This trips over a few other limitations, but in the interests of incremental development I'm starting here & I'll look at the issues with -verify and filesystem checks (the fact that the behavior depends on the existence of a 'foo' directory even though it shouldn't need it), etc.
Details
- Reviewers
rsmith - Commits
- rZORG691819924742: -frewrite-imports: Add support for wildcard rules in umbrella modules with
rZORG66a0d3c18649: -frewrite-imports: Add support for wildcard rules in umbrella modules with
rG691819924742: -frewrite-imports: Add support for wildcard rules in umbrella modules with
rG66a0d3c18649: -frewrite-imports: Add support for wildcard rules in umbrella modules with
rG89e58ddb2868: -frewrite-imports: Add support for wildcard rules in umbrella modules with
rL360195: -frewrite-imports: Add support for wildcard rules in umbrella modules with
rC360195: -frewrite-imports: Add support for wildcard rules in umbrella modules with
Diff Detail
- Repository
- rC Clang
Event Timeline
lib/Basic/Module.cpp | ||
---|---|---|
324 | Maybe findOrInferSubmodule; the current name suggests that this would always create a submodule if one doesn't exist (like ModuleMap::findOrCreateModule does). | |
lib/Lex/Pragma.cpp | ||
1586–1587 | Remove fixed FIXME please :) | |
test/Modules/preprocess-umbrella.cpp | ||
2–3 | FWIW, I think that probably happens when parsing the umbrella "foo" directive. |
test/Modules/preprocess-umbrella.cpp | ||
---|---|---|
2–3 | Thanks for the pointer - yeah, I'd come across that too. Both the ModuleMap and the Module ( https://clang.llvm.org/doxygen/classclang_1_1Module.html#acb7c8570da610974e567675303baa77b ) depend on a non-null DirectoryEntry to record the modules umbrella-ness. Any ideas on how I might go about removing that limitation? |
Maybe findOrInferSubmodule; the current name suggests that this would always create a submodule if one doesn't exist (like ModuleMap::findOrCreateModule does).