[Vimoutliner] Fwd: Lyx article on OSNews

[Steve Litt]

On Wednesday 28 February 2007 05:07, Lucas Gonzalez Santa Cruz wrote:
> Steve,
>
> I've done a little Cobol myself.  UK, circa 1979, no more than 200 lines in
> all, typing each line in a punch-card machine and then feeding it into the
> "mainframe" (bought 10 years earlier second hand).
>
> Your VTOC story makes me dizzy.  Wow!
>
> Do you keep the "source code" for that?  It belongs in a horror museum,
> much as my assembly code to turn an Apple II+ into a piano.

Unfortunately no. Back in the late 80's I still had a printout of the program, 
but I haven't seen it in years. In 1989 I moved from Venice California to 
Reseda California, then in 1998 to Apopka Florida, then to Longwood, Florida 
in 2000. I have a feeling it got thrown out or left behind during one of 
those moves.

I remember the basics. I called the main routine A. Then looked for all called 
subroutines *and loops* AA, AB, AC, AD, etc. Then I took all those 
subroutines and loops, parsed them for called subroutines and loops, and 
labeled them AAA, AAB, etc. In other words, I did it one level at a time 
rather than diving via recursion. My more experienced buddies (especially 
Jeff Jones :-) suggested I recurse, but I didn't know how, so I just kept 
shaving off a level at a time, keeping partially completed stuff in arrays.

What I don't remember is whether I made a single pass through the program, 
storing each subroutine and the loops and subroutines it called, or whether I 
repeatedly scanned for each level. Given my newbiism of the time, I'd guess 
the latter.

Finally I had a list of subroutines and loops, each with a name like ACBAD 
etc. I then sorted the names, and it produced the hierarchy.

From there on it was simply a matter of representing it graphically. Boxes 
were the width of the subroutine name plus 2.

STeveT

Steve Litt
Author: Universal Troubleshooting Process books and courseware
http://www.troubleshooters.com/