Understanding Packages

[Phillip Mills]

On Sunday, April 18, 2004, at 10:26 PM, Geoffrey Hunter wrote:

>  I guess that this could be the explanation
> for why applications come as a package of C functions, even though 
> only one
> (or a few) of the many functions that comprise the package are called
> (callable) by the user ?

It sounds to me as if you're mixing a number of unrelated concepts.
1) The use of packages has nothing to do with the source language of 
the packaged application.
2) If it's an application, the concept of a user calling any function 
is a bit of a stretch.  (Running the program "calls" main and clicking 
a button probably "calls" some handler function, but in my experience 
even people who write the applications don't think that way except when 
debugging.)
3) Packages are distribution mechanisms and generally have no relevance 
for application execution.

I'd also suggest clarifying what you mean by nesting.  When you're 
talking about scope of function/variable definitions, there's one 
answer.  If you consider the run-time functional hierarchy, however, a 
C stack trace is (little or) no different structurally from that of 
other languages.  My point being, any "virtues" that your paper 
discusses should likely have nothing to do with using the application 
-- unless you're dealing with languages where running an application is 
hard to distinguish from developing it (e.g. Forth, Prolog...).

For most uses of 'package', it's just a convenient way of accessing 
related files as a single entity.  Until you wander into the Java world 
where the word package opens a whole other can of worms.  :-)

........................
Phillip Mills
Multi-platform software development
(416) 224-0714

--
The Toronto Linux Users Group.      Meetings: http://tlug.ss.org
TLUG requests: Linux topics, No HTML, wrap text below 80 columns
How to UNSUBSCRIBE: http://tlug.ss.org/subscribe.shtml