Re: Re: cvs: pear-core /PEAR

Sergio Carvalho wrote:

Lukas Smith wrote:

Sergio Carvalho wrote:

Hmm, I can see how this warning discussion can get really interesting. This is potentially a nice idea. If I read you correctly, you'd do something like: PEAR_Warning::allow('Some_Package_Downcast'); Some_Package_Class::method(); And then, inside the method, you'd have the regular PEAR_Warning::add. If an unallowed warning is fired, it gets automatically upgraded to an exception, right?

err, it should be the other way around!

Greg's current model is designed the other way around. I personally don't have a formed opinion on warning handling -- other than treating PHP's warnings, I really don't use them in my code and actually find the current PHP4 model more of a nuisance than an useful tool. I don't know how are warnings currently used.

Well if you make every warning an exception they will be even more of a nuisance.

but isnt this another case of where the exception backtrace will get "polluted" with non relevant data?

Here, I don't find it a big problem. The backtrace will start in PEAR_Warning::add, but that's how things happened: a Warning was fired, but since it was disallowed, it was upgraded to an exception.

Well, kinda reminds me of the PEAR_Errors gets upgraded to PEAR_Exception thing we discussed a while back in order to make it possible to support Exceptions within PHP4 compatible code. Anyways seems like the same thing to me, but I am not going to argue over it.

seems like more and more this fact is becoming a show stopper (as it already was for the turn pear_errors into exceptions on demand php4/php5 compatibility thing). maybe its time we address this?

How is it a showstopper?

Showstopper in the sense that it effectively killed off the "Exceptions within PHP4 compatible code" idea. regards, Lukas