[breaking batch] don't glob export ast::UnOp variants #31487
Conversation
|
Thanks for the pull request, and welcome! The Rust team is excited to review your changes, and you should hear from @Manishearth (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 Due to the way GitHub handles out-of-date commits, this should also make it reasonably obvious what issues have or haven't been addressed.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. |
|
Feel free to just make a single PR which you keep adding commits to to avoid needing a rebase |
oli-obk
force-pushed
the
breaking_batch/ast/unop
branch
2 times, most recently
from
da37149
to
1bb12a6
Compare
|
Fyi the marker is exactly |
oli-obk
force-pushed
the
breaking_batch/ast/unop
branch
3 times, most recently
from
a0a5215
to
b177233
Compare
|
ok, this is ready, I'm done for this run. Before the rebase + squash it passed locally. Let's see what travis says. The only potentially controversial changes are in libsyntax/ast.rs, everything else is just fallout due to the renamings or removal of glob exports. |
oli-obk
force-pushed
the
breaking_batch/ast/unop
branch
from
b177233
to
acf3dc4
Compare
|
jup travis likes it :) |
|
Reviewed most of it, have marked places where I'm unsure of something. |
oli-obk
force-pushed
the
breaking_batch/ast/unop
branch
from
acf3dc4
to
74e6082
Compare
|
addressed comments |
|
the PR already bitrotted (see travis). rebasing... |
|
@petrochenkov I don't think it's too ugly, but I guess that's a matter of taste. |
|
Well, ok. I guess. |
|
@petrochenkov In the past we didn't have a concrete policy on breakages. Try to make a PR before this makes the nightly, basically. The window is flexible, though. |
|
Also, if someone wants to do the same reexport change to the HIR it would be nice. Don't worry about breakages there, since only clippy and some other lints depend upon the HIR and we can take care of that. |
btw, regarding this, there wasn't much discussion on this PR, but people affected by breaking changes were pinged a couple of days ago on the older version of this PR (which was split up). Usually I'd wait a few days so that further breaking changes can be collected; however this itself is a large collection of breakages and is large enough that it will bitrot almost every few hours. In the future, if you feel the need to make a breaking change, make a PR, get reviewed (without the r+), and when we have enough of them around we can roll them up. |
cc #31487 plugin-[breaking-change] The AST part of #30087 r? @Manishearth
|
@oli-obk perhaps you can say something about https://internals.rust-lang.org/t/nightly-broke-quote-tokens-1i32/3174/2 ? |
|
Is there some reasoning for this change? While I understand that we don't want libsyntax to be defacto stable, this seems like really unnecessary breakage that affects basically every crate doing anything with macros.
Nobody on the Diesel team was pinged. :\ |
rust-lang/rust#31487 included breaking changes which cause codegen to no longer compile on the latest nightly. It will likely be several days before I can synchronize everything to work on both nightly and stable against the latest versions.
This is due to some apparent changes in Rust nightly that is causing the `extern crate` definitions to conflict with each other. Unfortunately, we still don't compile on nightly as a result of rust-lang/rust#31487, but this is an easy win.
|
@sgrif Ah, sorry. I didn't know that Diesel uses syntax extensions a lot. The main breakage I was worried about was serde/aster (which is used by half the ecosystem). I'll ping y'all next time this happens, and keep a tracking issue where we can coordinate future libsyntax breaking changes. |
cc #31487 (comment) plugin-[breaking-change] The first commit renames `ast::Pat_` to `ast::PatKind` and uses its variants in enum qualified form. I've also taken the opportunity and renamed `PatKind::Region` into `PatKind::Ref`. The second commit splits `PatKind::Enum` into `PatKind::TupleStruct` and `PatKind::UnitStruct`. So, pattern kinds now correspond to their struct/variant kinds - `Struct`, `TupleStruct` and `UnitStruct`. @nikomatsakis @nrc @arielb1 Are you okay with this naming scheme? An alternative possible naming scheme is `PatKind::StructVariant`, `PatKind::TupleVariant`, `PatKind::UnitVariant` (it's probably closer to the common use, but I like it less). I intend to apply these changes to HIR later, they should not necessarily go in the same nightly with #31487 r? @Manishearth
Adapt to nightly changes introduced by rust-lang/rust#31487
It was pointed out in review that the glob-exported underscore-suffixed convention for `Spanned` HIR nodes is no longer preferred: see February 2016's rust-lang#31487 for AST's migration away from this style towards properly namespaced NodeKind enums. This concerns rust-lang#51968.
r? @Manishearth
I just noticed they can't be rolled up (often modifying the same line(s) in imports). So once I reach the critical amount for them to be merged I'll create a PR that merges all of them.