if you think people care about a maintainable code base are trolls what do you do on a devel list?
for the ordinary users this may not matter *until* things start to break because nobody konws what the code is suppsoed to doing Am 02.09.2013 22:03, schrieb stephane martin: > "Roundcube, code proudly cleant by argumenting trolls" > > > On Mon, Sep 2, 2013 at 9:43 PM, Thomas Bruederli <[email protected] > <mailto:[email protected]>> wrote: > > On Mon, Sep 2, 2013 at 9:20 PM, David Deutsch <[email protected] > <mailto:[email protected]>> wrote: > >> A proper and predictable process flow is something I learned to > appreciate > >> during my years as a software developer. > > > > Sure, although I don't understand how my approach makes it any less > > predictable. All I have done is fan out the if/elseif structures and, > > arguably, that makes it easier to check through them. Maybe changing it > to > > no-oneliner-ifs will make this a little clearer for you, but the return > > statements really stand out when you scroll through the code - whether > or > > not they are wrapped in if blocks or individually. > > > > > >> What's wrong about if clauses with proper indenting? These can even be > >> collapsed with a sexy IDE. > > > > Nothing wrong with them. It's just: Do you wrap the entire content of a > > function in an if or do you make a negated if -> return statement? The > first > > doesn't scale, the second makes things more readable - that was really > all I > > was trying to convey. > > For that specific "entire content of a function" case I mentioned the > exceptions that allow returns at the beginning of a function that > check preconditions. > > But I'm more referring to your initial refactoring in > > https://github.com/daviddeutsch/roundcubemail/blob/cleanup-3/plugins/archive/archive.php#L17 > > In such cases I think a sequence of if/else if/else blocks makes more > clear for the reader that these are three distinct cases handled in > the plugin init function.
signature.asc
Description: OpenPGP digital signature
_______________________________________________ Roundcube Development discussion mailing list [email protected] http://lists.roundcube.net/mailman/listinfo/dev
