I have been using objects for parameters for quite some time now, and I for myself cannot confirm your concern. Of course it requires some discipline when naming things, but as I have been a strong proponent of intelligent variable naming conventions since ages, that one comes naturally to me. Also you can search in content by searching für the key names. Using 1 letter variable names ($l $r $z) etc. obviously makes this a bit more difficult. I use objects for any method with more than 3 parameters where the parameters passed and left empty my vary greatly depending on the context of the call. I never experienced any problem doing it, and actually makes it a lot simpler to work with complex parameter lists. As Objects propagate themselves even across processes, you can also use them to communicate between processes, for example passing a stop command to a process you started by simply setting a parameter in the original parameter object, thus adapting a background processes flow according to changing circumstances in the foreground. Works a treat. Objects a a big fat drill, they allow you to make great holes, but when you slip, they can drill right into your foot :-) So handle with care Objects are the single greatest addition to the 4D language in the last years, I hope workers will be as important an addition.
Cheers Alex Am 03.10.2016 um 16:57 schrieb Chip Scheide <[email protected]>: > > Alex, > what I was trying to get at was, debugging the contents of an aboject: > > in a 'classic' parameter you can use internal 4D tools (find in design > for example) to locate instances of access to a variable; and yes in > the example you would need to transit through the call chain, but it > would be (or at last should be) obvious where an incorrect value/data > was set/remvoed. > > Where with an object, any of the 'constructed' manipulations could add > or remove large swaths of values (text) in the object, which to me, at > this point, makes tracking/debugging more difficult. > > > > On Mon, 3 Oct 2016 09:03:25 +0200, Herr Alexander Heintz wrote: >> Am 03.10.2016 um 00:04 schrieb Chip Scheide <[email protected]>: >>> >>> When $Object is finally used as a parameter list to Final_Method, >>> and the result/action(s) in FInal_Method are wrong due to bad >>> parameter list/values/data >> >> Sorry Chip, >> >> but this sounds like a quite constructed problem. >> >> If you program something that passes down a parameter several levels >> deep, you as the programmer of all methods involved must make sure >> this does not happen, as wether it happens with an element in an >> object parameter or a classic parameter variable does not really >> matter, the object is just the means of transport. >> >> If you fiddle with it in between stages, its your screwup, and it can >> happen with variables just as easily. >> >> The exact same discussion is currently running around on our german >> list, and the examples against using objects as parameters start to >> become more and more constructed, downright academic (silly?). >> >> Meaning no offense: >> >> „oh my god, something new! I am too experienced for that, I need an >> excuse not to bother with it“ >> >> Cheers >> Alex >> ********************************************************************** >> 4D Internet Users Group (4D iNUG) >> FAQ: http://lists.4d.com/faqnug.html >> Archive: http://lists.4d.com/archives.html >> Options: http://lists.4d.com/mailman/options/4d_tech >> Unsub: mailto:[email protected] >> ********************************************************************** > ********************************************************************** > 4D Internet Users Group (4D iNUG) > FAQ: http://lists.4d.com/faqnug.html > Archive: http://lists.4d.com/archives.html > Options: http://lists.4d.com/mailman/options/4d_tech > Unsub: mailto:[email protected] > ********************************************************************** ********************************************************************** 4D Internet Users Group (4D iNUG) FAQ: http://lists.4d.com/faqnug.html Archive: http://lists.4d.com/archives.html Options: http://lists.4d.com/mailman/options/4d_tech Unsub: mailto:[email protected] **********************************************************************
