i'd second this suggestion.
knowing the history of the code and the people
who have worked on it, i think it's relatively unlikely that
there are many completely bone-headed or irrational things
in the code. there may be code which is a reflection of
when it was first written or revised, but that's not
the same thing.
keep in mind that MH started life circa 1978 or so,
back when PDP-11s ruled the world and without split-I/D,
you had 64k bytes to work with as your total address space.
that's one of the big reasons MH is organized like it is -
you already have a fine command interpreter - you don't
need to write another one.
since "back then" there have been a lot of changes. Marshall Rose
and crowd being major contributors to a mid-life cleanup
and functional renovation. then much later, "N" got added to MH
along with welcome new features, some of which were less
natural in their fit with the rest of the code than others.
my first-order suggestion is that if you find something that
looks like it can't possibly make any sense, look again,
possibly further afield, because your understanding is
likely to be incomplete. just because MH compiles into
a plethora of little executables doesn't mean the parts
of any one of them can be understood individually.
the code in "sbr" is less like a pure library and more
like shared body parts.
whether this is a bug or a feature varies with the moment. (grin)
cheers,
-mo
Jeffrey C Honig wrote:
Jon Steinhart <jon(_at_)fourwinds(_dot_)com> wrote:
Are you using cscope? You should be using a tool like it to check for
side effects. I'm sure there are many in the sbr/ directory.
_______________________________________________
Nmh-workers mailing list
Nmh-workers(_at_)nongnu(_dot_)org
http://lists.nongnu.org/mailman/listinfo/nmh-workers