Okay, I think I've got a handle on things now. Thanks for taking the
time to elaborate.

Jamie

On Wed, 25 Feb 2004 14:53:46 -0800, in cf-talk you wrote:

>The way I described will work, and keep the typing as tight as possible,
>which is what you want.  You don't want to use any very often, because it's
>the same as not even specifying a type.  The point of typing is to catch
>errors in a structured way, rather than the "wait til it fails on first use"
>approach.
>
>The reason I don't like returning from init, is that it's elevates the init
>method to a status that's greater than 'normal' methods, but less than a
>real constructor, which is only confusing.  And at best, it'll save you the
>length of the variable name + 11 characters of typing, and cost you
>readability.
>
>Also, it's inconsistent with non-CFC objects, because their constructors
>CAN'T return anything, so you have to do the two line initialization.
>
>That's 100% personal taste, and there are many who disagree with me.  I'm
>not going to tell you not to do it that way (that's your decision to make),
>just that I don't do it.
>
>Cheers,
>Barneyb
>
>> -----Original Message-----
>> From: Jamie Jackson [mailto:[EMAIL PROTECTED]
>> Sent: Wednesday, February 25, 2004 2:41 PM
>> To: CF-Talk
>> Subject: Re: extends + init() = returntype confusion
>>
>> On Wed, 25 Feb 2004 14:17:35 -0800, in cf-talk you wrote:
>>
>> >First, IMHO, having init() return a reference to 'this' is
>> bad style.  It
>> >only makes things more confusing.
>>
>> Just when I thought I had one solid convention...
>>
>> So what's the preferred way to do it, and why is "return this" bad?
>>
>> One more thing about "return this": Why not just have the "any"
>> returntype in the super class, since "this" is guaranteed(?) to return
>> the current type?
>>
>> Thanks,
>> Jamie
>>
>>
>
>
[Todays Threads] [This Message] [Subscription] [Fast Unsubscribe] [User Settings]

Reply via email to