This new variant introduces finer-grain code extents, i.e. we now
track that a binding lives only for a suffix of a block, and
(importantly) will be dropped when it goes out of scope *before* the
bindings that occurred earlier in the block.
Both of these notions are neatly captured by marking the block (and
each suffix) as an enclosing scope of the next suffix beneath it.
This is work that is part of the foundation for issue #8861.
(It actually has been seen in earlier posted pull requests; I have
just factored it out into its own PR to ease my own rebasing.)
----
These finer grained scopes do mean that some code is newly rejected by
`rustc`; for example:
```rust
let mut map : HashMap<u8, &u8> = HashMap::new();
let tmp = Box::new(2);
map.insert(43, &*tmp);
```
This will now fail to compile with a message that `*tmp` does not live
long enough, because the scope of `tmp` is now strictly smaller than
that of `map`, and the use of `&u8` in map's type requires that the
borrowed references are all to data that live at least as long as the
map.
The usual fix for a case like this is to move the binding for `tmp`
up above that of `map`; note that you can still leave the initialization
in the original spot, like so:
```rust
let tmp;
let mut map : HashMap<u8, &u8> = HashMap::new();
tmp = box 2;
map.insert(43, &*tmp);
```
Similarly, one can encounter an analogous situation with `Vec`: one
would need to rewrite:
```rust
let mut vec = Vec::new();
let tmp = 'c';
vec.push(&tmp);
```
as:
```
let tmp;
let mut vec = Vec::new();
tmp = 'c';
vec.push(&tmp);
```
----
In some corner cases, it does not suffice to reorder the bindings; in
particular, when the types for both bindings need to reflect exactly
the *same* code extent, and a parent/child relationship between them
does not work.
In pnkfelix's experience this has arisen most often when mixing uses
of cyclic data structures while also allowing a lifetime parameter
`'a` to flow into a type parameter context where the type is
*invariant* with respect to the type parameter. An important instance
of this is `arena::TypedArena<T>`, which is invariant with respect
to `T`.
(The reason that variance is relevant is this: *if* `TypedArena` were
covariant with respect to its type parameter, then we could assign it
the longer lifetime when it is initialized, and then convert it to a
subtype (via covariance) with a shorter lifetime when necessary. But
`TypedArena` is invariant with respect to its type parameter, and thus
if `S` is a subtype of `T` (in particular, if `S` has a lifetime
parameter that is shorter than that of `T`), then a `TypedArena<S>` is
unrelated to `TypedArena<T>`.)
Concretely, consider code like this:
```rust
struct Node<'a> { sibling: Option<&'a Node<'a>> }
struct Context<'a> {
// because of this field, `Context<'a>` is invariant with respect to `'a`.
arena: &'a TypedArena<Node<'a>>,
...
}
fn new_ctxt<'a>(arena: &'a TypedArena<Node<'a>>) -> Context<'a> { ... }
fn use_ctxt<'a>(fcx: &'a Context<'a>) { ... }
let arena = TypedArena::new();
let ctxt = new_ctxt(&arena);
use_ctxt(&ctxt);
```
In these situations, if you try to introduce two bindings via two
distinct `let` statements, each is (with this commit) assigned a
distinct extent, and the region inference system cannot find a single
region to assign to the lifetime `'a` that works for both of the
bindings. So you get an error that `ctxt` does not live long enough;
but moving its binding up above that of `arena` just shifts the error
so now the compiler complains that `arena` does not live long enough.
SO: What to do? The easiest fix in this case is to ensure that the two
bindings *do* get assigned the same static extent, by stuffing both
bindings into the same let statement, like so:
```rust
let (arena, ctxt): (TypedArena, Context);
arena = TypedArena::new();
ctxt = new_ctxt(&arena);
use_ctxt(&ctxt);
```
Due to the new code rejections outlined above, this is a ...
[breaking-change]
In preparation for upcoming changes to the `Writer` trait (soon to be called
`Write`) this commit renames the current `write` method to `write_all` to match
the semantics of the upcoming `write_all` method. The `write` method will be
repurposed to return a `usize` indicating how much data was written which
differs from the current `write` semantics. In order to head off as much
unintended breakage as possible, the method is being deprecated now in favor of
a new name.
[breaking-change]
1. Produced more unique types than is necessary. This increases memory consumption.
2. Linking the type parameter to its definition *seems* like a good idea, but it
encourages reliance on the bounds listing.
3. It made pretty-printing harder and in particular was causing bad error messages
when errors occurred before the `TypeParameterDef` entries were fully stored.
followed by a semicolon.
This allows code like `vec![1i, 2, 3].len();` to work.
This breaks code that uses macros as statements without putting
semicolons after them, such as:
fn main() {
...
assert!(a == b)
assert!(c == d)
println(...);
}
It also breaks code that uses macros as items without semicolons:
local_data_key!(foo)
fn main() {
println("hello world")
}
Add semicolons to fix this code. Those two examples can be fixed as
follows:
fn main() {
...
assert!(a == b);
assert!(c == d);
println(...);
}
local_data_key!(foo);
fn main() {
println("hello world")
}
RFC #378.
Closes#18635.
[breaking-change]
in most cases, just the error message changed, but in some cases we
are reporting new errors that OUGHT to have been reported before but
we're overlooked (mostly involving the `'static` bound on `Send`).
(Previously, statically identifiable scopes/regions were solely
identified with NodeId's; this refactoring prepares for a future
where that 1:1 correspondence does not hold.)
I wanted to embed an `Rc<TraitRef>`, but I was foiled by the current
static rules, which prohibit non-Sync values from being stored in
static locations. This means that the constants for `ty_int` and so
forth cannot be initialized.
This adds a `Substs` field to `ty_unboxed_closure` and plumbs basic
handling of it throughout the compiler. trans now correctly
monomorphizes captured free variables and llvm function defs. This
fixes uses of unboxed closures which reference a free type or region
parameter from their environment in either their signature or free
variables. Closes#16791
- Unify the representations of `cat_upvar` and `cat_copied_upvar`
- In `link_reborrowed_region`, account for the ability of upvars to
change their mutability due to later processing. A map of recursive
region links we may want to establish in the future is maintained,
with the links being established when the kind of the borrow is
adjusted.
- When categorizing upvars, add an explicit deref that represents the
closure environment pointer for closures that do not take the
environment by value. The region for the implicit pointer is an
anonymous free region type introduced for this purpose. This
creates the necessary constraint to prevent unsound reborrows from
the environment.
- Add a note to categorizations to make it easier to tell when extra
dereferences have been inserted by an upvar without having to
perform deep pattern matching.
- Adjust borrowck to deal with the changes. Where `cat_upvar` and
`cat_copied_upvar` were previously treated differently, they are
now both treated roughly like local variables within the closure
body, as the explicit derefs now ensure proper behavior. However,
error diagnostics had to be changed to explicitly look through the
extra dereferences to avoid producing confusing messages about
references not present in the source code.
Closes issue #17403. Remaining work:
- The error diagnostics that result from failed region inference are
pretty inscrutible and should be improved.
Code like the following is now rejected:
let mut x = 0u;
let f = || &mut x;
let y = f();
let z = f(); // multiple mutable references to the same location
This also breaks code that uses a similar construction even if it does
not go on to violate aliasability semantics. Such code will need to
be reworked in some way, such as by using a capture-by-value closure
type.
[breaking-change]
The implementation essentially desugars during type collection and AST
type conversion time into the parameter scheme we have now. Only fully
qualified names--e.g. `<T as Foo>::Bar`--are supported.
lifetime bounds. This doesn't really cause any difficulties, because
we already had to accommodate the fact that multiple implicit bounds
could accumulate. Object types still require precisely one lifetime
bound. This is a pre-step towards generalized where clauses (once you
have lifetime bounds in where clauses, it is harder to restrict them
to exactly one).
[breaking-change]
1. The internal layout for traits has changed from (vtable, data) to (data, vtable). If you were relying on this in unsafe transmutes, you might get some very weird and apparently unrelated errors. You should not be doing this! Prefer not to do this at all, but if you must, you should use raw::TraitObject rather than hardcoding rustc's internal representation into your code.
2. The minimal type of reference-to-vec-literals (e.g., `&[1, 2, 3]`) is now a fixed size vec (e.g., `&[int, ..3]`) where it used to be an unsized vec (e.g., `&[int]`). If you want the unszied type, you must explicitly give the type (e.g., `let x: &[_] = &[1, 2, 3]`). Note in particular where multiple blocks must have the same type (e.g., if and else clauses, vec elements), the compiler will not coerce to the unsized type without a hint. E.g., `[&[1], &[1, 2]]` used to be a valid expression of type '[&[int]]'. It no longer type checks since the first element now has type `&[int, ..1]` and the second has type &[int, ..2]` which are incompatible.
3. The type of blocks (including functions) must be coercible to the expected type (used to be a subtype). Mostly this makes things more flexible and not less (in particular, in the case of coercing function bodies to the return type). However, in some rare cases, this is less flexible. TBH, I'm not exactly sure of the exact effects. I think the change causes us to resolve inferred type variables slightly earlier which might make us slightly more restrictive. Possibly it only affects blocks with unreachable code. E.g., `if ... { fail!(); "Hello" }` used to type check, it no longer does. The fix is to add a semicolon after the string.
This patch primarily does two things: (1) it prevents lifetimes from
leaking out of unboxed closures; (2) it allows unboxed closure type
notation, call notation, and construction notation to construct closures
matching any of the three traits.
This breaks code that looked like:
let mut f;
{
let x = &5i;
f = |&mut:| *x + 10;
}
Change this code to avoid having a reference escape. For example:
{
let x = &5i;
let mut f; // <-- move here to avoid dangling reference
f = |&mut:| *x + 10;
}
I believe this is enough to consider unboxed closures essentially
implemented. Further issues (for example, higher-rank lifetimes) should
be filed as followups.
Closes#14449.
[breaking-change]
This leaves the `Share` trait at `std::kinds` via a `#[deprecated]` `pub use`
statement, but the `NoShare` struct is no longer part of `std::kinds::marker`
due to #12660 (the build cannot bootstrap otherwise).
All code referencing the `Share` trait should now reference the `Sync` trait,
and all code referencing the `NoShare` type should now reference the `NoSync`
type. The functionality and meaning of this trait have not changed, only the
naming.
Closes#16281
[breaking-change]
except where trait objects are involved.
Part of issue #15349, though I'm leaving it open for trait objects.
Cross borrowing for trait objects remains because it is needed until we
have DST.
This will break code like:
fn foo(x: &int) { ... }
let a = box 3i;
foo(a);
Change this code to:
fn foo(x: &int) { ... }
let a = box 3i;
foo(&*a);
[breaking-change]
