This is the mail archive of the gcc-patches@gcc.gnu.org mailing list for the GCC project.


Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]
Other format: [Raw text]

Re: [C++ PATCH] [PR2204] Check for parameters of abstract types (partial fix)


"Zack Weinberg" <zack@codesourcery.com> writes:

| Gabriel Dos Reis <gdr@integrable-solutions.net> writes:
| > | Gabriel Dos Reis <gdr@integrable-solutions.net> writes:
| > | > The repetition oan information that is already there.  Any scheme that
| > | > requires repetition of information is wrong. 
| > | 
| > | Well, this is not consistent with what you have said before, and I
| >
| > Again, I don't understand what you're talking about.
| 
| You were quite adamant in the past that xxx_at() must go.  So adamant,
| in fact, that I thought it was already dead.
| 
| My position is:
| 
|  * locate_error is a nasty thing that should go away.
|  * ", decl, decl" isn't pretty but it isn't as nasty as locate_error.
|  * Consistency between the C and C++ front ends is desirable.
| 
| and most importantly
| 
|  * Why are we holding up a bug fix with this stupid argument anyway?

What I find stupid is your current position. Quite frankly, it is
incomprehensible to me. You came in keeping saying that I used to be
yyy in the past so zzz must be.  Even when I explained what I mean.

In fact, it is -you- holding up a patch with this stupid argument that
I used to be xxx.  
I think that xxx_at()  should be used until xxx() supports '%+', in
particular I recommend Giovanni makes the modification I suggested.

-- Gaby


Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]