owenpan added inline comments.
================ Comment at: clang/lib/Format/ContinuationIndenter.cpp:803-808 - } else if ((Previous.isOneOf(TT_BinaryOperator, TT_ConditionalExpr, - TT_CtorInitializerColon)) && + } else if (Previous.is(TT_CtorInitializerColon)) { + CurrentState.LastSpace = State.Column; + } else if ((Previous.isOneOf(TT_BinaryOperator, TT_ConditionalExpr)) && ((Previous.getPrecedence() != prec::Assignment && (Previous.isNot(tok::lessless) || Previous.OperatorIndex != 0 || Previous.NextOperator)) || Current.StartsBinaryExpression)) { ---------------- HazardyKnusperkeks wrote: > Before it was followed with an `&&`, I don't know which one was for this > case, and which one for the other, or for both. This is a bit hard, but maybe > this is just to few checks? IMO we should make an NFC patch to clean this up before attempting to fix the bug: e.g. `TT_ConditionalExpr` and `TT_CtorInitializerColon` don't have `prec::Assignment`, aren't `tok::lessless`, etc. CHANGES SINCE LAST ACTION https://reviews.llvm.org/D136154/new/ https://reviews.llvm.org/D136154 _______________________________________________ cfe-commits mailing list cfe-commits@lists.llvm.org https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits