le mercredi 11 novembre 2009 indent a écrit
> I haven't got around to merging it yet, but just looking at the patch
> lines it seemed acceptable - although you did not honour the coding style.
> I would point out that there are a number of problems with C++ code
> (which language "..." is part of and not C) that can break the code, so
> be careful using indent on C++.

I ran into this "bug" using indent on qemu source code which is not C++ 
(although it might certainly use some C++ style constructs supported by gcc).

That's why I tried to fix it.

JC

> Jean-Christophe Dubois wrote:
> > Just curious: Is this patch accepted or is it inadequate for some reason?
> >
> > I got no feed back on this proposed patch.
> >
> > JC
> >
> > le jeudi 1 octobre 2009 Jean-Christophe Dubois a écrit
> >
> >> indent is not handling correctly case statements dealing with values
> >> range.
> >>
> >> For example:
> >>
> >>         case 0x01 ... 0x0b:
> >>
> >> Will be transformed by indent in:
> >>
> >>         case 0x01...0 x0b:
> >>
> >> Which cannot compile.
> >>
> >> This patch tries to fix this issue. I am not sure this is the correct
> >> solution but it seems to work for me.
> >>
> >> Signed-off-by: Jean-Christophe Dubois <[email protected]>
> >>
> >> --- indent-2.2.10.org/src/lexi.c   2008-03-11 19:50:42.000000000 +0100
> >> +++ indent-2.2.10/src/lexi.c       2009-10-01 01:22:09.349653276 +0200
> >> @@ -938,28 +938,38 @@
> >>           break;
> >>
> >>     case '.':
> >> -      if (parser_state_tos->in_decl &&
> >> -          (buf_ptr[0] == '.') &&
> >> +      if ((buf_ptr[0] == '.') &&
> >>            (buf_ptr[1] == '.'))
> >>        {
> >> -        /* check for '...' in a declaration */
> >> +   /* We have a '...'. This is supposed to mean something */
> >>           if ((buf_ptr += 2) >= buf_end)
> >>           {
> >>              fill_buffer();
> >>           }
> >>
> >> -         unary_delim = true;
> >> -         code = decl;
> >> -         token_end = buf_ptr;
> >> -         break;
> >> -      }
> >> -      unary_delim = false;
> >> -      code = struct_delim;
> >> +         if (parser_state_tos->in_decl)
> >> +         {
> >> +            /* this is '...' in a declaration */
> >> +            unary_delim = true;
> >> +            code = decl;
> >> +            token_end = buf_ptr;
> >> +         } else {
> >> +            /* this is '...' somewhere else */
> >> +            /* for example: case 1 ... 9: */
> >> +            unary_delim = true;
> >> +            code = binary_op;
> >> +            token_end = buf_ptr;
> >> +         }
> >> +      } else {
> >> +
> >> +         unary_delim = false;
> >> +         code = struct_delim;
> >>
> >> -      if (*buf_ptr == '*')        /* object .* pointer-to-member */
> >> -      {
> >> -         ++buf_ptr;
> >> -         token_end = buf_ptr;
> >> +         if (*buf_ptr == '*')     /* object .* pointer-to-member */
> >> +         {
> >> +            ++buf_ptr;
> >> +            token_end = buf_ptr;
> >> +         }
> >>        }
> >>        break;
> >>
> >>
> >>
> >>
> >> _______________________________________________
> >> bug-indent mailing list
> >> [email protected]
> >> http://lists.gnu.org/mailman/listinfo/bug-indent
> >
> > _______________________________________________
> > bug-indent mailing list
> > [email protected]
> > http://lists.gnu.org/mailman/listinfo/bug-indent
> 



_______________________________________________
bug-indent mailing list
[email protected]
http://lists.gnu.org/mailman/listinfo/bug-indent

Reply via email to