[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: [IMP-dev] set_log_level



> thanks Daniel.
> Trying to debug some code, specifically now at the interface between  
> IMP::multifit and IMP::domino.
> So - added tons of IMP_LOG(VERBOSE,"")
> however - there are a lot of messages from core and kernel that i do  
> not really care about at this point.
If the core/kernel code is called from Model::evaluate and the domino code from optimize() then, just set the Model log level to NONE and the domino sampler level to verbose. If not, I'm curious what makes the needed cut through the call tree complicated. Perhaps there is a better way of grouping functionality to make it easier to control.

It is messier still when the base of the call stack is a non-member function since they have no internal state. Especially when those functions create and call Object methods. Per-module is probably the best we can do there, but getting free functions to use it will take a lot of work.


> It will be complicated to set log level for each object and seems not  
> to be the most elegant solution.

> 
> On Apr 20, 2010, at 8:28 AM, Daniel Russel wrote:
> 
>> No, currently there is just per-Object log levels. Adding per-module  
>> log levels would be easy, but I'm not sure how useful. It would make  
>> sense for, say, EM, but you would get more or less the same effect  
>> by setting it on the FitRestraint. Setting the general log level for  
>> core seems less useful. What are you trying to do?
>> 
>> 
>> On Apr 20, 2010, at 1:02 AM, Keren Lasker wrote:
>> 
>>> hi -
>>> is there a way to set different log levels to different modules ?
>>> thanks, keren.
>>> _______________________________________________
>>> IMP-dev mailing list
>>> 
>>> https://salilab.org/mailman/listinfo/imp-dev
>> 
>> 
>> _______________________________________________
>> IMP-dev mailing list
>> 
>> https://salilab.org/mailman/listinfo/imp-dev
> 
> _______________________________________________
> IMP-dev mailing list
> 
> https://salilab.org/mailman/listinfo/imp-dev