View Single Post
  #24   Report Post  
Mike Rivers
 
Posts: n/a
Default


In article writes:

Even without automated help, you might be surprised how mechanical a
process refactoring is. There is a procedure, and it's designed to
minimize errors. For instance, when you are removing code from a
function and making it its own function, you create the new function
and *copy* the code into it without deleting the original. Then you
compile, then you comment out the copied code and replace it with a
function call. Then you compile and test. Or something like that.


Other than the first compile (after which you do nothing but comment
out the code from the source file and recompile) it makes perfectly
good sense to me. I'm not a programmer, but that's how I'd do it.

When I'm writing something and decide to move a paragraph to a
different section to make it independent, that's sort of what I do.
First I copy it into the new section, see if it makes sense there, put
a reference in the original place (if one is needed, maybe even using
a sentence from the original paragraph) then, delete the orginal (now
redundant) section.

Thing is that with software and version controlled processes, you
probably keep the before and after versions.

The point is, the process is designed so that there are safeguards
to ensure you didn't screw something up, and even if you did, you
can easily put everything back like it was before.


That's why we don't throw away the original wiring diagram until we've
tested the new wiring.


--
I'm really Mike Rivers )
However, until the spam goes away or Hell freezes over,
lots of IP addresses are blocked from this system. If
you e-mail me and it bounces, use your secret decoder ring
and reach me he double-m-eleven-double-zero at yahoo