## Description
Fixes#5650.
`use` statements without a call path such as
```
use foo;
```
causes the compiler to crash during semantic analysis.
This PR causes the compiler to fail nicely instead.
These types of imports should be allowed, but I am in the process of
reworking the import logic, so I'm postponing an actual fix.
## Checklist
- [x] I have linked to any relevant issues.
- [x] I have commented my code, particularly in hard-to-understand
areas.
- [x] I have updated the documentation where relevant (API docs, the
reference, and the Sway book).
- [x] If my change requires substantial documentation changes, I have
[requested support from the DevRel
team](https://github.com/FuelLabs/devrel-requests/issues/new/choose)
- [x] I have added tests that prove my fix is effective or that my
feature works.
- [x] I have added (or requested a maintainer to add) the necessary
`Breaking*` or `New Feature` labels where relevant.
- [x] I have done my best to ensure that my PR adheres to [the Fuel Labs
Code Review
Standards](https://github.com/FuelLabs/rfcs/blob/master/text/code-standards/external-contributors.md).
- [x] I have requested a review from the relevant team or maintainers.
---------
Co-authored-by: IGI-111 <igi-111@protonmail.com>
## Description
`find_method_for_type` was throwing an error when multiple
implementations of a method exist receiving different kinds of unsigned
int. This was caused by the coercion comparison which returned true
while comparing any uint size to any other uint size.
The solution was to change the UnifyCheck on Coercion to only return
true to same size uints.
Fixes#5798
## Checklist
- [x] I have linked to any relevant issues.
- [x] I have commented my code, particularly in hard-to-understand
areas.
- [ ] I have updated the documentation where relevant (API docs, the
reference, and the Sway book).
- [ ] If my change requires substantial documentation changes, I have
[requested support from the DevRel
team](https://github.com/FuelLabs/devrel-requests/issues/new/choose)
- [x] I have added tests that prove my fix is effective or that my
feature works.
- [x] I have added (or requested a maintainer to add) the necessary
`Breaking*` or `New Feature` labels where relevant.
- [x] I have done my best to ensure that my PR adheres to [the Fuel Labs
Code Review
Standards](https://github.com/FuelLabs/rfcs/blob/master/text/code-standards/external-contributors.md).
- [x] I have requested a review from the relevant team or maintainers.
Co-authored-by: Joshua Batty <joshpbatty@gmail.com>
## Description
Some minor cleanups and refactors to module-related code.
## Checklist
- [ ] I have linked to any relevant issues.
- [ ] I have commented my code, particularly in hard-to-understand
areas.
- [ ] I have updated the documentation where relevant (API docs, the
reference, and the Sway book).
- [ ] If my change requires substantial documentation changes, I have
[requested support from the DevRel
team](https://github.com/FuelLabs/devrel-requests/issues/new/choose)
- [ ] I have added tests that prove my fix is effective or that my
feature works.
- [ ] I have added (or requested a maintainer to add) the necessary
`Breaking*` or `New Feature` labels where relevant.
- [x] I have done my best to ensure that my PR adheres to [the Fuel Labs
Code Review
Standards](https://github.com/FuelLabs/rfcs/blob/master/text/code-standards/external-contributors.md).
- [x] I have requested a review from the relevant team or maintainers.
## Description
This refactors the module evaluation order handling so we pre-compute it
for modules and store it for further usage by subsequent compiler
passes.
## Checklist
- [ ] I have linked to any relevant issues.
- [ ] I have commented my code, particularly in hard-to-understand
areas.
- [ ] I have updated the documentation where relevant (API docs, the
reference, and the Sway book).
- [ ] I have added tests that prove my fix is effective or that my
feature works.
- [ ] I have added (or requested a maintainer to add) the necessary
`Breaking*` or `New Feature` labels where relevant.
- [ ] I have done my best to ensure that my PR adheres to [the Fuel Labs
Code Review
Standards](https://github.com/FuelLabs/rfcs/blob/master/text/code-standards/external-contributors.md).
- [ ] I have requested a review from the relevant team or maintainers.
## Description
Fixes#1846Fixes#5691
## Checklist
- [x] I have linked to any relevant issues.
- [x] I have commented my code, particularly in hard-to-understand
areas.
- [ ] I have updated the documentation where relevant (API docs, the
reference, and the Sway book).
- [x] I have added tests that prove my fix is effective or that my
feature works.
- [x] I have added (or requested a maintainer to add) the necessary
`Breaking*` or `New Feature` labels where relevant.
- [x] I have done my best to ensure that my PR adheres to [the Fuel Labs
Code Review
Standards](https://github.com/FuelLabs/rfcs/blob/master/text/code-standards/external-contributors.md).
- [x] I have requested a review from the relevant team or maintainers.
---------
Co-authored-by: Vaivaswatha Nagaraj <vaivaswatha.nagaraj@fuel.sh>
Co-authored-by: xunilrj <xunilrj@hotmail.com>
## Description
This PR contains three fixes:
1 - Fix to partial monomorphization. The fix was just removing a
"quick-fix" that was merged and it is not needed anymore after
https://github.com/FuelLabs/sway/pull/5684;
2 - To play safe, `compile_fn_call` cache will not be used for
autogenerated code;
3 - `experimental-new-encoding` was not being copied in all cases.
Two other small changes are:
1 - `Buffer` struct does not offer a generic `push` anymore, because
this function was extremely deceiving for primitive datatypes which have
memory layouts very different from the encoding memory layout (u16, u32
etc...);
2 - Changing name of one test to make the test more obvious and easier
to run just one specific test.
## Checklist
- [ ] I have linked to any relevant issues.
- [ ] I have commented my code, particularly in hard-to-understand
areas.
- [ ] I have updated the documentation where relevant (API docs, the
reference, and the Sway book).
- [ ] I have added tests that prove my fix is effective or that my
feature works.
- [ ] I have added (or requested a maintainer to add) the necessary
`Breaking*` or `New Feature` labels where relevant.
- [ ] I have done my best to ensure that my PR adheres to [the Fuel Labs
Code Review
Standards](https://github.com/FuelLabs/rfcs/blob/master/text/code-standards/external-contributors.md).
- [ ] I have requested a review from the relevant team or maintainers.
## Description
## Checklist
- [x] I have linked to any relevant issues.
- [x] I have commented my code, particularly in hard-to-understand
areas.
- [x] I have updated the documentation where relevant (API docs, the
reference, and the Sway book).
- [x] I have added tests that prove my fix is effective or that my
feature works.
- [x] I have added (or requested a maintainer to add) the necessary
`Breaking*` or `New Feature` labels where relevant.
- [x] I have done my best to ensure that my PR adheres to [the Fuel Labs
Code Review
Standards](https://github.com/FuelLabs/rfcs/blob/master/text/code-standards/external-contributors.md).
- [x] I have requested a review from the relevant team or maintainers.
Signed-off-by: thirdkeyword <fliterdashen@gmail.com>
Co-authored-by: Sophie Dankel <47993817+sdankel@users.noreply.github.com>
## Description
This PR closes https://github.com/FuelLabs/sway/issues/5566.
Contracts now can have a special function decorated with `#[fallback]`
which is called when the contract method selection fails.
This function for all intents and purposes works as a standard contract
method, so:
- it cannot call others contracts methods;
- it has the same limitations of inputs and outputs.
This function can return a value like a normal contract would, or it can
use the `__contract_ret` intrinsics to return any value.
## Checklist
- [x] I have linked to any relevant issues.
- [x] I have commented my code, particularly in hard-to-understand
areas.
- [ ] I have updated the documentation where relevant (API docs, the
reference, and the Sway book).
- [x] I have added tests that prove my fix is effective or that my
feature works.
- [x] I have added (or requested a maintainer to add) the necessary
`Breaking*` or `New Feature` labels where relevant.
- [x] I have done my best to ensure that my PR adheres to [the Fuel Labs
Code Review
Standards](https://github.com/FuelLabs/rfcs/blob/master/text/code-standards/external-contributors.md).
- [x] I have requested a review from the relevant team or maintainers.
This PR adds support to express line number mapping from assembly to
source using DWARF object files. Since there are offset details that are
specific to the Fuel ISA, I'm using the existing `SourceMap` that's
already being built, rather than generating directly from the
instructions.
An important point to note is that DWARF uses line-col to express
positions rather than an absolute offset-from-start format. To avoid
computing line-col from absolute positions (which requires re-reading
the source files), I'm changing the source position format everywhere to
line-col. This also simplifies our code at many places (for example, in
`forc-debug`). The only exception is the `addr2line` tool, where the use
of absolute positions is in a function that seemed quite complex to me
to attempt changes. So I have, temporarily, added code to compute
absolute positions from line-col information and then reuse the existing
functions. This is inefficient, but I think that's _okay_ since it's a
standalone tool and the whole thing runs just once for every command
invocation.
The DWARF information is written to a file specified by the `-g` flag
and can be verified using `llvm-dwarfdump --debug-line ./test.obj`. If
the argument to the `-g` flag is a filename that ends with `.json`, then
the existing JSON format is written out, otherwise it's a DWARF object
in an ELF container.
---------
Co-authored-by: Joao Matos <joao@tritao.eu>
Co-authored-by: Sophie Dankel <47993817+sdankel@users.noreply.github.com>
## Description
This PR fix predicate entry function decoding arguments.
## Checklist
- [ ] I have linked to any relevant issues.
- [ ] I have commented my code, particularly in hard-to-understand
areas.
- [ ] I have updated the documentation where relevant (API docs, the
reference, and the Sway book).
- [ ] I have added tests that prove my fix is effective or that my
feature works.
- [ ] I have added (or requested a maintainer to add) the necessary
`Breaking*` or `New Feature` labels where relevant.
- [ ] I have done my best to ensure that my PR adheres to [the Fuel Labs
Code Review
Standards](https://github.com/FuelLabs/rfcs/blob/master/text/code-standards/external-contributors.md).
- [ ] I have requested a review from the relevant team or maintainers.
## Description
This PR makes `source_id` optional when auto generation code.
Specifically some contracts don't necessarily must have contract
methods.
## Checklist
- [ ] I have linked to any relevant issues.
- [ ] I have commented my code, particularly in hard-to-understand
areas.
- [ ] I have updated the documentation where relevant (API docs, the
reference, and the Sway book).
- [ ] I have added tests that prove my fix is effective or that my
feature works.
- [ ] I have added (or requested a maintainer to add) the necessary
`Breaking*` or `New Feature` labels where relevant.
- [ ] I have done my best to ensure that my PR adheres to [the Fuel Labs
Code Review
Standards](https://github.com/FuelLabs/rfcs/blob/master/text/code-standards/external-contributors.md).
- [ ] I have requested a review from the relevant team or maintainers.
## Description
In some cases args type checking is returning an error in the handler
but not in the result.
With this fix we check the handler for errors too.
Fixes#5659
## Checklist
- [x] I have linked to any relevant issues.
- [x] I have commented my code, particularly in hard-to-understand
areas.
- [ ] I have updated the documentation where relevant (API docs, the
reference, and the Sway book).
- [x] I have added tests that prove my fix is effective or that my
feature works.
- [x] I have added (or requested a maintainer to add) the necessary
`Breaking*` or `New Feature` labels where relevant.
- [x] I have done my best to ensure that my PR adheres to [the Fuel Labs
Code Review
Standards](https://github.com/FuelLabs/rfcs/blob/master/text/code-standards/external-contributors.md).
- [x] I have requested a review from the relevant team or maintainers.
Co-authored-by: João Matos <joao@tritao.eu>
## Description
When the where clause was used in the impl self the type params in the
added uses cases would not be properly unified.
This changes method application to also unify type parameters with impl
self type parameters.
Fixes#5639
## Checklist
- [x] I have linked to any relevant issues.
- [x] I have commented my code, particularly in hard-to-understand
areas.
- [ ] I have updated the documentation where relevant (API docs, the
reference, and the Sway book).
- [x] I have added tests that prove my fix is effective or that my
feature works.
- [x] I have added (or requested a maintainer to add) the necessary
`Breaking*` or `New Feature` labels where relevant.
- [x] I have done my best to ensure that my PR adheres to [the Fuel Labs
Code Review
Standards](https://github.com/FuelLabs/rfcs/blob/master/text/code-standards/external-contributors.md).
- [x] I have requested a review from the relevant team or maintainers.
## Description
This PR fixes a problem when decoding u16 and u32.
## Checklist
- [ ] I have linked to any relevant issues.
- [ ] I have commented my code, particularly in hard-to-understand
areas.
- [ ] I have updated the documentation where relevant (API docs, the
reference, and the Sway book).
- [ ] I have added tests that prove my fix is effective or that my
feature works.
- [ ] I have added (or requested a maintainer to add) the necessary
`Breaking*` or `New Feature` labels where relevant.
- [ ] I have done my best to ensure that my PR adheres to [the Fuel Labs
Code Review
Standards](https://github.com/FuelLabs/rfcs/blob/master/text/code-standards/external-contributors.md).
- [ ] I have requested a review from the relevant team or maintainers.
## Description
This PR implements `DebugWithEngines` and `DeclUniqueId` for
`ParsedDeclId`.
## Checklist
- [x] I have linked to any relevant issues.
- [x] I have commented my code, particularly in hard-to-understand
areas.
- [x] I have updated the documentation where relevant (API docs, the
reference, and the Sway book).
- [x] I have added tests that prove my fix is effective or that my
feature works.
- [x] I have added (or requested a maintainer to add) the necessary
`Breaking*` or `New Feature` labels where relevant.
- [x] I have done my best to ensure that my PR adheres to [the Fuel Labs
Code Review
Standards](https://github.com/FuelLabs/rfcs/blob/master/text/code-standards/external-contributors.md).
- [x] I have requested a review from the relevant team or maintainers.
## Description
This PR introduces a new symbol collection step to the compiler
pipeline, that is done after parsing but before type checking.
It collects identifiers for the top level entities in the AST and allows
referencing through a parsed declaration id.
These will be used in a subsequent PR that introduces a new AST order
resolving pass that runs before type checking, thus allowing us to get
rid of the deferred monormophization "hack", simplifying some things and
eliminating some bugs.
## Checklist
- [x] I have linked to any relevant issues.
- [x] I have commented my code, particularly in hard-to-understand
areas.
- [x] I have updated the documentation where relevant (API docs, the
reference, and the Sway book).
- [x] I have added tests that prove my fix is effective or that my
feature works.
- [x] I have added (or requested a maintainer to add) the necessary
`Breaking*` or `New Feature` labels where relevant.
- [x] I have done my best to ensure that my PR adheres to [the Fuel Labs
Code Review
Standards](https://github.com/FuelLabs/rfcs/blob/master/text/code-standards/external-contributors.md).
- [x] I have requested a review from the relevant team or maintainers.
## Description
This PR makes sure we assign specific module ids to the generated auto
impl decls.
Helps with https://github.com/FuelLabs/sway/issues/5698.
## Checklist
- [x] I have linked to any relevant issues.
- [x] I have commented my code, particularly in hard-to-understand
areas.
- [x] I have updated the documentation where relevant (API docs, the
reference, and the Sway book).
- [x] I have added tests that prove my fix is effective or that my
feature works.
- [x] I have added (or requested a maintainer to add) the necessary
`Breaking*` or `New Feature` labels where relevant.
- [x] I have done my best to ensure that my PR adheres to [the Fuel Labs
Code Review
Standards](https://github.com/FuelLabs/rfcs/blob/master/text/code-standards/external-contributors.md).
- [x] I have requested a review from the relevant team or maintainers.
---------
Co-authored-by: IGI-111 <igi-111@protonmail.com>
## Description
This PR implements references to mutable values, `&mut T`, as defined in
[references](https://github.com/FuelLabs/sway-rfcs/blob/ironcev/amend-references/files/0010-references.sw).
The overall effort related to references is tracked in #5063.
References to mutable values:
- can be created using `&mut`
- can be used in type system wherever a type is expected
- coerce to references (`&mut T` -> `&T`) and thus can be used wherever
"regular" references are expected
- can be passed to and returned from functions
The last point is a step in direction of replacing `ref mut` function
parameters with either `&` or `&mut`.
References to mutable values cannot be taken on constants and immutable
variables (see errors below). (Note that we slightly departure from the
Rust counterparts here. Rust allows `&mut` on constants but issues a
warning.)
Additionally, the PR implements `Diagnostic` for the
`ConflictingImplsForTraitAndType` and shows the already existing,
conflicting implementation of the trait.
## Demo
For more demos, see the tests in this PR.
```Sway
let mut x = 123u32;
let r_m_x: &mut u32 = &mut x;
fn fun(p: &mut T) -> &mut T { ... }
impl Foo for &mut &mut T { ... }
type RefMutRefMut = &mut &mut u64;
```
## Checklist
- [x] I have linked to any relevant issues.
- [x] I have commented my code, particularly in hard-to-understand
areas.
- [ ] I have updated the documentation where relevant (API docs, the
reference, and the Sway book).
- [x] I have added tests that prove my fix is effective or that my
feature works.
- [ ] I have added (or requested a maintainer to add) the necessary
`Breaking*` or `New Feature` labels where relevant.
- [x] I have done my best to ensure that my PR adheres to [the Fuel Labs
Code Review
Standards](https://github.com/FuelLabs/rfcs/blob/master/text/code-standards/external-contributors.md).
- [x] I have requested a review from the relevant team or maintainers.
## Description
This PR ports `TraitItem::TraitFn` to be stored in the parsed
declaration engine, thus allowing it to be referenced as a parsed
declaration id, which will be required by a future AST ordering pass.
## Checklist
- [x] I have linked to any relevant issues.
- [x] I have commented my code, particularly in hard-to-understand
areas.
- [x] I have updated the documentation where relevant (API docs, the
reference, and the Sway book).
- [x] I have added tests that prove my fix is effective or that my
feature works.
- [x] I have added (or requested a maintainer to add) the necessary
`Breaking*` or `New Feature` labels where relevant.
- [x] I have done my best to ensure that my PR adheres to [the Fuel Labs
Code Review
Standards](https://github.com/FuelLabs/rfcs/blob/master/text/code-standards/external-contributors.md).
- [x] I have requested a review from the relevant team or maintainers.
## Description
Enables a did_change stress test with random wait times between each
trigger to emulate random key typing. This then internally kicks off
compilation and triggers garbage collection every 3 keystrokes.
I had intended to include this test when garbage collection was
implemented in #5251. However, at that time, we were only performing GC
every 10th keystroke and it was overloading the stack in CI. Now that we
have reduced this to every 3rd keystroke it seems to be fine for CI.
Unfortunately, this test wasn't running to catch a bug that slipped
through in #5306. As such, this PR currently allows a way for us to
reproduce this error but won't be able to pass CI until the underlying
issue is resolved.
P.S.: This also adds a temporary fix where we retain elements without a
source id. This is tantamount to leaking them but seems preferable to
disabling the GC altogether.
---------
Co-authored-by: xunilrj <xunilrj@hotmail.com>
Co-authored-by: IGI-111 <igi-111@protonmail.com>
## Description
This PR contains the final bit of refactoring before I start fixing the
import problems mentioned in #5498.
The semantics of imports has so far been implemented in the `Module`
struct, and has assumed that the destination path of an import was
relative to `self`. In particular, this allows imports to any submodule
of the current module.
However, whenever that logic was triggered by the compiler the
destination path was always absolute. The only reason this worked was
that the logic was only ever triggered on the root module.
To reflect this behavior I have therefore moved the import semantics to
the `Root` struct, and made it clear in the comments that the paths are
assumed to be absolute.
This also reflects the fact that once the module structure has been
populated with the imported entities we don't actually need the import
logic anymore, so it shouldn't be located in the `Module` struct. (Name
resolution should obviously stay in the `Module` struct, so that logic
has not been moved).
I have left a couple of TODOs in the code to remind myself where I need
to make changes when I start implementing `pub use` (see #3113 ).
## Checklist
- [x] I have linked to any relevant issues.
- [x] I have commented my code, particularly in hard-to-understand
areas.
- [x] I have updated the documentation where relevant (API docs, the
reference, and the Sway book).
- [x] I have added tests that prove my fix is effective or that my
feature works.
- [x] I have added (or requested a maintainer to add) the necessary
`Breaking*` or `New Feature` labels where relevant.
- [x] I have done my best to ensure that my PR adheres to [the Fuel Labs
Code Review
Standards](https://github.com/FuelLabs/rfcs/blob/master/text/code-standards/external-contributors.md).
- [x] I have requested a review from the relevant team or maintainers.
---------
Co-authored-by: IGI-111 <igi-111@protonmail.com>
## Description
This fixes a bug introduced by
https://github.com/FuelLabs/sway/pull/5427, where the ABI was exporting
the "__entry" function instead of the "main".
When parsing and type checking I now mark function as "main" an only
later decide if the compilation will call the main or the entry.
I also moved the `is_entrypoint` method from the `TyAstNode` to the DCE
pass, because it is super specific to that pass.
## Checklist
- [x] I have linked to any relevant issues.
- [x] I have commented my code, particularly in hard-to-understand
areas.
- [x] I have updated the documentation where relevant (API docs, the
reference, and the Sway book).
- [x] I have added tests that prove my fix is effective or that my
feature works.
- [x] I have added (or requested a maintainer to add) the necessary
`Breaking*` or `New Feature` labels where relevant.
- [x] I have done my best to ensure that my PR adheres to [the Fuel Labs
Code Review
Standards](https://github.com/FuelLabs/rfcs/blob/master/text/code-standards/external-contributors.md).
- [x] I have requested a review from the relevant team or maintainers.
## Description
This PR implements the new encoding for contracts/scripts/predicates.
https://github.com/FuelLabs/sway/issues/5512 and
https://github.com/FuelLabs/sway/issues/5513
### Contract Calls
When the new encoding is turned on using `--experimental-new-encoding`,
contract calls like the example below will be "desugarized" differently
now.
```sway
let base_asset_id = BASE_ASSET_ID;
let other_contract_id = ContractId::from(0xa38576787f8900d66e6620548b6da8142b8bb4d129b2338609acd121ca126c10);
let test_contract = abi(ContextTesting, other_contract_id.into());
let returned_contract_id = test_contract.get_id { gas: gas, coins: 0, asset_id: BASE_ASSET_ID.value}(1, 2, 3);
```
and will be transformed to
```sway
let base_asset_id = BASE_ASSET_ID;
let other_contract_id = ContractId::from(0xa38576787f8900d66e6620548b6da8142b8bb4d129b2338609acd121ca126c10);
let test_contract = abi(ContextTesting, other_contract_id.into());
let returned_contract_id = contract_call::<ContractId, _>(other_contract_id.into(), "get_id", (1, 2, 3), coins, asset_id, gas);
```
And the important part is the `contract_call` function in the std
library. This function does all the encoding as necessary and delegates
the actual call to an intrinsic function `__contract_call`. Allowing the
protocol to evolve entirely in Sway.
```sway
pub fn contract_call<T, TArgs>(contract_id: b256, method_name: str, args: TArgs, coins: u64, asset_id: b256, gas: u64) -> T
where
TArgs: AbiEncode
{
let first_parameter = encode(method_name);
let second_parameter = encode(args);
let params = encode(
(
contract_id,
asm(a: first_parameter.ptr()) { a: u64 },
asm(a: second_parameter.ptr()) { a: u64 },
)
);
__contract_call::<T>(params.ptr(), coins, asset_id, gas)
}
```
### Contracts
On the other side, when the flag `--expiremental-new-encoding` is turned
on, the contract specification like the one below is being transformed
into all the decoding and encoding necessary.
The mains points are:
- The compiler generates a function called `__entry` that decodes the
method name and its arguments. The method is selected with a bunch of
`if`s at the moment, because we don´t have `match` for string slices.
Then we `decode` the arguments using the correct type, which is a tuple
with all the function arguments, and expand this tuple calling the
function;
- All the contract functions are converted to global functions prefixed
with `__contract_method`;
- Results are encoded and returned using the intrinsic call `__retd`.
Example:
```sway
abi SomeContract {
fn some_function(a: u64) -> u64;
}
impl SomeContract for Contract {
fn some_function(a: u64) -> u64 {
1
}
}
```
will be transformed into
```sway
fn __entry() {
let method_name = decode_first_parameter();
if method_name == "some_function" {
let args = decode_second_parameter::<(u64,)>();
let result = __contract_method_some_function(args.0);
__retd(encode(result));
}
__revert(0);
}
```
### Scripts and Predicates
The protocol to call scripts and predicates will also change and will be
very similar to contracts. See more above. Now when the flag is turned
on, the `main` function will not be entry point anymore. The compiler
will actually generate an `__entry` function that will decode arguments
and encode the result, like contracts.
For example:
```sway
fn main(a: u64) -> u64 {
1
}
```
will be transformed into
```sway
fn __entry() {
let args = decode_script_data::<(u64,)();
let result = main(args.0);
__retd(encode(result));
}
fn main(a: u64) -> u64 {
1
}
```
## Tests
To facilitate testing this PR introduces three changes to our test
harness:
1 - A new parameter can be called to update test output files (abi and
storage json). This facilitates when we only want to copy and paste
these output files to their respective oracles. Brings the framework
closer to a snapshot one.
```
> cargo r -p test --release -- --update-output-files
```
2 - Depending on the configuration at `test.toml` multiple executions of
the same test will be scheduled. At the moment, tests that depend on the
encoding will run with the `--experimental-new-encoding` flag
automatically. For example:
```
Testing should_pass/language/main_args/main_args_copy_copy ... ok
Testing should_pass/language/main_args/main_args_copy_copy (New Encoding) ... ok
```
3 - A new `script_data_new_encoding` was created because we want to
support and run tests with the two encoding for a time. This is also
what flags tests to run with the flag on automatically.
## Checklist
- [x] I have linked to any relevant issues.
- [x] I have commented my code, particularly in hard-to-understand
areas.
- [x] I have updated the documentation where relevant (API docs, the
reference, and the Sway book).
- [x] I have added tests that prove my fix is effective or that my
feature works.
- [x] I have added (or requested a maintainer to add) the necessary
`Breaking*` or `New Feature` labels where relevant.
- [x] I have done my best to ensure that my PR adheres to [the Fuel Labs
Code Review
Standards](https://github.com/FuelLabs/rfcs/blob/master/text/code-standards/external-contributors.md).
- [x] I have requested a review from the relevant team or maintainers.
## Description
When using nested generic methods, as in the use case added, the lower
method would not be properly replaced and would remain as a trait dummy
method.
The fix was to do find_method_for_type in the ReplaceDecls of
FunctionAplication, this allows the replacement of the function
application dummy method with a proper method implementation.
Fixes#5673
## Checklist
- [x] I have linked to any relevant issues.
- [x] I have commented my code, particularly in hard-to-understand
areas.
- [ ] I have updated the documentation where relevant (API docs, the
reference, and the Sway book).
- [x] I have added tests that prove my fix is effective or that my
feature works.
- [x] I have added (or requested a maintainer to add) the necessary
`Breaking*` or `New Feature` labels where relevant.
- [x] I have done my best to ensure that my PR adheres to [the Fuel Labs
Code Review
Standards](https://github.com/FuelLabs/rfcs/blob/master/text/code-standards/external-contributors.md).
- [x] I have requested a review from the relevant team or maintainers.
The new intrinsic jumps to an absolute address that is fetched from
`MEM[$hp]`. This is more versatile than the `__jmpb_ssp` that was
previously provided. An immediate advantage being that we can have a
working version of `ldc` even in debug
builds.
The test has been modified too, to demonstrate an LDC use that works. A
working version is at
309c79a8f7/test-contract/src/main.sw
Co-authored-by: Sophie Dankel <47993817+sdankel@users.noreply.github.com>
This PR solves a problem for https://github.com/FuelLabs/sway/pull/5427.
When compiling the Rust SDK, the type below hangs the compiler
inside`find_parent`.
```sway
#[allow(dead_code)]
struct MegaExample<T, U> {
a: ([U; 2], T),
b: Vec<([EnumWGeneric<StructWTupleGeneric<StructWArrayGeneric<PassTheGenericOn<T>>>>; 1], u32)>,
}
```
Now, propagating when a change is made, not only the example compiles,
but we should also be able to optimize compilation in general.
In small benchmarks I made here, compiling the Rust SDK had a 2%
improvement (from `0m8.404s` to `0m8.258s`).
## Checklist
- [x] I have linked to any relevant issues.
- [x] I have commented my code, particularly in hard-to-understand
areas.
- [x] I have updated the documentation where relevant (API docs, the
reference, and the Sway book).
- [x] I have added tests that prove my fix is effective or that my
feature works.
- [x] I have added (or requested a maintainer to add) the necessary
`Breaking*` or `New Feature` labels where relevant.
- [x] I have done my best to ensure that my PR adheres to [the Fuel Labs
Code Review
Standards](https://github.com/FuelLabs/rfcs/blob/master/text/code-standards/external-contributors.md).
- [x] I have requested a review from the relevant team or maintainers.
## Description
Fixes#5616 and #5622.
When generating the IR for constant and variable declarations the name
being declared is not added to the local environment if the initializer
cannot be compiled. This can happen when initializing a constant using a
non-constant expression.
Since the name isn't added to the environment, and since we attempt to
continue compilation after an error, any attempt to use the name later
in the code will cause the name to not be found, which in turn causes
the IR converter to throw an ICE because it assumes that all used names
are in scope.
This PR fixes the issue by not throwing the initializer error until the
name has been added to the environment. Note that
1. The initializer must be compiled before the name is added to the
environment. The name should not be in scope during its own
initialization.
2. #5602 is blocking a more elegant solution to this problem. The lack
of a `Never` type means that the call to `convert_resolved_typeid` may
fail if the initializer diverges, and since that call must be made
before the new name can be added to the environment we check for
termination first. The order of operations is therefore currently a)
compile the initializer, then b) check that the initializer did not
result in an error and did not diverge, then c) resolve the type of the
name being declared, then d) add the name to the environment, and
finally e) throw an error if the initializer resulted in an error. Once
the `Never` type is introduced we can move the terminator check to the
end of that sequence.
## Checklist
- [x] I have linked to any relevant issues.
- [x] I have commented my code, particularly in hard-to-understand
areas.
- [x] I have updated the documentation where relevant (API docs, the
reference, and the Sway book).
- [x] I have added tests that prove my fix is effective or that my
feature works.
- [x] I have added (or requested a maintainer to add) the necessary
`Breaking*` or `New Feature` labels where relevant.
- [x] I have done my best to ensure that my PR adheres to [the Fuel Labs
Code Review
Standards](https://github.com/FuelLabs/rfcs/blob/master/text/code-standards/external-contributors.md).
- [x] I have requested a review from the relevant team or maintainers.
The new intrinsic `__jmpb_ssp(offset: u64)` jumps backwards, from the
current value of `$ssp`, by a provided byte offset.
This intrinsic is necessary for dynamically loading and executing a
contract.
Here's an example contract (taken from
https://github.com/FuelLabs/ldc-testing) for dynamically loading another
contract and transferring control to it. The "loadable" contract is the
one used in the link above.
```
// to be compiled in release mode
contract;
abi MyContract {
fn test_function(code_id: ContractId);
}
impl MyContract for Contract {
fn test_function(code_id_p: ContractId) {
let length = asm(code_id, length, word, ssp_saved) {
// Allocate 32 bytes on the heap (we can't use the stack)
addi word zero i32;
aloc word;
lw code_id fp i74;
// Log the ContractID for debugging
logd zero zero code_id word;
// Load the entire contract with LDC
csiz length code_id;
// Save the old ssp
move ssp_saved ssp;
ldc code_id zero length;
// return the ssp difference, to feed __jmpb_ssp.
// This need not always be equal to `length` as `ldc` pads the `length`.
sub length ssp ssp_saved;
length: u64
};
__jmpb_ssp(length)
}
}
```
The intrinsic generates code to do the following:
1. Subract offset from `$ssp`.
2. Subtract `$is` since `$jmp` adds it back.
3. Divide by 4 since `Jmp` multiplies by 4
4. Jump to the computed value.
Closes#5563
## Description
This PR:
- fixes#5559, #5597, and #5492 by removing the `TypeInfo::Unknown` and
providing the required contextual information to type checking of
referencing, dereferencing, `if`, and `match` expressions respectively.
The contextual information provided is taken from the
`ctx.type_annotation()` but always adapted according to the semantics of
the type-checked expression.
- fixes#5583 and #5581 by combining the contextual information coming
from the `ctx.type_annotation()` with the one coming from the enum and
struct instantiation and declaration.
- fixes#5598 by forcing the name-based and not structure-based
identity. In other words, two enums or structs are considered equal only
if they whole `call_path`s are equal. Up to now, we were expecting only
the enum or struct _names_ to be equal, which was treating types with
same names and structures (variants or fields) as equal although they
were defined in different modules.
The PR also introduces parsing of references to mutable values (`&mut
T`). Since this addition does not overlap with the above bug fixes, it
was left as is, and can be fully ignored during the review. Other
changes related to references to mutable values are removed from the
code from type-checking onward to make this PR only fixing the issues.
Continuation on references to mutable values will be done in a separate
PR.
Closes#5559, #5597, #5492, #5583, #5581, #5598.
## Checklist
- [x] I have linked to any relevant issues.
- [x] I have commented my code, particularly in hard-to-understand
areas.
- [ ] I have updated the documentation where relevant (API docs, the
reference, and the Sway book).
- [x] I have added tests that prove my fix is effective or that my
feature works.
- [ ] I have added (or requested a maintainer to add) the necessary
`Breaking*` or `New Feature` labels where relevant.
- [x] I have done my best to ensure that my PR adheres to [the Fuel Labs
Code Review
Standards](https://github.com/FuelLabs/rfcs/blob/master/text/code-standards/external-contributors.md).
- [x] I have requested a review from the relevant team or maintainers.
## Description
So this is another incremental step towards a following PR with the
initial collection context migration.
[Refactor namespace access in type check
context.](a213a2d313)
[Remove clones from scoped namespace
usage.](db5140c2e2)
[Simplify type-check context creation from a
namespace.](e8eca9f082)
No functional changes intended.
## Checklist
- [x] I have linked to any relevant issues.
- [x] I have commented my code, particularly in hard-to-understand
areas.
- [x] I have updated the documentation where relevant (API docs, the
reference, and the Sway book).
- [x] I have added tests that prove my fix is effective or that my
feature works.
- [x] I have added (or requested a maintainer to add) the necessary
`Breaking*` or `New Feature` labels where relevant.
- [x] I have done my best to ensure that my PR adheres to [the Fuel Labs
Code Review
Standards](https://github.com/FuelLabs/rfcs/blob/master/text/code-standards/external-contributors.md).
- [x] I have requested a review from the relevant team or maintainers.
## Description
This small change fixes#5580 as described - only needed 1 additional
change and small test fix
Ping @xunilrj
## Checklist
- [x] I have linked to any relevant issues.
- [ ] I have commented my code, particularly in hard-to-understand
areas.
- [ ] I have updated the documentation where relevant (API docs, the
reference, and the Sway book).
- [ ] I have added tests that prove my fix is effective or that my
feature works.
- [x] I have added (or requested a maintainer to add) the necessary
`Breaking*` or `New Feature` labels where relevant.
- [x] I have done my best to ensure that my PR adheres to [the Fuel Labs
Code Review
Standards](https://github.com/FuelLabs/rfcs/blob/master/text/code-standards/external-contributors.md).
- [x] I have requested a review from the relevant team or maintainers.
## Description
While doing the argument type checking we set the type annotation to
Unknown because the method has not been resolved yet.
With this change when we fail to type check an argument expression with
the annotation Unknown we try again after resolving the method name.
The reason we do the double pass on argument type checking is because we
need the types from the arguments to resolve the method but sometimes we
need the method resolved for resolving the argument expressions with the
type parameters.
Fixes#5614
## Checklist
- [x] I have linked to any relevant issues.
- [x] I have commented my code, particularly in hard-to-understand
areas.
- [ ] I have updated the documentation where relevant (API docs, the
reference, and the Sway book).
- [x] I have added tests that prove my fix is effective or that my
feature works.
- [x] I have added (or requested a maintainer to add) the necessary
`Breaking*` or `New Feature` labels where relevant.
- [x] I have done my best to ensure that my PR adheres to [the Fuel Labs
Code Review
Standards](https://github.com/FuelLabs/rfcs/blob/master/text/code-standards/external-contributors.md).
- [x] I have requested a review from the relevant team or maintainers.
## Description
Related https://github.com/FuelLabs/sway-vscode-plugin/pull/166
Adds a [DAP](https://microsoft.github.io/debug-adapter-protocol//)
server as a new forc plugin, `forc-debug`. This enables a UI debugging
experience in IDEs such as VSCode.
For now, only debugging forc tests is supported. Users can:
- set breakpoints inside of forc tests
- step through the test, one VM instruction at a time, or continue to
the next breakpoint
- debug through multiple tests sequentially
- see the VM register values in the IDE while debugging
- see the current VM opcode, its inputs, and/or immediate value when
when the VM is stopped
### Screenshots
24e2016c-d96c-4ef6-931f-8a4ce4f1386b
5f0fed49-b278-4074-a1a1-d37de00776f8
### Local testing
1. Install `forc-debug`
2. Copy this sample launch.json in the VSCode workspace with sway code
to debug:
```json
{
"version": "0.2.0",
"configurations": [
{
"type": "sway",
"request": "launch",
"name": "Debug Sway",
"program": "${file}"
}]
}
```
3. Follow [the
steps](https://github.com/FuelLabs/sway-vscode-plugin/blob/master/docs/testing.md)
for testing the VSCode extension
### Limitations
- Breakpoints only work inside of the project/workspace being debugged
- Stack trace support is limited
- Not every line has source maps. Once debugging, "verified" breakpoints
will show red and "unverified" (no source maps) will be greyed out.
- Watch/repl expressions are not yet supported
- Step into/out of is not supported
- If you click "step over" many times in rapid succession, the server
takes a while to catch up.
Closes https://github.com/FuelLabs/sway/issues/5394
## Checklist
- [x] I have linked to any relevant issues.
- [x] I have commented my code, particularly in hard-to-understand
areas.
- [ ] I have updated the documentation where relevant (API docs, the
reference, and the Sway book).
- [ ] I have added tests that prove my fix is effective or that my
feature works.
- [ ] I have added (or requested a maintainer to add) the necessary
`Breaking*` or `New Feature` labels where relevant.
- [ ] I have done my best to ensure that my PR adheres to [the Fuel Labs
Code Review
Standards](https://github.com/FuelLabs/rfcs/blob/master/text/code-standards/external-contributors.md).
- [ ] I have requested a review from the relevant team or maintainers.
---------
Co-authored-by: Vaivaswatha Nagaraj <vaivaswatha.nagaraj@fuel.sh>
Co-authored-by: Vaivaswatha N <vaivaswatha@users.noreply.github.com>
Co-authored-by: IGI-111 <igi-111@protonmail.com>
Co-authored-by: João Matos <joao@tritao.eu>
Co-authored-by: Joshua Batty <joshpbatty@gmail.com>
Co-authored-by: Igor Rončević <ironcev@hotmail.com>
Co-authored-by: Sudhakar Verma <10460978+sudhackar@users.noreply.github.com>
Co-authored-by: Marcos Henrich <marcoshenrich@gmail.com>
Co-authored-by: jjcnn <38888011+jjcnn@users.noreply.github.com>
## Description
We now parse the `!` as a TypeInfo::Never, and remove the usage of empty
enums as Never type in our code.
This commit removes completely the DeterministicallyAborts and
TypeCheckUnificationContext.
The DeterministicallyAborts can be removed because the Never TypeInfo is
now propagated using the type checker. Code blocks that return, break,
continue, or call an expression that returns Never, are marked as Never.
Partially fixes#5562.
## Checklist
- [x] I have linked to any relevant issues.
- [x] I have commented my code, particularly in hard-to-understand
areas.
- [x] I have updated the documentation where relevant (API docs, the
reference, and the Sway book).
- [x] I have added tests that prove my fix is effective or that my
feature works.
- [x] I have added (or requested a maintainer to add) the necessary
`Breaking*` or `New Feature` labels where relevant.
- [x] I have done my best to ensure that my PR adheres to [the Fuel Labs
Code Review
Standards](https://github.com/FuelLabs/rfcs/blob/master/text/code-standards/external-contributors.md).
- [x] I have requested a review from the relevant team or maintainers.
## Description
Add a `#[namespace(my_namespace)]` annotation to allow developers to
avoid storage slot collisions when loading contract code using LDC.
Fixes#5490
## Checklist
- [x] I have linked to any relevant issues.
- [x] I have commented my code, particularly in hard-to-understand
areas.
- [x] I have updated the documentation where relevant (API docs, the
reference, and the Sway book).
- [x] I have added tests that prove my fix is effective or that my
feature works.
- [x] I have added (or requested a maintainer to add) the necessary
`Breaking*` or `New Feature` labels where relevant.
- [x] I have done my best to ensure that my PR adheres to [the Fuel Labs
Code Review
Standards](https://github.com/FuelLabs/rfcs/blob/master/text/code-standards/external-contributors.md).
- [x] I have requested a review from the relevant team or maintainers.
## Description
Closes https://github.com/FuelLabs/sway/issues/5607
Solves part [1] of https://github.com/FuelLabs/sway/issues/5607 by
making "release" the default build plan for forc-deploy. It can still be
overridden, i.e. for the test harness.
Other changes:
- removed the `--release` and `--error-on-warnings` options from
forc-deploy, since those aren't needed.
- refactored `BuildProfile` into its own file and added a test, since I
noticed some of the options weren't deserializing
## Checklist
- [x] I have linked to any relevant issues.
- [ ] I have commented my code, particularly in hard-to-understand
areas.
- [ ] I have updated the documentation where relevant (API docs, the
reference, and the Sway book).
- [ ] I have added tests that prove my fix is effective or that my
feature works.
- [ ] I have added (or requested a maintainer to add) the necessary
`Breaking*` or `New Feature` labels where relevant.
- [ ] I have done my best to ensure that my PR adheres to [the Fuel Labs
Code Review
Standards](https://github.com/FuelLabs/rfcs/blob/master/text/code-standards/external-contributors.md).
- [ ] I have requested a review from the relevant team or maintainers.
## Description
This PR introduces a hierarchy of lexical scopes to namespace modules.
This might still have some bugs to work out, but I don't expect the API
to need to change, so am offloading this right now to be able to unblock
further work we have pending on namespaces/modules.
## Checklist
- [x] I have linked to any relevant issues.
- [x] I have commented my code, particularly in hard-to-understand
areas.
- [x] I have updated the documentation where relevant (API docs, the
reference, and the Sway book).
- [x] I have added tests that prove my fix is effective or that my
feature works.
- [x] I have added (or requested a maintainer to add) the necessary
`Breaking*` or `New Feature` labels where relevant.
- [x] I have done my best to ensure that my PR adheres to [the Fuel Labs
Code Review
Standards](https://github.com/FuelLabs/rfcs/blob/master/text/code-standards/external-contributors.md).
- [x] I have requested a review from the relevant team or maintainers.
## Description
Solves [1] and [2] of https://github.com/FuelLabs/sway/issues/5607
fuel target
other targets
## Checklist
- [x] I have linked to any relevant issues.
- [ ] I have commented my code, particularly in hard-to-understand
areas.
- [ ] I have updated the documentation where relevant (API docs, the
reference, and the Sway book).
- [ ] I have added tests that prove my fix is effective or that my
feature works.
- [ ] I have added (or requested a maintainer to add) the necessary
`Breaking*` or `New Feature` labels where relevant.
- [ ] I have done my best to ensure that my PR adheres to [the Fuel Labs
Code Review
Standards](https://github.com/FuelLabs/rfcs/blob/master/text/code-standards/external-contributors.md).
- [ ] I have requested a review from the relevant team or maintainers.
## Description
This PR refactors the interface for changing the namespace in a type
checking context.
This change is preparatory work for the collecting context work and will
be used in upcoming PRs.
No functional changes intended.
## Checklist
- [x] I have linked to any relevant issues.
- [x] I have commented my code, particularly in hard-to-understand
areas.
- [x] I have updated the documentation where relevant (API docs, the
reference, and the Sway book).
- [x] I have added tests that prove my fix is effective or that my
feature works.
- [x] I have added (or requested a maintainer to add) the necessary
`Breaking*` or `New Feature` labels where relevant.
- [x] I have done my best to ensure that my PR adheres to [the Fuel Labs
Code Review
Standards](https://github.com/FuelLabs/rfcs/blob/master/text/code-standards/external-contributors.md).
- [x] I have requested a review from the relevant team or maintainers.
## Description
An issue was reported by @sdankel on #5477 caused by #5557.
A segmentation fault would occur when the for loop handling was
performed inside the match. This would occur even when that part of the
code was not called. Looks like a weird rust issue.
## Checklist
- [ ] I have linked to any relevant issues.
- [x] I have commented my code, particularly in hard-to-understand
areas.
- [ ] I have updated the documentation where relevant (API docs, the
reference, and the Sway book).
- [ ] I have added tests that prove my fix is effective or that my
feature works.
- [x] I have added (or requested a maintainer to add) the necessary
`Breaking*` or `New Feature` labels where relevant.
- [x] I have done my best to ensure that my PR adheres to [the Fuel Labs
Code Review
Standards](https://github.com/FuelLabs/rfcs/blob/master/text/code-standards/external-contributors.md).
- [x] I have requested a review from the relevant team or maintainers.
Co-authored-by: Igor Rončević <ironcev@hotmail.com>
## Description
This PR does three things:
1. use Rust 1.76.0
2. enable the new `iter_over_hash_type` Clippy lint on relevant crates
3. use `indexmap::IndexMap` to replace any equivalent `HashMap` being
iterated upon with a deterministically ordered alternative
## Checklist
- [x] I have linked to any relevant issues.
- [x] I have commented my code, particularly in hard-to-understand
areas.
- [x] I have updated the documentation where relevant (API docs, the
reference, and the Sway book).
- [x] I have added tests that prove my fix is effective or that my
feature works.
- [x] I have added (or requested a maintainer to add) the necessary
`Breaking*` or `New Feature` labels where relevant.
- [x] I have done my best to ensure that my PR adheres to [the Fuel Labs
Code Review
Standards](https://github.com/FuelLabs/rfcs/blob/master/text/code-standards/external-contributors.md).
- [x] I have requested a review from the relevant team or maintainers.
## Description
This implements an Iterator trait in std-lib, and adds iter() to Vec.
This also adds parsing and desugaring of for loops.
```
for pattern in iterator {
code_block
}
```
is desugared into:
```
let mut iterable = iterator;
while true {
let value_opt = iterable.next();
if value_opt.is_none() {
break;
}
let value = value_opt.unwrap();
code_block
}
```
This also adds for loops documentation to the control flow docs.
We still have to fix this issues:
- #5567
- #5568
- #5570
- #5571Closes#145
## Checklist
- [x] I have linked to any relevant issues.
- [x] I have commented my code, particularly in hard-to-understand
areas.
- [x] I have updated the documentation where relevant (API docs, the
reference, and the Sway book).
- [x] I have added tests that prove my fix is effective or that my
feature works.
- [x] I have added (or requested a maintainer to add) the necessary
`Breaking*` or `New Feature` labels where relevant.
- [x] I have done my best to ensure that my PR adheres to [the Fuel Labs
Code Review
Standards](https://github.com/FuelLabs/rfcs/blob/master/text/code-standards/external-contributors.md).
- [x] I have requested a review from the relevant team or maintainers.
## Description
This PR is the first step of a large refactoring of the `namespace`
module.
The PR removes the `DeRef` and `DeRefMut` implementations in the
`namespace` module, as they make refactoring more difficult to keep
track of. This change means that calls to `Root` and `Module` must now
explicitly go through `Namespace`, since that is the type that is passed
along during typechecking.
I'm submitting the PR now because there have been changes to `master`
while I've been working on this, and the changes are really difficult to
merge, so I want `master` to be relatively stable.
Subsequent PRs will:
- Eliminate the `dst` parameter from all import functions (the
destination is always the current module).
- Move all import functions to `Namespace` (they currently reside in
`Module` where they aren't needed, and where they require a clone the
module path to work).
- Move `Namespace::init` to `Root`, and change how it is used (it
currently holds a constant declaration of `CONTRACT_ID` for contract
modules, and that declaration is cloned into every submodule. The nice
way to do this is to declare it in the root module and implicitly import
it into each submodule).
- Eliminate the `Option` part of `Module::name` (modules are not allowed
to be nameless - the option only exists because they root module can in
principle be nameless, but it ought to be named based on the package the
module structure is in).
- Fix the import and name resolution functions so that they don't treat
absolute paths as relative (this should fix some of the performance
issues we're seeing).
- Make `Namespace::root` a reference so that the entire module structure
isn't cloned for every submodule (this requires a change in ownership of
the root module).
- Introduce reexports in the form of `pub use`.
This PR is the first step in fixing #5498.
## Checklist
- [x] I have linked to any relevant issues.
- [x] I have commented my code, particularly in hard-to-understand
areas.
- [x] I have updated the documentation where relevant (API docs, the
reference, and the Sway book).
- [x] I have added tests that prove my fix is effective or that my
feature works.
- [x] I have added (or requested a maintainer to add) the necessary
`Breaking*` or `New Feature` labels where relevant.
- [x] I have done my best to ensure that my PR adheres to [the Fuel Labs
Code Review
Standards](https://github.com/FuelLabs/rfcs/blob/master/text/code-standards/external-contributors.md).
- [x] I have requested a review from the relevant team or maintainers.
## Description
Explicit returns did not have any type annotation causing the method
disambiguation to fail.
The solution was to add a new field to the `TypeCheckContext` called
`function_type_annotation`. This field is initialized only in the
type_check of function declarations. The field is used to set the
type_annotation one while doing the type check of explicit return
expressions.
With this change `TypeCheckUnification` was also simplified as the
unification is now done with `function_type_annotation`. The result was
a few functions that are no longer required to be removed.
Closes#5518
## Checklist
- [x] I have linked to any relevant issues.
- [x] I have commented my code, particularly in hard-to-understand
areas.
- [ ] I have updated the documentation where relevant (API docs, the
reference, and the Sway book).
- [x] I have added tests that prove my fix is effective or that my
feature works.
- [x] I have added (or requested a maintainer to add) the necessary
`Breaking*` or `New Feature` labels where relevant.
- [x] I have done my best to ensure that my PR adheres to [the Fuel Labs
Code Review
Standards](https://github.com/FuelLabs/rfcs/blob/master/text/code-standards/external-contributors.md).
- [x] I have requested a review from the relevant team or maintainers.
---------
Co-authored-by: João Matos <joao@tritao.eu>
