cleanup: drop resolved-issue citations from src comments
Sweep all src/**.zig comments that cite resolved issues (issue NNNN / fix-NNNN / KB-N): the invariant or mechanism each comment states is kept; the historical citation is dropped, per the no-conclusion-comments rule. Pure-history parentheticals are removed outright. References to the 16 still-open issues (0030, 0041-0056) are untouched, as are test NAMES carrying regression provenance (matching the sanctioned "Regression (issue NNNN)" example-header convention). Also removes the issues/0019-import-non-transitive-c-scope/ fixture dir — the issue is superseded and its behavior is covered by examples/0706-modules-import-non-transitive.sx (the .md writeup stays). issues/0030's repro .sx stays: that issue is an open feature request. Gate: zig build OK; zig build test 426/426; run_examples 541/0; zero expected/ snapshot churn.
This commit is contained in:
@@ -1082,7 +1082,7 @@ test "lower: vectorLaneIndex maps swizzle components, colour aliases, rejects no
|
||||
try std.testing.expectEqual(@as(?u32, 3), Lowering.vectorLaneIndex("a"));
|
||||
// Any non-lane field is rejected (null) so the read and write paths share
|
||||
// one rule — a non-lane store no longer falls through to an .unresolved
|
||||
// pointee that panics at LLVM emission (issue 0086).
|
||||
// pointee that panics at LLVM emission.
|
||||
try std.testing.expectEqual(@as(?u32, null), Lowering.vectorLaneIndex("q"));
|
||||
try std.testing.expectEqual(@as(?u32, null), Lowering.vectorLaneIndex("xy"));
|
||||
try std.testing.expectEqual(@as(?u32, null), Lowering.vectorLaneIndex("len"));
|
||||
@@ -1301,10 +1301,10 @@ fn countRealBodies(module: *ir_mod.Module, name: []const u8) usize {
|
||||
return n;
|
||||
}
|
||||
|
||||
// fix-0102b: two flat-imported modules each author `greet`. The first-wins merge
|
||||
// two flat-imported modules each author `greet`. The first-wins merge
|
||||
// keeps a.sx's author in the merged decl list (the WINNER) and drops b.sx's,
|
||||
// which the `module_decls` raw facts still retain (0102a). `main` itself can't bare-call `greet`
|
||||
// — under fix-0102c two flat authors make that ambiguous — so it calls a.sx's
|
||||
// — with two flat authors this is ambiguous; two flat authors make that ambiguous — so it calls a.sx's
|
||||
// `use_greet` wrapper, whose own-author call to `greet` binds a.sx's winner.
|
||||
// BEFORE the identity-addressable pass, only the winner has a real body — the
|
||||
// shadowed author has no slot at all (the pre-fix symptom: one `greet`).
|
||||
@@ -1420,8 +1420,8 @@ test "lower: shadowed same-name author gets its own FuncId + real body (fix-0102
|
||||
// pointer for BOTH same-name authors — the exact pointers `fn_ast_map` and
|
||||
// the `module_decls` raw facts carry — not a per-iteration switch-capture
|
||||
// temporary. If the winner were keyed by `&fd` (the scanDecls bug), this
|
||||
// lookup by the stable `fn_ast_map` pointer would miss (null). fix-0102c
|
||||
// routes calls through exactly these pointers, so the round-trip must hold.
|
||||
// lookup by the stable `fn_ast_map` pointer would miss (null). Bare-call
|
||||
// routing goes through exactly these pointers, so the round-trip must hold.
|
||||
const winner_fd = lowering.program_index.fn_ast_map.get("greet").?;
|
||||
const winner_fid = lowering.fn_decl_fids.get(winner_fd);
|
||||
try std.testing.expect(winner_fid != null);
|
||||
@@ -1442,7 +1442,7 @@ test "lower: shadowed same-name author gets its own FuncId + real body (fix-0102
|
||||
try std.testing.expect(shadow_fid != null);
|
||||
try std.testing.expect(shadow_fid.? != winner_fid.?);
|
||||
|
||||
// fix-0102c / Phase C: THE bare-name selector routes per caller file over the
|
||||
// Phase C: THE bare-name selector routes per caller file over the
|
||||
// Phase A author collector. `main` flat-imports two `greet` authors and is its
|
||||
// own author of neither → a bare `greet()` from `main` is ambiguous. a.sx
|
||||
// authors the WINNER, so its bare `greet` resolves through the existing path
|
||||
|
||||
Reference in New Issue
Block a user