OpenEuphoria

Re: Store Includes

• Posted by DB James <larch at adelphia.net> Jul 19, 2005
• 533 views

Al Getz wrote:
BEGIN PREVIOUS
Hi again Quark,

Hee hee, YOWIE i guess is a good word for it too as there's a little
work involved as im sure you know :)
I think it would be very worthwhile however.

As far as your question goes, i can only see one area of caution:

A function call such as:
  x=GetValue()

could very well have been declared inside an include file that got
included via a namespace such as:

include GetValues.ew as GetStuff

Now in that previous line the user decided to skip the namespace prefix so
instead of writing:

  x=GetStuff:GetValue()

the previous line was used instead.

This works perfectly well in Euphoria, and as far as i know this
is the way it's always going to work.
Now the default is to use GetValue() without a namespace, but
from my experiance this wouldnt be the greatest idea because we
not only want our program to work now, we want it to work after
we add something else too, so a warning would be great, and
an option to either use the prefix or not use the prefix would
be nice too.



Another situation to be careful about is...

Since new names will appear as the mashing is in progress, they will
have to be checked to make sure there is no other name exactly the same.
Since you wont know all 'till the end of the whole mash process, you'll
have to keep some reference to occurances so you can 'rename' a second
or even third/fourth time if need be.  You probably thought about this
already though.



I'd be more than happy to test your program out fully once you think
you've got it to the point where it can handle namespace declares.



Take care,
Al
END PREVIOUS

REPLY
Hi Al,

I have printed out your caveats and will refer to them as I proceed.  Thanks
for offering to test the code (great to have that in writing :^D ).  I may
need a couple of programs plus their includes set from you too, or 
recommendations on posted programs I can practice with.  I need at least a
simple and a  medium-complexity program to mash so I can see how various
namespace issues play out.  I might even (coff) have to write some myself
as tests, but that is a bad idea because I don't really have a feel
for namespace yet.  This request is sent also to anyone who can suggest
a highly-namespaced program of not too great complexity. 

My preference/fantasy would be a duplicate-name-killer
program that would scan the known universe and annhillate all duplicate
names, for global items, to be run as part of an anti-virus program.  It
might be memory-resident too, and a voice would moan "No, no, no!" in a
sepulchural voice when the user attempts it.  On the other hand, it is
probably just an example of the gnarliness of the world outside the 
theoretical neatness of the programming mind.

--Quark
END REPLY