* All the data types from `@import("builtin")` are moved to
`@import("std").builtin`. The target-related types are moved
to `std.Target`. This allows the data types to have methods, such as
`std.Target.current.isDarwin()`.
* `std.os.windows.subsystem` is moved to
`std.Target.current.subsystem`.
* Remove the concept of the panic package from the compiler
implementation. Instead, `std.builtin.panic` is always the panic
function. It checks for `@hasDecl(@import("root"), "panic")`,
or else provides a default implementation.
This is an important step for multibuilds (#3028). Without this change,
the types inside the builtin namespace look like different types, when
trying to merge builds with different target settings. With this change,
Zig can figure out that, e.g., `std.builtin.Os` (the enum type) from one
compilation and `std.builtin.Os` from another compilation are the same
type, even if the target OS value differs.
- during diagnostics the string representation for root was empty
and now is `(root)`
- retrofitted all other namespace-qualified type naming to elide
prefixing with root
closes#2032
Previously if the type parameter was a pointer, it would assert that the
size of the type was resolved. It used to be that the size of pointers was
always resolved, however with lazy values, pointers gained the
possibility of not having their size resolved.
Now, type_allowed_in_extern triggers the resolution of whether a pointer
is zero bits, and returns a possible error if the resolution fails.
This fixes a compiler assertion when building the
[zootdeck project](https://github.com/donpdonp/zootdeck). I do not have
a test case reduction for the issue.
This breaks behavior tests as well as compile error notes for generic
function calls. However it introduces better circular dependency compile
errors.
The next step is to add Lazy Values to fix the regressions.
It used to be that usingnamespace was only allowed at top level. This
made it OK to put the state inside the AST node data structure. However,
now usingnamespace can occur inside any aggregate data structure, and
therefore the state must be in the TopLevelDeclaration rather than in
the AST node.
There were two other problems with the usingnamespace implementation:
* It was passing the wrong destination ScopeDecl, so it could cause an
incorrect error such as "import of file outside package path".
* When doing `usingnamespace` on a file that already had
`pub usingnamespace` in it would "steal" the usingnamespace, causing
incorrect "use of undeclared identifier" errors in the target file.
closes#2632closes#2580
because of this example:
```zig
export fn entry(b: bool) usize {
var runtime = [1]i32{3};
comptime var i: usize = 0;
inline while (i < 2) : (i += 1) {
const result = if (i == 0) [1]i32{2} else runtime;
}
comptime {
return i;
}
}
```
The problem is that the concept of "resetting" a result location,
introduced in the previous commit, cannot handle elision scopes.
This concept is inherently broken with inline loops.
Previously, the symbol name parameter of `@export` would be ignored for
variables, and the variable name would be used for the symbol name.
Now it works as expected.
See #2679
Not tested yet, but it builds.
This closes#761, and lays the groundwork for fixing the remaining
false positive "foo depends on itself" bugs, such as #624.
It also lays the groundwork for implementing ability to specify
alignment of fields (#1512).
