an indicator of non-root regressions, root, regressions and github, um like whatever I.

A

Don't know how many of the.

B

Okie-Dokie though it seems like we are so, it seems like the rough plan is to go ahead and make it harder for 2018.

B

Now and then maybe them are there for 2015 in the next version that we will like 14.

D

It and then we do the blog post.

B

A

A

Okay, I'm gonna run through a few small things very few more like FY eyes. One of them is. We got some regressions recently because we tweaked how the pretty printer works and people were pretty printing tokens and then relying on details like is there a space before or after comma and so forth, and didn't affect too much good? We mostly if it was a really small thing and it's reasonable.

A

We made it won't fix, but but if there were, there were some cases like I think we had started putting spaces in front of commas in that broke. A ton of stuff.

E

I mean we changed the whitespace formatting of debug output recently and that broke a whole ton of my tests that I had to go fix rigs. All my tests, workgroup, although usually unnormalized,.

B

Words and mostly thought the PR. This PR specifically like made the output nicer. So that's maybe why yeah.

A

I happen to think we should not have like putting a space before it comments and the priest unusual anyway, any case exploring people would know that was happening. We all agreed it like these details.

A

There's similarly, some regressions around but they'll kind of isolated, around tweeks resolution and hygiene to earlier topic, mostly in the what appears to be sort of bug, fixes like to look at this one here using um debug and derive debug there's a kind of cycle. Where is this debug like this refers to some macro bots in scope? Well, here's something that's in scope!

A

It could it be this this name like we don't. We can't resolve this, so we can't know if I guess somehow used to work.

B

E

Central you, this is predates you very slightly, but my firt one of my first big rusty projects was introducing a try-catch and we discovered that someone had a struct named lower case catch, and that is the reason why catch is not.

A

Jack similarly, there were some votes were sometimes be allowed. Shadow of lifetime names we didn't now. We just.

B

Can you just leave action items to actually becomes a lot? This.

B

E

Getting kicked out, I'll see you later yeah by.

B

The way I quickly check the number of regressions on on the crater um and it was 122.

A

Out of the 16, whenever yes, okay,.

B

By the way, it seems like, when you check your boxes, your box on the the then thing it didn't actually work, which one.

B

In the sterile and RFC review, I don't know if you did that if you review the attributes, thinking as well.

A

Inaudible are all recheck see you later yep.