[reSIProcate] Bug in Data::append()

["Brian Cuthie" <brian@xxxxxxxxxxxx>]

Hello,

I'm new to this list, and to using the Reciprocate stack. However, I
think I found a bug and wanted to hear what people think. I'm sure I'm
doing this wrong, so please don't be shy about helping me understand the
correct procedure.

Background:

While testing the stack under Windows, I had problems with a heap
corruption when writing debug log messages using the VSDebugWindow log
type.

BUG:

The following code in Data::append() appears to be in error. The NULL is
written past the end of the buffer, and furthermore I think it is
unnecessary, since c_str() adds a NULL.

Data&
Data::append(const char* str, size_type len)
{
   assert(str);
   if (mCapacity < mSize + len)
   {
      // .dlb. pad for future growth?
      resize(((mSize + len +16)*3)/2, true);
   }
   else
   {
      if (mMine == Share)
      {
         char *oldBuf = mBuf;
         mCapacity = mSize + len;
         mBuf = new char[mSize + len];
         memcpy(mBuf, oldBuf, mSize);
         mMine = Take;
      }
   }

   // could conceivably overlap
   memmove(mBuf + mSize, str, len);
   mSize += len;
   mBuf[mSize] = 0;   // ERROR

   return *this;
}

Finally, there appears to be a related bug fix to Data::c_str() (#5408).
I'm wondering if someone was tracking this same problem and tried to fix
it there, since resize() will copy the buffer anyway, making the check
in c_str() unecessary.

Cheers,

Brian