RE: [boost] Re: boost::execution_monitor impl under windows

From: carlos pizano (carlospizano_at_[hidden])
Date: 2003-09-21 18:57:11

> URL please.

Here:
http://aspn.activestate.com/ASPN/Mail/Message/boost/1811713

> What interaction?

See URL above and/or Robbins book chapter 13

>> I don't think so. Well, it can fail. 
>It only fails under restricted circumstances, which are documented.

That is the *interaction* I am talking about: it can fail. 

> which are documented.

Where is this documentation? In boost.test ? in the source?

I would like to learn where it can and cannot work. My current
understanding is that it works when the compiler does not optimize too
much :)

> I think you are seriously missing my point.  This is not a substitute
> for my technique because it causes unwinding and recovery instead of
> going immediately to JIT debugging.

a) I guess I am. You mentioned earlier that you don't know what
execution_monitor is doing, now is a good time to check it out because
we are not talking about your technique in the vacuum. 

b) My proposal can do the same thing(s) as your technique for
execution_monitor. Mine causes unwinding and yours not? You lost me
there. Please explain. 

c) causes recovery? I never said recovery... where do you get that from
my pseudo code?

> Interesting things are counter-productive at this point.  

Okeey.. How should I read that?

> "Invoke the
> debugger" is the right thing to do for developers, and "report
> failure and terminate" is the right thing to do for testers.

No problem there. Remember that we have other classes derived from
execution_monitor where behavior is specialized for production or
testing.

>> Think about it.
> No thanks! ;-)

I know, thinking sometimes is difficult. I hope I can get the attention
of the maintainer of boost.test and that he/she feels like thinking.

CPU

-----Original Message-----
From: boost-bounces_at_[hidden]
[mailto:boost-bounces_at_[hidden]] On Behalf Of David Abrahams
Sent: Sunday, September 21, 2003 1:16 PM
To: boost_at_[hidden]
Subject: [boost] Re: boost::execution_monitor impl under windows

"carlos pizano" <carlospizano_at_[hidden]> writes:

> Dave Abrahams wrote:
>> No, I'm disabling the warning with pragmas.  The technique works
fine.
>
> I don't think so. Well, it can fail. 

It only fails under restricted circumstances, which are documented.

> See Carl Daniel post on 9/21. 

URL please.

> I saw the thread(s) on comp.lang.c++.moderated about the subject but
> nobody mentioned the interaction of the *technique* with /EHs at
> that time.

What interaction?

> Here is where I am going with this. The current way, complied with
/EHs
> seems to me that can be better done with something along the lines of:
>
> execute() 
> {
>
> __try
> {
> 	try
> 	{
> 		user_function();
> 	}
> 	catch( specific_type1& )
> 	{
> 		report1(text);
> 	}
> 	catch( specific_type2& )
> 	{
> 		report2(text);
> 	}
> 	... other type specific catch(s) here..
> 	
> 	... do not put catch(...) here ever..
> }
> __except ( filter(GetExceptionInformation()) )
> {
> 	report_def()
> }
>
> } // end of function
>
> /* take the above code at the pseudo-code level */

I think you are seriously missing my point.  This is not a substitute
for my technique because it causes unwinding and recovery instead of
going immediately to JIT debugging.

> Now, inside filter() you can do interesting things for instance
> re-throwing the SE under certain conditions or checking for the magic
> number that says that the SE is really a C++ exception and maybe do
your
> trick...

Interesting things are counter-productive at this point.  "Invoke the
debugger" is the right thing to do for developers, and "report
failure and terminate" is the right thing to do for testers.

> Note that the code above will be done in two nested calls since VC
> does not Allow you to mix __try and try on the same function.
>
> There seems to be very little restrictions on what filter() can be,
> I wonder if we can make it dispatch to a virtual function so a
> derived class can customize the behaviors.
>
> Think about it.

No thanks! ;-)

-- 
Dave Abrahams
Boost Consulting
www.boost-consulting.com
_______________________________________________
Unsubscribe & other changes:
http://lists.boost.org/mailman/listinfo.cgi/boost