[boost] Re: Stack tracing (was: improved assertions?)

From: Ted Byers (r.ted.byers_at_[hidden])
Date: 2002-06-13 11:08:46

<Schalk_Cronje_at_[hidden]> wrote in message
news:C07DA1368274D3119DFC009027AA5B21048A5C01_at_AYL-135-211.nai.com...
> > I lean towards (b) until I see some indication that (a) is really
possible
> > for the popular platforms.

> I think one really needs both (a) and (b).
> (b) is good in that one can get a clear description as to the function
name,
> file etc. (assuming that info is added to the exception).
> (a) is good when one tries to trace hard to find bugs. It can also be used
> when replacing terminate() or in a signal handler to catch SEGV signals.
It
> can also be used to complement an exception being thrown, by showing the
> stack at that point in time.

I agree.  Nies addressed this also, albeit in a very briefly.  You can get
trace information on one of two ways; intrusively (Fernando's option b) or
not (Fernando's option a).

The question that needs to be addressed first is precisely what are the
functional requirements we hope to meet.  Then, when we know the target, we
would be in a better position to consider how we can produce a library that
meets most, of not all, of our percieved needs.

I have a prototype for a library that provides for option b, and Fernando
has provided the url where an early verion of this may be found, warts and
all.  ;-)  And my prototype library is quite ordinary standard C++ (or is
intended to be, barring error on my part), so it ought to be quite portable.
However, I do not see, at present, how option a can be done in a portable
manner; but then, I probably haven't been studying it long enough.

Cheers,

Ted