d8076b9333
Surface rename of the signed integer family: s1..s64 become i1..i64
(u1..u64, usize, isize unchanged). 'string' keeps the s-prefix arm in
name classification; width parsing moves to the i-prefix arm next to
isize.
Internal TypeId tags follow the surface (.s8/.s16/.s32/.s64 ->
.i8/.i16/.i32/.i64), as do mono-key mangle fragments (ptr_i64,
tu_i64_bool) and all display/diagnostic formatting (i{d}).
Migrated in the same sweep: stdlib + examples + issue repros + FFI C
companions (shared symbol names like ffi_id_i64), expected
stdout/stderr/ir snapshots, specs.md, readme.md, CLAUDE.md/AGENTS.md,
implementation_plan.md, docs/, issue writeups. Vendored stb_image and
historical flow state left untouched.
zig build test: 426/426; examples suite: 595/595.
49 lines
2.2 KiB
Plaintext
49 lines
2.2 KiB
Plaintext
// E6BR-5 regression — the open nested-pattern ambiguity hole that PAUSED E6b,
|
|
// re-filed under Fork C as a RESOLVER-TARGET regression (NOT an E6b attempt-6).
|
|
//
|
|
// Shape: a param-impl SOURCE pattern whose concrete `*Box` leaf is NESTED inside
|
|
// an inner parameterized form that itself carries unbound pack parts
|
|
// (`Closure(*Box, ..$inner) -> $IR`). On flow/stdlib/E6b the bug lived in
|
|
// `walkConcreteSigArgs` (lower.zig:14686): it SKIPS any direct arg that has an
|
|
// unbound part instead of RECURSING into it, so the nested concrete `*Box` leaf
|
|
// is never ambiguity-checked — `impl ... for Closure(Closure(*Box, ..)->.., ..)->..`
|
|
// compiled rc=0 while the DIRECT `Closure(*Box, ..)` form (0829) errored. It was
|
|
// the 3rd consecutive major on the parameterized-pattern surface
|
|
// (bare E6BR-2 -> wrapped E6BR-4 -> nested E6BR-5).
|
|
//
|
|
// CURRENT BASE (wt-stdlib-base, E6b unmerged): both the direct AND the nested
|
|
// concrete leaf silently resolve via the pre-E6BR-4 no-author wrapper path
|
|
// (`type_bridge.resolveTemplateSignatureType`, global last-wins) — this program
|
|
// exits 0 with NO diagnostic. Run it to see the fail-before.
|
|
//
|
|
// TARGET (Fork C, flips active+green at S3.9): the whole-AST resolver walks EVERY
|
|
// reference position — including nested parameterized-pattern leaves — so the
|
|
// nested `Box` is ambiguity-checked like any other reference and the build emits
|
|
// the loud "type 'Box' is ambiguous" diagnostic and exits 1. The exact golden
|
|
// bytes are produced by the resolver at S3.9 (no oracle produces them today);
|
|
// see ../expected/e6br5-nested-pack-source-ambiguous.target.md for the spec.
|
|
//
|
|
// WHY SUBSUMED (one line): a resolver that resolves every reference position has
|
|
// no notion of "skip a nested arg with an unbound sibling" — the nested `*Box`
|
|
// leaf is resolved by construction, so the hole cannot exist.
|
|
|
|
#import "modules/std.sx";
|
|
#import "e6br5-nested-pack-source-ambiguous/a.sx";
|
|
#import "e6br5-nested-pack-source-ambiguous/b.sx";
|
|
|
|
Block :: struct { tag: i32; }
|
|
|
|
Sink :: protocol(T: Type) {
|
|
convert :: () -> T;
|
|
}
|
|
|
|
impl Sink(Block) for Closure(Closure(*Box, ..$inner) -> $IR, ..$args) -> $R {
|
|
convert :: (self: Closure(Closure(*Box, ..$inner) -> $IR, ..$args) -> $R) -> Block {
|
|
.{ tag = 0 }
|
|
}
|
|
}
|
|
|
|
main :: () -> i32 {
|
|
0
|
|
}
|