[Bioperl-l] Grrrr. Another subtle overloading bug ...

[Ewan Birney]

On Sun, 6 Jan 2002, Hilmar Lapp wrote:

> > 
> > This is a nasty bug and something we have to get right. It also emphasises
> > how important it is to test the 0.9 series before we call it 1.0 otherwise
> > we will just trigger alot of bugs world-wide...
> > 
> 
> I essentially agree with Ewan. I'd call it mandatory that $@
> evaluates to TRUE upon an exception having been thrown. If this
> cannot be achieved, I suggest to rollback to the previous way of
> exception throwing.
> 

Reading you loud and clear. I need to either up my knowledge of Perl
overloading systems or - as you say - the rather dramatic step of rolling
back out steve's exception objects....


> I'd actually suspect that 99% of bioperl users don't need an
> exception object. Remember that one of the main use cases of
> bioperl is scripting, not only ensembl-scale applications. If an
> exception occurs you just want to bail out with a nice message
> that aids debugging. The worst scenario is if the kind of
> exception throwing affects the bioperl learning curve in any
> (negative) way.
> 

agreed. It is another one of those areas where Perl should have had a
good, established exception throwing behaviour a good five years ago ---
now we are getting annoyed by the multiple details you have to get
right when you roll-your-own. 


Hey-ho. Part and parcel of life. I hope they get this right in Perl6 ;)



> 
> 	-hilmar
>