I've committed this as r105815. Let me know if anything needs to change.
http://codereview.appspot.com/1591041/diff/8001/3006 File include/clang/Basic/Diagnostic.h (right): http://codereview.appspot.com/1591041/diff/8001/3006#newcode179 include/clang/Basic/Diagnostic.h:179: On 2010/06/10 22:58:47, chandlerc wrote: > Doc for the enum proper? Done. http://codereview.appspot.com/1591041/diff/8001/3006#newcode328 include/clang/Basic/Diagnostic.h:328: /// default, we omit candidates when they're spammy. On 2010/06/10 22:58:47, chandlerc wrote: > Is this comment about the default still accurate? Also, do we want to specify > the default here, or with the flag's docs? Oops, updated. It seems to need to be in multiple places. Diagnostics has one default when you fail to set it. DiagnosticOptions has another. The flag inherits the default from DiagnosticOptions. http://codereview.appspot.com/1591041/diff/8001/3004 File include/clang/Frontend/DiagnosticOptions.h (right): http://codereview.appspot.com/1591041/diff/8001/3004#newcode39 include/clang/Frontend/DiagnosticOptions.h:39: /// Diagnostic::OverloadsShown On 2010/06/10 22:58:47, chandlerc wrote: > Is one bit always going to be enough then? Who knows. We'll have to remember to expand it if we need more bits. Is that a problem? http://codereview.appspot.com/1591041/diff/8001/3012 File lib/Sema/SemaOverload.cpp (right): http://codereview.appspot.com/1591041/diff/8001/3012#newcode5708 lib/Sema/SemaOverload.cpp:5708: "Condition when adding candidates should have prevented this"); On 2010/06/10 22:58:47, chandlerc wrote: > This shouldn't refer to the other code, but just the invariant imo: "Non-viable > builtin operators are not added to our reported set." or some such... Done. http://codereview.appspot.com/1591041/show _______________________________________________ cfe-commits mailing list [email protected] http://lists.cs.uiuc.edu/mailman/listinfo/cfe-commits
