Remove all remaining uses of TypeIds as instruction operands. (#5280)

In preparation for shifting from `TypeId`s potentially representing
attached types to always representing unattached types, using
[terminology suggested on
Discord](https://discord.com/channels/655572317891461132/963846118964350976/1359286326779973712).
This change causes us to track slightly more type spelling information
through SemIR.

One change that has significant impact on the SemIR output is that we
now build a `struct_type` instruction in each class representing the
types of the fields, including the spelling used for those types. This
is now no longer always identical to the corresponding canonical
`struct_type` for the object representation, so it's built separately
and owned by the class.

Also remove `TypeBlock` support entirely, as its only use was
representing `TupleType`s, which now use an `InstBlock`.
This commit is contained in:
Richard Smith
2025-04-10 20:53:42 +00:00
committed by GitHub
parent 1445ec9e4f
commit a74ca9071b
350 changed files with 1880 additions and 1189 deletions
+5 -4
View File
@@ -75,12 +75,13 @@ auto HandleParseNode(Context& context, Parse::TuplePatternId node_id) -> bool {
.PopAndDiscardSoloNodeId<Parse::NodeKind::TuplePatternStart>();
const auto& inst_block = context.inst_blocks().Get(refs_id);
llvm::SmallVector<SemIR::TypeId> type_ids;
type_ids.reserve(inst_block.size());
llvm::SmallVector<SemIR::InstId> type_inst_ids;
type_inst_ids.reserve(inst_block.size());
for (auto inst : inst_block) {
type_ids.push_back(context.insts().Get(inst).type_id());
type_inst_ids.push_back(
context.types().GetInstId(context.insts().Get(inst).type_id()));
}
auto type_id = GetTupleType(context, type_ids);
auto type_id = GetTupleType(context, type_inst_ids);
context.node_stack().Push(
node_id,
AddPatternInst<SemIR::TuplePattern>(