Interesting case. I'm not sure if I would overload functions or just accept a non-typed param and then check it's type as a wrapper. Seems like a lot of duplicate code to have each function exist separately, but maybe not.
I guess it would just be another language feature, like singletons that some circles of devs tell each other not to use if they want to be 'cool' and not get bit later. :) brought to you by the letters A, V, and I and the number 47 On Sun, Feb 3, 2013 at 3:48 PM, Nicholas Kwiatkowski <nicho...@spoon.as>wrote: > One quick example -- in my ArduinoANE (an AIR ANE that allows you to send > data over a serial port), I have to have at least 5 functions that do the > same thing -- they just accept different variable types. > > Ultimately, I'd like my API to be > > serial.send(var); > > but I have to have : > > serial.sendAsInt(int); > serial.sendAsObject(object); > serial.sendAsArray(array); > serial.sendAsString(string); > serial.sendAsByteArray(ba); > serial.sendAsByte(int); > serial.sendAsFloat(float); > .... > .... > > This means that the end-developer needs to know all these different > function names instead of one. Sure, a good IDE helps with that, but it > seems unnecessary. It also prevents me from allowing better code-reuse, > etc. > > -Nick > > > On Sun, Feb 3, 2013 at 7:13 AM, Avi Kessner <akess...@gmail.com> wrote: > > > I can not for the life of me understand the desire for overloading > > functions. If it has different behavior give it a different name. > > > > brought to you by the letters A, V, and I > > and the number 47 > > > > > > On Sun, Feb 3, 2013 at 1:53 PM, Roland Zwaga <rol...@stackandheap.com > > >wrote: > > > > > +100.000 for generics (although I fully understand that this is > probably > > > one of the most difficult features to implement) > > > > > > +1 for lamba expressions > > > > > > On 3 February 2013 12:48, christofer.d...@c-ware.de < > > > christofer.d...@c-ware.de> wrote: > > > > > > > +1 for method overloading from me too > > > > > > > > And: > > > > > > > > +1 for private/protected constructors :-) > > > > > > > > > > > > > > > > -----Ursprüngliche Nachricht----- > > > > Von: Frédéric THOMAS [mailto:webdoubl...@hotmail.com] > > > > Gesendet: Sonntag, 3. Februar 2013 05:16 > > > > An: dev@flex.apache.org > > > > Betreff: Re: Language features > > > > > > > > Nick, +1 or even 10 > > > > > > > > -Fred > > > > > > > > -----Message d'origine----- > > > > From: Nicholas Kwiatkowski > > > > Sent: Saturday, February 02, 2013 6:58 PM > > > > To: dev@flex.apache.org > > > > Subject: Re: Language features > > > > > > > > I'd be fairly excited to see method overloading. It's one of the > > things I > > > > miss from Java... > > > > > > > > -Nick > > > > > > > > On Sat, Feb 2, 2013 at 12:17 PM, Avi Kessner <akess...@gmail.com> > > wrote: > > > > > > > > > If it was up to me, I would vote against method overloading. I > think > > > > > that's a code smell personally. > > > > > > > > > > brought to you by the letters A, V, and I and the number 47 > > > > > > > > > > > > > > > On Sat, Feb 2, 2013 at 7:05 PM, Frédéric THOMAS < > > > webdoubl...@hotmail.com > > > > > >wrote: > > > > > > > > > > > Hi Gordon, > > > > > > > > > > > > > > > > > > Adding abstract classes and private constructors to Falcon > should > > be > > > > > easy > > > > > >> > > > > > > > > > > > > That's a good news, at this point protected constructor would be > > > > > > welcomed > > > > > > as well as private constructors are commonly used in classes that > > > > > > contain > > > > > > static members only. > > > > > > > > > > > > And I voting +1 for the rest :-) you gonna make happy a lot of > > people > > > > > > who > > > > > > wait for a long time for these features. > > > > > > > > > > > > -Fred > > > > > > > > > > > > -----Message d'origine----- From: Gordon Smith > > > > > > Sent: Friday, February 01, 2013 7:38 PM > > > > > > To: dev@flex.apache.org > > > > > > Subject: RE: Language features > > > > > > > > > > > > > > > > > > Adding abstract classes and private constructors to Falcon should > > be > > > > > easy. > > > > > > Adding generics and method overloading would be considerably > harder > > > but > > > > > > probably doable after a lot of design. Two other features worth > > > > > considering > > > > > > are strong function types (i.e., a type like (int, int):String > for > > a > > > > > > function that takes two ints and returns a String) and > > strongly-typed > > > > > fixed > > > > > > arrays (i.e., int[]). > > > > > > > > > > > > I'm going to continue to focus on MXML. Until it is finished, we > > > can't > > > > > > move from the old compiler to the new one. I don't recommend > making > > > any > > > > > > modifications to the old compiler. > > > > > > > > > > > > - Gordon > > > > > > > > > > > > -----Original Message----- > > > > > > From: Frédéric THOMAS [mailto:webdoublefx@hotmail.**com< > > > > > webdoubl...@hotmail.com> > > > > > > ] > > > > > > Sent: Friday, February 01, 2013 3:07 AM > > > > > > To: dev@flex.apache.org > > > > > > Subject: Re: Language features > > > > > > > > > > > > +1 Nick > > > > > > > > > > > > May be possible, I don't know, time ago, I looked at adding the > > > > > > possibility to have the constructor accepting other NS than > public > > to > > > > > > simulate abstract classes and seen 2 places where it was checked > > but > > > > > didn't > > > > > > dare to change it besause I didn't know the impacts, I hope > someone > > > > > better > > > > > > than me here can take care of it, compiler geeks, are you here ? > > > > > > > > > > > > -Fred > > > > > > > > > > > > -----Message d'origine----- > > > > > > From: Nick Collins > > > > > > Sent: Friday, February 01, 2013 11:24 AM > > > > > > To: dev@flex.apache.org > > > > > > Subject: Language features > > > > > > > > > > > > With the cancellation of AVM next, should we perhaps look at > adding > > > > some > > > > > > additional language features to our compiler? > > > > > > > > > > > > As I think about some of the features I would like to see, such > as > > > > > > abstract classes, generics, method overloading, etc. it seems to > me > > > > that > > > > > at > > > > > > least some of them could be implemented into our compiler? > > > > > > > > > > > > Nick > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > -- > > > regards, > > > Roland > > > > > > -- > > > Roland Zwaga > > > Senior Consultant | Stack & Heap BVBA > > > > > > +32 (0)486 16 12 62 | rol...@stackandheap.com | > > > http://www.stackandheap.com > > > > > > http://zwaga.blogspot.com > > > http://www.springactionscript.org > > > http://www.as3commons.org > > > > > >
