Details
- Reviewers
alexfh aaron.ballman NoQ dcoughlin - Commits
- rG88a16a07909b: [ASTMatchers] Introduce a matcher for matching any given Objective-C selector
rC328747: [ASTMatchers] Introduce a matcher for matching any given Objective-C selector
rL328747: [ASTMatchers] Introduce a matcher for matching any given Objective-C selector
Diff Detail
Event Timeline
include/clang/ASTMatchers/ASTMatchers.h | ||
---|---|---|
2737 | I'd suggest to do the same as with hasAnyName: leave just the std::vector<llvm::StringRef> overload. WDYT? |
include/clang/ASTMatchers/ASTMatchers.h | ||
---|---|---|
2737 | OK. |
Not yet. A few more comments.
lib/ASTMatchers/ASTMatchersInternal.cpp | ||
---|---|---|
327 | What's the point of this variable? Maybe just use its initializer directly? | |
334 | const std::string & / const auto & ? | |
335 | S == SelString? | |
342 | What's the point of this variable? Maybe just use its initializer directly? | |
416 | What was wrong with the partially qualified name here? |
lib/ASTMatchers/ASTMatchersInternal.cpp | ||
---|---|---|
327 | The principle of not having too many things magically happening in one line would be IMO violated if we write internal::Matcher<NamedDecl>(new internal::HasNameMatcher(vectorFromRefs(NameRefs))) | |
416 | It does not work. |
LG
lib/ASTMatchers/ASTMatchersInternal.cpp | ||
---|---|---|
327 | I wouldn't consider any of this stuff magical, but I also won't object to retaining this variable. |
I'd suggest to do the same as with hasAnyName: leave just the std::vector<llvm::StringRef> overload. WDYT?