On Fri, 2008-07-04 at 17:48 +1000, Simon Coulter wrote:
[snip substantive comment on coding issue]
Last, and only, comment on this ridiculous thread.
I agree that this thread has generated more heat than light.
Still, for the second time, I find myself drawn in. Is
there a lesson here? Let me try.
The arguments presented show that either form of the
file-test bifs would be sufficient. The disagreement is
about which form is preferable for readability and
maintainability.
I propose the general guiding principle that similar things
should look similar and different things should look
different. My previous post, with its proposed coding
style, should have said that the rationale presented is not
a definitive argument (well, obviously <duh />) but rather a
mnemonic for my almost arbitrary choice of how to code the
simple--and very common--case.
Avoiding things just a little bit different from each other
helps in following the guiding principle. Thus, I would be
happier had we only one form of each file-test bif. I don't
care *which* one, just that there were but one. (Look,
Mommy! I used the subjuntive twice in two sentences. Note
to self: cut the coffee and get some solid food.)
This is probably enough rambling for one morning.
Cheers,
Terry.
Regards,
Simon Coulter.
midrange.com
