[Bioperl-l] Bio::WebAgent sleep warning

Sendu Bala bix at sendu.me.uk
Mon Oct 16 13:13:53 EDT 2006


Hilmar Lapp wrote:
> It depends. What triggers the sleeping? If it's part of every request 
> that it processes then I'd agree. If it is triggered by failure to 
> precede the next try then the failure is probably not expected (though 
> possible), and hence should be reported by warn().
> 
> If it is just part of the polling cycle then there should probably be a 
> limit up to which the time waited is considered 'normal' and after which 
> it is considered 'excessive' and hence should be reported through warn().

=head2 sleep

  Title   : sleep
  Usage   : $self->sleep
  Function: sleep for a number of seconds indicated by the delay policy
  Returns : none
  Args    : none

NOTE: This method keeps track of the last time it was called and only
imposes a sleep if it was called more recently than the delay_policy()
allows.

=cut

It issues a warning every time it actually sleeps. I find it 
inappropriate that a method warns me that it did what I asked it to do.


More information about the Bioperl-l mailing list