it's interesting, we've gotten used to Flash being extremely tolerant of all sorts of bad behavior, calling null as a function, indexing off the end of strings, etc. I can't think of any system code we ever used that actually throws an exception.
On 10/29/07, Donald Anderson <[EMAIL PROTECTED]> wrote: > > That doesn't seem to phase it -- at least in SWF, > no error. I do get an error in dhtml. > > - Don > > On Oct 29, 2007, at 3:39 PM, P T Withington wrote: > > > You can't get divide by zero errors because it is not an error. > > You get NaN. > > > > Try something like: > > > > 'x' instanceof 3 > > > > if the compiler lets that through, instanceof is supposed to signal > > an error if the rhs is not a Function (aka class). > > > > On 2007-10-29, at 15:17 EDT, Donald Anderson wrote: > > > >> I have an implemention of try/catch/finally that matches what the > >> Adobe > >> compiler produces. However, I cannot produce any code that fires > >> an exception and I have not yet implemented 'throw'. Is there a > >> way to get an exception - I've tried division by zero, opening > >> nonexistent > >> files, etc. Throw shouldn't be difficult to implement - is it > >> useful to have > >> try/catch that only responds to throw? > >> > >> - Don > >> > >> > >> On Oct 17, 2007, at 6:01 PM, Max Carlson wrote: > >> > >>> This is fairly important - it would be great to have exceptions > >>> in SWF! > >>> > >>> P T Withington wrote: > >>>> Well, if you get bored with other stuff... > >>>> On 2007-10-17, at 13:48 EDT, Donald Anderson wrote: > >>>>> I think it's a good idea. > >>>>> I assume there is some support for exceptions in SWF > >>>>> for use in the future. I bet it's not too hard to do. > >>>>> > >>>>> - Don > >>>>> > >>>>> On Oct 17, 2007, at 1:07 PM, P T Withington wrote: > >>>>> > >>>>>> What do you think about at least permitting try/catch in swf > >>>>>> for now, even if we just pass the try block through as a > >>>>>> block, ignore any catch and pass through any finally. It's > >>>>>> lame, but it would let us unify a bunch of the LFC code base. > >>>>>> > >>>>>> Then when we have time, we can actually implement the correct > >>>>>> code generation for it... > >>>>> > >>>>> > >>>>> -- > >>>>> > >>>>> Don Anderson > >>>>> Java/C/C++, Berkeley DB, systems consultant > >>>>> > >>>>> Voice: 617-547-7881 > >>>>> Email: [EMAIL PROTECTED] > >>>>> WWW: http://www.ddanderson.com > >>>>> > >>> > >>> -- > >>> Regards, > >>> Max Carlson > >>> OpenLaszlo.org > >> > >> > >> -- > >> > >> Don Anderson > >> Java/C/C++, Berkeley DB, systems consultant > >> > >> Voice: 617-547-7881 > >> Email: [EMAIL PROTECTED] > >> WWW: http://www.ddanderson.com > >> > > > > > -- > > Don Anderson > Java/C/C++, Berkeley DB, systems consultant > > Voice: 617-547-7881 > Email: [EMAIL PROTECTED] > WWW: http://www.ddanderson.com > > -- Henry Minsky Software Architect [EMAIL PROTECTED]
