Re: Fortran Refactoring for Legacy Systems

On Tuesday, October 25, 2011 1:32:54 PM UTC-7, Robert Miles wrote:
That program didn't try to access the higher
precisions in a standard-conforming way, partly
because it was started before there was any
standard offering those higher precisions yet.
Instead, it called a single assembler-language
subroutine, with the arguments including one
specifying which operation to perform, and one
specifying the precision level. That program
used old-style Fortran source, on punched cards
most of the time I was working with that program.

I missed the statement in your reply that you were
referring to a specific program you had worked on.
Sorry about that...

While that is clearly a problem to be dealt with in
updating old code, I still stand by my opinion that
a refactoring program shouldn't be expected to
rewrite that sort of thing...