Re: some questions re. Ada/GNAT from a C++/GCC user

[dewar@cs.nyu.edu (Robert Dewar)]

Mike says:

"It's rare to find even poorly written C++ code "several pages long," let
alone nested scopes that size. You can find poor programmers in any
discipline, apparently."

Odd! Many people say this sort of thing about both C++ and Ada, but in
my actual experience, real life large scale C++ and Ada programs do not
conform to this dictum. Very long files and procedures are common.

Sometimes, this is inevitable. Both Ada and C++ provide no way of separately
compiling pieces of a large switch/case statement. So if your application
has a case statement with hundreds of branches, then you are going to have
a long procedure, even if you write externs and calls for procedures for
each branch. 

Sometimes, it is true that long = bad code, but this is by no means always
the case. I have often seen short bad code, and good long code, the
correlation is only vague. What makes code easy to read is not shortness
per se, but good structure and organization, and logical thinking. 

I am actually surprised how often we run into the case of very large
single procedures, but it happens in both the Ada and C++ code I have
seen (you really would not expect much difference between the two
languages in this particular respect!)