Re: Re: namespaces?

Erik Hjortsberg wrote: > > At 15:27 2001-07-15 +0200, you wrote: > >Erik Hjortsberg wrote: > > > > > > At 14:07 2001-07-15 +0200, you wrote: > > > >- Return in constructors. For me seems a little bit silly the need off > > > >providing an extra method to do the work and get the error back. > > > >$foo = new bar(); > > > >$err = $foo->build(); //or something similar > > > >if (PEAR::isError($error)) { .. } > > > > > > > >while you could simply do: > > > >$foo = new bar(); // or perhaps something like $foo = new::bar(); > > > >if (PEAR::isError($foo)) { .. } > > > > > > > > > > > >Tomas V.V.Cox > > > > > > Hmm, you can always do: > > > $foo = & bar::new(); > > > if (PEAR::isError($foo)) { .. } > > > > > > and do all of your init in new(). > > > Kinda like Smalltalk does it. > > > > > > >With you idea I should do: > > > >class Bar { > > function new($file) { > > // do things > > if ($error) { > > return $error; > > } > > return & new Bar; > > } > >} > > > >Instead of this strange thing I would probably prefer to provide a > >build() method. > > > >Someone some day, posted a curious system to do what I say: > >class Bar { > > function Bar() { > > if ($error) { > > // overwrite the object > > $this = $error; > > } > > } > >} > > > >But's ok, no more discuss on this topic as it has been clearly rejected > >:) > > Nah, I can't drop it yet. :) > > True, it looks kinda awkward, but it has it's points. > I've done a system for accessing a database through objects (similiar to > DB_Storage, but with some major differences). > One of the differences is that getting and creating new objects is made > through static methods, resulting in that an object with a specific id will > only be fetched from the database once per request. (The system keeps > tracks of which objects are "checked" out, thus relieving the sql-server.) > <code> > $person1 = & Person::Get(1); > $person2 = & Person::Get(1); > </code> > Both $person1 and $person2 will point to the same object. > With creating entries it's unlogical to first make an empty object and then > issuing it a create message. The logical way is to send a Create message to > the class and let it return a new object if successful. Most or all of the pear community agrees with you, the problem is that Zeev doesn't want to make this a feature. There's a thread about it from two months or so back. > On another note, for my system I've tried to arrange it for the objects to > be able to wait with updating the database and do it first when their > constructors are called (ie when the user already has gotten the page). The > problem is that database-connections doesn't seem to survive through to the > shutdown_function. They look alright but fails without an error being > raised when used. Sure, I can create a new database-connection in the > shutdown_function, but for a number of reason I don't want that. I sent a > message to the list a couple of weeks ago, perhaps it was swallowed by the > list. I'll send it again. Please do. - Stig