On Wed, 2006-09-27 at 00:47 -0700, James Dennett wrote: > There is a miscommunication here: changing the allocator interface > allows for realloc-like semantics.
The are no methods in C++ which can support relocating an object. That is required to use 'realloc'. > Quite a bit of works was done with what one developer called > 'vampiric clones' which is essentially a move > > [snip] > > > The point is your 'essentially negligible' is my > 'essential makes the application impossible to implement'. > > The std::string implementation on Solaris is the most inefficient > I have ever encountered, by orders of magnitude, and I've also > expended significant effort in working around it. We had our own string types -- and most of the code was actually built with gcc I think. Multiple string types: some were part of ACN 1.1 package. > In no way does > it reflect on issues in resizing vectors, which (as noted here, in > more detail than before) should give no more than a factor of 2 > slowdown and even that only on code inserting into a vector > without using reserve. Sigh -- I was trying to show that arguments that 'its only a small number of cases' are useless. -- John Skaller <skaller at users dot sf dot net> Felix, successor to C++: http://felix.sf.net ------------------------------------------------------------------------- Take Surveys. Earn Cash. Influence the Future of IT Join SourceForge.net's Techsay panel and you'll get the chance to share your opinions on IT & business topics through brief surveys -- and earn cash http://www.techsay.com/default.php?page=join.php&p=sourceforge&CID=DEVDEV _______________________________________________ Felix-language mailing list Felix-language@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/felix-language
