[boost] Math toolkit - revision in progress using policy for fine-grained control of error handling etc.

From: Paul A Bristow (pbristow_at_[hidden])
Date: 2007-07-23 09:18:40

Just a heads up to warn that, in response to review requires, John Maddock has done some more work on using policies to provide
fine-grained control of what do when things go wrong (exceptions, NaNs, errno...), and whether mathematically undefined things
should compile and return NaN, or not compile at all.

This stuff is now in the sandbox and testing, examples and documentation are in progress, but not yet complete. The docs may not yet
be fully in sync with the code - the changes have triggered a zillion minor changes in the docs :-)

Using the current sandbox or final release (1.35?) code *may* require some minor changes to your code if you are/were using the
existing code (reviewed but un-released).  But the underlying math functions & distributions are essentially unchanged.

I will post again when a version that we believe is 'release-ready' is in the sandbox.

Paul

---
Paul A Bristow
Prizet Farmhouse, Kendal, Cumbria UK LA8 8AB
[phone] & SMS, Mobile [phone] & SMS
pbristow_at_[hidden]