[CAS] Cache symbol graph outputs #77935
Conversation
The symbol graph output from a module can contain an arbitrary number of files, depending on what extensions it contains, so cache a list of symbol graph files with their base name and contents so that they can be replayed. rdar://140286819
benlangmuir
requested review from
artemcm,
tshortli and
DougGregor
as code owners
|
@swift-ci please smoke test |
| initBackend(InputsAndOutputs); | ||
| } | ||
|
|
||
| llvm::Expected<std::unique_ptr<llvm::vfs::OutputFileImpl>> | ||
| createFileImpl(llvm::StringRef ResolvedPath, | ||
| std::optional<llvm::vfs::OutputConfig> Config) { | ||
| auto ProducingInput = OutputToInputMap.find(ResolvedPath); | ||
| if (ProducingInput == OutputToInputMap.end() && |
There was a problem hiding this comment.
Instead of doing this special condition, I prefer add symbol graph to SupplementaryOutputPaths and compute the path in ArgsToFrontendOutputsConverter.cpp.
Hope it is not too bad to pass the output path to the location where the file is generated.
There was a problem hiding this comment.
I prefer add symbol graph to SupplementaryOutputPaths and compute the path in ArgsToFrontendOutputsConverter.cpp.
The symbol graph contains an arbitrary number of files, and I think you can't know how many files there will be until you typecheck, because if you have module A that contains import B; extension B.Foo { public func bar() {} } there will be A.symbols.json and [email protected]. That's why I didn't model it as a "normal" output.
The symbol graph output from a module can contain an arbitrary number of files, depending on what extensions it contains, so cache a list of symbol graph files with their base name and contents so that they can be replayed.
rdar://140286819