Am Freitag, 18. August 2006 17:07 schrieb Jean-Marc Lasgouttes: > >>>>> "Michael" == Michael Gerz <[EMAIL PROTECTED]> writes: > > Michael> Are there any reasons why we cannot update to boost 1.3.4 for > Michael> 1.4.3? Michael > > Maybe Because it is not released yet? And what is it going to give us?
Too risky IMO. 1.34 has still a high number of regressions. BTW, speaking of boost, the long awaited boost::process exists, implemented as a Google Summer of Code project: http://www.netbsd.org/~jmmv/process/. It is not completed yet, but it can be very useful in the future. > Actually, in an ideal world, upgrading to 1.33.1 should not be needed > either. Upgrading a full library because of what is probably a few > lines worth of patch is lazy. But I do not think we have a choice. IMO investigating what exactly is broken in 1.32 is wasted time, and upgrading to 1.33 is straightforward. Jean-Marc, I can do it if you want. Georg
