Boost logo

Boost Users :

Subject: Re: [Boost-users] [Log] How to use?
From: Olivier Tournaire (olitour_at_[hidden])
Date: 2010-08-20 13:47:52


Thank you Oskar!

2010/8/20 Oskar Nordquist <oskar.nordquist_at_[hidden]>

> 2010/8/20 Olivier Tournaire <olitour_at_[hidden]>
>
> 2010/8/20 Oskar Nordquist <oskar.nordquist_at_[hidden]>
>>
>> Hi,
>>>
>>> Just a wild guess:
>>>
>>> Since the difference between your test program and the Boost example is
>>> that you are doing all the preparations in the constructor, whilst Boost
>>> example has everything in the main() scope. My guess is something goes out
>>> of scope at the end of the constructor, and some attribute is unregistered
>>> and thus generates "missing attribute" exception.
>>>
>>
>> Maybe it is a problem with the singleton? No one experienced such a
>> problem?
>> I tried a really simple program found in the documentation (
>> http://boost-log.sourceforge.net/libs/log/doc/html/log/detailed/attributes.html#log.detailed.attributes.timer)
>> which uses a class. The result is the same: it throws the same exception in
>> the destructor.
>>
>
>
> I tried it on MSVC 10 as well and it throws the same exception. I even did
> a boolean formatter checking if the attribute exists:
>
> backend->set_formatter(
> fmt::stream
> << fmt::if_(flt::has_attr("Duration"))
> [
> fmt::stream << "Duration: " << fmt::attr("Duration")
> ]
> << "\tMsg: " << fmt::message());
>
> Still throws the same exception.
>
> I switched the attribute to a attrs::counter<int> and it works as expected.
>
> Seems like attrs::timer is broken!
>

Seems you are right. Commenting the line

<< "] [" << fmt::time_duration< boost::posix_time::time_duration >("Uptime")

makes the sample work.
Should a bug report be open?

Regards,

Olivier

>
> _______________________________________________
> Boost-users mailing list
> Boost-users_at_[hidden]
> http://lists.boost.org/mailman/listinfo.cgi/boost-users
>



Boost-users list run by williamkempf at hotmail.com, kalb at libertysoft.com, bjorn.karlsson at readsoft.com, gregod at cs.rpi.edu, wekempf at cox.net