This is the mail archive of the gcc@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]

Re: [C++] deprecate overload resolution extension?



  Making it permissive is fine by me, but ...

  What is the intended difference between deprecated things and permissive
  things?

  * deprecated features are warned[1] and that inhibited with -fdeprecated

  * permissive features are errored and that can be warned with -fpermissive

  * deprecated features will be removed 'in some future release'

  Are we also intending to support permissive features forever?

I think so -- or at least until we deprecate them first.

The idea is that -fpermissive is for accepting code that we think is
really out there and really matters.  As Joe said, this is often code
that was legal by the ARM, or by Cfront, or some such.

Deprecated things are largely G++ extensions that we don't want to
support any more.  Generally, we feel these aren't as important or as
prevelant.  That's a subjective statement, of course...

I think the reason to not give errors for deprecated features is that
they're supposed to still work.  The warning is to say "last chance --
fix this NOW!".  

All a little muddy, I know...

--
Mark Mitchell                   mark@codesourcery.com
CodeSourcery, LLC               http://www.codesourcery.com


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