Re: [boost] Re: future of boost.threads

From: Peter Dimov (pdimov_at_[hidden])
Date: 2004-02-10 11:00:35

Alexander Terekhov wrote:
> Peter Dimov wrote:
> [...]
>> One case where "propagating" exceptions is wrong is when the thread
>> is not supposed to be joined.
> 
> When the thread is not supposed to be joined, then all expected
> exceptions [e.g. thread cancelation or exit] should be caught
> (but "ignored") and all unexpected exceptions should still
> terminate the entire process

Yes, that's exactly what I had in mind.

> at throw point

We can't do this from the library side.

> (that's not the way how exception specification work currently).

We can't change this from the library side either. ;-)