On Wed, Apr 17, 2013 at 9:00 AM, Alexander Kornienko <[email protected]>wrote:
> > On Apr 17, 2013 9:35 AM, "Manuel Klimek" <[email protected]> wrote: > > > > > > > > ================ > > Comment at: lib/Format/UnwrappedLineParser.h:84 > > @@ +83,3 @@ > > + /// \brief Returns actual token start location without leading escaped > > + /// newlines. > > + SourceLocation getRealLocation() const { > > ---------------- > > It's not only escaped newlines, but all whitespace, right? > > > > > > http://llvm-reviews.chandlerc.com/D682 > > Whitespace is not part of a token anyway. > Yes, whitespace is not part of the Token, but it's part of the FormatToken. Since we refer to both as "token" in speech, this comment (inside the context of FormatToken) is highly misleading. FormatToken always contained all preceeding whitespace. I think the comment needs more context, and I'd rename it to something different. The problem is again that getTokenLocation() is ambiguous, as it's unclear whether it's the format token or the lexer's token.
_______________________________________________ cfe-commits mailing list [email protected] http://lists.cs.uiuc.edu/mailman/listinfo/cfe-commits
