On Fri, 2008-09-12 at 15:15 -0400, Eric Gorr wrote: > On Sep 12, 2008, at 2:51 PM, Robert Cummings wrote: > > > On Fri, 2008-09-12 at 14:33 -0400, Eric Gorr wrote: > >> On Sep 12, 2008, at 2:14 PM, Robert Cummings wrote: > >> > >>> On Fri, 2008-09-12 at 11:47 -0400, Eric Gorr wrote: > >>>> On Sep 12, 2008, at 11:39 AM, Andrew Ballard wrote: > >>>> > >>>>> On Fri, Sep 12, 2008 at 9:52 AM, Jochem Maas > >>>>> <jochem@xxxxxxxxxxxxx> > >>>>> wrote: > >>>>>> Luke schreef: > >>>>>>> > >>>>>>> I wonder if this is a shared trait between C and PHP (since I > >>>>>>> understand > >>>>>>> PHP > >>>>>>> is written in C) that the break; and the default: are placed for > >>>>>>> good > >>>>>>> practice in all switch statements since they prevent memory > >>>>>>> leaks? > >>>>>> > >>>>>> default is not required, never heard it was good practice to > >>>>>> always > >>>>>> put it > >>>>>> in. > >>>>> > >>>>> I can't say I've ever heard it recommended as good practice from > >>>>> the > >>>>> standpoint of performance in any specific language I've ever > >>>>> worked > >>>>> with, but I have heard people suggest that you always include an > >>>>> explicit default case in any kind of branching logic. It does seem > >>>>> useless to say > >>>>> > >>>>> default: // do nothing > >>>>> break; > >>>>> > >>>>> in a switch block, but I imagine the reasoning behind it is so > >>>>> that > >>>>> anyone who reads your code can see that you actually thought about > >>>>> what should/would happen if none of the other conditions were true > >>>>> rather than ignoring those conditions. > >>>> > >>>> It is always useful for a 'default:' case, which would normally do > >>>> nothing, to include some debug-only code so you can be notified if > >>>> the > >>>> default case is ever hit. Whenever I see an empty or just missing > >>>> 'default:' case, I always cringe. > >>> > >>> Whenever I see pointless debug fluff just for the sake of not having > >>> an > >>> empty or omitted default statement I cry. > >> > >> It's only pointless debug fluff until it saves untold numbers of > >> hours > >> which would have been spent attempting to track down a bug that would > >> have been caught by the now critical test which took only about 10 > >> seconds to write. > > > > Maybe for complex cases... but you used a broad brush when you > > advocated > > always doing this and so I most point out the pointlessness of debug > > fluff for simple use cases. > > Even in the simple cases, it is always useful to employ defensive > programming techniques because it may catch a bug which would have > otherwise taken hours to track down. > > > Since case statements are for the most part > > equivalent to if/else if/else statements, I must ask if you also > > enforce > > an else statement on all such conditionals. > > Who said anything about enforcement? > > I simply said it make me cringe and stated an opinion which is derived > from an accepted ideal that defensive programming is a good thing. I presumed that you wold attempt to prevent yourself from cringing and would thusly enforce such methodology upon yourself. I'm all for defensive programming where appropriate. I guess I just never realized some people need to defend against themselves at all times. > > Such debugging should be > > done on a case by case basis depending on the particular scenario at > > hand. If the code is simple and obvious then debugging serves little > > purpose but to convolute the code. > > "Convoluted" is a relative and subjective term. Quite true. > Once one gets used to writing and maintaining code written this way, > it is no longer convoluted. Furthermore, every decent editor these > days would allow one to collapse such code for those who find it > difficult to interpret...providing even a stronger reason to write it > in the first place. I don't see how that in any way makes an argument for or against. Once still must spend client's money wasting time on code that has questionable merit. Yes, some debugging code is a great boon in any application, but littered everywhere to fulfill someone's subjective philosophical ideal when sometimes it's just plain unnecessary... wasteful IMHO. Cheers, Rob. -- http://www.interjinn.com Application and Templating Framework for PHP -- PHP General Mailing List (http://www.php.net/) To unsubscribe, visit: http://www.php.net/unsub.php
