In article <zI-dnc1MsPMR--vVnZ2dnUVZ_gmdnZ2d@comcast.com>,
Michael J. Mahon <mjmahon@aol.com> wrote:
mdj wrote:
On Jul 10, 11:15 pm, pau...@saaf.se (Paul Schlyter) wrote:
Apple DOS didn't even have a decent command line! Only a very restricted
syntax of the command line was allowed by Apple DOS - other needed input
parameters had to be obtained though a user dialogue by the application
itself, or perhaps through some data (text) file. Compare this with the
very flexible command line interface of CP/M (which was directly inherited
by MS-DOS): after the name of the executable file you could type almost
anything you wanted. CP/M and MS-DOS didn't object, it simply passed it on
to the application and then it was up to the application to try to make sense
of the command line, and to determine whether it was legal or illegal.
It's a shame this was never fixed - 10 years later the Apple II
'prompt' had much the same limitations.
I agree. It would have been easy to accept, say, a quoted string,
and pass it into the invoked program using something like the
"startup" protocol.
Of course, one can still re-scan the input buffer... ;-)
If the application gets executed it can do so, yes. However, Apple DOS
won't even execute the application if it finds what it considers a
syntax error in the command line. So the syntax check by Apple DOS
must then be changed, so it just passes on to the application whatever
it finds unintelligible.