Disdep
reads each
file,
which must be a Dis object file,
and finds all unique strings in it that end in
.dis.
It takes each such string as the name of a Dis file, and
if the file exists, it does the same for it, and so on, recursively.
It writes each unique name to the standard output.
The result is a list of all statically-named Dis files that might be referenced by
an application, typically as the operand of a Limbo
load
operator.
Several options change or extend the output:
-a
Print all names as they are encountered in the search, including duplicates.
-d
Indent to show the dependency structure.
-o
Show only the immediate (outermost) dependencies of each
file.
-p
Print the dependency relation as pairs:
a file, a space, and the name of a file on which it depends.
Only the the first name is printed when a file depends on no other.
This format is useful as input to
mk(10.1)
dependency generators, or dependency graphing programs.
-s
Include strings of the form
"$[A-Z].*"
on the assumption
they are the names of system modules loaded by the application.