Hi,

I am concerned that the copyfmt() routine in basic_ios may have a bug in
its specification.  The specification is:

<quote>
  basic_ios& copyfmt(const basic_ios& rhs);

  Effects:
    Assigns to the member objects  of  *this  the  corresponding  member
    objects of rhs, except that:

  --rdstate() and rdbuf() are left unchanged;

  --exceptions() is altered last by calling exceptions(rhs.except).

  --The  contents of arrays pointed at by pword and iword are copied not
    the pointers themselves.8)

1 If any newly stored pointer values in *this point  at  objects  stored
  outside  the  object  rhs, and those objects are destroyed when rhs is
  destroyed, the newly stored pointer values are  altered  to  point  at
  newly constructed copies of the objects.

2 Before  copying any parts of rhs, calls each registered callback  pair
  (fn,index) as (*fn)(erase_event,*this,index).   After  all  parts  but
  exceptions()  have  been  replaced,  calls each callback pair that was
  copied from rhs as (*fn)(copy_event,*this,index).
  Notes:
    The second pass permits a copied pword value to be  zeroed,  or  its
    referent  deep  copied  or  reference  counted or have other special
    action taken.
  Returns:
    *this.
<end quote>

If I'm not mistaken, the intent is to copy the format information from rhs
to *this, but not the stream (and associated info).

According to the spec above, the information that tie() returns is
copied.  This means that copyfmt() will change format information, but not
the actual rdbuf, BUT it will change what the rdbuf is tied too!  I am not
an I/O expert, but that just seems too weird.  Is this intentional?

By the way, copy_event is misspelled in the spec.

Thanks,
-Howard
---
[ comp.std.c++ is moderated.  To submit articles, try just posting with ]
[ your news-reader.  If that fails, use mailto:std-c++@ncar.ucar.edu    ]
[              --- Please see the FAQ before posting. ---               ]
[ FAQ: http://reality.sgi.com/austern_mti/std-c++/faq.html              ]