I said simple variable initialization is fine. My experience is that logic mostly appears in non-default constructors.Well then when it appears in default constructors don't you think it is better suited there then in the pseudo-constructor area?
Our experiences differ. I can only speak to my experience in C++ (since early '92) and in Java (since around '96/'97) - and a smattering of Smalltalk - and that is that non-default constructors are usually where the complex logic is and also that non-default constructors are more common than default constructors. We'll probably just have to agree to differ on this one...There is certainly a number of default constructors in the Java hierarchy. Certainly, less than the number that take arguments, but significant nevertheless.
I don't believe there is a precedent for such a thing in CFML. CFML uses strings to specify paths - all bare identifiers are evaluated (and are therefore not types, by definition). I think it would very strange syntactically in the context of CFML to add an expression like:Clearly the following wouldn't fit.
new fully.qualified.identifier
I could, however, see an argument for:
new("fully.qualified.identifier")
new /fully/qualified/identifier
Where as the following would.
new fully.qualified.identifier
Which I think this a great reason to lose the ability to use a path to specify a CFC.
-Matt
----------------------------------------------------------
You are subscribed to cfcdev. To unsubscribe, send an email
to [EMAIL PROTECTED] with the words 'unsubscribe cfcdev' in the message of the email.
CFCDev is run by CFCZone (www.cfczone.org) and supported by Mindtool, Corporation (www.mindtool.com).
An archive of the CFCDev list is available at www.mail-archive.com/[EMAIL PROTECTED]
