Brent, Thanks for the various patches. I've included them in my CVS version. I wonder if the easiest way to silence the messages from the ML part would be to have the ML call a function in the RTS (driver) to find out whether the "quiet" switch had been set. It should be fairly easy to add.

David.

Brent Fulgham wrote:

The attached patch adds a couple of flags to the poly driver program:

-q = "Quiet Mode", which just silences some of the self-reporting. -v = "Version", prints the version information and exits.

To get things to work the way I wanted, I had to modify the ML sources and rebuild my Database. This obviously works, but it would be much cleaner if I could somehow do this "on-the-fly".

If I could override the "startupFn" in the INITIALISE_ struct, I think this would work, but it doesn't seem to be accessible from a running system.

Thanks,

-Brent

--- David Matthews wrote:

Brent Fulgham wrote:

...

Also, can I get PolyML to stop reporting "The standard function is OS.Process.exit" when I call "PolyML.quit()" ?

Use "OS.Process.exit OS.Process.success".

I did try this, but unfortunately the signature for the "onEntry" function wants a return value of {}, while the OS.Process.exit method returns a success status.

Furthermore, if you do exit the process using the OS.Process.exit function, you get additional runtime system output:

" Warning: The type of (it) contains a free type variable. Setting it to a unique monotype. "

This would also be nice to silence in "quiet" mode.

I didn't want to remove PolyML.quit completely but once there was a standard basis function to do the same job it seemed unnecessary to keep a non-standard function. I guess PolyML.quit is shorter to type. Does anyone have any feelings either way?

I would prefer to use the basis function. But some of the PolyML plumbing needs to be changed slightly to allow this to happen (as I described above).

Thanks,

-Brent